This is a repository copy of Statutory homelessness in England : the experience of familiesand 16-17 year olds.White Rose Research Online URL for this paper:https://eprints.whiterose.ac.uk/73926/Version: Published VersionMonograph:Pleace, Nicholas orcid.org/0000-0002-2133-2667, Fitzpatrick, Suzanne, Johnsen, Sarah etal. (2 more authors) (2008) Statutory homelessness in England : the experience of familiesand 16-17 year olds. Research Report. Department for Communities and Local Government , London. eprints@whiterose.ac.ukhttps://eprints.whiterose.ac.uk/Reuse Items deposited in White Rose Research Online are protected by copyright, with all rights reserved unless indicated otherwise. They may be downloaded and/or printed for private study, or other acts as permitted by national copyright laws. The publisher or other rights holders may allow further reproduction and re-use of the full text version. This is indicated by the licence information on the White Rose Research Online record for the item. Takedown If you consider content in White Rose Research Online to be in breach of UK law, please notify us by emailing eprints@whiterose.ac.uk including the URL of the record and the reason for the withdrawal request. Statutory Homelessness in England: The experience of families and 16-17 year oldswww.communities.gov.ukcommunity, opportunity, prosperityStatutory Homelessness in England: The experience of families and 16-17 year oldsNicholas PleaceSuzanne FitzpatrickSarah JohnsenDeborah QuilgarsDiana SandersonCentre for Housing Policy, University of YorkMarch 2008Department for Communities and Local GovernmentCommunities and Local GovernmentEland HouseBressenden PlaceLondon SW1E 5DUTelephone: 020 7944 4400Website: www.communities.gov.uk© Queen’s Printer and Controller of Her Majesty’s Stationery office, 2008.Copyright in the typographical arrangement rests with the Crown.This publication, excluding logos, may be reproduced free of charge in any format or medium for research, private study or for internal circulation within an organisation. This is subject to it being reproduced accurately and not used in a misleading context. The material must be acknowledged as Crown copyright and the title of the publication specified.Any other use of the contents of this publication would require a copyright licence. Please apply for a Click-Use Licence for core material at www.opsi.gov.uk/click-use/system/online/pLogin.asp, or by writing to the Office of Public Sector Information, Information Policy Team, St Clements House, 2-16 Colegate, Norwich, NR3 1BQ. Fax: 01603 723000 or email: HMSOlicensing@opsi.x.gsi.gov.ukIf you require this publication in an alternative format please email alternativeformats@communities.gsi.gov.ukCommunities and Local Government PublicationsPO Box 236WetherbyWest YorkshireLS23 7NBTel: 08701 226 236Fax: 08701 226 237Textphone: 08701 207 405Email: communities@capita.co.ukor online via the Communities and Local Government website: www.communities.gov.uk75%March 2008Reference number: 07 ACST 05147ISBN: 9781851128938Acknowledgements 1AcknowledgementsThe authors would like to express their gratitude to BMRB Social Research, and in particular to Joel Williams and Angela Charlton, for the highly professional manner in which they executed their work on this complex and demanding project and for all their continued support in bringing it to a successful completion. Jane Allen and Nicola Hawthorne, of the Centre for Housing Policy, University of York, provided invaluable assistance in preparing the manuscript for submission to Communities and Local Government. We would like to thank Helen Smith, who managed the research on behalf of Communities and Local Government, for her advice and support throughout the research process. In addition, we are grateful to Keith Kirby and Penny Withers for their extensive input, and to a range of policy and analytical colleagues in Communities and Local Government for their helpful comments. We would also like to thank all of the local authorities who agreed to participate in the research. Above all, we are very grateful to the parents, children and young people who took the time to take part in the research. All views and any errors contained in the report are the sole responsibility of the authors. 2 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsContentsAcknowledgements 1Tables and figures 7Glossary of terms 15Families accepted as homeless 15Young people accepted as a homeless 16-17 year old 19List of acronyms 22Summary 23Introduction 23Key Points 24Characteristics and support needs 25The causes of statutory homelessness 28The experience of temporary accommodation and the provision of settled housing 30The impacts of homelessness and temporary accommodation 33Conclusion 36Chapter 1: Introduction 39Introduction 39Previous research on family homelessness and homelessness amongst 16-17 year olds 41Research aims 44Research methods 44Data analysis and presentation 47Investigating the impacts of homelessness and temporary accommodation 48Structure of report 49Chapter 2: The characteristics of families accepted as homeless 51Introduction 51Key points 53The characteristics of families accepted as homeless 53The characteristics of adult respondents in families accepted as homeless 58Where the families were accepted as homeless 62Contents 3Families in temporary accommodation for more than one year 67Conclusions 72Chapter 3: Personal and housing history of adult respondents 73Introduction 73Key points 75Personal history of adult respondents 76Housing history of adult respondents 81Adult respondents in temporary accommodation for more than one year 84Conclusions 85Chapter 4: Seeking help from a local authority 87Introduction 87Key points 89The type of last settled accommodation occupied by families 89The type of accommodation that families were living in when accepted as homeless 90Seeking alternative help before approaching the local authority 93Awareness of the statutory homelessness arrangements 95Concerns about applying as homeless 96Information on assessment process 98Families in temporary accommodation for more than one year 98Conclusions 100Chapter 5: Reasons for applying as homeless 102Introduction 102Key points 104An overview of reasons for applying as homeless 105Comparing reasons for applying as homeless to reasons for leaving last settled accommodation 108The reasons for applying as homeless amongst those seeking help from different types of accommodation setting 109The reasons for applying as homeless amongst different demographic groups 111The reasons for applying as homelessness in different areas of England 112Independent influences on the principal reasons for applying as homeless 1134 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFamilies in temporary accommodation for more than one year 115Conclusions 116Chapter 6: Families’ experience of temporary accommodation 118Introduction 118Key points 121Housing situation at point of survey 122Length of time in temporary accommodation 124Types of temporary accommodation experienced 129Moves between temporary accommodation addresses 131Pathways through temporary accommodation 133Families in temporary accommodation for more than one year 135Conclusions 139Chapter 7: Conditions in temporary accommodation 141Introduction 141Key points 143The current or most recent temporary accommodation occupied by families 144Space, rooms and sharing in temporary accommodation 146Satisfaction with facilities in temporary accommodation 154Access to household items and amenities in temporary accommodation 156Physical conditions within temporary accommodation 158Safety in temporary accommodation 162Overall satisfaction with temporary accommodation 165Families in temporary accommodation for over one year 166Conclusions 171Chapter 8: Conditions in settled housing 173Introduction 173Key points 175The families in settled housing 175Tenure and choice in settled housing 176Living space in settled housing 177Satisfaction with facilities in settled housing 178Access to household items and amenities in settled housing 180Contents 5Physical conditions in settled housing 181Safety in settled housing 183Overall satisfaction with settled housing 184Conclusions 185Chapter 9: Health and social support 186Introduction 186Key points 188General health 189Changes in diet since last settled accommodation 193Mental health problems 194Drug and alcohol problems 196Support needs and service use 197Use of the NHS and care and support services 202Access to social support 204Adult respondents in temporary accommodation for more than one year 208Conclusions 213Chapter 10: Employment, income and expenditure 215Introduction 215Key points 217Economic status, worklessness and barriers to work 218Qualifications and literacy 224Income, debt, and major expenditure 225Financial exclusion 235Overall financial circumstances 235Families in temporary accommodation for more than one year 240Conclusions 247Chapter 11: Children’s experiences 249Introduction 249Key points 251Background 251Children’s views on leaving their last settled accommodation 254Children’s views on their current accommodation 2556 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsHealth 260Support networks 265School and educational performance 269Service use 272Children within families in temporary accommodation for more than one year 274Conclusions 275Chapter 12: The experience of young people accepted as homeless 16-17 year olds 277Introduction 277Key points 280Demographic profile 281Personal history 284Seeking help from a local authority 288Reasons for applying as homeless 293Accommodation experiences since being accepted as homeless 295Health 301Social and professional support 306Economic status and financial circumstances 313Conclusions 324Chapter 13: Overall quality of life 326Introduction 326Key points 327Adults 329Children 333Young people 338Conclusions 340Appendix 1: Technical report summary 342Sample design 342Questionnaire development and fieldwork measures (all surveys) 348Field outcomes 353Weighting and design factors 356Appendix 2: The analysis process 359Tables and figures 7Tables and figuresTable 2.1: Survey 1 families in comparison with the general population 54Table 2.2: Size of lone women parent and couple households 55Table 2.3: The age and gender of parents in families accepted as homeless 55Table 2.4: The age of parents in Survey 1 families compared to the age of parents in the general population in England 56Table 2.5: The ethnicity of adult respondents compared to ethnicity of applicants’ households accepted as homeless in first two quarters of 2005 59Table 2.6: Ethnic background of adult respondents compared with adults with dependent children in the general population 60Table 2.7: Regions where Survey 1 families were accepted compared to all acceptances of families accepted as homeless during the first two quarters of 2005 63Table 2.8: DEFRA classification of areas where Survey 1 families were accepted compared to pattern of acceptances for all families accepted as homeless 66Table 2.9: Household composition of Survey 4 families in comparison with Survey 1 families 67Table 2.10: Household size of Survey 4 families in comparison with Survey 1 families 68Table 2.11: Age and gender of parents in Survey 4 families compared to Survey 1 families 68Table 2.12: Number of children in households, Survey 4 compared to Survey 1 69Table 3.1: Childhood experiences of Survey 4 and Survey 1 adult respondents 84Table 4.1: Tenure or accommodation type from which sought help from local authority, by age of adult respondent 93Table 4.2: Sources of help or assistance sought prior to approaching a local authority for Survey 4 and Survey 1 adult respondents 99Table 5.1: Accommodation setting from which adult respondents sought help from a local authority, by most frequently reported reasons for applying as homeless 109Table 5.2: Reasons for homelessness reported by Survey 4 and Survey 1 adult respondents 115Table 6.1: Housing situation of families at point of survey, by percentage of all families in each region 122Table 6.2: Current living situation of families in temporary accommodation at point of survey 1238 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 6.3: Current living situation of families in temporary accommodation at point of survey, by region 124Table 6.4: Time spent in temporary accommodation by those families in settled housing at point of survey 125Table 6.5: Time spent in temporary accommodation by those families in temporary accommodation at point of survey 126Table 6.6: Types of temporary accommodation experienced by families (families who had stayed in temporary accommodation) 129Table 6.7: Number of different temporary accommodation addresses reported by families (families who had stayed in temporary accommodation) 132Table 6.8 Types of temporary accommodation experienced by Survey 4 families and by Survey 1 families (with a temporary accommodation stay) 136Table 6.9: Number of temporary accommodation addresses experienced by Survey 4 families and by Survey 1 families (with a temporary accommodation stay) 136Table 6.10: Pathways through temporary accommodation for Survey 4 families 137Table 7.1: Type of temporary accommodation (current or last occupied temporary accommodation) 144Table 7.2: Type of temporary accommodation reported on by families by broad region (current or last occupied temporary accommodation) 145Table 7.3: Extent of problematic bedroom sharing among families sharing bedrooms, by type of current/ last temporary accommodation 147Table 7.4: Access to kitchens, bathrooms, living rooms (current or last occupied temporary accommodation) 151Table 7.5 Satisfaction with living space by temporary accommodation type (current or last occupied temporary accommodation) 153Table 7.6: Satisfaction with facilities (current or last occupied temporary accommodation) 155Table 7.7: Access to household items and amenities (percentage of current or last occupied temporary accommodation that had facilities) 157Table 7.8: Problems with conditions in temporary accommodation, by temporary accommodation type (current or last occupied temporary accommodation) 159Table 7.9: Problems with conditions in temporary accommodation, by broad region (current or last occupied temporary accommodation) 160Table 7.10: Frequency with which problems with physical conditions tended to be reported in temporary accommodation (current or last occupied temporary accommodation) 161Table and figures 9Table 7.11: Frequency with which problems with physical conditions were reported in temporary accommodation, by broad region (current or last occupied temporary accommodation) 161Table 7.12: Frequency with which three or more physical problems with temporary accommodation were reported, by temporary accommodation type and broad region (current or last occupied temporary accommodation) 162Table 7.13: Safety within temporary accommodation (current or last occupied temporary accommodation) 163Table 7.14: Factors influencing satisfaction with temporary accommodation (current or last occupied temporary accommodation) 166Table 7.15: Satisfaction with facilities (Survey 4 adult respondents and Survey 1 adult respondents reporting on self-contained temporary accommodation) 168Table 7.16 Physical problems with temporary accommodation (Survey 4 adult respondents and Survey 1 adult respondents reporting on self-contained temporary accommodation) 170Table 7.17: Factors associated with satisfaction with temporary accommodation (Survey 4 adult respondents and Survey 1 respondents reporting on self-contained temporary accommodation) 171Table 8.1: Location of families in settled housing at point of survey 176Table 8.2: Satisfaction with facilities in settled housing, compared to current or last temporary accommodation that was self-contained housing 179Table 8.3: Access to household items and facilities in settled housing, compared to current or last temporary accommodation that was self-contained housing 180Table 8.4: Problems with conditions in settled housing, compared to current or last occupied temporary accommodation that was self-contained housing 182Table 8.5: Frequency with which problems with physical conditions were reported in settled housing, compared to current or last occupied temporary accommodation that was self-contained housing 183Table 8.6: Why respondents felt unsafe in the area surrounding their settled housing, compared to current or last occupied temporary accommodation that was self-contained housing 184Table 9.1: Self-assessed general health of Survey 1 adult respondents at point of survey by age, compared with the general population (aged 16-54) 191Table 9.2: Approximate comparison of self-reported current mental health problems among adult respondents with the general population, by gender (aged 16-54) 194Table 9.3: Self-reported drug and alcohol problems 19610 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 9.4: Families’ self-reported receipt and need for practical support 199Table 9.5: Families’ self-reported need for personal support 201Table 9.6: Use of NHS, care and support services and other formal support services in current accommodation, by temporary and settled accommodation 202Table 9.7: Changes in contact with family between last settled housing and current accommodation 207Table 9.8: Self-assessed general health of Survey 4 adult respondents at point of survey, compared with Survey 1 adult respondents and general population (aged 16-54) 208Table 9.9: Survey 4 families’ self -reported need for practical support 210Table 9.10: Families’ self-reported need for personal support 211Table 9.11: Use of NHS and other care and support services, for Survey 4 and Survey 1 212Table 10.1: Economic status of all adults in families accepted as homeless, compared with all adults living in households containing children in general population 218Table 10.2: Barriers to employment reported by adult respondents in families accepted as homeless and by adults in families containing children in general population 223Table 10.3: Academic qualifications held by adult respondents in families accepted as homeless in comparison with adults in families in the general population 224Table 10.4: Total family income per week (excludes Housing Benefit) 226Table 10.5: Total family income per week compared with families in general population (excludes Housing Benefit) 227Table 10.6 Benefits received by families accepted as homeless and families in general population 229Table 10.7: Grants and loans received by families accepted as homeless, by whether in temporary or settled housing 230Table 10.8: Outstanding loans and repayment problems among families accepted as homeless 231Table 10.9: Regular utility and housing-related bills paid by families accepted as homeless 232Table 10.10: Expenditure on specific items as a result of leaving last settled accommodation 234Table 10.11: How families accepted as homeless were managing financially compared with families in the general population 236Tables and figure 11Table 10.12: How families accepted a homeless were managing financially at point of survey, by whether in temporary or settled accommodation 237Table 10.13: How families accepted as homeless were managing financially in their last settled home compared to their accommodation at point of survey 238Table 10.14: Economic status of all adults in Survey 4 families compared with all adults in Survey 1 families 240Table 10.15: Barriers to employment reported by adult respondents in Survey 4 families and Survey 1 families 242Table 10.16: Qualifications held by Survey 4 adults, in comparison with Survey 1 adult respondents and adults in families in the general population 243Table 10.17: Total family income per week for Survey 4 families (excludes Housing Benefit) 243Table 10.18 Benefits received by Survey 4 families, compared to Survey 1 families 244Table 10.19: Grants and loans received by Survey 4 families, compared to Survey 1 families 245Table 10.20: How Survey 4 families were managing financially at point of survey, compared to Survey 1 families 246Table 11.1: Child age band, as reported by parents 252Table 11.2: People, places and things missed by children since moving from their last settled accommodation, as reported by children (8-15 years) 255Table 11.3: Child’s (5-17 years) school attendance and experience, as reported by parents 269Table 11.4: Services used by children (0-17 yrs), as reported by parents 273Table 12.1: Accommodation Survey 3 young people were living in when sought help from the local authority 289Table 12.2: Reasons for applying as homeless given by Survey 3 young people 293Table 12.3: Who Survey 3 young peoples’ relationships had broken down with 294Table 12.4: Reasons for relationship breakdown of Survey 3 young people with parent(s) or step-parent(s) 295Table 12.5: Current accommodation type of Survey 3 young people, by whether in settled or temporary accommodation 296Table 12.6: Time spent in temporary accommodation at point of survey by Survey 3 young people, by whether in settled or temporary accommodation 297Table 12.7: Prevalence of current poly-substance problems amongst Survey 3 young people 305Table 12.8: Survey 3 young people receiving and needing support services 30712 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 12.9: Use of formal support services by Survey 3 young people 309Table 12.10: Barriers to education, employment and training identified by NEET Survey 3 young people 317Table 12.11: Income per week (excluding Housing Benefit) of Survey 3 young people, by whether in temporary or settled accommodation 319Table 12.12: Comparisons of current income and income in last settled accommodation received by Survey 3 young people, by whether in temporary or settled accommodation 320Table 12.13: Benefits currently received by Survey 3 young people, by whether in temporary or settled accommodation 320Table 12.14: Grants and loans received by Survey 3 young people, by whether in temporary or settled accommodation 321Table 12.15: Survey 3 young peoples’ assessment of their financial management in current accommodation and last settled accommodation 323Table 13.1: How adult respondents felt about life overall 332Table 13.2: How children feel about life overall, as reported by children (8-15 yrs) 336Table 13.3: How young people feel about life overall 340Table A1.1: Number of local authorities/families sampled by BMRB (across sampling strata) 345Table A1.2: Field outcomes (LAs) 353Table A1.3: Cumulative response rate (unweighted) 354Table A1.4: P1E data distributions and (design-weighted) family survey data distributions 356Table A1. 5: Sampling frame distribution against design-weighted distribution (Survey 3) 358Table A2.1: Groups of binary variables commonly employed for regression and bivariate analysis by the shorthand terms used to describe them (families accepted as homeless) 361Table A2.2: Groups of binary variables commonly employed for regression and bivariate analysis by the shorthand terms used to describe them (16-17 year-olds accepted as homeless) 365Tables and figure 13Figure 2.1: Number of dependent children in lone women parent and couple households 57Figure 2.2: Distribution of families accepted as homeless compared to distribution of national population, according to DEFRA classification of local authorities 65Figure 3.1: Childhood experiences reported by adult respondents 76Figure 3.2: Experiences as an adult reported by adult respondents 79Figure 3.3: Adult respondents’ experience of homelessness or insecure housing circumstances, from age 16 onwards, prior to the place from which accepted as homeless 82Figure 4.1: Type of last settled accommodation occupied by families 90Figure 4.2: Where families were staying when accepted as homeless 91Figure 4.3: Assistance sought by adult respondents before approaching a local authority for help 94Figure 4.4: How adult respondents found out about applying as homeless (adult respondents who knew they were going to apply before they approached a local authority) 96Figure 4.5: Concerns that adult respondents had about applying as homeless 97Figure 5.1: Reasons for applying as homeless 106Figure 6.1: Distribution of temporary accommodation stays across regions 128Figure 6.2: Summary of temporary accommodation pathways 134Figure 6.3: Reasons why survey 4 respondents thought they had had to wait so long in temporary accommodation 139Figure 7.1: Problems associated with sharing bedrooms in current/last temporary accommodation 148Figure 7.2: Problems reported with sharing one or more rooms with other households in current or last temporary accommodation (any sharing of kitchen, living room, bathroom) 152Figure 7.3: Problems reported with lack of space (current/last temporary accommodation) 153Figure 7.4: Reasons why families felt surrounding area was unsafe (current or last temporary accommodation) 164Figure 9.1: Access to practical and emotional support for Survey 1 adult respondents at point of survey by age, compared with responses to British Household Panel Survey, 2004 204Figure 9.2: Sources of emotional and instrumental support 206Figure 9.3: Sources of emotional and instrumental support 213Figure 10.1: Changes in families’ working status since their last settled accommodation 22114 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 11.1: Places outside available for play or relaxation, as reported by children (8-15 years) 256Figure 11.2: Perceptions of current accommodation, as reported by children (8-15 years) 258Figure 11.3: Perceptions of safety when walking on own or with friends in the local neighbourhood, as reported by children (8-15 years) 259Figure 11.4: General physical health status of children (0-17 years), as reported by parents (with national comparison) 261Figure 12.1: Household type of Survey 3 young people 282Figure 12.2: Ethnic origin of Survey 3 young people 283Figure 12.3: Personal history of Survey 3 young people 286Figure 12.4: The actions undertaken by Survey 3 young people before approaching a local authority for help 291Figure 12.5: Survey 3 young peoples’ concerns about making a homelessness application 292Figure 12.6: Type of temporary accommodation experienced by Survey 3 young people 298Figure 12.7: Current accommodation conditions experienced by Survey 3 young people, by whether in temporary or settled accommodation 300Figure 12.8: Current general health status of Survey 3 young people, with national comparison (16-18 year olds) 302Figure 12.9: Proportion of Survey 3 young people with access to emotional and instrumental support, with national comparison (16-18 years) 311Figure 12.10: Sources of emotional and instrumental support for Survey 3 young people 312Figure 12.11: Current economic activity of Survey 3 young people, with national comparators 314Figure 12.12: Changes in economic status of Survey 3 young people since leaving last settled accommodation 316Figure 12.13: Qualifications of Survey 3 young people 318Figure 12.14: Survey 3 young peoples’ assessment of their current financial situation 322Figure 13.1: Parental assessments of changes in the quality of life of their children 335Figure 13.2: Children’s feelings about aspects of life, as reported by children (12-15 yrs) 337Glossary of terms 15Glossary of termsFamilies accepted as homelessAccepted as homeless• : accepted as ‘eligible for assistance’, ‘unintentionally’ ‘homeless’ and in ‘priority need’, and therefore owed the ‘main homelessness duty’ Accommodation conditions• : this group of variables used for analysis included: whether three or more physical conditions problems reported (damp, infestation, conditions for children, was in poor repair when first arrived, dirty when first arrived, insufficient control over heating, difficult to enter with pram or buggy, not well decorated when first arrived); perception of safety inside accommodation; perceptions of safety in neighbourhood in which accommodation is located; perception of sufficiency of living space; sharing of kitchen, bathroom or living rooms (temporary accommodation only); whether sharing perceived as a problem (temporary accommodation only)Adult respondent• : the adult in each family who completed the questionnaire. These adult respondents were purposively selected as the person best placed to comment on the position and experiences of the family as a whole (and was usually the mother) Causes of homelessness• : this group of variables used for analysis included the four major ‘immediate’ reasons given by adult respondents for applying as homeless, namely, relationship breakdown (both violent and non-violent); overcrowding; eviction/threatened with eviction; overstayed welcome/could no longer be accommodated Child respondent• : the child in each family who completed the questionnaire. These child respondents were randomly selected from the children aged between 8-15 years old in each family which had a child in that age groupCurrent accommodation type• : this group of variables used for analysis included: whether family was still in temporary accommodation or settled housing at point of survey; and, where relevant, type of temporary accommodation living in at point of survey (self-contained; friends and relations; or hostels and B&B hotels)Demographic characteristics• : this group of variables used for analysis included: age of adult respondent; ethnicity of adult respondent; household type; household size; whether the adult respondent had ever sought asylum in the UK 16 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsEligible for assistance• : some persons from abroad are ineligible for any assistance under the homelessness legislation, except advice and information about homelessnessEmotional support• : whether respondent had someone to count on to listen when they needed to talk Ethnic minority background• : all non-White ethnic backgroundsGeographical variables• : this group of variables used for analysis included: whether accepted in a rural or urban local authority; whether accepted in London; whether accepted in the South (South West, South East, East of England); whether accepted in the ‘North and Midlands’ (North West, North East, West Midlands, East Midlands, Yorkshire/Humber); housing stress in local authority area; deprivation levels in local authority area1Housing history variables• : this group of variables used for analysis included: whether had had a self-defined ‘settled home’ as adult; whether had ever had own independent accommodation as a tenant or owner occupier; whether had made any previous homelessness applications Immediate family• : an adult respondents’ partner, child(ren) and any other members of their immediate household with whom they intended to live in their ‘settled’ accommodation Instrumental support• : whether respondent had someone to count on to help out in a crisis Last settled accommodation• : ‘ordinary housing’ (see below) lived in before acceptance as homeless that was stable enough and recent enough to be used as a comparison point to investigate changes in health, well-being and economic circumstances that may be associated with homelessness and staying in temporary accommodation2 Main homelessness duty• : the duty that requires a local authority to ensure that a person has suitable temporary accommodation available for his or her household until suitable settled accommodation becomes available Managed accommodation• : hostels, B&B hotels, and any other form of managed or supported accommodation not located in ordinary housingNorth and Midlands broad region• : North West, North East, West Midlands, East Midlands, Yorkshire/Humber1 See Appendix 2 for details on how housing stress, deprivation and rurality were measured.2 See Appendix 1 for a full explanation.Glossary of terms 17Ordinary housing• : a residential house or flatP1E data• : statistical returns made by local authorities on decisions made under the homelessness legislation, and households in temporary accommodation at the end of each quarter, collated and published by central government as the official homelessness statistics. See www.communities.gov.uk – housing/housing research and statistics/housing statistics/housing statistics by topic/homelessness statistics Personal support• : advice or help with mental health, drug or alcohol problems, or parenting issuesPractical support• : advice or help with financial matters, with furnishing or maintaining a home, getting a job, accessing health or other services, etc.Priority need• : housing applicants who are homeless or threatened with homelessness and whose households include any of the following have ‘priority need’ status under the homelessness legislation: dependent children; a pregnant woman; a person who is ‘vulnerable’ due to old age, disability, or some other reason. Applicants also have priority need if they are a young person aged 16-17 years (or aged 18-20 years and were formally in local authority care) or if they are vulnerable as a result of having spent time in care, custody or the armed forces, or vulnerable because they have fled their home because of violence Self-contained temporary accommodation• : a house or flat used only by the adult respondent and their immediate familySettled accommodation• : accommodation offered to someone accepted as homeless that discharges the local authority’s duty owed to them under the homelessness legislation. In most cases, the local authority will arrange the offerShared forms of temporary accommodation• : non-self contained forms of temporary accommodation (i.e. temporary arrangements with friends or relatives, or hostels and B&B hotels)South broad region• : South West, South East, East of EnglandStatutory homelessness• : (for the purposes of this report) the experience of having been accepted as owed the main homelessness duty and assisted under the homelessness legislationSurvey 1• : a survey of families accepted as owed the main homelessness duty, with data collected from a (purposively selected) adult respondent (see Appendix 1)18 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsSurvey 2• : a survey of children in families accepted as owed the main homelessness duty, with data collected from a (randomly selected) child respondent aged 8-15 years old in all families sampled for Survey 1 which had a child of this age (see Appendix 1)Survey 4• : a survey of families accepted as owed the main homelessness duty and who had stayed in temporary accommodation for more than one year, with data collected from a (purposively selected) ‘adult respondent’ (see Appendix 1)Survey 5• : a survey of children in families accepted as owed the main homelessness duty and who had stayed in temporary accommodation for more than one year, with data collected from a (randomly selected) ‘child respondent’ aged 8-15 years old in all families sampled for Survey 4 which had a child of this age (see Appendix 1)Temporary accommodation• : accommodation secured by a local authority for someone accepted as homeless – and his or her family – until settled accommodation becomes available Temporary accommodation experiences• : this group of variables used for analysis included: whether spent any time in temporary accommodation; length of time in temporary accommodation; moves between temporary accommodation addresses; whether ever experienced specific types of temporary accommodation (self-contained; friends and relatives; or hostels and B&B hotels) Type of temporary accommodation• : temporary accommodation was divided into three types for most analytical purposes: self-contained temporary accommodation; temporary arrangements with friends and relatives; and hostels and B&B hotelsUnintentionally homeless• : where someone becomes homeless through no fault of their ownVulnerability clusters• : vulnerability clusters of adult respondents were generated by the K-means method (see Appendix 2). This method grouped adult respondents together if they shared particular personal history characteristics. The ‘more problems in adulthood’ vulnerability cluster was characterised by multiple social deprivation in adult life, and the ‘few problems as an adulthood’ vulnerability cluster by fewer such deprivations. Likewise, two vulnerability clusters of adult respondents were generated on the basis of the degree of multiple social deprivation experienced in childhood (up to the age of 16)Glossary of terms 19Young people accepted as a homeless 16-17 year old Accepted as homeless• : accepted as ‘eligible for assistance’ ‘unintentionally’ ‘homeless’ and in ‘priority need,’ (and therefore owed the ‘main homelessness duty’) where the primary reason for priority need was because the applicant was aged 16 or 17Accommodation conditions• : this group of variables used for analysis included: whether accommodation damp; whether sufficient control over heating; perception of safety inside accommodation; perceptions of safety in neighbourhood in which accommodation is located; and perception of sufficiency of living spaceCurrent accommodation type• : this group of variables used for analysis included: whether in temporary or settled housing; where appropriate, whether in self-contained temporary accommodation (flat/house used only for the young person/their immediate family group) or shared temporary accommodation (hostel, B&B hotel, supported lodgings, or the homes of friends, relatives or parents)Demographic characteristics• : this group of variables used for analysis included: age; gender; ethnicityEmotional support• : whether respondent had someone to count on to listen when they needed to talkEthnic minority background• : all non-White ethnic backgroundsImmediate family• : a young person’s partner, child(ren) and any other members of their immediate household with whom they intended to live, and can be reasonably expected to live, in their settled accommodation3Instrumental support• : whether respondent had someone to count on to help out in a crisisLast settled accommodation• : ‘ordinary housing’ (see below) lived in before acceptance as homeless that was stable enough and recent enough to be used as a comparison point to investigate changes in health, well-being and economic circumstances that may be associated with the experience of homelessness and living in temporary accommodation43 Survey 3 was restricted to those 16-17 year olds who were accepted as having priority need (and therefore owed the main homeless duty) primarily because of their age, and so those young people with a child at the time of acceptance would not be in Survey 3. However, Survey 3 young people may have had a child subsequently, and as in all surveys in this study, eligibility for inclusion in the sample was defined at the point of acceptance, and so their having a child by point of interview would not affect their eligibility for inclusion in Survey 3. 4 See Appendix 1 for a full explanation.20 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsOrdinary housing• : a residential house or flatNEET• : a young person not in education, employment or trainingPersonal support• : advice or help with mental health, or drug or alcohol problemsPractical support• : advice or help with financial matters, with furnishing or maintaining a home, getting a job, accessing health or other services, etc.Self-contained temporary accommodation• : ordinary flat/house used only for the young person/their family groupSettled accommodation• : accommodation offered to someone accepted as a homeless 16-17 year old that discharges the local authority’s duty owed to them under the homelessness legislation. In most cases, the local authority will arrange the offerShared forms of temporary accommodation• : hostel, B&B hotel, supported lodgings, or the homes of friends, relatives or parentsSocial networks• : this group of variables used for analysis included: whether see family more/less since last settled accommodation; whether see friends more/less since last settled accommodationSupported accommodation• : supported lodgings, hostels, and other forms of accommodation with onsite management/supportSupported lodgings• : a lodgings arrangement within the home of another family, where young people are provided with support to develop the skills necessary to live independentlySurvey 3• : a survey of young people accepted as 16-17 year-olds owed the main homelessness duty (see Appendix 1)Temporary accommodation• : accommodation secured by a local authority as a temporary measure for a young person accepted as a homeless 16-17 year old until settled accommodation becomes available Temporary accommodation experiences• : this group of variables used for analysis included: whether spent any time in temporary accommodation; length of time in temporary accommodation; moves between temporary accommodation addresses; whether ever experienced shared temporary accommodationGlossary of terms 21Vulnerability clusters• : vulnerability clusters were generated by the K-means cluster analysis method (see Appendix 2). This method grouped young people together if they shared particular personal history characteristics. The four vulnerability clusters generated were: Cluster One – ‘multiple problems’; Cluster Two – ‘mental health and other problems’; Cluster Three – ‘offending and other problems’; Cluster Four – ‘fewest problems’22 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsList of acronymsASB: Anti-social behaviourB&B hotel: Bed & Breakfast hotel BHPS: British Household Panel Survey BMRB: BMRB Social Research (part of BMRB Limited, British Market Research Bureau)CHP: Centre for Housing Policy, University of YorkCPR: Child Protection RegisterDCLG: Department for Communities and Local GovernmentDEFRA: Department for Environment, Food and Rural Affairs DfES: Department for Education and Skills DoH: Department of HealthDTLR: Department of Transport, Local Government and the RegionsDWP: Department for Work and PensionsEHCS: English House Conditions Survey FACS: Families and Children Study HSE: Health Survey for EnglandLA: Local authorityNASS: National Asylum Support ServiceNEET: not in employment, education or trainingODPM: Office of the Deputy Prime MinisterP1E: Statistical returns made by local authorities to central government on their decisions and actions under the homelessness legislationSEH: Survey of English Housing SEN: Special Educational Needs statement TA: Temporary accommodationYCS: Youth Cohort StudySummary 23SummaryIntroductionThe main aim of this study was to provide robust statistical evidence on families and 16-17 year olds accepted as owed the main homelessness duty5 by English local authorities, in order to inform effective policy interventions. The study focused on the characteristics and support needs of families and 16-17 year olds accepted as homeless; the causes of statutory homelessness; the experience of temporary accommodation; and the impacts of homelessness and stays in temporary accommodation. It drew on data from five linked surveys:• Surveys 1 and 2: parents and children in families accepted as homeless;• Survey 3: young people accepted as homeless 16-17 year olds; and• Surveys 4 and 5: parents and children in families accepted as homeless who had spent more than a year in temporary accommodation.The findings of this study could be viewed as largely a ‘good news’ story with regards to families accepted as homeless, in that these families appeared in the main not to be extremely vulnerable, and the provision of assistance under the homelessness legislation had apparently secured a substantial overall net improvement in their quality of life. Key points of concern include the lengthy periods spent in temporary accommodation by families in London and the South, and the reported deterioration in many families’ (already weak) economic position. In contrast, young people accepted as homeless 16-17 year olds appeared to be an extremely vulnerable group, for whom (supported) temporary accommodation could be viewed as a helpful transitional intervention. As with families, the provision of assistance under the homelessness legislation had apparently brought about a substantial overall net improvement in their quality of life, but seemed to have a negative net effect on their economic circumstances (a very high proportion were not in education, employment or training at point of interview). 5 Hereafter generally referred to as ‘accepted as homeless’.24 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsKey PointsFamilies accepted as homeless were mainly young, headed by lone women parents, and workless. • Adult respondents6 (usually the mothers) in these families seemed to be a relatively disadvantaged group with respect to their health and access to social support, and many had experienced domestic violence. However, only a minority appeared extremely vulnerable and very few self-reported current drug or alcohol problems. Children in these families were generally happy at home and at school and were reportedly in good health. • Young people accepted as homeless 16-17 year olds were, in contrast, an extremely vulnerable group, who had often experienced educational and/or family disruption, violence at home, and mental health and/or substance misuse problems. A very high proportion were not in education, employment or training. • The main reasons for applying as homeless amongst families were relationship breakdown, eviction, overcrowding, or overstaying welcome (although the latter two reasons often seemed to reflect a breakdown in informal arrangements entered into after losing settled accommodation). • For young people, the overwhelming reason for applying as homeless was relationship breakdown with parents or step-parents. • The great majority of both families and young people had sought at least one form of alternative help with their housing problems before seeking assistance from a local authority.• Families’ experience of temporary accommodation was largely determined by where they were accepted as homeless. In particular, those accepted in London, and to a lesser extent in the South, were likely to experience much longer periods in temporary accommodation than those in the North and Midlands. • Self-contained temporary accommodation was the most common form of provision for families with children. Overall levels of satisfaction varied little between this and other forms of temporary accommodation – namely, hostels and B&B hotels, or temporary arrangements with friends or relatives. • Much of the temporary accommodation experienced by young people was ‘supported’ accommodation of various kinds. Most young people seemed to appreciate the company of other young people and the help from staff in such accommodation.66 An ‘adult respondent’ was purposively selected in each couple-headed or multiple adult family as the person best placed to comment on the position and experiences of the whole family (usually this was the mother).Summary 25• Families who had been provided with settled housing (almost always social housing) were markedly more satisfied with their accommodation than those still in (any form of) temporary accommodation. In contrast, young people in settled housing were only marginally more satisfied with their accommodation than those still in temporary accommodation.• For parents, children and young people assisted under the homelessness legislation, life was far more likely to be reported as better rather than worse than when they lived in their last settled accommodation. • The overall (net) impact of homelessness and temporary accommodation on the health and social support circumstances of families and young people seemed largely negligible, or marginally positive, and improvements were often reported in children’s relationships with their parent(s) and in their school performance (since leaving their last settled accommodation). However, there was a substantial (net) negative impact on these families’ and young people’s economic position (since leaving their last settled accommodation), and in children’s participation in clubs/activities. • Families in settled housing reported a consistently better quality of life than those still in temporary accommodation. Whether they were living in temporary or settled accommodation seemed less critical to the quality of life of young people. • The great majority of families in temporary accommodation for over one year had been accepted as homeless in London. These families tended to be larger than other families accepted as homeless, and were more likely to be headed by an adult respondent who had an ethnic minority background and/or who was a former asylum seeker. In most respects the circumstances and quality of life of both adults and children in these families was very similar to that of adults and children in families who had spent shorter periods in temporary accommodation. However, families in temporary accommodation for over one year were less satisfied with the living space and facilities in their accommodation; more likely to report that they were struggling financially; and were very often frustrated at the length of wait for settled housing.7Characteristics and support needs Families accepted as homelessMost families accepted as homeless were headed by a lone woman parent (65 per cent), and usually contained one or two children. The other main household type was couples with children (30 per cent). Very few families were in other sorts of household arrangements. Both parents and children in families accepted as homeless tended to be young (32 per cent of parents were under 25 years old; 50 per cent of children were pre-school age). 7 Adult respondents and young people accepted as homeless 16-17 year olds were asked a range of questions about their circumstances and well-being in their ‘last settled accomodation’ prior to acceptance as homeless, as a means of investigating whether there was evidence of changes that could be associated with the experience of homelessness and temporary accomodation.26 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAdult respondents in families accepted as homeless were overwhelmingly women (84 per cent). While the great majority were White (76 per cent), adult respondents with a Black/Black British background (at 12 per cent) were over-represented as compared with parents in the wider population. Ethnic minority adult respondents were heavily concentrated in London. One in ten (11 per cent) of all adult respondents had, at some point, sought asylum in the UK – these former asylum seekers8 were mainly from ethnic minority groups, tended to be older than other adult respondents, and most were living in London. Families accepted as homeless were far less likely to contain a working member than families with dependent children in the general population: 64 per cent were ‘workless’, compared with 14 per cent of all families with children. Most were in receipt of Income Support and other means-tested benefits or tax credits. They were much more likely to self-report difficulties managing financially (34 per cent) than families with children nationally (10 per cent).Many adult respondents had experienced some family or educational disruption in childhood, and two in five (41 per cent) reported being a victim of domestic violence at some point during their adult lives. However, while they were a relatively disadvantaged group with regards to their health and access to social support, adult respondents did not appear in the main to be extremely vulnerable. Half (52 per cent) self-reported experience of anxiety, depression or other mental health problems, but the proportion who said that they had current mental health problems was much lower (27 per cent) (although this was still somewhat higher than the rate found in the general population (18 per cent)). The proportion of adult respondents with some (self-reported) experience of drug or alcohol problems was 11 per cent, and current drug or alcohol problems were reported by only 3 per cent. White lone parents were the group most likely to have experienced multiple personal problems in childhood and/or adulthood. There was some evidence of stability in many adult respondents’ housing histories, and two thirds (65 per cent) had lived independently in their own mainstream (rented or owned) housing at some point prior to their acceptance as homeless. However, around half had experienced at least one episode of homelessness or insecure housing before the circumstances which led to their acceptance as homeless; most commonly, they had stayed with friends or relatives because they had no home of their own (41 per cent). A much smaller number (8 per cent) had at some point slept rough or in a car or a squat (almost none of whom had had their children with them when they experienced these scenarios). A similar proportion (7 per cent) reported that their family had experienced homelessness when they were a child. In total, one quarter (26 per cent) of all adult respondents reported that they had never had a ‘settled home’ as an adult. Two-thirds (63 per cent) of adult respondents reported that they had received one or more forms of ‘practical support’ from service providers since being accepted as homeless 8 Current asylum seekers are ineligible for assistance under the homelessness legislation.Summary 27(very often help with repairs to their accommodation or with acquiring furniture or other household equipment). However, 35 per cent reported at least one current unmet need for practical support; this was mainly related to practical or financial help with getting furniture or with money management. A much lower proportion (16 per cent) said that they had received help with ‘personal support’ needs since acceptance as homeless (such as with mental health or substance misuse problems or with parenting skills); only 4 per cent reported an unmet personal support need. The children in families accepted as homeless generally appeared happy at school and at home, and were reportedly in good health. Only a small minority seemed to have extremely difficult or fractured family relationships. The majority (77 per cent) of all child respondents reported being very or fairly happy with life, with the youngest children interviewed (8-11 year olds) tending to be happiest overall. Families in temporary accommodation for over a yearFamilies in temporary accommodation for more than one year had a quite distinct profile from that of other families accepted as homeless. The great majority (82 per cent) were accepted as homeless in London, and they had a larger average household size than other families accepted as homeless, both because they were more often headed by couples, and because they tended to have more children. Over half the adult respondents in all families in temporary accommodation for over one year had an ethnic minority background (59 per cent), and one third (33 per cent) reported that they had, at some point, sought asylum in the UK. The personal characteristics and support needs of adult respondents in families in temporary accommodation for over one year were in the main very similar to those of other adult respondents, although they were somewhat less likely to report troubled childhoods (their experience of personal problems in adulthood was very similar). The characteristics and experiences of children in these families reflected those of children in other families accepted as homeless. Young people accepted as homeless 16-17 year oldsTwo thirds of young people accepted as homeless 16-17 year olds were young women, and the remaining third were young men. These young people, over half of whom had turned 18 by point of interview, were a much more vulnerable group than adult respondents in families accepted as homeless. Many had suffered violence at home and other forms of childhood trauma, as well as severe disruption to their education. A far higher proportion of young respondents (37 per cent) had experienced drug or alcohol problems than adult respondents in families accepted as homeless (11 per cent); and 16 per cent had a current substance misuse problem (compared to 3 per cent of adult respondents). Current substance misuse problems were more common amongst the young men (22 per cent) than amongst the young women (12 per cent). Half of all young respondents (52 per cent) had experienced 28 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsdepression, anxiety or other mental health problems; and 33 per cent had current mental health problems (a rate approximately three times that of young people in the general population). Current mental health problems were more common amongst the young women (40 per cent) than amongst the young men (24 per cent).Young people accepted as homeless 16-17 year olds were five times more likely not to be in employment, education or training than young people in the general population (57 per cent as compared with 11 per cent). They were living on very low incomes (median of £45 per week, excluding Housing Benefit), and 35 per cent reported difficulties managing financially (this was similar to the proportion of adult respondents who reported financial problems). A much greater proportion of these young people were in receipt of practical forms of support from service providers than were adult respondents: for example, 43 per cent reported getting assistance with filling in official forms or claiming benefits, as compared with 21 per cent of adult respondents. In addition, they were far likelier than adult respondents to be in receipt of help to facilitate their access to employment, education or training. They were also marginally more likely to be receiving help with mental health and/or drug problems. As with adult respondents, levels of self-identified unmet personal support needs were low. The causes of statutory homelessness Families accepted as homelessThe ‘immediate’ causes of statutory homelessness were predominantly disintegrating social relationships on the one hand, and housing pressures on the other – with most adult respondents identifying only one or the other as the reason they had applied as homeless.Around half (55 per cent) of families applied as homeless from somewhere other than their last settled accommodation. This suggests that many families make short-term accommodation arrangements before applying to a local authority for help.Approximately one quarter of all families accepted as homeless applied from each of the following settings: the private rented sector; the parental home; and friends’ or (other) relatives’ homes. The remaining families applied as homeless directly from a social rented tenancy (11 per cent); owner-occupation (5 per cent); ‘managed’ accommodation (such as hostels or B&B hotels) (10 per cent); or ‘other’ settings (such as tied housing) (3 per cent). Summary 29The most prevalent reason for applying as homeless, cited by 38 per cent of adult respondents, was relationship breakdown (usually, but not necessarily, with a partner). Violent relationship breakdown with a partner affected 13 per cent of all adult respondents. The other major reasons that adult respondents gave for applying as homeless were eviction or being threatened with eviction (usually because a private sector fixed-term tenancy had come to an end) (26 per cent); overcrowding (24 per cent); and ‘outstaying their welcome/could no longer be accommodated’ (20 per cent). However, it should be noted that both overcrowding and overstaying welcome as reasons for applying as homeless sometimes seemed to reflect the breakdown or expiry of informal ‘emergency’ arrangements with friends or relatives, rather than the ‘original’ cause of homelessness.All of the other potential reasons for applying as homeless were identified by only a small minority, including those relating to ‘individual’ personal problems such as drug, alcohol or mental health problems (2 per cent in total). At the same time, purely ‘financial’ reasons, such as the inability to pay the mortgage or rent (7 per cent), were also rarely mentioned. Leaving National Asylum Support Service (NASS) accommodation was mentioned as a reason for applying as homeless by 2 per cent. Only small numbers of adult respondents reported that they had applied as homeless because they perceived this to be the ‘quickest’ (3 per cent) or ‘only’ (6 per cent) way to gain access to social housing. This evidence, coupled with the fact that the great majority (85 per cent) of adult respondents had made efforts to gain alternative help with their housing problems before approaching the council for assistance (usually by asking to stay with friends or relatives or by trying to acquire a private or social tenancy), weighs against suggestions of widespread ‘abuse’ of the homelessness legislation. For 87 per cent this was their first homelessness application, and the majority (70 per cent) reported at least one concern about making a homelessness application (most commonly that they would have to live in a ‘rough’ area).These findings on the immediate causes of homelessness lend some support to arguments for a ‘structural’ understanding of family homelessness, insofar as eviction or being threatened with eviction was more commonly reported as a reason for applying as homeless in the areas of highest housing stress. There is certainly little support for an ‘individual’ analysis of the causes of family homelessness, given the small numbers reporting health problems or substance misuse as contributing to their reasons for applying as homeless. Families in temporary accommodation for over a yearFamilies in temporary accommodation for over one year mainly reported similar reasons for applying as homeless as other families, but they were less likely to say that relationship breakdown had contributed to their homelessness. 30 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsYoung people accepted as homeless 16-17 year oldsFor young people accepted as homeless 16-17 year olds, relationship breakdown (almost always with parents or step-parents) was the overwhelming reason for applying as homeless (70 per cent). Two in five of young people (41 per cent) affected by relationship breakdown with their parents or step-parents reported that violence had been involved. As noted above, these young people had often had traumatic childhoods and frequently had a range of support needs which may well have contributed to the relationship breakdown or other circumstances that led to their homelessness.As with adult respondents, most young people (85 per cent) had tried to do something to address their housing problem before approaching the council for help (most commonly they had asked family or friends to let them stay). Two thirds (64 per cent) reported at least one concern about making a homelessness application, and, as with families, this was most often that they would have to live in a ‘rough’ area. The experience of temporary accommodation and the provision of settled housing Families accepted as homelessThe overall experience of temporary accommodation – including length of stay, type of temporary accommodation experienced, and number of moves between temporary accommodation addresses – was largely determined by where families were accepted as homeless.For example, one fifth (21 per cent) of all families accepted as homeless had been moved directly into settled accommodation without a stay in temporary accommodation. However, this included only 6 per cent of families accepted in London, compared to 30 per cent of families accepted in the North and Midlands. At point of survey (on average 9 months after acceptance as homeless), 55 per cent of families had been provided with settled housing, and 45 per cent were still living in temporary accommodation. However, in London only 18 per cent of families had moved into settled housing, compared to 76 per cent in the North and Midlands. Those accepted in London, and to a lesser extent in the South, were likely to experience prolonged stays in temporary accommodation, and to spend much of their time in self-contained temporary accommodation. Families in the North and Midlands typically experienced a relatively short stay in temporary accommodation (very often temporary arrangements with parents, friends, or (other) relatives) before being moved on to settled housing. Summary 31Amongst those still in temporary accommodation at point of survey, 78 per cent were in self-contained temporary accommodation, and only 2 per cent were in B&B hotels (6-7 per cent were in each of hostels; parents’ houses; and staying with friends or (other) relatives)9. However, some families had stayed in more than one type of temporary accommodation and overall experience of shared forms of temporary accommodation was somewhat higher than was suggested by where families in temporary accommodation were living at point of survey. In all, 59 per cent of families with a temporary accommodation stay had experienced self-contained temporary accommodation (including 84 per cent of those in London); 24 per cent had stayed in a hostel; 15 per cent had stayed in a B&B hotel; 25 per cent had lived in temporary arrangements with parents’; and 27 per cent in temporary arrangements with friends or (other) relatives. Multiple moves between temporary accommodation addresses were rare: only 35 per cent of all families accepted as homeless had stayed at more than one temporary accommodation address, and only 8 per cent had stayed at more than two such addresses. Moves between temporary accommodation addresses were likeliest in London and the South, and the purpose of most of these moves appeared to be to relocate families from shared forms of provision – including B&B hotels, hostels and temporary arrangements with friends and relatives – into self-contained settings whenever it seemed likely that they would be subject to sustained stays in temporary accommodation.Overall satisfaction levels varied little between temporary accommodation types. When adult respondents were asked to rank their temporary accommodation using a score of between 1 and 10 (where 10 was ‘excellent’), the median for all forms of provision was 6. However, different forms of temporary accommodation were perceived to offer distinct advantages and disadvantages. Thus self-contained temporary accommodation was reported to offer better space standards than other forms of provision, and was rated most highly with regards to cooking, sleeping, bathroom and other facilities. On the other hand, and perhaps surprisingly, this type of provision was often reported to have worse physical conditions than other forms of temporary accommodation, particularly with respect to damp, décor and state of repair.Temporary arrangements with friends and relatives, on the other hand, appeared to offer families the best physical conditions and access to the widest range of household items and amenities. Families also felt safest when in this form of temporary accommodation. However, concerns about space and privacy were at their most acute in these arrangements. 9 The disparity with the quarterly P1E statistics published by Communities and Local Government (which reported 84 per cent of households in self-contained temporary accommodation at end June 2005) is mainly attributable to the fact that P1E statistics treat those in temporary arrangements with friends and relatives as an entirely separate category, whereas they are considered alongside all other households in temporary accommodation in this analysis. See http://www.communities.gov.uk/ for information on the P1E statistics.32 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAccess to household items and amenities (including kitchens and living rooms) was often more restricted in hostels and B&B hotels than in other forms of temporary accommodation. However, the worst physical conditions and space standards were not generally found in these forms of temporary accommodation. Families accepted in the North and Midlands tended to report better physical conditions in their temporary accommodation than those accepted elsewhere. This was in part attributable to the relatively low use of self-contained temporary accommodation in this broad region. However, another important factor was that conditions in all forms of temporary accommodation were reported to be better in the North and Midlands than in their equivalents elsewhere. Almost all of the 55 per cent of families in settled housing by point of interview had been provided with social rented housing. Only 25 per cent of these families reported being given any choice over this settled housing. Overall living space and access to gardens was reported as better in settled housing than in self-contained temporary accommodation, and problems with several physical conditions (such as damp, infestation and risks to child safety) were less commonly reported. However, satisfaction with cooking, laundry and, especially, bathroom facilities was much lower amongst adult respondents in settled housing than amongst those reporting on self-contained temporary accommodation. Despite these mixed results with regards to accommodation conditions, adult respondents’ overall satisfaction with settled housing was markedly higher than with self-contained (or indeed any other form of) temporary accommodation. Likewise, children in settled housing were happier with their accommodation and less likely to want to move elsewhere than those still living in temporary accommodation.Families in temporary accommodation for over a yearAll families in temporary accommodation for over one year (average stay at point of survey was 2.9 years) were staying in self-contained temporary accommodation when interviewed. These families were more likely than other families accepted as homeless to have made multiple moves between temporary accommodation addresses (43 per cent had stayed in three or more temporary accommodation addresses). However, most of these moves seemed to have happened early in these families’ temporary accommodation experience, as the average length of time they had spent in their current temporary accommodation address at point of survey was 2.5 years. As with other families accepted as homeless, the purpose of many of these moves appeared to be to relocate families from shared forms of provision into self-contained settings in situations where they were likely to spend a prolonged period in temporary accommodation. Lack of space was more of a problem for families in temporary accommodation for over one year than for other families accepted as homeless (58 per cent were satisfied with their living space, compared with 69 per cent of other families). They were also less satisfied with Summary 33bathroom, cooking and sleeping arrangements than other families accepted as homeless. These findings appeared to be associated with the larger average household size of families in temporary accommodation for over one year, and in particular to the substantial proportion of these families (26 per cent) with five or more members (only 12 per cent of other families accepted as homeless had five or more members). Levels of frustration at the length of wait for settled housing were high amongst the adult respondents in families in temporary accommodation for over one year: more than half (59 per cent) reported that they were ‘very frustrated’ and 28 per cent that they were a ‘bit frustrated’ with the wait. Young people accepted as homeless 16-17 year oldsAt point of survey, 40 per cent of young people had moved into settled housing, and 60 per cent were still living in temporary accommodation. Almost all of these young people had spent some time in temporary accommodation (only 6 per cent had moved directly into settled housing), and half (47 per cent) had experienced some form of ‘supported’ accommodation (that is, a hostel, other managed/supported accommodation, or supported lodgings). While some problems were reported with sharing in these forms of provision, 71 per cent of young people felt that the other young people they shared with were ‘good company’, and 77 per cent felt that the staff in their accommodation were ‘helpful’. Unlike adult respondents, young people in settled housing were only marginally more satisfied with their accommodation than those still in temporary accommodation. This finding, together with the data on overall quality of life (see below), indicates that the meaning and significance of temporary accommodation may well be very different for young people than for families accepted as homeless. For young people, it is perhaps more accurate and helpful to view such accommodation as ‘transitional’ rather than simply as ‘temporary’. The impacts of homelessness and temporary accommodation Families accepted as homelessEncouragingly, those adult respondents who reported that life was now better than in their ‘last settled accommodation’10 heavily outnumbered those for whom it was perceived to be worse (57 per cent as compared with 19 per cent). Likewise, they were far likelier to report an improvement (57 per cent) than a decline (12 per cent) in their child(ren)’s overall quality of life. While positive changes were commonest amongst those families in settled 10 Adult respondents were asked about a range of their family’s circumstances in their ‘last settled accommodation’ prior to acceptance as homeless, as a means of investigating whether there was evidence of changes that could be associated with the experience of homelessness and temporary accommodation.34 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldshousing, a substantial net improvement was also reported for adults and children still in temporary accommodation. Positive change was particularly associated with families for whom violent relationship breakdown had been a cause of homelessness, but was far from limited to this group. Across a range of measures, adult respondents in settled housing reported a better current quality of life than those still living in temporary accommodation. In particular, while the majority of adult respondents in temporary accommodation considered their lives to be ‘on hold’ (64 per cent), this was true of only 18 per cent of those in settled housing. Adult respondents in temporary accommodation were also more likely than those in settled housing to worry about the future (55 per cent compared to 36 per cent), and to report lower levels of overall happiness (44 per cent were very or fairly happy, compared to 68 per cent of adult respondents in settled housing). Quality of life was consistently reported to be poorer amongst adult respondents staying in temporary arrangements with friends or relatives, or in hostels and B&B hotels, than amongst those in self-contained temporary accommodation. Poorer quality of life was also often associated with mental health problems, financial difficulties, feeling unsafe in accommodation or neighbourhood, and having insufficient living space.The impacts of homelessness and spending time in temporary accommodation on the health and social support circumstances of adult respondents seemed largely negligible, or marginally positive. Thus most adult respondents (66 per cent) reported no change in their health status since leaving their last settled accommodation, and where it had changed, their health was more likely to have improved than deteriorated. Their access to emotional support (someone to listen if they needed to talk) and instrumental support (someone to help out in a crisis) had seldom changed since their last settled accommodation. Likewise, there was no net change reported with respect to adult respondents’ contact with relatives since leaving their last settled accommodation, although a net drop in contact with friends was reported (36 per cent had less contact with friends, and 20 per cent had more). Very few had no contact at all with friends or relatives at point of survey. However, there appeared to be a net deterioration in these families’ economic position as compared with when they were living in their last settled accommodation. In particular, while 74 per cent of families had not seen any changes in their working status since their last settled accommodation, 21 per cent had moved from a working to workless status, and this was offset to only a small degree by the 6 per cent of families who had experienced the reverse. Despite this finding, ‘homelessness-specific’ barriers to employment – such as ‘living in temporary accommodation’ or ‘the disruption caused by homelessness’ – were very seldom cited by adult respondents.Overall, 47 per cent of adult respondents reported that their financial circumstances had worsened since leaving their last settled accommodation, while only 18 per cent said that Summary 35they had got better. Families living in self-contained temporary accommodation appeared more likely to struggle financially than those in other forms of accommodation (a finding that was only partly accounted for by their concentration in London, where families in general were more likely to report financial difficulties). Expenses directly associated with moves due to homelessness, such as the purchase of new furniture and household goods, seemed to be relatively minor problems in the context of the overall weak economic position of these families. Some positive (net) changes were reported for children (as compared with their last settled accommodation), particularly with regards to improvements in their school performance and their relationship with their parents. However, some negative (net) changes were also apparent in relation to loneliness and reduced participation in clubs/activities. One third of school-aged children had changed school as a direct result of homelessness. There was evidence that changing schools could have powerful positive as well as negative impacts on children. The perception of parents was that any negative impacts on their children were largely attributable to the initial disruption and uncertainty caused by leaving their last settled accommodation. Likewise, positive changes were generally attributed by parents to moving away from former ‘family problems’ in their last settled accommodation, and the establishment of a more stable home environment.Families in temporary accommodation for over a yearIn most respects, the circumstances and quality of life of both adults and children in temporary accommodation for over one year were very similar to those of adults and children in families who had spent shorter periods in temporary accommodation. This suggested that the length of time spent in temporary accommodation was not generally the key influence with respect to the impacts of homelessness on families.Families in temporary accommodation for over one year were more likely than other families accepted as homeless to report that, overall, they were struggling financially (49 per cent). However, this finding seemed related to their concentration in London, and the form of accommodation in which these families were living (self-contained temporary accommodation), rather than to the length of time they had stayed in temporary accommodation. Young people accepted as homeless 16-17 year oldsYoung people accepted as homeless 16-17 year olds, like families accepted as homeless, were much more likely to report that life was better (52 per cent) rather than worse (25 per cent) than it had been in their last settled accommodation. For the minority of young people who perceived their quality of life to have declined, this was associated with feeling unsafe in their current neighbourhood, and also with deteriorations in their ability to cope financially (see below). Notably, it was not associated with whether they were living in settled or temporary accommodation, nor with temporary accommodation type.36 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsYoung people still living in temporary accommodation were, as with adult respondents, much more likely than those in settled housing to perceive their life to be ‘on hold’ (57 per cent as compared with 18 per cent). However, in contrast to adult respondents, neither worrying about the future nor general levels of (un)happiness were associated with living in temporary accommodation amongst young people. For these young people, the key negative influences on quality of life appeared to be feeling unsafe in their accommodation and/or neighbourhood. It was also notable that a smaller proportion of young people reported being very or fairly happy (47 per cent) than either adult respondents (57 per cent) or child respondents (77 per cent). There was an overall net reduction in young people’s contact with family and friends since leaving their last settled accommodation. Nonetheless, their access to emotional support (someone to listen if they needed to talk) and instrumental support (someone to help out in a crisis) appeared to have improved overall (primarily because of increased professional support), albeit that this was still poorer than the level of support available to young people in the general population. There was, as with families, evidence of a substantial overall (net) deterioration in the economic position of these young people (since leaving their last settled accommodation). Thus approximately one third (34 per cent) had discontinued participation in education, employment or training, and this was offset to only a very small degree by the 4 per cent who had taken up one of these activities. Moreover, 56 per cent of young people reported that their ability to manage financially had declined since leaving their last settled accommodation, and only 12 per cent said that it had improved. Conclusion This study sought to provide robust statistical evidence on families and 16-17 year olds accepted as homeless by English local authorities, drawing on data from five linked surveys which covered parents, children and young people assisted under the homelessness legislation (including parents and children who had spent more than one year in temporary accommodation). The findings of this study could be viewed as largely a ‘good news’ story with regards to families accepted as homeless. These families appeared in the main not to be extremely vulnerable, but rather were generally low income households who found themselves unable to secure alternative housing when they were confronted with a crisis such as relationship breakdown or eviction which caused them to lose their settled accommodation. The provision of statutory homelessness assistance seemed to have secured a substantial overall net improvement in the quality of life for both adults and children in these families. Moreover, those families (mainly in the North and Midlands) who had moved on to settled housing by point of interview appeared reasonably satisfied with their accommodation. However, the long waits for settled housing in London and the Summary 37South were a source of considerable frustration. Another key note of concern has to be the apparent negative impact of homelessness on families’ (already weak) economic position. For young people accepted as homeless 16-17 year olds, the data tells quite a different ‘story’. This is an extremely vulnerable group, in need of extensive support, for whom (supported) temporary accommodation could be viewed as a helpful transitional intervention. As with families, the provision of statutory homelessness assistance appeared to have brought about a substantial overall net improvement in young people’s quality of life, and had also increased their access to professional sources of support. However, the pronounced negative (net) impact on these young people’s economic position, and in particular the very high proportion who were not in education, employment or training at point of interview, is clearly a cause for concern. 38 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsChapter 1 Introduction 39Chapter 1:IntroductionIntroductionThe Housing (Homeless Persons) Act 1977 first established a statutory safety 1.1 net for certain groups of homeless people11. This statutory framework provides that ‘eligible’12 households found to be ‘homeless’13, to be in ‘priority need’14, and to be ‘unintentionally homeless’15 are entitled to be ‘accommodated’ by the local authority to which they applied as homeless16. Strictly speaking, this ‘main homelessness duty’, is to secure temporary accommodation until suitable settled housing becomes available, found either by the household itself or by the local authority. However, in practice settled housing is almost always secured by the local authority that owes a duty under the homelessness legislation.Households with dependent children and pregnant women are amongst the 1.2 key groups with ‘priority need’ status, and from the outset have comprised the majority of households accepted as owed the main homelessness duty in England. The numbers of households accepted as homeless rose during the late 1990s, and reached a peak in 2003 when 135,590 households were accepted, of whom 82,790 were families with dependent children or an expectant mother17. In parallel with this, the number of households in temporary accommodation awaiting settled housing also increased in the latter part of the 1990s, reaching 101,300 in 200418.11 With respect to England, these rights were subsequently re-enacted in the Housing Act 1985, then replaced with some modifications by Part 7 of the Housing Act 1996. Part 7 has also been amended by the Homelessness Act 2002.12 Certain categories of ‘persons from abroad’, including asylum seekers, are ‘ineligible’ for assistance under the homelessness legislation. 13 The statutory definition of ‘homelessness’ includes those without any accommodation in the UK which they have a legal right to occupy and in which they can live together with their whole household. It also includes those who cannot gain access to their accommodation, or cannot reasonably be expected to live in it (for example because of a risk of violence.)14 The ‘priority need’ groups include applicants whose household contains a dependent child; a pregnant woman; or someone who is vulnerable because of age, disability or some other reason. They also include young people aged 16 or 17 (or 18-20 years old if formerly in local authority care). They also include adults who are ‘vulnerable’ because of time spent in care, custody or the armed forces or because of having fled their homes because of violence. 15 A homeless applicant who has deliberately done or failed to do something in consequence of which they have lost accommodation which was available and reasonable for them to occupy is ‘intentionally’ homeless. An act in good faith in ignorance of a material fact is not considered to be ‘deliberate’. 16 The only exception to this is that, if the applicant’s household has no ‘local connection’ with the district of the local authority to whom they applied, and does have a local connection with the district of another local authority in England, Wales or Scotland, then, subject to certain conditions (e.g. no risk of violence in the other area) responsibility can be transferred to the latter local authority. 17 Information on ‘decisions made’ under the homelessness legislation, including households accepted as owed the main homelessness duty, is recorded in local authority ‘P1E’ returns, which are published by central government. For information on the P1E returns from local authorities to central government see: http://www.communities.gov.uk/18 The number of households in temporary accommodation at the end of each quarter is recorded in local authority P1E returns, which are published by central government. For information on the P1E returns from local authorities to central government see: http://www.communities.gov.uk/40 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsIn response to these rising levels of statutory homelessness and increased 1.3 use of temporary accommodation, Government placed a new emphasis on homelessness prevention19. A statutory duty was introduced for local authorities to produce a strategy for preventing homelessness and for ensuring that sufficient accommodation and support are available for people who are homeless or at risk of homelessness20, and dedicated central government funding was provided to support homelessness prevention activities. There has been a subsequent decline in the numbers of households accepted as homeless in England, down to 64,970 during 2007, of whom 45,070 were families with children or expectant mothers. Likewise, by the end of December 2007, the numbers of households in temporary accommodation had dropped to 79,500, of whom three quarters (59,990) were families with dependent children and/or a pregnant woman21. Since January 2005 there has been an official target to halve the total number of households in temporary accommodation, from the December 2004 level, by 201022. Since 2002, when the priority need categories under the homelessness 1.4 legislation were extended23, all homeless 16 and 17 year olds have had a priority need for accommodation on the grounds of their age alone24. Previously, homeless 16 and 17 year olds were not accepted as being owed the main homelessness duty unless the local authority was satisfied that they were ‘vulnerable’ for some special reason. Very few, if any,were accepted. In 2006/7, 5,650 households were accepted as owed the main homelessness duty primarily because the applicant was in priority need through being aged 16 or 17. This represents a substantial reduction in absolute numbers from 10,060 in 2003/4, but this group has consistently accounted for around 8 per cent of total acceptances since the legal change to their priority need status in 2002. There has been longstanding policy interest in the causes of statutory 1.5 homelessness amongst families with children and 16-17 year olds. Linked with this, there have been persistent concerns about the potential impact of temporary accommodation, particularly prolonged stays in such accommodation, on the health, well-being and economic circumstances of parents, children and young people accepted as homeless. 19 Pawson, H., Netto, G. and Jones, C. (2006) Homelessness Prevention: A guide to good practice, London: Communities and Local Government; Pawson, H,, Netto, G., Jones, C., Wager, F., Fancy, C. and Lomax, D. (2007) Evaluating Homelessness Prevention, London: Communities and Local Government. 20 Homelessness Act 2002.21 Source: Statutory Homelessness: 4th Quarter 2007, England (Communities and Local Government) http://www.communities.gov.uk/22 Office of the Deputy Prime Minister (2005) Sustainable Communities: Homes for all. A five year plan from the Office of the Deputy Prime Minister, London: ODPM.23 The Homelessness (Priority Need for Accommodation) (England) Order 2002. (SI 2002/2051). 24 The only exception is 16-17 year olds who are either a ‘relevant child’ or a ‘child in need’ in terms of the Children Act 1989, where responsibility for arranging suitable accommodation rests with the children’s services authority. Chapter 1 Introduction 41This report presents the findings of the first major quantitative study of the 1.6 causes, experiences and impacts25 of statutory homelessness amongst families with children and 16-17 year olds in England. The study was commissioned in December 2004 by the then Office of the Deputy Prime Minister (ODPM) (now Communities and Local Government). It was conducted by the Centre for Housing Policy (CHP) at the University of York (www.york.ac.uk/chp) and BMRB Social Research (British Market Research Bureau) (www.bmrb.co.uk). This introductory chapter provides a brief overview of previous research on 1.7 family homelessness and homelessness amongst 16-17 year olds, before describing the aims and methods of the present study. The data analysis process is then outlined, with particular attention paid to the approach taken to investigating the ‘impacts’ of homelessness and temporary accommodation. Finally, the structure and content of the remainder of the report are described. Previous research on family homelessness and homelessness amongst 16-17 year oldsThere is a substantial volume of previous research on family homelessness 1.8 in the UK, particularly in London, but this body of literature suffers from significant limitations26. First, the great majority of this research is qualitative and/or small-scale 1.9 in nature. Debate on the ‘causes’ of homelessness in particular has thus tended to be either theoretical in its orientation or based largely on such qualitative research27. While this literature provides rich insights into the experiences of families, and highlights their possible support needs, it is not designed to assess the overall scale or pattern of such needs or experiences. Meanwhile, the more statistically robust studies have tended to focus on very narrow topics, such as the use of hospital services by families accepted as homeless28. A broader overview report on the support needs of homeless people, published in 2003, drew attention to concerns such as social isolation, poverty, substance misuse and mental health problems amongst 25 An attempt was made in this research to measure the Exchequer costs (or savings) associated with changes in service use by families and 16-17 year olds accepted as homeless, but the data proved insufficiently robust to be published in this report. 26 This brief overview of the UK and US literature is based on a substantial international literature review undertaken as part of the feasibility study for this research project. 27 For an overview, see Fitzpatrick, S. (2005) ‘Explaining homelessness: a critical realist perspective’, Housing, Theory and Society, 22, (1) 1-17.28 Victor, C.R., Connelly, J., Roderick, P. Cohen, C. (1989) ‘Use of hospital services by homeless families in an inner London health district’ British Medical Journal, 229, 725-7; Richman, S., Roderick, P., Victor, C.R. and Lissauer, T. (1991) ‘Use of acute hospital services by homeless children’, Public Health, 105 (4) 297-302; Lissauer, T., Richman, S., Tempia, M., Jenkins, S., Taylor, B. and Spencer, N.J. (1993) ‘Influence of homelessness on acute admissions to hospital’, Archives of Disease in Childhood. 93 (4) 423-429.42 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldshomeless parents, and behavioural, educational and health needs amongst their children29. But, again, this research was not designed to estimate the prevalence of such needs. Without robust data on the scale of support needs amongst families accepted as homeless, the level of support provision required for these families (for example, under the Supporting People programme30) remains unclear. There is a particular danger that perceptions derived entirely from qualitative studies, which tend to focus on the neediest groups with high levels of service intervention, may lead to an exaggerated sense of the extent of support needs amongst families accepted as homeless.Second, previous research on family homelessness has tended to be very 1.10 narrow in scope, with attempts to examine the impacts of homelessness heavily skewed towards examinations of the experiences of families placed in Bed & Breakfast hotels (B&B hotels), in London31. These studies all report the now familiar findings of inappropriately small rooms; poor, shared, bathrooms and inadequate cooking facilities; alongside general disrepair and poor hygiene. However, Government (P1E) statistics indicate that this popular image of families living in B&B hotels no longer reflects the reality of most temporary accommodation experience in England, with self-contained temporary accommodation and ‘homeless at home’ living arrangements with friends or relatives now far commoner, but much less well researched32. The last systematic work on conditions and experiences across different forms of temporary accommodation in England was conducted in the late 1980s33. This leads on to the third main weakness, which is the 1.11 very dated nature of much of the research on family homelessness. This is already clear from the reference just noted to key research on temporary accommodation dating back to the 1980s, and much of the work on London’s B&B hotel situation (see para 1.10 above) was likewise conducted in the late 1980s or early 1990s. It is also worth noting that the only previous large-scale survey of households experiencing the statutory homelessness system was published in 199634. In any case, this (longitudinal) research, which included both families 29 Randall, G. and Brown, S. (2003) The Support Needs of Homeless People and their Families. London: ODPM 30 Pleace, N. and Quilgars, D. (2003) Supporting People: Guide to accommodation and support options for homeless households, London: ODPM, Homelessness Directorate.31 Bayswater Hotel Homelessness Project (1987) Speaking for Ourselves: Families in Bayswater B&B, London: Bayswater Hotel Homelessness Project; Howarth, V. (1987) A Survey of Families in Bed And Breakfast Hotels: Report to the governors of the Thomas Coram Foundation for Children, London: Thomas Coram Foundation for Children; Murie, A. and Jeffers, S. (1987) Living in Bed and Breakfast: the experience of homelessness in London, Bristol: University of Bristol, School for Advanced Urban Studies; Crane, H. (1990) Speaking from Experience: Working with homeless families, London: Bayswater Hotel Homelessness Project; Carter, M. (1995) Out of Sight: London’s Continuing B&B Crisis, London: London Homelessness Forum; Carter, M. (1997) The Last Resort: Living in bed and breakfast in the 1990s, London: Shelter; Sawtell, M. (2002) Lives on Hold: Homeless families in temporary accommodation, London: The Maternity Alliance; Kings Cross Homelessness Project (2002) Report on Life in Bed and Breakfast Accommodation, London: Kings Cross Homelessness Project. 32 See Chapter 6 of this report for a detailed analysis of temporary accommodation use.33 Thomas, A. and Niner, P. (1989) Living in Temporary Accommodation: A survey of homeless people, London: HMSO.34 O’Callaghan, B., Dominion, L., Evans, A., Dix, J., Smith, R., Williams, P. and Zimmeck, M. (2006) Study of Homeless Applicants, London: HMSO.Chapter 1 Introduction 43and single people, was concerned mainly with the process and (housing) outcomes of the statutory system, rather than with broader issues such as the causes of statutory homelessness, the support needs of applicant households, or the impacts of temporary accommodation on parents and children. Much of the existing research available on homeless 16-17 year olds has 1.12 been collected as part of the wider body of research on ‘youth homelessness’ among 16-25 year olds in the UK. There is a very large volume of literature on youth homelessness in the UK, which highlights the extreme vulnerability of some of these young people due to their background35, as well as associated factors such as poor physical and mental health36, risky behaviour37, and wider social exclusion issues38. But again, with only a few exceptions, this research is qualitative rather than statistically robust. Whilst there have been a number of recent evaluations of specific youth homelessness prevention initiatives39, there has been little research on young people’s experience of temporary accommodation (other than specialist provision such as foyers40). Moreover, almost all of the youth homelessness literature has focused on ‘non-statutory’ homelessness rather than the experience of those accepted as homeless; this is unsurprising given that, until 2002, most young homeless people, including 16-17 year olds, were not considered to be in priority need (see para 1.4 above)41. Thus the current body of research leaves us with important gaps, particularly 1.13 with respect to statistically rigorous data on the characteristics and support needs of families accepted as homeless, the experiences that led them in to homelessness, and the impacts of temporary accommodation on both parents and children. Likewise, the evidence on 16-17 year olds is mainly confined to qualitative insights, and does not focus on those accepted within the statutory framework. 35 Hutson, S. and Liddiard, M. (1994) Youth Homelessness: The construction of a social issue, London: Macmillan; Smith, J., Gilford, S. and O’Sullivan, A. (1998) The Family Background of Homeless Young People, York: Joseph Rowntree Foundation; Bruegel, I. and Smith, J. (1999) Taking Risks: An analysis of the risks of homelessness for young people in London, London: Safe in the City; Fitzpatrick, S. (2000) Young Homeless People, London: Macmillan; Randall, G. and Brown, S. (2002) Trouble at Home: Family conflict, young people and homelessness. London: Crisis. 36 Craig, T. K. J., Hodson, S., Woodward, S. and Richardson, S. (1996) Off to a Bad Start: A longitudinal study of homeless young people in London, London, Mental Health Foundation; Commander, M., Davis, A., McCabe, A. and Stayner, A. (2002) ‘A comparison of homeless and domiciled young people’ Journal of Mental Health, 11, 557-564; Vasiliou, C. (2006) Making the Link Between Mental Health and Youth Homelessness: A pan-London study, London, Mental Health Foundation. 37 Klee, H. and Morris, J. (1997) Risk Behaviour Among Young Homeless Users. Coping strategies and self-protection, Swindon: Economic and Social Research Council; Wincup, E., Buckland, G. and Bayliss, R. (2003) Youth Homelessness and Substance Use: Report to the Drugs and Alcohol Research Unit, London: Home Office. 38 Social Exclusion Unit (2005) Transitions: Young adults with complex needs, London, Office of the Deputy Prime Minister; York Consulting Limited (2005) Literature Review of the NEET Group, Edinburgh: Scottish Executive. 39 Dickens, S. and Woodfield, K. (2004) New Approaches to Youth Homelessness Prevention: A qualitative evaluation of the Safe in the City cluster scheme, York, Joseph Rowntree Foundation; Quilgars, D., Jones, A. and Pleace, N. (2005) Safe Moves: An evaluation, York, Centre for Housing Policy, University of York; Smith, J. and Ravenhill, M. (2006) What is Homelessness?, London: Cities Institute, London Metropolitan University. 40 Anderson I. and Quilgars, D. (1995) Foyers for Young People: Evaluation of a pilot project, York: Centre for Housing Policy; Maginn, A., Frew, R., O’Regan, S. and Kodz, J. (2000) Stepping Stones: An evaluation of foyers and other schemes serving the housing and labour market need of young people, London: Department of the Environment, Transport and the Regions. 41 The exception to this was an early Shelter study on the priority need changes. (Anderson, I. and Thomson, S. (2005) More Piority Needed: The impact of legislative change on young homeless people’s access to housing and support, London: Shelter.)44 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsResearch aimsThe main aim of this study was to provide nationally representative statistical 1.14 evidence on families and 16-17 year olds accepted as owed the main homeless duty by English local authorities42, in order to accurately inform effective policy interventions. The key areas on which the study focused were as follows:• the characteristics of families and 16-17 year olds accepted as homeless, including the nature and extent of support needs amongst these households;• the causes of statutory homelessness amongst families and 16-17 year olds;• the experience of temporary accommodation; • the impacts of homelessness on parents, children and 16-17 year olds, and in particular the impacts (both positive and negative) of stays in temporary accommodation, differentiated by: – the type of temporary accommodation experienced; – the duration of stay in temporary accommodation; and – the number of moves between different temporary accommodation addresses.• the role that demographic characteristics (such as age, ethnicity, gender and household type) and geographical variables (such as local housing market pressures, levels of deprivation, and the extent of rurality) may play in the causation, experiences and impacts of family homelessness and homelessness among 16-17 year olds. Research methodsThe study comprised five linked surveys, all conducted by BMRB and analysed 1.15 by CHP: • Survey 1: a survey of adults in families accepted as homeless. • Survey 2: a survey of children (aged 8-15 years old) in families accepted as homeless. • Survey 3: a survey of young people accepted as homeless 16-17 year-olds.42 Hereafter generally referred to as ‘accepted as homeless’Chapter 1 Introduction 45• Survey 4: a survey of adults in families accepted as homeless and who had stayed in temporary accommodation for more than a year.• Survey 5: a survey of children in families accepted as homeless and who had stayed in temporary accommodation for more than a year.The sample population for ‘Survey 1’ was all families with dependent 1.16 children or an expectant mother accepted as homeless by English local authorities over a six month period between 1 January 2005 and 30 June 2005. One adult was selected for interview from each of these families (‘the adult respondent’). These adult respondents were ‘purposively’ selected as the person best able to comment on the position and experiences of the whole family (and was usually the mother). As a result, these adult respondents are not (and were not intended to be) statistically representative of all adults in families accepted as owed the main homelessness duty over the relevant time period. In all households sampled for Survey 1, one child aged between 8 and 15 years 1.17 (where there was such a child in the household) was randomly selected for interview. These ‘child respondents’ provided the data reported as ‘Survey 2’. The sample population for ‘Survey 3’ was all young people accepted by 1.18 English authorities as homeless 16-17 year olds over the same time period as Survey 1 (i.e. 1 January 2005 to 30 June 2005). This included not only 16-17 year olds living alone, but also those applying as homeless with other household members, so long as the reason for priority need status recorded by the local authority was that the applicant was aged 16 or 1743. This ‘time window’ approach (i.e. based on homelessness acceptances over a 1.19 given period) to defining the sample populations for Surveys 1, 2 and 3 was adopted instead of a simple ‘cross-sectional’ survey of those in temporary accommodation at a certain point in time because the latter would over-represent the experiences of families and 16-17 year olds with prolonged stays in temporary accommodation at the expense of those with a shorter (or no) temporary accommodation stay44. This means that the samples 43 Survey 3 was restricted to those 16-17 year olds who were accepted as owed the main homeless primarily on the grounds of their age. There will have been other 16-17 year olds accepted as homeless over the relevant time period whose priority need status will have been based on other grounds (e.g. having a dependent child). As in all surveys in this study, eligibility for inclusion in the sample was defined at the point of acceptance as statutory homeless. Thus if families, or 16-17 year olds, fitted the relevant criteria when they were accepted by a local authority, their changed circumstances by point of interview was no bar to their inclusion in the survey samples. 44 The reason for this is the greater probability that a household has of being selected for a cross-sectional study the longer that it stays in temporary accommodation (see Wong, Y-LI. (1997) ‘Patterns of Homelessness: A review of longitudinal studies’, in D.P. Culhane and S.P. Hornburg (eds) Understanding Homelessness: New policy and research perspectives, Washington DC: Fannie Mae Foundation.) Regional differences within England were an important influence on the decision to use a time-window approach, as families accepted as homeless in the North and Midlands are much less likely to experience protracted stays in temporary accommodation than those in many parts of the South and (especially) in London, and their experiences would therefore have received insufficient coverage in a cross-sectional survey. 46 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsselected for the study reflected, and were intended to reflect, the profile of families and 16-17 year olds accepted as homeless over time (an ‘inflow’ sample) rather than the profile of families or 16-17 year olds in temporary accommodation at any given point in time (a ‘stock’ sample) (see Chapter 2)45. While the time-window approach was effective in delivering a nationally 1.20 representative sample of those accepted as homeless over a period of time, it did have the important limitation that, by definition, it excluded those families in temporary accommodation for extended periods. Clearly, the experiences and impacts associated with prolonged stays in temporary accommodation are of great policy interest. Thus two further surveys were conducted to deliver nationally representative data on families accepted as homeless and in temporary accommodation for more than one year46: ‘Survey 4’ (data collected from adult respondents in these families) and ‘Survey 5’ (data collected from child respondents in these families)47. Across Surveys 1, 2 and 3, disproportionate sampling of local authority areas 1.21 which made the heaviest use of temporary accommodation was employed; this was to ensure sufficient representation of key minority groups, including ethnic minorities, and those in less commonly used forms of temporary accommodation. A design weight was subsequently applied to correct for this intentional bias in the sampling, and thus deliver a nationally representative sample for analysis. Disproportionate sampling did not need to be employed for Surveys 4 and 5 because the ‘key minority’ groups of Surveys 1-3 represented the majority in Surveys 4 and 5. All of the analysis presented in this report is based on weighted data, so that 1.22 the results can be taken as nationally representative of the sample populations defined for this study. Actual sample numbers (‘base sizes’) are provided on tables and graphics to indicate the robustness of each finding48. The total number of usable interviews from each survey was: 2053 (Survey 1), 1.23 450 (Survey 2), 350 (Survey 3), 571 (Survey 4), 180 (Survey 5)49. For statistical tests of difference we have used an estimated effective sample size rather than the actual sample size. The cluster sample design used for all five surveys 45 Chapter 2 provides comparisons between the demographic profile of our sample and the (limited) P1E data available on the demographic profile of all families accepted as homeless over the time period when the research was carried out (first six months of 2005).46 According to P1E homelessness returns, around a quarter (26 per cent) of those who left temporary accommodation in England in 2005 had been staying in it for more than one year. As this data only relates to those who have managed to move on from temporary accommodation, it is likely that the overall proportion of households in temporary accommodation who will eventually spend more than a year there will be higher. 47 BMRB attempted a survey of young people accepted as homeless and in temporary accommodation for at least one year. However, the achieved sample size was too small for robust statistical analysis. 48 The ratio of one sub-group’s base size to another may be quite different from what would be expected given the estimated prevalence of each sub-group. This is normal with disproportionate sampling and should not cause concern.49 These base figures exclude a very small number of ineligible households that were interviewed in error.Chapter 1 Introduction 47and the disproportionate sample design used for Surveys 1-3 means that the effective sample size is smaller than the actual sample size. This ‘design effect’ varies from statistic to statistic and from sub-group to sub-group but we have assumed a conservative standard in which the effective sample size is half that of the actual sample size. This conservative approach enhances the statistical reliability of the findings presented in this report. Further details of the sampling and fieldwork strategy, response rates, and 1.24 weighting applied are provided in Appendix 1 of this report. Data analysis and presentationMuch of the analysis presented in this report has been undertaken using 1.25 frequency counts, bivariate analysis (in the form of crosstabulations employing chi-square tests50) and various measures of central tendency (average and median) and dispersion (standard deviation). Where appropriate, however, binary logistic regression has been employed for multivariate analysis. Such regression analysis allows for exploration of which variables have an independent effect in determining the likelihood of a given finding, when a range of other factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected using regression techniques. See Appendix 2 for further details on the bivariate and regression analysis undertaken, including a list of the standard variables routinely used in both types of analysis and identified by shorthand descriptions in the text. To further assist interpretation, BMRB used the K-means method to 1.26 cluster together those with similar life histories. This method was mainly employed to enhance Survey 3 analyses but a simpler version was also used with Survey 1 adult respondents (see Appendix 2 for details of this cluster analysis).When interpreting the statistics reported, readers should bear in mind that:1.27 • all associations reported are statistically significant at the 95 per cent level of confidence or above (i.e. p<0.05)51. Thus when it is reported that there was ‘no association’ between certain variables, what is meant is that there was no association that reached this level of statistical significance52. 50 See Appendix 2.51 There are a very limited number of exceptions to this, all of them in Chapter 13, where it is explicitly noted that the reported associations do not quite reach statistical significance at the 95 per cent confidence level.52 It is therefore possible that on some occasions where it is reported that there is no association between variables, there may in fact be a relationship, but our sample may not be large enough to detect it at the required level of statistical significance. 48 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds• the ‘margins of error’ on all percentages (‘point estimates’) discussed in the text are within a +/-10% boundary (though individual cells in tables may on occasion exceed this margin of error). Where statistically significant differences have been identified (at the 95 per cent confidence level), but the relevant point estimates exceed a +/-10% margin of error, the results are described but the actual percentages are not stated in the text. • all percentages are rounded up or down to the nearest whole number.• base sizes will often be less than overall actual sample sizes (see para 1.23 above) for each survey, due to missing data.We have, wherever possible, provided broad contextual data in which to 1.28 situate the results from our sample. Where national data sets (e.g. the British Household Panel Survey and Health Survey for England) have been used for comparative purposes we have adhered to weighting guidance published by the data providers. Where appropriate, we have also filtered the data by age, gender, household type and other variables to make it as closely comparable to our survey data as possible. A guide to the terminology used in this report can be found in the ‘Glossary 1.29 of terms’ (see above).Investigating the impacts of homelessness and temporary accommodationAs noted above, one of the key aims of this research was to investigate 1.30 the impacts of homelessness and stays in temporary accommodation on adults, children and young people accepted as homeless. Adult respondents were therefore asked about various aspects of their health, well-being and economic circumstances, and those of their children, in their ‘last settled accommodation’ prior to acceptance as homeless, as a means of investigating whether there was evidence of any changes that could be associated with the experience of homelessness and temporary accommodation. Likewise, young people accepted as homeless 16-17 year olds, were asked about their ‘last settled accommodation’ to identify changes that may be attributable to homelessness and temporary accommodation.This ‘last settled accommodation’ could be self-defined by the respondent, 1.31 or, failing that, could be a questionnaire-defined last settled accommodation. In either case, to be used as a ‘valid comparator’ for the purposes of the analysis of possible impacts of homelessness and temporary accommodation, a respondent’s last settled accommodation had to fulfil the following criteria:Chapter 1 Introduction 49• it had to be ‘ordinary’ housing rather than ‘managed’ accommodation (such as a hostel, supported accommodation, or a B&B hotel); • the family/young person had to have lived there within the last 2 years;• the family/young person had to have lived there for at least 6 months; • it had to be situated within the UK; and• it could not be the family/young person’s current accommodation.Most (71 per cent) families in Survey 1 had a last settled accommodation that 1.32 fulfilled the criteria to be deemed a valid comparison point for the purposes of this research, and 29 per cent did not. Likewise, two thirds (66 per cent) of young people accepted as homeless 16-17 year olds had a last settled accommodation that was valid for comparison purposes (in most cases their parental home). All analysis of ‘changes since last settled accommodation’ is restricted to the families/young people who reported a last settled accommodation that provided such a valid comparison point53. See Appendix 1 for a definition and fuller explanation of the concept of last 1.33 settled accommodation as employed by the questionnaires in this study. Structure of reportThe remaining chapters of this report cover the following topics1.34 54:• Chapter 2: the demographic characteristics of families accepted as homeless, and their distribution across England;• Chapter 3: the personal and housing history of adult respondents in families accepted as homeless; • Chapter 4: the circumstances and intentions of adult respondents when they sought help from a local authority, and their experience of the application process;• Chapter 5: the reasons why families applied as homeless;• Chapter 6: families’ ‘pathways’ through temporary accommodation after acceptance as homeless, including length of time (if any) spent in temporary accommodation, the types of temporary accommodation experienced, and moves between temporary accommodation addresses;53 For almost all (99 per cent) of adult respondents in families accepted as homeless, either a respondent-defined or a questionnaire-defined last settled accommodation could be identified, even though in a proportion of (respondent-defined) cases this was not valid as an (objective) comparison point. This broader definition of last settled accommodation (i.e. not restricted to those cases where it provided a valid comparison point) was used to investigate the ‘origins’ of family homelessness (see Chapter 4).54 Please note that the analysis of the position of families (adults and/or children as appropriate) in temporary accommodation for more than one year is included at the end of all relevant chapters.50 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds• Chapter 7: families’ experiences of the physical and other conditions in temporary accommodation;• Chapter 8: families’ experience of the physical and other conditions in settled housing provided after acceptance as homeless; • Chapter 9: the health and social support needs of adult respondents in families accepted as homeless, and any changes in these that may be attributable to the experience of homelessness and temporary accommodation; • Chapter 10: the employment and financial circumstances of families accepted as homeless, and any changes in these that may be attributable to the experience of homelessness and temporary accommodation;• Chapter 11: the needs and experiences of children in families accepted as homeless;• Chapter 12: the needs and experiences of young people accepted as homeless 16-17 year olds; and• Chapter 13: the overall quality of life of adults and children accepted as homeless, and young people accepted as homeless 16-17 year olds, and any changes in their quality of life that may be attributable to the experience of homelessness and temporary accommodation. Chapter 2 The characteristics of families accepted as homeless 51Chapter 2:The characteristics of families accepted as homelessIntroduction The overall objective of this study was to understand the causes, experiences 2.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data was collected in five separate surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys). This chapter describes the characteristics of families that participated in the 2.2 main survey conducted for this study – a survey of families accepted between 1st January 2005 and 30 June 2005 as being owed the main homelessness duty (‘Survey 1’). As explained in Chapter 1 (see paras 1.19 and 1.20), a ‘time-window’ 2.3 sampling design was adopted for Survey 1 because a simple cross-sectional survey of those in temporary accommodation at a certain point in time would fail to capture the the full range of statutory homelessness experiences, and in particular would neglect the experiences of those with shorter stays in temporary accommodation. Thus the profile of families in Survey 1 reflected (and was intended to reflect) those families accepted as homeless over a period of time (i.e. an ‘inflow’ sample) rather than the ‘stock’ of those in temporary accommodation at any given point in time55. However, this survey design has the important limitation that, by definition, it excludes those families in temporary accommodation for extended periods. Because the experience of households in temporary accommodation for longer periods was of particular policy concern, a separate survey of families accepted as being owed the main homelessness duty and in temporary accommodation for more than a year (‘Survey 4’) was also conducted. The characteristics of these Survey 4 families and adult respondents are compared to those in Survey 1 in the last section of this chapter. 55 This means that the comparisons made throughout this chapter are to local authority (P1E) data on all homeless acceptances (of families with children) in the relevant ‘time window’ (1st January 2005 and 30 June 2005), rather than to the ‘stock’ of families in temporary accommodation at the end of each quarter. 52 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsVarious aspects of disadvantage (such as lone parenthood, worklessness and 2.4 an ethnic minority background) have long been thought to be associated with family homelessness56. This chapter documents whether these characteristics were in fact disproportionately present amongst a nationally representative sample of families accepted as homeless57, and provides an important context for interpreting the findings of the remaining chapters of this report. The topics covered in this chapter (for both Survey 1 and Survey 4 families) 2.5 include: • the characteristics of families accepted as homeless: household size, type and composition, and working status; • the characteristics of the adult respondents within these families: including age, gender and ethnicity, and whether they had ever sought asylum in the UK; and • the geographical distribution of these families across England. As in other chapters, we present two types of statistical analysis here: 2.6 bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a range of other factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected58.This survey evidence confirms the results of earlier research in many respects. 2.7 In particular it reinforces the ‘gendered’ nature of statutory homelessness, in that it is experienced mainly by lone mothers and their children. Another key point is the young age of both these female lone parents and their children. Their economic disadvantage is demonstrated through the low levels of paid work within these households. Also consistent with earlier research is the over-representation of Black and Black British people amongst adult respondents in families accepted as homeless. Perhaps more surprising is the study’s finding that one in ten of all adult respondents (rising to one third of adult respondents in temporary accommodation for over one year) were former asylum seekers59. 56 For example: Gervais, M.C. and Rehman, H. (2005) Causes of Homelessness Amongst Ethnic Minority Households, London: ODPM; Randall, G. and Brown, S. (2003) The Support Needs of Homeless People and Their Families, London: ODPM; Fitzpatrick, S., Pleace, N. and Jones, A. (2006) The Support Needs of Homeless Families: An audit of provision for families affected by homelessness in Scotland, Edinburgh: NHS Health Scotland.57 This chapter demonstrates the representativeness of Survey 1 by comparing the profile of these families to that of all families accepted as homeless in England in the relevant ‘time window’ (1st January 2005 and 30 June 2005) using local authority homelessness (P1E) data (wherever this is available).58 See Chapter 1 and Appendix 2 for more detail on analysis. 59 It is worth noting here that the survey of single homeless people conducted in 1991 found that 7 per cent of those staying in hostels and B&B hotels had left their last home because of the political situation in a country outside the UK (and this included 24 per cent of those with an ethnic minority background) (Anderson, I., Kemp, P. and Quilgars, D. (1993) Single Homeless People, London: HMSO). Chapter 2 The characteristics of families accepted as homeless 53Key points• The majority of families accepted as homeless were female lone parents (65 per cent). Most of the remainder were couples with a child or children (30 per cent of the total). • Parents within families accepted as homeless tended to be younger than parents with dependent children in the general population, and their children were also generally young (50 per cent of children were pre-school age). • Approximately one third (36 per cent) of families accepted as homeless contained someone in paid work. • There was an over-representation of Black or Black British people amongst adult respondents as compared to the general population of parents with dependent children in England. All ethnic minority groups were highly concentrated in London. • One in ten (11 per cent) of all adult respondents reported being a former asylum seeker. This group were also highly concentrated in London. • Families accepted as homeless in rural areas had broadly similar characteristics to those in urban areas, but they were far less likely to contain an adult respondent with an ethnic minority background or who had sought asylum in the UK.• Four-fifths (82 per cent) of families in temporary accommodation for more than one year (Survey 4 families) had been accepted as homeless in London. Survey 4 families were more likely than other families accepted as homeless to have an adult respondent who was older, from an ethnic minority background, and/or who was a former asylum seeker. Survey 4 families were typically larger than other families accepted as homeless, both because they were more often headed by couples, and because they tended to have more children.The characteristics of families accepted as homelessHousehold type and sizeLone women parent households represented 65 per cent of all families 2.8 accepted as homeless, compared to 25 per cent of families with dependent children in the general population of England (see Table 2.1). Two parent households, conversely, were much less common amongst families accepted as homeless (30 per cent) than amongst families in the general population (72 per 54 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldscent). Other types of household were as uncommon amongst families accepted as homeless as they were among families in the general population60. Table 2.1: Survey 1 families in comparison with the general populationFamily type Survey 1 Families with dependent children in England DifferenceLone woman parent 65% 25% +40%Couple with children 30% 72% -42%Lone male parent 4% <1% +3%Lone pregnant woman* 1% <1% 0%Couple containing pregnant woman* 1% 2% +1%Other 0% <1% 0%Base 2,053 6,500Source: Survey 1 and Families and Children Survey (FACS) (2005) (England only). CHP analysis. *Household contained no other children at point of surveyFamilies accepted as homeless contained an average of 3.1 people including 2.9 both adults and children, compared to an average of 3.4 people in families in the general population of England (the median was three people among both families accepted as homeless and families in England)61. Families headed by lone women were, on average, smaller than families 2.10 headed by couples. As can be seen in Table 2.2, only a small proportion (7 per cent) of lone women households contained five or more people, compared to 21 per cent of families headed by couples. Families headed by a lone woman had an average size of 2.8 people (median of two people), and families headed by couples had an average size of 3.8 people (median of four people).60 For the purposes of analysis (of both survey data and national compararitive statistics), lone pregnant women are henceforth merged into the ‘women lone parents’ category. Similarly, couples who have no children but one of whom was a pregnant woman, are merged into the ‘couples with children’ category. In some instances, all ‘lone parents’ (both women and men) are considered in the analysis and where this is the case it is specified. For the most part, however, ‘women lone parents’ are examined as a discrete category, and are often compared to ‘couples with children’, the other main category. 61 The comparison is with FACS (2005) (England only) CHP analysis. Chapter 2 The characteristics of families accepted as homeless 55Table 2.2: Size of lone women parent and couple householdsFamily size Lone women parents Couples with childrenAll families*Two people 50% 0% 35%Three people 29% 50% 36%Four people 13% 30% 18%Five people 4% 11% 7%Six or more people 3% 10% 5%Total 100% 100% 100%Base 1,300 640 2,045Source: Survey 1 * Includes other household types as well as lone women parents and couples with children. The parents in families accepted as homelessWomen outnumbered men amongst parents in families accepted as homeless 2.11 (76 per cent were female)62. This reflected the finding that the majority of households were headed by lone women parents (see Table 2.1). Women in these families tended to be younger than men. As Table 2.3 2.12 demonstrates, 40 per cent of women were under 25, compared to 6 per cent of men. Men were more likely to be in the older age ranges (26 per cent were aged over 40, compared to 12 per cent of women). Table 2.3: The age and gender of parents in families accepted as homelessFemale Male AllUnder 25 40% 6% 32%25-29 18% 23% 19%30-34 17% 25% 19%35-39 13% 21% 15%40-44 7% 13% 8%45+ 5% 13% 6%Total 100% 100% 100%Base 2,386 730 3,116Source: Survey 162 This analysis excludes the small number of ‘other’ adults in families accepted as homeless who were not the ‘parents’ of any of the child(ren) in the household. The term ‘parent’ is used here to denote both parents and step-parents. 56 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsParents in families accepted as homeless were younger than parents in the 2.13 general population (see Table 2.4). In particular, Survey 1 parents were much more likely to be aged under 25 than parents in England in general (32 per cent compared to 5 per cent). Table 2.4: The age of parents in Survey 1 families compared to the age of parents in the general population in EnglandAge Band Survey 1 Parents with dependent children in England Difference Under 25 32% 5% +27%25-29 19% 9% +10%30-34 19% 16% -3%35-39 15% 22% -7%40-44 8% 22% -14%45 plus 6% 26% -20%Total 100% 100% –Base 3,116 11,251 –Source: Survey 1 and FACS (2005) (England only) CHP analysisThere was evidence of some very young parents among families accepted as 2.14 homeless: overall, 15 per cent of parents in families accepted as homeless were aged 16 to 19, with another 17 per cent aged 20-24.Children and pregnancy in families accepted as homelessFamilies with large numbers of dependent children were firmly in the 2.15 minority63. One half of families (54 per cent) contained one child, 29 per cent had two children, and 16 per cent three or more children. These proportions closely reflect those reported for the first two quarters of 2005 by local authorities (at 53 per cent with one child, 28 per cent with two children, and 19 per cent with three or more children)64. The overall average number of children per family accepted as homeless 2.16 was 1.7, while the median was one child per family. This compares to an average of 1.6 children, and a median of one child, amongst all families with dependent children in England65. 63 All analysis in this chapter on dependent children is confined to children aged under 18.64 Unverified data reported by local authorities to CLG on the P1E form, Quarters 1 and 2, 2005. 65 The comparison is with FACS (2005) (England only) CHP analysis. Chapter 2 The characteristics of families accepted as homeless 57Figure 2.1 demonstrates that lone women parents and families headed by 2.17 couples tended to have similar numbers of children. Figure 2.1: Number of dependent children in lone women parent and couple households0%10%20%30%40%50%60%Four or more childrenThree childrenTwo childrenOne childCouplesLone woman parent55%28%11%5%52%31%9% 8%Source: Survey 1 Base: 1,300 lone women parent households and 640 households containing a couple and children.Children in families accepted as homeless were generally young. In total, 2.18 30 per cent were aged one or under at point of survey, and 20 per cent were aged between two and four. A further 23 per cent were between five and nine years old, and the remaining 27 per cent were aged 10 or over66 (see Chapter 11 for more details on the profile and characteristics of children in families accepted as homeless). In total, 8 per cent of families accepted as homeless contained a pregnant 2.19 woman. This included the small number of lone pregnant women and couples containing a pregnant woman without children noted above; in addition, a small proportion (6 per cent) of lone women who already had one or more children were pregnant, alongside women in a minority (10 per cent) of couple households. During the first two quarters of 2005, households in which there was 2.20 a pregnant woman and no children accounted for 12 per cent of all homelessness acceptances67. The discrepancy with Survey 1 (where only 2 per cent of families contained a pregnant woman but no children) is most likely accounted for by pregnant women giving birth between acceptance as homeless and point of survey68. 66 This group included teenagers aged 16-17 who were still living with their parent(s).67 P1E does not collect data on pregnancy in families accepted as homeless which already contain children.68 On average, 9 months had elapsed between acceptance as homeless and interview. See Chapter 6.58 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsWorking status of families accepted as homelessMost families accepted as homeless (64 per cent) were ‘workless’, that is, 2.21 they contained no adults in employment. Families comprising couples with children were more likely to contain a working adult than lone female parent households (53 per cent and 28 per cent respectively). By way of comparison, in the general population, 96 per cent of couples with children and 60 per cent of lone female parents were in households containing someone in work in 200569. As discussed in detail in Chapter 10, the low proportion of lone women parents in work is explained in part by the young age of their children, as this is an important factor associated with propensity to be in paid work amongst the general population of lone parents. The characteristics of adult respondents in families accepted as homelessAs noted in Chapter 1 (see para 1.16), the adult respondents from families 2.22 accepted as homeless were selected purposively on the basis that they were best placed to provide information on the household as a whole. However, we also collected a considerable amount of individual-level data on the personal history and circumstances of these adult respondents which it was not practical to collect for all adult members of the household. For this reason, we present some basic information here on the characteristics of these adult respondents – their age, gender, ethnicity and whether they had ever sought asylum in the UK – to provide context for the findings in later chapters regarding their personal histories and experiences. The age and gender of adult respondentsWomen far outnumbered men among adult respondents (84 per cent were 2.23 female). Again, this reflected the finding that the majority of families accepted as homeless were headed by lone women parents (see Table 2.1), although even in couple households the female rather than male partner tended to be the adult respondent. Women adult respondents were younger than their male equivalents: 44 per cent of women adult respondents were aged under 25, compared to only 12 per cent of the male adult respondents (see also Table 2.3 on the age and gender of all parents in families accepted as homeless). The ethnicity of adult respondentsThree-quarters (76 per cent) of adult respondents were White2.24 70 and 12 per cent had a Black or Black British origin (see Table 2.5). Asian and Asian British groups accounted for 7 per cent of adult respondents. Other ethnic groups 69 Source: FACS (2005) (England only). See Chapter 10.70 As Table 2.5 indicates, ‘White’ is used in this report to denote not only White British adult respondents, but also those from an Irish or ‘other’ White ethnic minority background (1 per cent of all Survey 1 adult respondents were from a White Irish background, and 5 per cent were from an ‘other’ White ethnic minority background). When the term ‘ethnic minority’ is used in this report it refers only to those from a non-White ethnic background. Chapter 2 The characteristics of families accepted as homeless 59comprised 5 per cent of all adult respondents71. As Table 2.5 indicates, this ethnic profile of adult respondents closely tallied with the recorded ethnicity of applicants within all households accepted as homeless in England in the first two quarters of 200572. Table 2.5: The ethnicity of adult respondents compared to ethnicity of applicants’ households accepted as homeless in first two quarters of 2005Ethnicity Survey 1 adult respondentsEthnicity of applicant in households accepted as homeless in England during quarters 1-2 of 2005White: British, Irish or any other White background76% 75%Black or Black British: Caribbean, African, any other Black background12% 10%Asian or Asian British: Indian, Pakistani, Bangladeshi, any other Asian background7% 5%Other ethnic groups 5% 5%Ethnicity not recorded 0% 5%Total 100% 100%Base 2,053 54,230Source: Survey 1 and P1E returns for 2005 (Statutory Homelessness: 2nd Quarter 2007, England (Communities and Local Government) http://www.communities.gov.uk/news/housing/468373) CHP analysisAs Table 2.6 demonstrates below, Black or Black British people were over-2.25 represented amongst adult respondents in families accepted as homeless compared with their presence amongst parents with dependent children in the general population in England (12 per cent compared with 2 per cent) 73. Conversely, White people were under-represented (76 per cent of 71 Further detail on ‘other ethnic groups’ is not available in the P1E data for the first two quarters of 2005, but can be given for Survey 1 adult respondents. ‘Mixed’ ethnic origins (including White and Black Caribbean, White and Black African, White and Asian or any other mixed background) comprised 2 per cent of all adult respondents to Survey 1, and another 3 per cent of all adult respondents had a ‘Chinese or other’ ethnic background. See also Table 2.6.72 The comparison is not an exact one, as the P1E homelessness statistics for England on ethnicity are not broken down by household type (i.e. they include lone people and couples without children as well as applicants within families with children). Also, as can be seen in Table 2.5, the P1E homelessness statistics did not record the ethnicity of 5 per cent of applicants during the first six months of 2005.73 Over half of this Black/Black British group amongst adult respondents in families accepted as homeless was Black African (7 per cent of all adult respondents had a Black African ethnic origin). 60 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsadult respondents were White compared with 94 per cent of parents with dependent children in the general population). There is longstanding evidence of disproportionate Black experience of ‘non-statutory’ homelessness74. These survey results, together with other recent analysis of the local authority homelessness (P1E) data75, make clear that there is a similar pattern among families accepted as homeless. Table 2.6: Ethnic background of adult respondents compared with adults with dependent children in the general populationEthnicity of adult respondentSurvey 1 adult respondentsEthnicity of respondent parent in FACS (2005)DifferenceWhite: British, Irish or any other White background76% 94% -18%Black or Black British: Caribbean, African, any other Black background12% 2% +10%Asian or Asian British: Indian, Pakistani, Bangladeshi, any other Asian background 7% 2% +5%Mixed: White and Black Caribbean, White and Black African, White and Asian, or any other Mixed background 2% 1% +1%Chinese or other ethnic group: Chinese, or any other ethnic group 3% 2% +1%Total 100% 100% –Base 2,053 7,658 –Source: Survey 1 and FACS (2005) (England only) CHP analysisThere were some variations in respondent characteristics by ethnic 2.26 background. For example, White adult respondents tended to be younger than those from other ethnic groups: 43 per cent were under 25, as compared with only 26 per cent of Black or Black British adult respondents. Adult respondents from Asian or Asian British backgrounds were less likely to be lone parents than those from other ethnic groups. 74 Anderson, I., Kemp, P. and Quilgars, D. (1993) Single Homeless People, London: Department of the Environment.75 Gervais, M.C. and Rehman, H. (2005) Causes of Homelessness Amongst Ethnic Minority Households, London: ODPM. Chapter 2 The characteristics of families accepted as homeless 61Families in which the adult respondent was from an ethnic minority 2.27 background were slightly larger on average. White adult respondents were in households with an average size of 3 persons, compared to adult respondents with an ethnic minority background whose average household size was 3.3 persons. The median for both groups was three people. Former asylum seekers amongst adult respondentsThe survey asked adult respondents if they had ever sought asylum in the UK. 2.28 One in ten of all adult respondents reported that they had claimed asylum at some point (11 per cent) 76.Three-quarters (73 per cent) of those who had sought asylum in the UK 2.29 had an ethnic minority background. In all, one third (35 per cent) of adult respondents with an ethnic minority background had sought asylum, compared to 4 per cent of White respondents77. Black or Black British adult respondents were the most likely ethnic group to be former asylum seekers (41 per cent had claimed asylum). Former asylum seekers were less likely to be lone parents (45 per cent) 2.30 than other adult respondents (68 per cent). Former asylum seekers were correspondingly more likely to be in couple households (50 per cent) than other adult respondents (27 per cent). Former asylum seekers were also, on average, five years older than other 2.31 adult respondents (34 compared to 2978). Families in which the adult respondent was a former asylum seeker were 2.32 marginally larger than other families who had been accepted as homeless. Thus the average size of households in which the adult respondent was an asylum seeker was 3.5 people, whereas for other households it was 3.1 people. However, the median value for both groups was three people. It should be noted that no information was collected on the period of time 2.33 that had elapsed since adult respondents who were former asylum seekers had been granted refugee status or exceptional leave to remain, and in some cases this may have been some considerable time ago. It cannot therefore be assumed that former asylum seekers had become homeless immediately following discharge from National Asylum Support Service (NASS) accommodation. In fact, as Chapter 5 demonstrates, discharge from NASS accommodation accounted for only a very small proportion of the reasons given for applying as homeless by Survey 1 adult respondents79. 76 These were all ‘former asylum seekers’. Current asylum seekers are ineligible for assistance under the homelessness legislation. A person granted refugee status is eligible for assistance, as is someone granted exceptional leave to remain (without the condition that they make no recourse to public funds).77 Almost all of the White adult respondents who had claimed asylum were from White ethnic minority backgrounds (other than Irish). 78 Median ages were 33 compared to 26.79 However, see Gervais, M.C. and Rehman, H. (2005) Causes of Homelessness Amongst Ethnic Minority Households, London: ODPM, for qualitative evidence that loss of NASS accommodation is a major cause of homelessness amongst former asylum seekers in some areas.62 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThe relatively high incidence of former asylum seekers among adult 2.34 respondents from families accepted as homeless should be considered in the context of existing evidence of sustained socio-economic disadvantage among refugee groups and those given leave to remain in the UK80. A heightened risk of homelessness may be one particular element of the disadvantage experienced by former asylum seekers. Where the families were accepted as homelessThis section reports on the regions in which families were accepted as 2.35 homeless, and on any regional variations in household size and type, ethnicity, and propensity to have claimed asylum in the UK. It also considers the breakdown between urban and rural areas amongst families accepted as homeless, and any differences in profile between those accepted in these two kinds of area. The regions in which families were accepted as homelessTable 2.7 shows the distribution of families accepted as homeless by the 2.36 region in which they were accepted. As can be seen, the largest group of families were within London (25 per cent), followed by the North West (16 per cent). The South East, South West and North East were all represented at a similar level (11 per cent). The smallest proportions of families were accepted in the East of England (8 per cent), Yorkshire and the Humber (7 per cent), West Midlands (6 per cent), and East Midlands (5 per cent). Table 2.7 also demonstrates that this regional distribution broadly matched that found in the local authority homelessness (P1E) statistics for the first six months of 200581. 80 See Quilgars, D., Burrows, R. and Wright, K. (2004) Refugee Housing and Neighbourhood Issues: A scoping review, York: Centre for Housing Policy. 81 However, the representation of the West Midlands was lower in Survey 1 than should have been the case. The explanation for this is the non-participation in the survey of a major city in the West Midlands.Chapter 2 The characteristics of families accepted as homeless 63Table 2.7: Regions where Survey 1 families were accepted compared to all acceptances of families accepted as homeless during the first two quarters of 2005Region Survey 1 families Families accepted as homeless in England during quarters 1-2 of 2005DifferenceLondon 25% 23% +2%North West 16% 14% +2%South East 11% 10% -1%South West 11% 8% +3%North East 11% 6% +5%East of England 8% 9% -1%Yorkshire and Humber7% 9% -2%West Midlands 6% 13% -7%East Midlands 5% 8% -3%England 100% 100% –Base 2,053 34,580 –Source: Survey 1 and P1E returns for 2005 (Statutory Homelessness: 2nd Quarter 2007, England (Communities and Local Government) http://www.communities.gov.uk/news/housing/468373) CHP analysisIn the analysis that follows, we normally combine these regions into broader 2.37 sub-national regions, comprising ‘London’, ‘the South’ (the East of England, the South East and the South West), and ‘the North and Midlands’ (North West, North East, Yorkshire and the Humber, West Midlands and East Midlands) to aid statistical reliability.Household type and size by broad regionHouseholds headed by lone parents were more common amongst families 2.38 accepted as homeless in the northern than in the southern parts of England. Thus, 57 per cent of families in London, and 58 per cent in the South, were headed by lone parents, compared to 74 per cent in the North and Midlands. This contrasts with the position in the general population where the incidence of lone parents (as a proportion of all families with dependent children) is highest in London (30 per cent) and lowest in the South (20 per cent), with the North and Midlands approximately midway between (26 per cent)82. 82 Source: Labour Force Survey (Spring 2006). Updates Table 3.17 in Regional Trends No 39.64 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThere was no evidence found of a difference in average household size across 2.39 broad regions. The ethnicity of adult respondents by broad regionThere was a striking difference between London and the rest of England 2.40 with regards to the proportion of adult respondents with an ethnic minority background. In London, more than half of adult respondents (59 per cent) had an ethnic minority background, and 41 per cent were White. Elsewhere in England, 10 per cent of adult respondents had an ethnic minority background and 90 per cent were White. Overall, 66 per cent of all families in which the adult respondent had an ethnic minority background were accepted in London.These findings indicate an even greater concentration of ethnic minority 2.41 adult respondents in London than would be predicted by the generally strong representation of ethnic minorities in the capital. According to the 2001 Census83, ethnic minorities comprised 29 per cent of London’s total population (as compared with 59 per cent of adult respondents from families accepted as homeless in the capital). Outside of London, the proportion of adult respondents from ethnic minorities (at 10 per cent) was very similar to their proportion in the general population (9 per cent). Former asylum seekers by broad region Adult respondents who were former asylum seekers were highly concentrated 2.42 in London: they comprised 29 per cent of all adult respondents in London, as compared to 5 per cent of those living elsewhere in England. In total, 69 per cent of all former asylum seekers among the adult respondents were in London. Most of the remaining former asylum seekers were in the South (28 per cent), with only 13 per cent living in the North and Midlands. There is evidence that former asylum seekers tend to gravitate towards and 2.43 stay within the capital because there are established formal and informal support networks there that do not exist outside London84. One of the aims of the UK dispersion policy for asylum seekers has been to attempt to counter this tendency85. Urban and rural areas Figure 2.2 shows the local authority areas where families were accepted as 2.44 homeless according to the Department for Environment, Food and Rural Affairs (DEFRA) Rural Definition and Local Authority Classification86. 83 See: http://www.statistics.gov.uk/census2001/census2001.asp 84 See, for example, Burnett, A. and Peel, M. (2001) ‘Asylum seekers and refugees in Britain: What brings asylum seekers to the United Kingdom?’ British Medical Journal, 322, 485-488. 85 See Quilgars, D; Burrows, R. and Wright, K. (2004) Refugee Housing and Neighbourhood Issues: A scoping review, York: Centre for Housing Policy. 86 See http://www.defra.gov.uk/rural/ruralstats/rural-definition.htm. See also Appendix 2. Running Head Chapter title with Chapter in bold 65Figure 2.2: Distribution of families accepted as homeless compared to distribution of national population, according to DEFRA classification of local authoritiesNational populationSurvey 1 families0%10%20%30%40%50%80% ofpopulationrural50% ofpopulationruralSignificantRuralOther urbanLarge urbanMajor urban42%34%28% 28%7%15% 15% 14% 13% 13%7%12%Sources: Survey 1, DEFRA and ONS mid year population estimates for England (2004)77 Base: 2,053 families (Survey 1).87 In total, 77 cent of families had been accepted as homeless in ‘urban’ areas2.45 88: this is somewhat higher than the 64 per cent of the population in England as a whole defined by DEFRA as living in urban areas. One in six (15 per cent) were accepted in areas that were defined as having ‘Significant Rural’ populations (close to the 13 per cent living in such areas in the general population). However, only 8 per cent of families were accepted in areas in which more than half the population were defined as living in rural settings (as compared to 24 per cent of the general population whom DEFRA defines as living in such areas). The families were thus somewhat more ‘urban’ than the general population. Comparing Survey 1 families to all families accepted as homeless in the first 2.46 six months of 2005, it is clear that Survey 1 families were more likely to be accepted in ‘Large Urban’ areas (28 per cent compared to 14 per cent of all families accepted as homeless reported in P1E statistics) (see Table 2.8). However, this was almost balanced by their being somewhat less likely to be accepted in ‘Major Urban’ or ‘Other Urban’ areas. Likewise, while a marginally higher proportion of Survey 1 families was accepted in ‘Significant Rural’ areas than was the case in the P1E acceptances (15 per cent as compared to 10 per cent), this was more than balanced by fewer Survey 1 families being 87 See http://www.statistics.gov.uk/88 As is explained in Appendix 2, for analytical purposes families were classified as having been accepted in an ‘urban’ area if they were accepted in a ‘Major Urban’, ‘Large Urban’ or ‘Other Urban’ local authority area. Families were considered to have been accepted in a ‘rural’ area if they were accepted within ‘Significant Rural’, ‘Rural-50’ or ‘Rural-80’ areas. 66 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsaccepted in the most rural areas (more than 50 per cent of population rural) than P1E data would predict (8 per cent, compared to 17 per cent)89. This pattern meant that, overall, Survey 1 families broadly matched the urban/rural split of all families accepted as homeless in the first two quarters of 2005. Table 2.8: DEFRA classification of areas where Survey 1 families were accepted compared to pattern of acceptances for all families accepted as homelessRegion Survey 1 families Families accepted as homeless in England during quarters 1-2 of 2005 DifferenceMajor urban 42% 47% -5%Large urban 28% 14% +14%Significant rural 15% 9% +6%Other urban 7% 14% -7%Rural 50 % 7% 9% -2%Rural 80 % 1% 7% -6%England 100% 100% –Base 2,053 35,478 –Source: Survey 1 and P1E returns for 2005 (Statutory Homelessness: 2nd Quarter 2007, England (Communities and Local Government) http://www.communities.gov.uk/news/housing/468373.)CHP analysisThere was found to be no difference between the size and types of families 2.47 when urban areas were compared with more rural areas90. The proportion of lone women parents in urban and more rural areas was almost identical, as was the proportion of couples with children. Adult respondents were also of very similar ages in urban and rural areas. However, there were very few adult respondents with an ethnic minority 2.48 background in more rural areas. In urban areas (including London), 29 per cent of adult respondents had an ethnic minority background. In more rural areas, by comparison, only 3 per cent had an ethnic minority background. Likewise, adult respondents were far more likely to be former asylum 2.49 seekers in urban areas than in more rural areas. Fourteen per cent of adult respondents were former asylum seekers in urban areas (including London); this contrasts with less than 2 per cent in more rural areas. 89 The explanation for this last point is that the numbers of acceptances in some very rural areas were so low that it was not feasible to include those areas in the fieldwork (see Appendix 1).90 See footnote 88 above for explanation of ‘urban’ and ‘rural’ as used in this report. Running Head Chapter title with Chapter in bold 67Families in temporary accommodation for more than one year As noted in the introduction (see para 2.3), a separate survey (‘Survey 2.50 4’) of those families in temporary accommodation for more than a year was required to capture the experience of sustained stays in temporary accommodation91. This section reviews the characteristics of Survey 4 families and adult respondents, and details where they were accepted as homeless, comparing these findings to those for Survey 1. Characteristics of families in temporary accommodation for more than one yearSurvey 4 families were less likely to be lone women parents and more likely to 2.51 be couples with children than Survey 1 families (Table 2.9). Nonetheless, lone women parents were the predominant group of families in both surveys (56 per cent and 65 per cent respectively).Table 2.9: Household composition of Survey 4 families in comparison with Survey 1 familiesHousehold type Survey 4 familiesSurvey 1 familiesDifferenceLone woman parent 56% 65% -9%Couple with children 38% 30% +8%Lone male parent 5% 4% +1%Couple (pregnant woman) <1% <1% 0%Lone pregnant woman <1% 1% 0%Total 100% 100% –Base 571 2,053 –Source: Survey 4 and Survey 1Survey 4 families were typically larger than those in Survey 1. As Table 2.10 2.52 shows, Survey 4 families were more likely to contain five or more people (37 per cent) than were Survey 1 families (12 per cent). Conversely, they were less likely to contain two persons (21 per cent) than Survey 1 families (35 per cent). The average size of Survey 4 families was 3.7 persons, as compared with 3.1 persons in Survey 1.91 See Chapter 1 (para 1.20) and Appendix 1.68 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 2.10: Household size of Survey 4 families in comparison with Survey 1 familiesHousehold size Survey 4 families Survey 1 familiesDifferencePregnant woman only <1% <1% <1%Two persons 21% 35% -14%Three persons 28% 36% -8%Four persons 14% 18% -4%Five or more persons 37% 12% +25%Total 100% 100% –Base 571 2,053 –Source: Survey 4 and Survey 1As with Survey 1, most of the parents in Survey 4 families were women 2.53 (69 per cent) (see Table 2.11). Survey 4 parents tended to be older, however, with 52 per cent being aged 35 or over, compared to 29 per cent of Survey 1 parents92. Parents in Survey 4 families were much less likely to be aged 16-19 (2 per cent) than Survey 1 parents (15 per cent). Parents aged 20-24 formed a similar proportion of Survey 4 parents (13 per cent) and Survey 1 parents (17 per cent). Table 2.11: Age and gender of parents in Survey 4 families compared to Survey 1 familiesFemale Male AllUnder 25 21% [40%] 3% [6%] 15% [32%]25-29 14% [18%] 12% [23%] 14% [19%]30-34 21% [17%] 19% [25%] 20% [19%]35-39 19% [13%] 24% [21%] 21% [15%]40-44 14% [7%] 15% [13%] 15% [8%]45+ 11% [5%] 28% [13%] 16% [6%]Total 100% [100%] 100% [100%] 100% [100%]Base 544 [2,386] 253 [730] 797 [3,116]Source: Survey 4 Figures in brackets are for Survey 1 adult respondents92 As with Survey 1, this analysis excludes the small number of ‘other’ adults in families accepted as homeless who were not the ‘parents’ of any of the child(ren) in the household. The term ‘parent’ is used here to denote both parents and step-parents. Running Head Chapter title with Chapter in bold 69While Survey 4 parents were typically older than Survey 1 parents, they were 2.54 nevertheless younger than parents in the general population of England. Thus, 15 per cent of parents were aged under 25 amongst Survey 4 families, while the figure for the general population was just 5 per cent93. Table 2.12 contrasts the number of children in households in Survey 1 with 2.55 the number of children in households in Survey 4. As can be seen, families in Survey 4 were substantially less likely than those in Survey 1 to contain one child (33 per cent did so as compared with 54 per cent), and more likely to contain three or more children (32 per cent as against 17 per cent). Table 2.12: Number of children in households, Survey 4 compared to Survey 1Number of childrenSurvey 4 Survey 1 Difference 1 33% 54% -21%2 35% 29% +6%3 20% 11% +9%4 or more 12% 6% +6%Total 100% 100% –Base 568 2,053 –Source: Survey 4The children in Survey 4 families were generally older than those in Survey 1; only 2.56 32 per cent of the children were aged under five in these families, compared to 50 per cent of the children in Survey 1. Dependent children aged 10 or over accounted for 42 per cent of children among Survey 4 families, compared to 27 per cent of children in families in Survey 194. The proportion of Survey 4 families containing a pregnant woman, at 9 per 2.57 cent, was very similar to the level found in Survey 1.Patterns of work and worklessness were nearly identical for Survey 1 and 2.58 Survey 4 families. Thus, 36 per cent of Survey 4 families had at least one household member in paid work, as did 36 per cent of Survey 1 families95. Characteristics of adult respondents in temporary accommodation for more than one yearAs with Survey 1, most of the Survey 4 adult respondents were women (76 2.59 per cent). Again as with Survey 1, these female adult respondents were generally younger than their male counterparts: 25 per cent of Survey 93 FACS (2005) (England only) CHP analysis.94 See Chapter 11 for a detailed analysis of the characteristics of children in both Survey 1 and Survey 4 families. 95 See Chapter 10 for a detailed analysis of employment patterns amongst both Survey 1 and Survey 4 families.70 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds4 female respondents, as compared with 1 per cent of Survey 4 male respondents, were aged under 25. However, Survey 4 adult respondents were in general older than those in Survey 1: only 19 per cent were aged under 25, as compared with 40 per cent being in this age group in Survey 1. There was a very strong representation of ethnic minority groups among adult 2.60 respondents in Survey 4. More than half (59 per cent) had an ethnic minority background (compared with 24 per cent of adult respondents in Survey 1). As with Survey 1, the largest ethnic minority group in Survey 4 was Black or Black British (comprising 28 per cent of all adult respondents in Survey 4, compared with 12 per cent in Survey 1). Asian and Asian British groups accounted for a further 18 per cent of Survey 4 adult respondents (compared with 7 per cent in Survey 1), with other ethnic minority groups collectively forming another 14 per cent of all Survey 4 adult respondents (compared with 5 per cent in Survey 1). As compared with the general population of parents with children in 2.61 England, people with a Black or Black British ethnic minority background were even more over-represented among adult respondents in Survey 4 than in Survey 196. Moreover, both Asian and ‘other’ ethnic groups were over-represented in Survey 4. Within London, 70 per cent of adult respondents in Survey 4 had an ethnic 2.62 minority background, compared to 59 per cent in Survey 1. Outside London, the proportion of Survey 4 adult respondents with an ethnic minority background, at 10 per cent, was close to that for Survey 1 (and the general population). One third (33 per cent) of adult respondents to Survey 4 were former asylum 2.63 seekers; a much higher proportion than was the case among Survey 1 adult respondents (11 per cent). This group were even more heavily concentrated within London than was the case in Survey 1 (97 per cent were in the capital, as compared with 69 per cent in Survey 1). Within London, 39 per cent of adult respondents to Survey 4 were former 2.64 asylum seekers (29 per cent in Survey 1). Outside London, there was no greater a propensity for Survey 4 adult respondents to be former asylum seekers than in Survey 1. As was the case with Survey 1 (see para 2.29), most Survey 4 former asylum seekers (77 per cent) had an ethnic minority background97. 96 See Table 2.6 above.97 Almost all of the White Survey 4 adult respondents who had claimed asylum were from a White ethnic minority background (other than Irish).Chapter 3 Personal and housing history of adult respondents 71Former asylum seekers were less likely than other Survey 4 adult respondents 2.65 to be a lone woman parent (38 per cent compared to 60 per cent), and correspondingly more likely to be living as a couple with children (58 per cent compared to 32 per cent of other Survey 4 adult respondents). Families headed by former asylum seekers were larger than other Survey 4 families (an average of 4.4 people per household compared to an average of 3.4 for other households in Survey 4). Former asylum seekers in Survey 4 also tended to be older (average age of 39) than other adult respondents in Survey 4 (average age of 33). Where families in temporary accommodation for more than one year were accepted as homeless The great majority (82 per cent) of Survey 4 families were accepted in 2.66 London. A further 18 per cent were accepted in the South. None of these families were accepted in the North and Midlands.This geographical distribution of families in Survey 4 reflects the known 2.67 concentrations of temporary accommodation use by local authorities in England, and in particular the exceptionally high use in London (see Chapter 6). In the North and Midlands, and to a lesser extent the South, temporary accommodation use is generally for much shorter periods than in London98. Regression analysis confirmed that, other things being equal2.68 99, being accepted as homeless in London was by far the most important independent factor which increased the likelihood of a family being within Survey 4 rather than Survey 1. The next most important (though much weaker) independent influence was being accepted in an area of ‘higher housing stress’100. The other families which had a slightly heightened chance of being in Survey 4 rather than Survey 1 were large families containing four or more people and those in which the adult respondent was from an ethnic minority background. Conversely, families accepted in ‘more deprived’ local authority areas were 98 See Appendix 1 for the methodology adopted in generating a nationally representative sample of those in temporary accommodation for over one year. While some families are in temporary accommodation for longer than a year in the North and Midlands broad region, they were too few in number to make survey fieldwork viable within Survey 4.99 Factors controlled for in this regression analysis included: whether accepted in London; whether accepted in an area of ‘higher housing stress’ (Appendix 2); whether accepted in a ‘more deprived’ area (see Appendix 2); whether adult respondent was former asylum seeker; whether household contained four or more people; household type; and whether the adult respondent was from an ethnic minority group.100 The local authorities in which adult respondents lived were assigned a ranking based on the affordability ratio of owner occupation for people aged 20-39 in that area (gross average house price in relation to gross average household income). The 70 local authorities which participated in Survey 1 were divided into quartiles, ‘most affordable’ (a ratio of less than 3.7), the ‘next most affordable’ (3.7 – 4.1), ‘less affordable’ (4.2 – 4.8) and the ‘least affordable’ (a ratio of more than 4.8). Those referred to in this report as living in areas in the areas of ‘higher housing stress’ were living in the two least affordable quartiles of local authorities. See: Wilcox, S. (2005) Affordability And The Intermediate Housing Market: Local measures for all local authority areas in Great Britain, York: Joseph Rowntree Foundation. See also Appendix 2.72 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsless likely than other families to be found in Survey 4101. Once these factors were taken into account, neither household type nor whether or not the adult respondent was a former asylum seeker were independently associated with being in Survey 4. ConclusionsIn many respects these findings confirm the results of earlier research, 2.69 and reinforce the picture of family homelessness suggested by existing statistical data. While the extent to which lone parents are at particular risk of homelessness was broadly understood, the survey results draw further attention to the highly ‘gendered’ nature of acceptance as statutory homeless, in that it is, in most cases, an experience that lone mothers and their children go through. Another key point is the generally young age of both these female lone parents and their children. Their relative disadvantage is demonstrated through the low levels of paid work within these households. Also consistent with earlier research is the over-representation of Black and 2.70 Black British people amongst adult respondents in families accepted as homeless. Perhaps more surprising is the study’s finding that one in ten of all adult respondents (rising to one third of adult respondents in temporary accommodation for over one year) had at some stage sought asylum in the UK.Survey 4 families and adult respondents in temporary accommodation for 2.71 more than one year had a quite distinct profile from those in Survey 1. Most crucially, the great majority (82 per cent) of families in Survey 4 were accepted in London, as compared to only 25 per cent accepted in London in Survey 1. Survey 4 families were more likely than Survey 1 families to have an adult respondent who was older, from an ethnic minority background, and/or who had sought asylum in the UK. Survey 4 families also tended to be larger than Survey 1 families. These demographic and locational distinctions between Survey 1 and Survey 4 families have important implications for comparisons between the two groups throughout this report102. The next chapter moves on to examine the personal and housing history of 2.72 adult respondents. 101 The local authorities where Survey 1 adult respondents lived were also categorised according to their rank in the 2004 Indices of Deprivation for England. Four quartiles were created confined to the 70 local authority areas which participated in Survey 1. One quartile was of ‘very deprived’ local authorities (those within the 29 most deprived authorities in England), ‘deprived’ (ranked between 30 and 79), ‘affluent’ (80 to 177) and ‘very affluent’ (ranked 178 or lower). Source: ODPM (2005) The English Indices of Deprivation 2004 (Revised), London: ODPM. The reference here to the ‘more deprived’ local authority areas denotes the ‘deprived’ and ‘very deprived’ quartiles combined. See also Appendix 2.102 We use regression techniques throughout the remainder of this report to check whether any differences in the findings for Survey 1 and Survey 4 adult respondents/families were attributable to their distinct demographic and geographical profiles (see Appendix 2).Chapter 3 Personal and housing history of adult respondents 73Chapter 3:Personal and housing history of adult respondents IntroductionThe overall objective of this study was to understand the causes, experiences 3.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data was collected in five separate surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys).Personal characteristics and experiences – such as mental health problems, 3.2 substance misuse, or abusive childhood experiences – are often thought to be ‘risk factors’ which increase an individual’s chances of becoming homeless103 (see also Chapter 5). These personal characteristics and experiences are also important because they may influence a person’s support needs and the interventions required to resolve their homelessness (see also Chapter 9). However, almost all of the existing UK-based evidence profiling homeless 3.3 adults’ characteristics, and certainly all of the robust statistical evidence104, has focused on ‘single homeless people’ (that is, childless households who have not been assisted under the homelessness legislation) rather than adults within families accepted as homeless105. There is some relevant US data; for example, one major quantitative study suggested that families experiencing recurrent homelessness were more likely to be led by a woman who had a diagnosed mental health problem, who was drug dependent, and who had experienced abuse as a child, than was the case for ‘first time’ families accepted as homeless106. But the social and economic context in the US is very different to that of England and the wider UK, which is likely to impact substantially on the characteristics and experiences of homeless adults107. 103 Randall, G and Brown, S. (1999) Prevention is Better than Cure: New solutions to street homelessness from Crisis, London: Crisis; Fitzpatrick, S., Kemp, P. A. and Klinker, S. (2000) Single Homelessness: An overview of research in Britain, Bristol: The Policy Press.104 Anderson, I., Kemp, P.A. and Quilgars, D. (1993) Single Homeless People, London: HMSO.105 A general point to bear in mind in this chapter that the data relates only to adult respondents within families accepted as homeless (who were purposively selected as the person best able to comment on the whole household (see para 2.22 and Appendix 1)), rather than to all adults in these families. In a minority of families accepted as homeless there is at least one other adult in the family whose longer-term history may also influence circumstances for the family as a whole, but we do not possess personal or housing history data for them. 106 Bassuk, E.L., Perloff, J.N. and Dawson, R. (2001) ‘Multiply homeless families: the insidious impact of violence’, Housing Policy Debate, 12, 2, 299-320.107 Fitzpatrick, S. and Christian, J. (2006) ‘Comparing research on homelessness in the United Kingdom and United States: What lessons can be learned?’, European Journal of Housing Policy, 6 (3) 313-333. 74 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThe housing as well as personal histories of adults in families accepted 3.4 as homeless may be important in setting the context for their families’ route in to statutory homelessness. In particular, it is of interest to find out whether their recent acceptance as homeless is part of a long-term, or recurrent, history of homelessness, or rather a singular event in an otherwise stable housing history. Given persistent concerns about intergenerational homelessness, any experience of homelessness as a child amongst adults in families accepted as homeless is especially important. This chapter therefore draws on data from Survey 1 – a survey of families 3.5 accepted as being owed the main homeless duty between 1st January 2005 and 30 June 2005 – to explore adult respondents’:• personal history: both as children and as adults; and• housing history: including experience of living independently in mainstream housing; previous experience of homelessness or housing insecurity; whether had had a ‘settled home’ as an adult; and previous homelessness applications. It should be noted that self-completion questions were used to gather the 3.6 most sensitive material reported in this chapter, including that on sexual matters, involvement in crime, and experiences of violence. The last section of the chapter compares the personal and housing history of 3.7 Survey 1 adult respondents to that of adult respondents in families accepted as owed the main homelessness duty and in temporary accommodation for more than a year (Survey 4)108.As in other chapters, we present two types of statistical analysis here: 3.8 bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a range of other factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected109.This survey evidence indicates that amongst adult respondents, there was 3.9 fairly widespread experience of family and school disruption in childhood, and mental health problems and domestic violence in adulthood. However, other personal problems in adulthood, including substance misuse, were reported 108 Survey 4 was required because the ‘time-window’ design for Survey 1, while delivering a representative sample of families accepted as homeless over a six month period, by definition excluded those with prolonged stays in temporary accommodation. See Chapter 1 (para 1.20) and Appendix 1 for a full explanation. 109 See Chapter 1 and Appendix 2 for more detail on analysis. Chapter 3 Personal and housing history of adult respondents 75by relatively small numbers. White women lone parents accepted outside of London were the most likely group to report experience of personal problems. Most adult respondents had experience of independent mainstream housing prior to acceptance as homeless, but one quarter reported having never had a settled home as an adult. Key points • Experience of family and school disruption was quite widespread amongst adult respondents in families accepted as homeless. A small minority (7 per cent) reported being homeless as a child.• Half (52 per cent) of all adult respondents had experience of anxiety, depression or other mental health problems, and four in ten (41 per cent) reported having been a victim of domestic violence. However, other personal problems in adulthood, such as drug or alcohol problems, were noted by much smaller proportions. • White women lone parents accepted outside of London were the group most likely to report multiple problems in childhood and in adulthood.• Overall, 65 per cent of adult respondents to Survey 1 had lived independently in their own rented or owner-occupied housing at some point since age 16, including the great majority of those aged over 25. • Half (51 per cent) of all adult respondents had experienced at least one episode of homelessness or insecure housing before the homelessness that led to their current situation (most commonly they had stayed with friends and relatives as a result of having no home of their own). A small minority (8 per cent) had experience of sleeping rough and/or sleeping in a car and/or squatting, but had almost never had their children with them in these situations.• One quarter (26 per cent) of adult respondents reported that they had never had a settled home as an adult. • For the great majority of adult respondents (87 per cent), this was their first homelessness application. • Adult respondents who had lived in temporary accommodation for more than one year (Survey 4 adult respondents) were less likely than other adult respondents to have had problematic personal histories as children. This was attributable to their distinct demographic profile. 76 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsPersonal history of adult respondents Experiences during childhoodFigure 3.1 provides an overview of the childhood experiences reported by 3.10 adult respondents in families accepted as homeless110. Figure 3.1: Childhood experiences reported by adult respondents0% 10% 20% 30% 40% 50%Spent time in careExperienced homelessnessExperienced sexual abuseExperienced domestic violenceFamily moved house a lotRan away from home and stayedaway for more than one nightSuspended or excluded from school at least onceParents violent towards each otherA step-parent moved into our homeMissed a lot of schoolParents separated or divorced 45%33%28%24%24%23%21%20%16%7%6%Source: Survey 1 Base: 2,053 adult respondents Multiple responses were possible.Experience of disruption to family life during childhood was quite widespread: 3.11 45 per cent of adult respondents reported parental separation or divorce when they were a child, and 28 per cent experienced living with a step-parent. Both of these figures appear far higher than amongst the general population111. One quarter (24 per cent) of adult respondents reported that one or both of 3.12 their parents had been violent towards the other during their childhood, and 20 per cent said that they had experienced violence at home, directed against them, as a child. Sixteen per cent of adult respondents reported experience of sexual abuse when they were a child. The rate of physical abuse reported by adult respondents appears similar to that amongst children within the general population, but the incidence of sexual abuse is higher112. 110 We asked about these particular experiences because all have been suggested in research and/or in policy debates as potentially associated with a heightened vulnerability to homelessness.111 Mayhew, E. (2005) ‘Demography of childhood’, in J. Bradshaw and E. Mayhew The Well-being of Children in the UK, London: Save the Children. 112 Cawson, P., Wattam, C., Brooker, S. and Kelly, G. (2000) Child Maltreatment in the United Kingdom: A Study of the prevalence of child abuse and neglect, London: NSPCC. Chapter 3 Personal and housing history of adult respondents 77The proportion of adult respondents who reported experience of homelessness 3.13 as a child was 7 per cent113. This finding suggests that fears that family homelessness may be largely intergenerational are misplaced. Nevertheless, as Figure 3.1 shows, 23 per cent of adult respondents reported running away from home as a child, and 21 per cent said that their family had ‘moved around a lot’ – both experiences which may indicate a degree of family and/or housing instability. Six per cent had spent time in local authority care. One third of adult respondents reported having ‘missed a lot of school’ 3.14 (33 per cent), and a quarter had been suspended or excluded from school (24 per cent)114.Experiences of multiple problems as a childOverall, 31 per cent of adult respondents did not report any of the problems 3.15 in childhood shown in Figure 3.1. A further 17 per cent reported one of the problems, with another 26 per cent reported two or three of the problems. One quarter (26 per cent) of adult respondents reported experiencing four or more of the childhood problems shown in Figure 3.1. The distribution of problems experienced by adult respondents as children 3.16 was further explored through K-means cluster analysis (see Appendix 2). This resulted in two ‘vulnerability clusters’: those with ‘few’ (if any) problems during childhood, and those with ‘more’ problems115. Around one third of adult respondents (32 per cent) were in the ‘more 3.17 childhood problems’ vulnerability cluster, and the remaining two thirds (68 per cent) were in the ‘few childhood problems’ vulnerability cluster. Regression analysis confirmed that, other things being equal3.18 116, the following factors were independently associated with being in the ‘more problems in childhood’ vulnerability cluster:• age: 40 per cent of those under 25, compared to 22 per cent of those over this age, were in the ‘more problems in childhood’ vulnerability cluster.• being a woman lone parent: 34 per cent of women lone parents, as compared to 27 per cent of other adult respondents, were in this vulnerability cluster.113 No definition of homelessness was given in this question so this is based on the respondents’ own interpretation.114 As a very broad point of reference, 7 per cent of pupils in secondary education were defined as ‘persistent absentees’ in 2005/6. Source: DfES. (March 2007) Pupil Absence in Secondary Schools in England 2005/6 115 These vulnerability clusters were based on the experiences listed in Figure 3.1, and also whether their ‘family had financial difficulties’ when they were a child. 116 Factors controlled for in this regression analysis included demographic characteristics and geographical variables. 78 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsConversely, the factors negatively associated with being in the ‘more 3.19 problems in childhood’ vulnerability cluster were:• being accepted as homeless in London: only 20 per cent of those accepted in the capital were in this vulnerability cluster, as compared with 35 per cent of those accepted elsewhere.• being from an ethnic minority background: only 16 per cent of ethnic minority adult respondents were in this vulnerability cluster, as compared with 36 per cent of other adult respondents.This division between those in the ‘few’ and ‘more’ childhood problems 3.20 vulnerability clusters is used in subsequent analyses in this report. Experiences as an adultFigure 3.2 shows a range of experiences that adult respondents reported as 3.21 adults117. Half of all adult respondents (52 per cent) said that they had experienced 3.22 anxiety, depression or other mental health problems at one time or another. The proportion reporting current mental health problems at point of survey was much lower (27 per cent), but this was still somewhat higher than amongst the general population (albeit that this difference was partly attributable to the predominance of women amongst adult respondents, see Chapter 9 for detailed comparisons). Two in five (41 per cent) of all adult respondents reported having been a 3.23 victim of violence from a partner as an adult. As one would expect, this was a predominantly female experience, affecting 44 per cent of women respondents as compared with only 15 per cent of male respondents. The proportion of female adult respondents reporting experience of domestic violence is considerably higher than in the general population (it is estimated that one in four women in Britain will experience domestic violence during their lifetime)118. However, only around one third (36 per cent) of those who reported experiencing domestic violence from a partner cited this as a reason for applying as homeless (accounting for 13 per cent of all adult respondents). Sexual assault as an adult was reported by 14 per cent of adult respondents 3.24 (these were almost all female respondents). Involvement in prostitution (defined in the survey as exchanging sex for money, food or shelter) was reported by 2 per cent of all adult respondents. 117 We asked about these particular experiences because all have been suggested in research and/or in policy debates as potentially associated with a heightened vulnerability to homelessness. It is worth bearing in mind, in considering the material in this section, that these are generally young adults, with 40 per cent under 25 (see Chapter 2). 118 British Crime Survey (2005/06) Home Office, p.23.Chapter 3 Personal and housing history of adult respondents 79One third of adult respondents reported that they had ‘lived on benefits for 3.25 most of my adult life’ (33 per cent). This finding was age-related: 42 per cent of adult respondents under 25, as compared to 25 per cent of those over this age, reported that they had relied on benefits for most of their adulthood (see also Chapter 10)119. Figure 3.2: Experiences as an adult reported by adult respondents 52%41%33%25%14%10%9%6%4%3%2%2%Have had sex in exchange formoney, food, drugs, shelterHave lived as a 'traveller'Have been in the armed forcesHave been in prison or youth offender institutionExperience of problems with alcoholHave been involved incrime or anti-social behaviourExperience of problems with drugsHave been sexually assaulted as an adultHave never had a settled home as an adultHave lived on benefits most of my adult lifeHave been in a violent relationship (as abused)Experience of mental health problems0% 10% 20% 30% 40% 50% 60%Source: Survey 1 Base: 2,053 adult respondents Multiple responses were possible One in ten adult respondents (10 per cent) self-reported having had a drug or 3.26 solvent problem at some point in their lives, and 6 per cent said that they had experienced problems with alcohol. In all, 11 per cent of adult respondents self-reported having ever had any kind of substance misuse problem. The proportion who self-report a current substance misuse problem at point of survey was very low (3 per cent) (see Chapter 9 for details).Very few adult respondents had spent time in the armed forces or living as 3.27 a traveller. They were also unlikely to report having spent time in prison or a young offenders’ institution and, while they were somewhat more likely to have been involved in crime or anti-social behaviour, this was still reported by only around one in ten (9 per cent) of all adult respondents. 119 As a broad point of comparison, in 2005, approximately 14 per cent of people of working age were claiming a ‘key benefit’ (bear in mind that this national figure is a snapshot so not directly comparable). Source: DWP http://www.dwp.gov.uk/asd/asd1/stats_summary/Stats_Summary_June_2005.pdf. Key benefits include Jobseeker’s Allowance, Incapacity Benefit, Severe Disablement Allowance, Disability Living Allowance and Income Support.80 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsExperiences of multiple problems as an adultMost adult respondents had had at least one of the experiences shown 3.28 in Figure 3.2. Only one quarter (26 per cent) did not report any of these experiences; 60 per cent reported between one and three of the experiences; and 14 per cent four or more. The distribution of personal experiences of adult respondents as adults 3.29 were, as with their childhood experiences, subjected to further exploration using K-means cluster analysis (see Appendix 2). Again, this resulted in adult respondents being placed in one of two groups, the ‘few’ (or no) problems group and a group with ‘more problems’. Most adults were within the ‘few problems in adulthood’ group (65 per cent), with 35 per cent being in the ‘more problems in adulthood’ cluster120. There was an association between being in the ‘more problems’ in childhood 3.30 vulnerability cluster and being in the ‘more problems’ in adulthood cluster. Thus, only 26 per cent of adult respondents who were in the ‘few problems’ in childhood vulnerability cluster were in the ‘more problems’ in adulthood cluster, compared to 54 per cent of those in the ‘more problems’ in childhood cluster who were also in the ‘more problems’ in adulthood cluster.Regression analysis indicated that, other things being equal3.31 121, the following additional factors had an independent influence on the likelihood of being in the ‘more problems in adulthood’ vulnerability cluster: • being a woman lone parent: 42 per cent of women lone parents were in the ‘more problems’ vulnerability cluster, as compared to only 19 per cent of other adult respondents.• age: younger respondents (under 25) were underrepresented in the ‘more problems’ vulnerability cluster (27 per cent were in this cluster, compared to 36 per cent of older adult respondents).• being accepted in London: those accepted in London were also underrepresented in the ‘more problems’ vulnerability cluster (22 per cent were in this cluster, as compared with 39 per cent of other adult respondents).• having sought asylum in the UK: former asylum seekers likewise were underrepresented in the ‘more problems’ vulnerability cluster (15 per cent were in this cluster, compared with 37 per cent of other adult respondents).120 The vulnerability clusters were based on benefit dependency, drug or alcohol dependency, mental health problems, experience of prison, involvement in crime or anti-social behaviour, experience of domestic violence, experience of sexual assault, and the presence of support needs (see Appendix 2). Please note that the experiences noted in Figure 3.2 which may not necessarily be viewed as negative, i.e. being in the armed forces or being a traveller, were not included in the K-means cluster, nor was prostitution as it was so rare. 121 Factors controlled for in this regression analysis included demographic characteristics and geographical variables. Chapter 3 Personal and housing history of adult respondents 81• being from an ethnic minority background: only 20 per cent of ethnic minority adult respondents were in the ‘more problems’ group, as compared with 39 per cent of White adult respondents.Housing history of adult respondents Experience of living independently Overall, 65 per cent of adult respondents reported living in their own 3.32 accommodation which they rented and/or owned at some point between age 16 and prior to their acceptance as homeless. One third (35 per cent) had had no such experience of independent living.The largest proportion (45 per cent) had rented, or jointly rented, from the 3.33 private rented sector; and 31 per cent had been social rented sector tenants. A smaller proportion of adult respondents (17 per cent) had been owner occupiers at some point. Age had by far the strongest impact on whether an adult respondent had 3.34 experience of living in their own accommodation. Four in five (82 per cent) of adult respondents aged over 25 had lived in their own independent accommodation, compared to 38 per cent of those under 25.Adult respondents who cited violent relationship breakdown as a reason for 3.35 applying as homeless were more likely than other adult respondents to have lived in their own independent accommodation (78 per cent had done so as compared to 63 per cent of other respondents), and in particular were more likely to have been owner occupiers (27 per cent had been, compared to 15 per cent of other respondents)122. Prior experience of homelessness and housing insecurityFigure 3.3 summarises adult respondents’ experiences of homelessness and 3.36 insecure housing settings between the age of 16123 and prior to the place from which they were accepted as homeless 124. In total, half (51 per cent) of all adult respondents reported experience of one 3.37 or more of the scenarios shown in Figure 3.3. 122 Overall, 13 per cent of adult respondents reported violent relationship breakdown as a reason for applying as homeless, and a further 3 per cent reported other types of violent relationship breakdown as contributing to their homelessness. See Chapter 5 for a full discussion of the reasons for applying as homeless, and in particular the role played by violent relationship breakdown. 123 See para 3.13 above for homelessness experiences prior to age 16.124 The circumstances from which adult respondents were accepted as homeless are excluded here because this section is seeking to explore adult respondents’ longer-term housing history to see if their acceptance as homeless was a one-off event or formed part of a recurrent pattern of homelessness or living in insecure housing. The immediate circumstances which led to their acceptance as homeless are explored in Chapters 4 and 5.82 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 3.3: Adult respondents’ experience of homelessness or insecure housing circumstances, from age 16 onwards, prior to the place from which accepted as homeless41%12%9%6%4%4%3%1%Live in a squatLive in a caravanSleep in a carSleep roughStay in a women's refugeStay in a hostelStay in a B&BStay with friends or relatives0% 10% 20% 30% 40% 50%Source: Survey 1 Base: 2,053 adult respondents Multiple responses were possible. Please note that when asked about staying with friends and relatives, in B&B hotels, or in a caravan or car, adult respondents were asked to only include those experiences that arose because they had ‘no home of their own’.Experience of staying with friends and relatives, because a respondent had 3.38 no home of their own, was relatively widespread (41 per cent of all adult respondents reported this) and accounted for most adult respondents’ experiences of homelessness or housing insecurity prior to the place from which they were accepted as homeless. More limited experience of B&B hotels (12 per cent) and hostels (9 per cent) was also reported. A small minority of adult respondents (8 per cent) had experience of sleeping 3.39 rough and/or sleeping in cars and/or squatting. It was extremely rare for adult respondents to have children with them when they were in these situations. Regression analysis indicated that, other things being equal3.40 125, being in the ‘more problems in childhood’ vulnerability cluster exerted an independent (positive) effect on the likelihood of having experienced one of the scenarios shown in Figure 3.3. Being a former asylum seeker also had an independent effect which increased the likelihood of having experienced one of these scenarios. 125 Factors controlled for in this regression analysis included demographic characteristics, geographical variables, and adult and child vulnerability clusters.Chapter 3 Personal and housing history of adult respondents 83Respondents who had never had a settled home as an adultAdult respondents were asked whether or not it was the case that they ‘had 3.41 never had a settled home as an adult’. The definition of ‘settled home’ was left wholly to the adult respondent. One quarter of adult respondents (26 per cent) reported that they had never had a settled home as an adult126. Regression analysis indicated that, other things being equal3.42 127, the only independent effect on the likelihood of having never had a settled home was age (32 per cent of those under 25, as compared with 22 per cent of those over this age, reported never having had a settled home as an adult). Previous homelessness applications Previous homelessness applications (to any UK local authority) were reported 3.43 by 13 per cent of adult respondents128; and the great majority of these adult respondents had made only one previous application. Thus the numbers reporting multiple previous homelessness applications were very small (3 per cent of all adult respondents). Regression analysis confirmed that, other things being equal3.44 129, being in the adult or child ‘many problems’ vulnerability clusters had an independent effect which made it more likely that an adult respondent would have made more than one homelessness application. Being young (under 25) made previous applications less likely. Around half (47 per cent) of adult respondents who had made a previous 3.45 homelessness application had been provided with settled accommodation as a result of that application (accounting for 6 per cent of all adult respondents).126 This is an entirely separate concept from that of ‘last settled accommodation’ that is used for comparative purposes in subsequent chapters of this report (see Chapter 1 and Appendix 1 for a full explanation). 127 Factors controlled for in this regression analysis included demographic characteristics, geographical variables, and adult and child vulnerability ‘clusters’.128 This is around half the 27 per cent level found by Scottish research on repeat homelessness (Pawson, H., Third, H. and Tate, J. (2001) Repeat Homelessness in Scotland, Edinburgh: Scottish Homes 2001). Likewise a major report on the experience of homeless applicants in England found that 28 per cent had made a previous homelessness application (O’Callaghan, B., Dominion, L., Evans, A., Dix, J., Smith, R., Williams, P. and Zimmeck, M. (1996) Study of Homeless Applicants, London: HMSO.). Both of these studies included single people as well as families, and it is known that the former are more likely to make repeat applications than the latter, so this probably explains some of the discrepancy. It is also possible that the rate of repeat homelessness applications may have reduced in England in recent years as a result of the prevention measures which been adopted (see para 1.3)129 Factors controlled for in this regression analysis included demographic characteristics, geographical variables, and adult and child vulnerability ‘clusters’.84 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAdult respondents in temporary accommodation for more than one yearAs Table 3.1 demonstrates, adult respondents in temporary accommodation 3.46 for more than one year (Survey 4 adult respondents) were less likely to have had troubled childhoods than Survey 1 adult respondents. Thus fewer Survey 4 than Survey 1 adult respondents had parents who had divorced or separated (24 per cent as compared with 44 per cent), or had a step-parent move in with the family when they were a child (16 per cent as compared with 28 per cent). Survey 4 adult respondents were also less likely to have been suspended or excluded from school (10 per cent had been as compared with 23 per cent of Survey 1 adult respondents); to have missed a lot of school (24 per cent as compared with 33 per cent); or to have run away from home (13 per cent as compared with 22 per cent). There were no other clear distinctions in the childhood experiences of Survey 4 and Survey 1 adult respondents. Table 3.1: Childhood experiences of Survey 4 and Survey 1 adult respondentsSurvey 4 Survey 1 DifferenceParents separated or divorced 24% 44% -20%Suspended or excluded from school at least once10% 23% -13%Step-parent moved into home 16% 28% -12%Missed a lot of school 24% 33% -9%Ran away from home and stayed away for more than one night 13% 22% -9%Experienced violence at home 17% 20% -3%Family moved house a lot 16% 19% -3%Experienced homelessness 5% 6% -1%Experienced sexual abuse 16% 16% 0%Spent time in care 6% 6% 0%Parents violent towards each other 25% 24% 1%Base 571 2,048Sources: Survey 1 and Survey 4 Multiple responses were possibleChapter 3 Personal and housing history of adult respondents 85The experience of personal problems in adulthood amongst Survey 4 adult 3.47 respondents was very similar to that of Survey 1 adult respondents. However, Survey 4 adult respondents were more likely to feel that they had ‘never had a settled home’ as an adult (45 per cent felt that they had not had a settled home as compared with 26 per cent of those in Survey 1). They were also very unlikely to have made a previous homelessness application (only 5 per cent had done so, as compared with 13 per cent of Survey 1 adult respondents). Their housing histories were otherwise very similar. Regression analysis indicated that, insofar as the personal and housing 3.48 histories of Survey 4 adult respondents differed from that of Survey 1 adult respondents, this was accounted for by their distinct demographic composition (see Chapter 2)130. Conclusions This chapter has reviewed the personal and housing history of adult 3.49 respondents in families accepted as homeless. It demonstrates that there was widespread experience of family and school disruption in childhood, and mental health problems and domestic violence in adulthood, amongst these adult respondents. However, other personal problems in adulthood, including substance misuse, were reported by much smaller numbers. White women lone parents accepted outside of London were most likely to report experience of personal problems in both childhood and adulthood.It seems that most adult respondents had experience of independent housing 3.50 prior to acceptance as homeless, albeit in around half of cases their housing history was punctuated with at least one prior experience of homelessness or housing insecurity (mainly staying with friends and relatives as a result of having no home of their own). For a minority, most notably the quarter of adult respondents who reported never having had a settled home as an adult, their recent statutory homeless episode appears to form part of a longer history of homelessness and housing insecurity (albeit that these were mainly younger adult respondents). 130 The factors controlled for in this regression analysis included: demographic characteristics; geographical variables; and whether a Survey 1 or Survey 4 adult respondent. See Appendix 2 for an explanation of the regression analysis on the merged Survey 1 and Survey 4 dataset.86 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsSurvey 4 adult respondents in temporary accommodation for over one 3.51 year had less troubled personal histories in childhood than Survey 1 adult respondents, but were more likely not to have had a settled home as an adult. These distinctions were accounted for by the distinct demographic profile of Survey 4 adult respondents.The next chapter explores adult respondents’ experiences of seeking help 3.52 from a local authority and being accepted as homeless. Chapter 4 Seeking help from a local authority 87Chapter 4:Seeking help from a local authorityIntroductionThe overall objective of this study was to understand the causes, experiences 4.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data was collected in five separate surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys). A range of research reports have commented on homeless applicants’ 4.2 experiences of the statutory process, but most of these are now somewhat dated131. This existing research has identified variations and problems in how applicants are treated by local authority staff132, but some reports have also highlighted positive experiences of some applicants, and attempts by staff to be as ‘fair’ and ‘sympathetic’ as possible133. At the same time, there have been longstanding concerns about possible ‘perverse incentives’ generated by the homeless persons legislation, whereby some applicants (particularly young women still living in the family home) may ‘engineer’ their homelessness in order to gain unwarranted priority in access to social housing134. There have been suggestions that these potential ‘moral hazards’ are particularly acute in London and other areas of high pressure on the social rented stock135. However, the routes that households take in to statutory homelessness are not currently well understood, and hitherto little has been known about the steps these households take (if any) to address their housing problems before approaching local authorities for help. 131 Niner, P. (1989) Homelessness in Nine Local Authorities: Case studies of policy and practice, London: HMSO; Lidstone, P. (1994) ‘Rationing housing to the homeless applicant’, Housing Studies, 9(4): 459-472; O’Callaghan, B., Dominion, L., Evans, A., Dix, J., Smith, R., Williams, P. and Zimmeck, M. (1996) Study of Homeless Applicants, London: HMSO; Evans, A. (1999) ‘Rationing Device Or Passport To Social Housing? The operation of the homelessness legislation on Britain in the 1990s’, in S. Hutson and D. Clapham (eds), Homelessness: Public Policies and Private Troubles, London: Cassell.132 Lidstone, P. (1994) ‘Rationing housing to the homeless applicant’, Housing Studies, 9(4): 459-472; Niner, P. (1989) Homelessness in Nine Local Authorities: Case studies of policy and practice, London: HMSO; Evans, A. (1999) ‘Rationing Device Or Passport To Social Housing? The operation of the homelessness legislation on Britain in the 1990s’, in S. Hutson and D. Clapham (eds), Homelessness: Public policies and private troubles, London: Cassell.133 Niner, P. (1989) Homelessness in Nine Local Authorities: Case studies of policy and practice, London: HMSO; Lidstone, P. (1994) ‘Rationing housing to the homeless applicant’, Housing Studies, 9(4): 459-472; O’Callaghan, B., Dominion, L., Evans, A., Dix, J., Smith, R., Williams, P. and Zimmeck, M. (1996) Study of Homeless Applicants. London: HMSO.134 Robson, P. and Poustie, M. (1996) Homelessness and the Law in Britain, 3rd Ed, London: Butterworths/Planning Exchange; Fitzpatrick, S. and Stephens, M. (1999) ‘Homelessness, need and desert in the allocation of council housing’, Housing Studies, 14(4) 413-431. 135 Fitzpatrick, S. and Pawson, H. (2007) ‘Welfare safety net or tenure of choice? The dilemma facing social housing policy in England’ Housing Studies, 22(2)163-182. 88 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThis chapter therefore draws on data from Survey 1 – a survey of families 4.3 accepted as being owed the main homeless duty between 1st January 2005 and 30 June 2005 – to explore: • the type of last settled accommodation occupied by families accepted as homeless;• the accommodation circumstances of families when they were accepted as homeless;• the efforts they made (if any) to resolve their accommodation problems before approaching the local authority;• their intentions when they approached the local authority, and the sources of their knowledge (if any) about the statutory homelessness arrangements before they approached the local authority; • any concerns they had about applying as homeless; and• how well informed they felt while their homelessness application was being assessed. The last section of the chapter compares the responses of Survey 1 adult 4.4 respondents to those of adult respondents in families accepted as owed the main homelessness duty and in temporary accommodation for more than a year (Survey 4)136.As in other chapters, we present two types of statistical analysis here: 4.5 bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a range of other factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected137.This survey evidence demonstrates that the great majority of adult 4.6 respondents sought at least one form of alternative help before approaching the local authority for assistance. Moreover, two in five did not know they were going to apply as homeless when they initially approached the local authority for help, and most had concerns about applying as homeless. Adult respondents (including those in temporary accommodation for over one year) were evenly split over whether they were kept well or poorly informed while their application was being assessed.136 Survey 4 was required because the ‘time-window’ design for Survey 1, while delivering a representative sample of those accepted as homeless over a six month period, by definition excluded those with prolonged stays in temporary accommodation. See Chapter 1 and Appendix 1 for a full explanation. 137 See Chapter 1 and Appendix 2 for more detail. Chapter 4 Seeking help from a local authority 89Key points• Half (55 per cent) of all adult respondents were accepted as homeless whilst living in a place other than their last settled accommodation. This suggests that many families make short-term accommodation arrangements before approaching local authorities for help.• Approximately one quarter of adult respondents had been accepted as homeless when living in each of: their parents’ homes; other friends and relatives’ homes; and the private rented sector. Around one in ten (11 per cent) approached a local authority for help directly from a social rented tenancy, and 5 per cent from owner occupancy. • The remaining 13 per cent of adult respondents had been accepted as homeless while living in managed forms of accommodation, such as B&B hotels or hostels, or in other forms of accommodation, such as tied housing. • Young women lone parents approaching a local authority from the parental home, having never lived anywhere else, constituted only a very small proportion (7 per cent) of all adult respondents.• The great majority of families (85 per cent) had sought at least one form of alternative help before seeking assistance from a local authority. • Two in five (42 per cent) adult respondents did not know they were going to apply as homeless when they approached the local authority. Most of this group approached the local authority because they ‘needed help with their housing situation but did not know what to do’. • The majority (70 per cent) of adult respondents reported at least one concern about making a homelessness application, most commonly that they would have to live in a ‘rough’ area.• Adult respondents in temporary accommodation for over one year (Survey 4 adult respondents) were less likely than other adult respondents to have sought alternative help to address their housing situation before approaching the council. This was accounted for by their distinct demographic and geographical profile.The type of last settled accommodation occupied by familiesAs is shown in Figure 4.1, the three main types of ‘last settled 4.7 accommodation’138 identified for adult respondents in families accepted as 138 This broad definition of ‘last settled accommodation’ used to identify the ‘origins’ of families’ homelessness could be respondent-defined or questionnaire-defined. See Appendix 1 for a full explanation. 90 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldshomeless were the parental home (26 per cent), a private sector tenancy (also 26 per cent), and social rented housing (18 per cent). Arrangements with friends or relatives were identified as the last settled accommodation for 16 per cent. One in ten (9 per cent) of adult respondents had a last settled accommodation that was owner-occupied housing. Managed settings, such as hostels or B&B hotels, or other arrangements (such as tied housing), were unlikely to be identified as last settled accommodation.Figure 4.1: Type of last settled accommodation occupied by families9%18%OtherManaged settingsParental homeFriends or relativesPrivate rented sectorSocial rentedOwner occupied26%16%26%3% 2%Source: Survey 1, Base: 2,051 householdsThe type of accommodation that families were living in when accepted as homelessAround half of all adult respondents (55 per cent) were accepted as homeless 4.8 whilst living in a place other than their last settled accommodation. This most likely reflects the fact that many families had entered in to short-term accommodation arrangements of various kinds after losing their last settled accommodation but before approaching a local authority for help.Figure 4.2 details the type of accommodation families were living when they 4.9 were accepted as homeless. Chapter 4 Seeking help from a local authority 91Figure 4.2: Where families were staying when accepted as homeless5%11%OtherManaged settingsParental homeFriends or relativesPrivate rented sectorSocial rentedOwner occupied23%25%23%10%3%Source: Survey 1 Base: 2,053 familiesThe private rented sector (at 23 per cent)4.10 139 and the parental home (also at 23 per cent) were almost as prominent as places from which families were accepted as homeless as they were as types of last settled accommodation. However, friends’ and (other) relatives’ houses (at 25 per cent) was more common as a place from which families were accepted as homeless than it was as a type of last settled accommodation. This suggests that such settings may often represent emergency arrangements rather than the place from which homelessness ‘originated’140. The proportion of families approaching a local authority for help directly 4.11 from a social rented tenancy141 was around one in ten overall (11 per cent) (somewhat lower than the 18 per cent for whom such accommodation represented their last settled accommodation), although higher for those adult respondents accepted in London (at 20 per cent). Being accepted as homeless while living in owner-occupied housing was rare 4.12 (5 per cent).139 Where the adult respondent was the tenant, joint tenant or partner of the tenant of the property. 140 Half (48 per cent) of those applying from friends or family had a last settled accommodation which was of a different type (usually their parents’ home or their own tenancy). This was also true of 27 per cent of those applying from their parents’ house, but of only 15 per cent of those applying from the private rented sector.141 Where the adult respondent was the tenant, joint tenant or partner of the tenant of the property. 92 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsOnly a very small number of families (2 per cent) were accepted as 4.13 homeless while living in a B&B hotel; a further 8 per cent were staying in some other form of managed accommodation, such as a hostel or supported housing. These managed settings appeared mainly to represent short-term arrangements entered into after families lost their last settled accommodation142.There were some patterns evident with regards to demographic characteristics 4.14 and the type of place from which families had sought help from the local authority. Thus, couple households were more likely to seek help from a private 4.15 sector tenancy than were women lone parents (31 per cent of the former were living in the private rented sector when they approached a local authority, as compared with only 20 per cent of the latter)143. There were no other associations between household type and place from which adult respondents had sought help. As Table 4.1 demonstrates, and as might be expected, there were clear age-4.16 related patterns, with adult respondents under 25 far likelier to seek help from their parents than those over this age (39 per cent of younger adult respondents were living in the parental home when they approached a local authority, as compared with 13 per cent of those aged over 25). Younger adult respondents were also more commonly living with friends and relatives (31 per cent, as compared with 22 per cent of older adult respondents). Conversely, respondents under 25 were far less likely to have been private sector tenants when they were accepted as homeless than were older respondents (11 per cent were in private tenancies, as compared with 31 per cent of those over this age).142 The majority (79 per cent) of this group has a last settled accommodation that was of a different type – for around half (56 per cent) their last settled accommodation was mainstream rented or owned accommodation; in 13 per cent of cases it was their parent’s home; and in 9 per cent it was the home of (other) family and friends.143 As is described in Chapter 5, eviction as a cause of homelessness was particularly associated with this tenure.Chapter 4 Seeking help from a local authority 93Table 4.1: Tenure or accommodation type from which sought help from local authority, by age of adult respondentAged under 25 Aged over 25 AllOwner occupied 1% 7% 5%Social rented 9% 13% 11%PRS 11% 31% 23%Friends or relatives 31% 22% 25%Parental home 39% 13% 23%Managed accommodation10% 15% 13%Total 100% 100% 100%Base 788 1,265 2,053Source: Survey 1Given the controversy highlighted above regarding potential manipulation of 4.17 the homeless persons legislation by young people wishing to leave home, it is worth noting that young (under 25) female lone parents, applying from the parental home, having never lived anywhere else, were a very small minority (7 per cent) of all adult respondents. Former asylum seekers were more likely have sought help from managed 4.18 forms of accommodation than were other adult respondents (21 per cent were living in this type of accommodation when they approached a local authority for help, as compared with only 12 per cent of other adult respondents), and from a social rented tenancy (26 per cent had done so as compared with 10 per cent of other adult respondents). As former asylum seekers were concentrated in London (see para 2.42), this last point is consistent with the association noted above between seeking help as a social tenant and being accepted in London (see para 4.11 above): adult respondents accepted in the capital who were not former asylum seekers were also more likely than those accepted outside the capital to apply from a social tenancy. Seeking alternative help before approaching the local authorityThe great majority (85 per cent) of all adult respondents reported undertaking 4.19 one or more actions to try to prevent or address their homelessness. Overall, 30 per cent of adult respondents had sought help from one source prior 94 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsto seeking help from the local authority; 55 per cent had sought assistance from two or more sources; and only 15 per cent had not sought any of the specified forms of help. Figure 4.3 summarises the assistance adult respondents had sought before 4.20 approaching a local authority. They were most likely to try to fall back on friends or family (43 per cent). Adult respondents’ next most common response was to try to secure housing in the private rented sector (33 per cent), or to attempt to gain access to the social rented sector (30 per cent). One quarter of adult respondents (23 per cent) had gone to a housing advice centre. Figure 4.3: Assistance sought by adult respondents before approaching a local authority for help43%33%30%23%15%7%4%2%1%0%Looked into buyingSpeak to a family mediation serviceAsked for professional help dealingwith harassmentAsked for professional help dealingwith domestic violenceTried to get help from a rent deposit schemeSpoke to a support workerabout housing problemsSpoke to a housing advice centreJoined waiting list or housing registerTried to get a flat or house to rentfrom a private landlordTried to get family or friends to let you stay0% 10% 20% 30% 40% 50%Source: Survey 1 Base: 2,053 adult respondents Multiple responses were possible * Less than 1 per cent.Regression analysis indicated that, when a range of variables were held 4.21 constant144, there were no independent effects on the likelihood of having sought at least one form of help prior to approaching a local authority. However, demographic factors did impact on the type of help sought. Thus, 4.22 younger adult respondents were more likely than older respondents to have tried to stay with friends or relatives (50 per cent of under 25s had attempted this, as compared to 38 per cent of those over this age). Conversely, they 144 The factors controlled for in this regression analysis included: demographic characteristics; geographical variables; child and adult vulnerability clusters. Chapter 4 Seeking help from a local authority 95were less likely to have spoken to someone at a housing advice centre (17 per cent of under 25s had done so, as compared with 27 per cent of over 25s). Also, ethnic minority adult respondents were less likely to seek a number of forms of help than White adult respondents, including joining a waiting list (18 per cent of ethnic minority adult respondents had done this, as compared with 33 per cent of White adult respondents), seeking a private tenancy (20 per cent as compared with 36 per cent), and asking family and friends to accommodate them (34 per cent as compared with 45 per cent)145.Adult respondents in London were less likely to seek a private sector tenancy 4.23 than those accepted elsewhere (only 17 per cent of those accepted in the capital had attempted to secure a private tenancy, as compared with 38 per cent elsewhere). This ‘London effect’ may relate to the high level of private sector rents in the capital. No other geographical associations were identified with respect to the type of help sought. Two in five families (41 per cent) reported that they had been on a waiting 4.24 list or the housing register in their area prior to being accepted as homeless. There was no variation in this by any demographic or geographical variables. Awareness of the statutory homelessness arrangementsJust over half (58 per cent) of adult respondents had known they were 4.25 going to apply as homeless when they approached a local authority. There was no relationship with any demographic or geographical variables, nor, interestingly, with whether the adult respondent had ever been a social rented tenant. The minority of adult respondents who had made a previous homeless application (13 per cent of all adult respondents146) were, however, marginally more likely to have known that they were going to apply as homeless (70 per cent of this group had known they were going to apply as homeless). This was the only variation detected. Figure 4.4 shows how those adult respondents who knew they were going 4.26 to apply as homeless had found out about the statutory homelessness arrangements. 145 See also Gervais, M.C. and Rehman, H. (2005) Causes of Homelessness Amongst Ethnic Minority Households, London: ODPM.146 See para 3.43 for details on previous homelessness applications. 96 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 4.4: How adult respondents found out about applying as homeless (adult respondents who knew they were going to apply before they approached a local authority)39%38%7%5%3%1%0%0% 10% 20% 30% 40%From the internet*From the landlordHave been homeless beforeFound out through leaflet/booklets etcCouncilProfessional (eg. GP,social worker, key worker)Friends family told meSource: Survey 1 Base: 1,231 adult respondents who reported that they knew they were going to apply as homeless before they approached a council. Multiple responses were possible. * Less than 1 per cent.Adult respondents clearly relied heavily on informal sources and on 4.27 information from professionals. Around four in ten (39 per cent) reported that they had found out from friends or family, and a similar proportion (38 per cent) found out via a key-worker, GP, social worker or similar professional. Few adult respondents reported having found out about applying as homeless directly from the council they approached (7 per cent), and even fewer from booklets or pamphlets. The very small proportion of adult respondents who found out about homelessness services through the Internet (less than 1 per cent) should be noted.Among the 42 per cent of adult respondents who did not know they were 4.28 going to apply as homeless, the great majority approached the council because they ‘needed help with their housing situation but did not know what to do’ (73 per cent). Far less commonly, they approached the council specifically to get onto the waiting list (23 per cent). Concerns about applying as homelessFigure 4.5 summarises the concerns that adult respondents reported about 4.29 applying to a council as homeless. In total, 70 per cent of adult respondents reported at least one concern about applying as homeless.Chapter 4 Seeking help from a local authority 97Figure 4.5: Concerns that adult respondents had about applying as homelessGenerally worriedConcern about long waitConcern about being put in a B&BConcern about location of new accommodationConcern about children futureConcern about being put in a hostel/shelterBeing labeled 'homeless'Would not be accepted as homelessWould have to live far from friends/familyWould have to accept first place that was offeredWould be given a poor quality flat/houseWould have to live in a 'rough' area0% 10% 20% 30% 40% 50%42%34%33%25%15%15%3%2%1%1%1%1%Base: 2,053 adult respondents Source: Survey 1 Multiple responses were possible.The most common concern (reported by 42 per cent of all adult respondents) 4.30 was that they would have to live in a ‘rough’ area if they applied as homeless. One third (34 per cent) were concerned that they would be given poor quality accommodation, and again one third (33 per cent) were worried that they would have to accept the first offer they were made and would not have a choice. One quarter (25 per cent) were concerned that the location would be a long way from friends and family. Smaller numbers were worried about being ‘labelled’ homeless, or about 4.31 not being accepted as homeless (15 per cent in both cases). A few adult respondents had specific concerns, such as having to have a long wait in temporary accommodation or being put in a hostel or B&B hotel. There were no variations in these concerns with respect to geographical or 4.32 demographic variables, except that, interestingly, former asylum seekers were less concerned than other adult respondents (46 per cent of former asylum seekers, as compared with only 28 per cent of other adult respondents, had no concerns about making a homelessness application) (see also para 4.35 below). 98 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsInformation on assessment processAdult respondents were evenly divided about how well they had been kept 4.33 informed while their application was being assessed. Approximately, one half said they had been kept ‘very well informed’ or ‘fairly well informed’ (20 per cent and 32 per cent respectively), while the other half said they had ‘not been very well informed’ or ‘not informed at all’ (27 per cent and 22 per cent respectively). Adult respondents who were still in temporary accommodation were no more or 4.34 less likely than those who had been provided with settled housing to report that they were well or badly informed by their local authority about the process147.There were no geographical or demographic variations in how well informed 4.35 adult respondents felt except that, again, former asylum seekers were more positive than other respondents. Thus, 70 per cent of former asylum seekers reported being very or fairly well informed, as compared with only 50 per cent of other adult respondents. It is possible that this result is explained by former asylum seekers having relatively low expectations with regards to what a reasonable level of information might be. Families in temporary accommodation for more than one yearThe type of accommodation that families in temporary accommodation 4.36 for more than one year (Survey 4 families) were living in when accepted as homeless largely reflected the pattern for Survey 1 families, except that those in Survey 4 were less likely to have been living with the adult respondents’ parents (15 per cent were, as compared to 23 per cent of Survey 1 adult respondents). Regression analysis indicated that this difference was accounted for by the distinct demographic profile of Survey 4 families, and in particular to the older average age of Survey 4 adult respondents (see para 2.59)148. The proportion of Survey 4 adult respondents who knew that they were 4.37 going to apply as homeless when they approached a local authority was, at 61 per cent, very similar to that of Survey 1 adult respondents (58 per cent). Amongst those Survey 4 adult respondents who knew that they were going to apply as homeless, their sources of knowledge about the statutory homelessness arrangements closely resembled that of Survey 1 147 Almost half (45 per cent) of adult respondents were still in temporary accommodation at point of survey (see Chapter 6 for details). 148 The factors controlled for in this regression analysis included: demographic characteristics; geographical variables; and whether a Survey 1 or Survey 4 adult respondent. See Appendix 2 for an explanation of the regression analysis on the merged Survey 1 and Survey 4 dataset.Chapter 4 Seeking help from a local authority 99adult respondents, except that they were less likely to have found out about applying as homeless from a professional (26 per cent had found out from this source, compared with 38 per cent in Survey 1). Survey 4 adult respondents were less likely than Survey 1 families to have 4.38 sought alternative help to address their housing situation before approaching the council: 66 per cent in Survey 4 had done so, as compared with 85 per cent in Survey 1. Table 4.2: Sources of help or assistance sought prior to approaching a local authority for Survey 4 and Survey 1 adult respondentsSurvey 4 Survey 1 Difference Tried to get friends or family to let them stay24% 43% -19%Tried to get a flat or house to rent from a private landlord21% 33% -12%Joined waiting list or housing register 24% 41% -17%Spoke to housing advice centre 17% 23% -6%Spoke to a support worker about housing problems10% 15% -5%Tried to get help from a rent deposit scheme4% 7% -3%Asked for professional help dealing with domestic violence2% 4% -2%Spoke to a family mediation service 2% 1% 1%Looking into buying <1% <1% <1%Total 100% 100% –Base 571 2,053 –Source: Survey 1As Table 4.2 above demonstrates, a smaller proportion of Survey 4 adult 4.39 respondents had asked family or friends to let them stay (24 per cent had done so, as compared with 43 per cent of Survey 1 families); had joined the housing waiting list or register (24 per cent as compared with 41 per cent); or had tried to secure a flat/house from the private rented sector (21 per cent as compared with 33 per cent). Regression analysis indicated that these differences were accounted for by the distinct demographic and geographical profile of Survey 4 adult respondents149.149 The factors controlled for in this regression analysis included: demographic characteristics; geographical variables; and whether a Survey 1 or Survey 4 adult respondent. See Appendix 2 for an explanation of the regression analysis on the merged Survey 1 and Survey 4 dataset.100 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsSurvey 4 adult respondents were less likely than Survey 1 adult respondents to 4.40 have joined a waiting list or housing register prior to having been accepted as homeless (24 per cent had done so as compared to 41 per cent in Survey 1). Again, regression analysis indicated that this was attributable to demographic differences150. The proportion of Survey 4 adult respondents who reported having had 4.41 concerns about applying as homeless was very similar to the proportion in Survey 1. As with Survey 1, respondents’ key concern related to having to live in a rough area (42 per cent of all adult respondents to both surveys mentioned this as a concern).Finally, and again as with Survey 1 (see para 4.33), Survey 4 adult respondents 4.42 were almost evenly split on how well informed they felt when their application was being assessed: 48 per cent felt very or fairly well informed, while 52 per cent felt not very well informed or not informed at all151. Conclusions This chapter has reviewed the circumstances in which families sought 4.43 assistance from a local authority with their housing situation. It has indicated that many adult respondents appeared to have made short-term, informal accommodation arrangements after losing their settled housing and before approaching the local authority for assistance, and the great majority sought at least one form of alternative help before applying as homeless. Moreover, two in five did not know they were going to apply as homeless when they initially approached the local authority, and most had concerns about applying as homeless. Young female lone parents applying as homeless from the parental home, without ever having lived elsewhere, constituted only a very small minority of all adult respondents.While the research was not designed to address directly the issue of ‘moral 4.44 hazards’ within the homelessness legislation, these findings do weigh against suggestions of widespread manipulation of the statutory homelessness arrangements in order to gain priority access to social housing152 . Given the particular concerns about ‘perverse incentives’ in areas of high housing pressure (see para 4.2), it is also worth noting that no differences were 150 The factors controlled for in this regression analysis included: demographic characteristics; geographical variables; and whether a Survey 1 or Survey 4 adult respondent. See Appendix 2 for an explanation of the regression analysis on the merged Survey 1 and Survey 4 dataset.151 See Chapter 6 (para 6.56) for an analysis of how these adult respondents felt about the information provided by local authorities regarding progress made towards offering them settled housing. 152 See also Figure 5.1 which demonstrates the very low proportions who report that their reasons for applying as homeless were because this was the ‘quickest’ or ‘only’ way to get rehoused. Chapter 4 Seeking help from a local authority 101detected between adult respondents in London and those accepted elsewhere with respect to their behaviour or intentions when they approached a local authority (except that those accepted in London were less likely to first seek private rented accommodation, possibly because of the high rents in the capital). The next chapter turns to consider the reasons why families applied as 4.45 homeless. 102 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsChapter 5:Reasons for applying as homelessIntroductionThe overall objective of this study was to understand the causes, experiences 5.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data was collected in five separate surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys).The reasons why families, and also single people, become homeless have 5.2 been the subject of extensive debate in both the UK and US153. Explanations of homelessness have traditionally been divided into two broad categories: ‘individual’ and ‘structural’154. Broadly speaking, individual explanations focus on the personal characteristics, behaviours and support needs of homeless people. Structural explanations, on the other hand, locate the causes of homelessness in external social and economic factors, such as housing market conditions, poverty and unemployment. However a ‘new orthodoxy’ seems now to have been established in both the UK and US that posits that structural factors create the conditions within which homelessness will occur and determine its overall extent, but also that people with support needs are particularly susceptible to these adverse social and economic conditions, and that this susceptibility explains any concentration of vulnerable people in the homeless population155.Most of the debate on the causes of homelessness, at least within the UK, 5.3 has been either theoretical in its orientation or based largely on qualitative research156. This study of families accepted as homeless in England brings a new dimension to our understanding of the causes of homelessness by providing a detailed analysis of nationally representative data on the reasons for applying as homeless given by adult respondents in these families. This 153 Fitzpatrick, S. and Christian, J. (2006) ‘Comparing research on homelessness in the United Kingdom and United States: What lessons can be learned?’ European Journal of Housing Policy, 6(3)313-333. 154 Neale, J. (1997) ‘Theorising Homelessness: contemporary sociological and feminist perspectives’, in R. Burrows, N. Pleace and D. Quilgars (eds) Homelessness and Social Policy, London: Routledge; Pleace, N. and Quilgars, D. (2003) ‘Led rather than leading? Research on homelessness in Britain’, Journal of Community and Applied Social Psychology. 13, 87-196. 155 Pleace, N. (2000) ‘The new consensus, the old consensus and the provision of services for people sleeping rough’, Housing Studies, 15(4) 581-594. 156 For an overview see Fitzpatrick, S. (2005) ‘Explaining homelessness: a critical realist perspective’, Housing, Theory and Society, 22(1)1-17. Chapter 5 Reasons for applying as homeless 103chapter draws on data from Survey 1 – a survey of families accepted as being owed the main homeless duty between 1st January 2005 and 30 June 2005 – to explore:• the reasons why adult respondents had applied as homeless, and how these compared to reasons for leaving last settled accommodation;• the reasons for homelessness amongst those who were accepted as homeless whilst living in different accommodation settings;• the reasons for applying as homeless amongst different demographic groups; • the reasons for applying as homeless in different parts of England; and• the independent influence (if any) of demographic, geographical, housing history, and personal vulnerability factors157 on the likelihood of reporting particular reasons for applying as homeless.The last section of the chapter compares the reasons for applying as homeless 5.4 given by Survey 1 adult respondents to those given by adult respondents in families accepted as owed the main homelessness duty and in temporary accommodation for more than a year (Survey 4)158.It must be borne in mind that what is reported here is largely the ‘immediate’ 5.5 reasons (‘triggers’) for applying as homeless, rather than the underlying structural factors which may create the conditions for homelessness (such as housing or labour markets), or the longer-term personal history factors that may increase a person or household’s vulnerability to homelessness (such as mental health or substance misuse problems)159. That said, some insight as to the impact of these wider contextual factors is attempted by examining geographical variations in the immediate causes of homelessness, and the associations between the immediate causes of homelessness and ‘vulnerability clusters’ is also explored below (see also Chapter 3)160. As noted in Chapter 4, many families applied as homeless from places which 5.6 appeared to represent short-term or emergency arrangements entered into after loss of their ‘last settled accommodation’161. As such, it is possible that their reasons for applying as homeless will reflect the breakdown in these short-term arrangements rather than ‘originating’ causes of their 157 See Chapter 3 and Appendix 2 for an explanation of the adult and child vulnerability ‘clusters’ used to investigate the influence of personal vulnerability factors.158 Survey 4 was required because the ‘time-window’ design for Survey 1, while delivering a representative sample of those accepted as homeless over a six month period, by definition excluded those with prolonged stays in temporary accommodation. See Chapter 1 and Appendix 1 for a full explanation. 159 Fitzpatrick, S., Kemp, P. and Klinker, S. Single Homelessness: An overview of research in Britain, Bristol: The Policy Press.160 See Chapter 3 and also Appendix 2 for an explanation of these adult and child vulnerability ‘clusters’.161 See Chapter 1 and Appendix 1 for an explanation of ‘last settled accommodation’ as used in this research. It should just be noted here that this broadly defined last settled accommodation can be either respondent-defined or questionnaire-defined. 104 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldshomelessness. Consequently, we also present data in this chapter on why families had left their last settled accommodation162.As in other chapters, we present two types of statistical analysis here: 5.7 bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a range of other factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected163.This survey evidence confirms that relationship breakdown is the commonest 5.8 ‘trigger’ for statutory homelessness amongst families. It lends some support to arguments for a structural understanding of family homelessness, insofar as eviction/threatened with eviction (usually because a private sector tenancy had come to an end) was a particularly important reason for homelessness in areas of higher housing stress. There is little support for an ‘individual’ analysis of the causes of family homelessness in these findings: only very small numbers report health or substance misuse as contributing to their reasons for applying as homeless. Key points• Relationship breakdown (usually, but not necessarily, with a partner) was the commonest reason for applying as homeless, with the other major reasons being: overcrowding; eviction/threatened with eviction; and overstaying welcome/could no longer be accommodated.• Only very small numbers reported that physical or mental ill-health, drug or alcohol problems, or anti-social behaviour, had contributed to their reasons for applying as homeless. • Relationship breakdown as a reason for applying as homeless (especially violent relationship breakdown with a partner) was, as one would expect, positively associated with being a woman lone parent. It was also associated with being accepted as homeless in the North and Midlands. Former asylum seekers were less likely than other adult respondents to report relationship breakdown as a reason for applying as homeless.162 However, this data was only available for a minority of adult respondents, for reasons which are explained below (see para 5.17). 163 See Chapter 1 and Appendix 2 for more detail. Chapter 5 Reasons for applying as homeless 105• There was a very strong relationship between eviction/threatened eviction as a reason for applying as homeless and being accepted directly from the private rented sector (usually because a fixed-term tenancy had come to an end). Adult respondents accepted in areas of higher housing stress, and in rural areas, were more likely than other adult respondents to report eviction as a reason for applying as homeless.• The two other principal reasons for applying as homeless – overcrowding and overstayed welcome/could no longer be accommodated – were most common amongst those approaching a local authority for help from friends, relatives or parents’ houses, and amongst adult respondents aged under 25. These two reasons often seemed to reflect a breakdown in short-term or emergency arrangements rather than the ‘originating’ cause of homelessness.• Adult respondents who had lived in temporary accommodation for more than one year (Survey 4 adult respondents) were less likely than other adult respondents to report relationship breakdown as a reason for applying as homeless. This discrepancy was accounted for by the distinct demographic and geographical profile of Survey 4 families.An overview of reasons for applying as homelessFigure 5.1 provides an overview of all of the reasons for applying as homeless 5.9 reported by Survey 1 adult respondents. Respondents were prompted with a list of possible reasons for homelessness and asked to choose all of those which they thought had contributed to their application as homeless (additional/alternative reasons could also be indicated). 106 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 5.1: Reasons for applying as homeless38%26%24%20%7%6%4%4%3%2%2%0%Drug or alchohol problemsHad to leave NASS accommodationMental health problems orphysical health problemsApplying as homeless was thequickest way to get re-housedHarassment, anti-social behaviour or crimeHousing was in poor conditionApplying as homeless was theonly way to get re-housedProblems with paying the mortgage or rentOverstayed welcome or couldno longer be accommodatedHousing was overcrowdedEviction/tenancy endedRelationship breakdown0% 10% 20% 30% 40%Source: Survey 1 Base: 2,053 respondents. Multiple responses were possible. * Less than 1 per cent.Relationship breakdown was the most prevalent reason for applying as 5.10 homeless, cited by two in five adult respondents (38 per cent). Of this group, 62 per cent reported that their relationship had broken down with a partner (and so relationship breakdown with a partner affected 23 per cent of all adult respondents164), while 30 per cent said that their relationship breakdown had been with a parent, step-parent and/or foster parent (11 per cent of all adult respondents). The remaining 8 per cent of those who experienced relationship breakdown said this was with other relatives or friends (4 per cent of all adult respondents). Of those whose relationship breakdown was with a partner, 57 per cent 5.11 said violence was involved (thus violent relationship breakdown with a partner affected 13 per cent of all adult respondents). Amongst those whose relationship breakdown was with parents or other relatives or friends, 16 per cent said that violence was involved (and so violent relationship breakdown with someone other than a partner affected 3 per cent of all adult respondents). 164 This is very close to the P1E figures for the first two quarters of 2005 (the period over which the Survey 1 sample had been accepted as homeless), which reported that relationship breakdown with a partner accounted for 20 per cent of all acceptances. It should be noted that the P1E data relate simply to the primary reason for loss of last settled accommodation whereas the survey statistics included all reported contributory factors to making a homelessness application. Chapter 5 Reasons for applying as homeless 107Eviction/threatened with eviction (including fixed-term tenancy coming 5.12 to an end) was the second most commonly identified reason for applying as homeless, cited by a quarter (26 per cent) of all adult respondents165. Almost three-quarters (70 per cent) of adult respondents who applied as homeless due to eviction/threatened with eviction were accepted directly from the private rented sector. Within this group, by far the most common reason given for eviction/threatened eviction was that the ‘landlord wanted property back’ or that ‘tenancy had come to an end’ (83 per cent); it was uncommon for rent arrears to be cited as the reason for eviction/threatened eviction (13 per cent). Only a small minority of those who reported eviction/threatened with eviction as their reason for applying as homeless sought help directly from a social rented tenancy (9 per cent of adult respondents who were evicted/threatened with eviction, accounting for just 2 per cent of all adult respondents)166. Amongst this small group, rent arrears was the most frequently cited reason for applying as homeless. Reported eviction for anti-social behaviour was very unusual amongst adult 5.13 respondents. Only 2 per cent of adult respondents who reported eviction or threatened eviction cited anti-social behaviour as the cause (representing less than 1 per cent of all adult respondents). Turning to other causes, overcrowding5.14 167, like eviction, was identified by approximately one quarter (24 per cent) of all adult respondents as a reason for their application as homeless. ‘Overstayed welcome or could no longer be accommodated’ was the only other category of a substantial size, reported by one fifth of respondents (20 per cent).All of the other suggested reasons for applying as homeless were identified 5.15 by fewer than 10 per cent of the sample. Thus problems with paying rent or mortgage, and housing being in poor condition, were cited by only 7 and 4 per cent of adult respondents respectively, while even smaller proportions identified with the ‘social’ categories of harassment, crime or anti-social behaviour (4 per cent); health problems (2 per cent); and drug/alcohol problems (less than 1 per cent). Two per cent reported having to leave NASS accommodation as a reason for their homelessness. There were also 165 This is close to P1E statistics for the first two quarters of 2005 (the period over which the Survey 1 sample had been accepted as homeless), where 22 per cent of all acceptances were accounted for by end of assured shorthold tenancy, rent arrears or other loss of rented housing. Again, it should be noted that the P1E data relate simply to the primary reason for loss of last settled accommodation whereas the survey statistics included all reported contributory factors to making a homelessness application.166 This small proportion is to be expected given that tenants in the social rented sector, unlike in the private rented sector, generally have security of tenure unless they can be evicted on ‘behavioural’ grounds, such as rent arrears. In such circumstances they would be likely to be found ‘intentionally homeless’ (see Chapter 1, para 1.1 and footnote 5), and thus not form part of Survey 1. Some of those whose homelessness was caused by eviction/threatened eviction from either the private or social rented sectors will have made short-term accommodation arrangements before approaching a local authority for help, so the circumstances in which they were accepted as homeless will not necessarily reflect the tenure from which they were evicted/threatened with eviction. 167 It should be noted that this is the adult respondents’ definition of overcrowding and does not necessarily imply that they were statutorily overcrowded.108 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsonly small numbers reporting that, amongst their reasons for applying as homeless, was a perception that this was the ‘only’ (6 per cent) or ‘quickest’ (3 per cent) way to get rehoused. Three quarters (75 per cent) of adult respondents reported only one reason 5.16 for applying as homeless. This meant that that when we asked adult respondents to identify their ‘main reason’ for applying as homeless, the pattern closely resembled that for ‘all reasons’. Thus the most commonly cited ‘main’ reasons for homelessness were relationship breakdown (32 per cent), followed by eviction/threatened with eviction (26 per cent) and overcrowding (16 per cent). A further 10 per cent of respondents reported that they had outstayed their welcome or could no longer be accommodated. All other main reasons were reported by 4 per cent or fewer adult respondents. Comparing reasons for applying as homeless to reasons for leaving last settled accommodationAs noted in Chapter 4 (see para 4.8), 55 per cent of adult respondents 5.17 (particularly those accepted from managed accommodation or from friends’ and relatives’ houses) had a ‘last settled accommodation’168 that was a different place from where they were living when they were accepted as homeless. For some of this group (accounting for 29 per cent of all adult respondents), we have information on why they left this last settled accommodation169. While the response categories differed somewhat from those given with respect to reasons for applying as homeless, a broad comparison of the two sets of data (for the minority of adult respondents for whom it was available) indicates that relationship breakdown was equally important as a reason for leaving last settled accommodation (46 per cent) as it was as a reason for applying as homeless (48 per cent). There is a similar consistency on eviction: 14 per cent cited this as a reason for leaving their last settled accommodation, and 17 per cent said this was a reason for applying as homeless. On the other hand, overstayed welcome/could no longer be accommodated 5.18 was less prominent as a reason for leaving last settled accommodation (8 per cent) than it was as a reason for applying as homeless (19 per cent). Likewise with overcrowding (only 8 per cent gave this as a reason for leaving last settled accommodation as compared with 24 per cent who have it as 168 See Appendix 1 for an explanation of the concept of ‘last settled accommodation’ and as it is used in this report. 169 While around half of adult respondents (55 per cent) had either a respondent-defined or a questionnaire-defined last settled accommodation that was different from the place from which they were accepted as homeless, we only asked why they left this accommodation if it satisfied all of the criteria to be a ‘valid’ comparison point for the purposes of this research (see Chapter 1 and Appendix 1). This is why we have this data for only 29 per cent of the sample.Chapter 5 Reasons for applying as homeless 109a reason for applying as homeless). The reason for these discrepancies are explored further below (see para 5.24). The reasons for applying as homeless amongst those seeking help from different types of accommodation settingAs Table 5.1 demonstrates, there was evidence of a relationship between 5.19 particular reasons for applying as homeless and specific settings from which respondents had approached a local authority for help. Table 5.1: Accommodation setting from which adult respondents sought help from a local authority, by most frequently reported reasons for applying as homelessSetting from which sought help from local authority Relationship BreakdownEviction/ tenancy endedOver- crowdedOverstayed or could no longer be accommodatedBase Social rented sector37% 20% 19% ** 231Private rented sector13% 72% 6% ** 465Owner occupation65% 10%* 9% ** ***91Friends or relatives47% 10% 32% 35% 488Parental home43% 5% 49% 32% 480Managed settings44% 23% 7% 12% 290All 38% 26% 24% 20% 2,045Source: Survey 1 Multiple responses were possible. *A few owner occupiers appear to have used the term ‘eviction’ to describe repossession, this was probably the result of a design flaw in the questionnaire that meant the option to report repossession was omitted. **A few households reported that they had outstayed their welcome or could no longer be accommodated from these tenures, but this is likely to have been the result of misunderstanding the relevant response categories. ***The small sample size here means that all percentages should be treated with caution.110 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThe most striking finding here is the strong association between approaching 5.20 a local authority from the private rented sector and eviction as a reason for applying as homeless (72 per cent of those applying from a private tenancy, as compared with 26 per cent of all adult respondents, reported eviction (usually because a fixed-term tenancy had come to an end) as a cause of homelessness). All of the other principal causes, including relationship breakdown (at only 13 per cent), were very under-represented amongst those applying from a private rented tenancy. By contrast, as Table 5.1 indicates, the pattern of reasons for applying as 5.21 homeless amongst those approaching a local authority from a social rented tenancy largely matched that for Survey 1 respondents as a whole170.With regards to the small number of Survey 1 respondents who approached 5.22 a local authority from owner occupation (5 per cent of all adult respondents), clearly the overriding cause of homelessness, reported by approximately two-third of these respondents, was relationship breakdown171. The next largest cause amongst this small group was “difficulties in paying the mortgage” (this is not included in Table 5.1 as was not a frequently cited reason overall). As one might expect, a substantial proportion of those accepted as homeless 5.23 from friends and relatives houses, and from the parental home, said that a reason why they applied as homeless was that they had overstayed their welcome/could no longer be accommodated (35 per cent and 32 per cent did so respectively). They were also more likely than other adult respondents to report overcrowding as a reason for applying as homeless, especially those approaching a local authority for help from the parental home (49 per cent of adult respondents accepted as homeless from the parental home, compared with only 24 per cent of all adult respondents, reported overcrowding as a cause of homelessness). However, as noted above (see para 5.18), amongst the minority of adult 5.24 respondents for whom relevant data was available, it was evident that both overcrowding and overstaying welcome were less prominent as reasons for leaving last settled accommodation than they were as reasons for applying as homeless. These discrepancies were mainly accounted for by the much lower propensity of those who were accepted as homeless from parents’, friends’ or (other) relatives’ houses to cite these as reasons for leaving their last settled accommodation, as compared with the reasons they gave for applying as homeless. 170 We also investigated whether the reasons for homelessness amongst all former social tenants (i.e. the 31 per cent of all adult respondents had ever been a social tenant before acceptance as homeless, see Chapter 3, para 3.33) were distinctive in any way. Again, however, the reasons they gave for homelessness mirrored that for all Survey 1 adult respondents. 171 Please note that the small sample size of owner occupiers means that all percentages related to this group should be treated with caution. Chapter 5 Reasons for applying as homeless 111Adult respondents who had applied as homeless from managed settings (such 5.25 as hostels and B&B hotels) most commonly reported relationship breakdown as the reason why they applied as homeless (44 per cent), with ‘eviction’ (23 per cent) also being relatively prominent. Amongst those in this group for whom relevant data was available, there was seldom any difference between the reasons they gave for applying as homeless and the reasons they gave for leaving their last settled accommodation. The reasons for applying as homeless amongst different demographic groups As would be expected, women lone parents were likelier than couple 5.26 households to report relationship breakdown as a reason for applying as homeless (47 per cent compared to 17 per cent)172. Conversely, couples with children were more likely than woman lone parents to report eviction (36 per cent compared to 26 per cent) and overcrowding (31 per cent compared to 21 per cent) as amongst the reasons why they had applied as homeless. Woman lone parents and couples with children were equally likely to report that they had overstayed their welcome/could no longer be accommodated. There were no associations between reported reasons for approaching local 5.27 authorities and the size of families. Age, however, was strongly associated with reasons for applying as homeless. 5.28 Thus 30 per cent of all adult respondents over 25 reported relationship breakdown with a partner, as compared with only 13 per cent of adult respondents under 25, as a reason for applying as homeless. Conversely, 23 per cent of adult respondents under 25 reported that relationship breakdown with parents was a reason for approaching the local authority for help, as compared with only 3 per cent of adult respondents aged over 25. Overcrowding and outstayed welcome/could no longer be accommodated were also more common amongst younger adult respondents (37 per cent and 29 per cent respectively amongst those under 25 reported these reasons, as compared with 16 per cent and 14 per cent of those over this age). Conversely, adult respondents aged under 25 were less likely to report eviction/threatened eviction as a cause of homelessness than were older respondents (16 per cent as compared to 33 per cent). 172 Please bear in mind that relationship breakdown can be with parents, other relatives or friends, as well as with a partner (see para 5.10). 112 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThe reasons for applying as homeless in different areas of England Adult respondents accepted in areas of higher housing stress were likelier to 5.29 report eviction as a cause of homelessness (31 per cent) than those living in areas of lower housing stress (21 per cent) 173. Conversely, adult respondents accepted within the more affordable areas were more likely to report relationship breakdown as a cause of their homelessness (43 per cent) than those living in areas of higher housing stress (32 per cent). Adult respondents accepted in ‘more affluent’ local authority areas were 5.30 likelier to report eviction/threatened with eviction as a cause of homelessness (31 per cent) than were adult respondents accepted in ‘more deprived’ areas (20 per cent)174. As deprivation levels tend to be inversely associated with housing affordability indicators, this pattern on eviction is in keeping with the findings on housing stress just noted. However, there was no discrepancy with respect to other causes, including relationship breakdown, between adult respondents living in more affluent or deprived areas. Within ‘rural’ areas, eviction/threatened with eviction was more commonly 5.31 cited as a reason for applying as homeless (36 per cent), than was the case amongst those adult respondents accepted in ‘urban’ areas (22 per cent)175. However, there was no difference in respect of other causes of homelessness, including relationship breakdown, between rural and urban areas. One particular hypothesis that the research was designed to test was whether 5.32 adult respondents without support needs might be more commonly found in the areas of higher housing stress, as it is in these locations that one might expect homelessness to most often result from simple financial inability to compete in the housing market. However, as Chapter 9 demonstrates, the proportion of adult respondents with personal support needs was generally very low, and no geographical distinctions were found with respect to the contribution of such needs to reasons for applying as homeless. London in comparison with the rest of England London is singled out in this section for separate analysis because, unlike 5.33 most other areas of England, high housing stress co-exists with high levels of deprivation in the capital, and so the patterns identified elsewhere may not pertain here. 173 Local authorities with relatively lower and higher levels of housing stress were identified by employing the Wilcox affordability ratio. See Appendix 2 for a full explanation.174 See Appendix 2 for an explanation of how ‘more deprived’ and ‘more affluent’ areas were defined in this research.175 See Appendix 2 for our definition of ‘rural’ and ‘urban’. Chapter 5 Reasons for applying as homeless 113It was found that adult respondents accepted in London were less likely to 5.34 report that relationship breakdown was a reason for applying as homeless than adult respondents accepted elsewhere in England (28 per cent as compared to 42 per cent). This was linked to differences between families accepted in London and those accepted elsewhere, in particular the high proportion of former asylum seekers and relatively small proportion of lone parents accepted in the capital (see Chapter 2). London closely paralleled elsewhere in England with respect to the other 5.35 reasons for applying as homeless, including eviction/threatened eviction. Given the concerns outlined in Chapter 4 (see para 4.2), it is also worth noting that adult respondents accepted in London were no more likely than those accepted elsewhere to report that applying as homeless was the “only” or “quickest” way to get rehoused. Independent influences on the principal reasons for applying as homeless We undertook regression analysis to investigate the independent influence (if 5.36 any) of a range of demographic, geographical, housing and personal history factors on the likelihood of adult respondents reporting each of the four principal reasons for applying as homeless.Relationship breakdown As one would expect, other things being equal5.37 176, the main association with relationship breakdown as a reason for homelessness was being a woman lone parent. Being accepted as homeless in the North and Midlands was also (positively) associated with relationship breakdown as a reason for applying as homeless, as was being in the ‘more problems in adulthood’ vulnerability cluster177. The factors which had an independent negative effect on the likelihood of reporting relationship breakdown included being a former asylum seeker, and approaching a local authority for help from a private sector tenancy178.176 Factors controlled for in this regression analysis included demographic characteristics; geographical variables; housing history variables; accommodation settings from which accepted as homeless; and adult and child ‘vulnerability clusters’. This regression analysis was repeated without household type to identify whether the very strong association with lone parents was masking other effects. When household type was excluded, being accepted in the North and Midlands emerged as an independent effect. 177 See Chapter 3 and also Appendix 2 for an explanation of the adult and child ‘vulnerability clusters’ used to investigate the influence of personal vulnerability factors.178 This negative association with applying as homeless from the private rented sector probably reflects the dominance of eviction/threatened eviction as a reason for applying as homeless amongst this group (see para 5.39). 114 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsLooking at violent relationship breakdown with a partner specifically, other 5.38 things being equal179, this reason for applying as homeless was even more strongly associated with women lone parents. The other independent (positive) relationships were approaching a local authority for help from managed forms of accommodation (such as hostels or B&B hotels), and being accepted as homeless in the North and Midlands. Conversely, the strongest independent negative effect on violent relationship breakdown with a partner as a cause of homelessness was being a former asylum seeker. Those approaching a local authority from the parental home or the private rented sector, and adult respondents aged under 25, were also less likely than other adult respondents to attribute their homelessness to violent relationship breakdown. Eviction/threatened with evictionOther things being equal5.39 180, by far the most powerful independent influence on eviction as a reason for applying as homeless was being accepted from a private sector tenancy, though adult respondents accepted in rural areas181, and in areas of higher housing stress182, were also more likely to report eviction as a cause of homelessness. Eviction was negatively associated with women lone parents, and adult respondents aged under 25. OvercrowdingWith regards to overcrowding, the main independent influence, holding 5.40 other factors constant183, was approaching a local authority for help when living with friends or relatives (other than parents)184. Overcrowding as a reason for applying as homeless was also likelier amongst those who had approached the local authority when living with their parents, and amongst adult respondents under 25. Conversely, being a woman lone parent had an independent negative effect on the likelihood of overcrowding being a reason for homelessness. 179 Factors controlled for in this regression analysis included demographic characteristics; geographical variables; housing history variables; accommodation settings from which accepted as homeless; and child vulnerability clusters (adult vulnerability cluster not included as includes experience of domestic violence).180 Factors controlled for in this regression analysis included demographic characteristics; geographical variables; housing history variables; accommodation settings from which accepted as homeless; and adult and child vulnerability clusters.181 See Appendix 2 for explanation of definition of rural. 182 Local authorities with relatively lower and higher levels of housing stress were identified by employing the Wilcox affordability ratio. See Appendix 2 for a full explanation. 183 Factors controlled for in this regression analysis included demographic characteristics; geographical variables; housing history variables; accommodation settings from which accepted as homeless; and adult and child vulnerability clusters.184 As discussed above (para 5.18), overcrowding appeared to be less important as a reason for loss of last settled accommodation than it was as reason for applying as homeless, and this discrepancy seemed linked to the use of friends’ and relatives houses as short-term or emergency accommodation.Chapter 5 Reasons for applying as homeless 115Overstayed welcome/could no longer be accommodatedThere was a similar pattern with outstayed welcome/could no longer be 5.41 accommodated in that, other things being equal185, approaching a local authority when living with friends or relatives (other than parents) was by far the strongest independent (positive) influence on this as a reason for applying as homeless186. Being aged under 25 also made reporting overstaying welcome/could no longer be accommodated more likely, as did being accepted in an area of higher housing stress187. As with overcrowding, there was an independent (negative) association with being a woman lone parent and reporting overstaying welcome/could no longer be accommodated as a reason for applying as homeless. Families in temporary accommodation for more than one yearAs Table 5.2 demonstrates, the reasons for applying as homeless given by adult 5.42 respondents in temporary accommodation for more than one year (Survey 4 adult respondents) largely reflected the pattern for Survey 1 adult respondents, except that those in Survey 4 were less likely to report relationship breakdown (20 per cent did so, as compared with 38 per cent in Survey 1). Table 5.2: Reasons for homelessness reported by Survey 4 and Survey 1 adult respondentsReason for homelessness Survey 4 Survey 1 Difference Relationship breakdown 20% 38% -18%Eviction/tenancy ended 27% 26% +1%Overcrowding 26% 24% +2%Outstayed welcome/could no longer be accommodated15% 20% -5%Problems with paying rent or mortgage4% 7% -3%Applying as homeless “only way to get re-housed”6% 6% 0%Housing in poor condition 3% 4% -1%185 Factors controlled for in this regression analysis included demographic characteristics; geographical variables; housing history variables; accommodation settings from which accepted as homeless; and adult and child vulnerability clusters.186 As discussed above (para 5.18), overstaying welcome appeared to be less important as a reason for loss of last settled accommodation than it was as reason for applying as homeless, and this discrepancy seemed linked to the use of friends’ and relatives houses as short-term emergency accommodation. 187 Local authorities with relatively lower and higher levels of housing stress were identified by employing the Wilcox affordability ratio. See Appendix 2 for a full explanation.116 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 5.2: Reasons for homelessness reported by Survey 4 and Survey 1 adult respondents (continued)Reason for homelessness Survey 4 Survey 1 Difference Applying as homeless “quickest way to get re-housed”3% 3% 0%Mental or physical health problems1% 1% 0%Had to leave NASS accommodation4% 2% -2%Drug or alcohol problems 1% <1% +<1%Base 571 2,053 –Sources: Survey 1 and Survey 4. Multiple responses were possibleRegression analysis indicated that this difference was attributable to 5.43 the distinct demographic and geographical profile of Survey 4 adult respondents188. Conclusions This chapter reported on the reasons for applying as homeless amongst adult 5.44 respondents. Whilst bearing in mind the caveat that this chapter reports mainly on the ‘immediate’ rather than ‘underlying’ causes of homelessness, it does lend some support to arguments for a structural understanding of family homelessness, and points in particular to the importance of housing market conditions. This is most clearly the case with regards to eviction in areas of higher housing stress, usually because a private sector tenancy has come to an end. As was already known, relationship breakdown, usually, though not 5.45 necessarily, with a partner, was the commonest ‘trigger’ for family homelessness. Violent relationship breakdown with a partner was an important though minority element within this. Relationship breakdown has always sat uneasily in the conventional individual/structural dichotomy of causes of homelessness outlined above189. There is certainly little support for an ‘individual’ analysis of the causes of 5.46 family homelessness in these findings: only very small numbers reported 188 The factors controlled for in this regression analysis included: demographic characteristics; geographical variables; and whether a Survey 1 or Survey 4 adult respondent. See Appendix 2 for an explanation of the regression analysis on the merged Survey 1 and Survey 4 dataset. 189 Fitzpatrick, S. (1998) ‘Homelessness in the European Union’, in M. Kleinman, M. Stephens and W. Mattzenetter (eds), European Integration and Housing Policy (pp.197-214), London and New York: Routledge. Chapter 5 Reasons for applying as homeless 117health problems or substance misuse as contributing to their reasons for applying as homeless (see also Chapter 3 (on personal history) and Chapter 9 (on personal support needs)). The next chapter moves on to consider the experience of temporary 5.47 accommodation amongst families accepted as homeless.118 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsChapter 6:Families’ experience of temporary accommodationIntroductionThe overall objective of this study was to understand the causes, experiences 6.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data was collected in five surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys).This chapter considers experience of temporary accommodation amongst 6.2 families accepted as homeless, and it is important to provide some broader context for the findings presented within it. The numbers of households in temporary accommodation at the end of each 6.3 quarter is recorded in local authority ‘P1E’ returns, which are published by central Government190. These statistics indicate that the number of households in temporary accommodation in England rose during the late 1990s, and reached a peak in 2004, when 101,300 households were in temporary accommodation. There has been a subsequent decline, at least in part as the result of an increased policy emphasis by Government on homelessness prevention191. By end December 2007, the numbers of households in temporary accommodation had dropped to 79,500, of which three quarters (59,990) were families with dependent children192. Since January 2005 there has been an official target to halve the total number of households in temporary accommodation, from the December 2004 level, by 2010193. There has long been a concern about the impact of prolonged stays in 6.4 temporary accommodation on families with children, with a number of studies highlighting the sense of uncertainty and loss of control experienced by parents in these families, many of whom describe feeling that their 190 For information on the P1E returns from local authorities to central government see: http://www.communities.gov.uk/ 191 Pawson, H., Netto, G. and Jones, C. (2006) Homelessness Prevention: A guide to good practice, London: Communities and Local Government; Pawson, H,, Netto, G., Jones, C., Wager, F., Fancy, C. and Lomax, D. (2007) Evaluating Homelessness Prevention, London: Communities and Local Government. 192 Communities and Local Government Statistical Release. Statutory Homelessness: 4th quarter 2007. England. Table 6. 193 Office of the Deputy Prime Minister (2005) Sustainable Communities: Homes for all. A five year plan from the Office of the Deputy Prime Minister, London: ODPM.Chapter 6 Families’ experience of temporary accommodation 119lives are ‘on hold’194. It has also been suggested that multiple moves between temporary accommodation addresses can exacerbate the sense of unsettledness and dislocation felt by families in temporary accommodation195. Specific concerns about the use of B&B hotels as temporary accommodation 6.5 have persisted since the late 1980s and early 1990s, when numerous research reports documented the very poor standards faced by the families accommodated in these hotels, especially in London196. There has been a recent concerted attempt to end the use of this form of provision. By the end of December 2007, only 1 per cent of all families with dependent children or an expectant mother in temporary accommodation were in B&B hotels197. It is now prohibited for local authorities to accommodate homeless families in B&B hotels except in where no other accommodation is available, and occupation must not exceed six weeks198. The use of hostels as temporary accommodation has also declined in recent years; at end December 2007 only 5 per cent of all families in temporary accommodation were in hostels. The bulk of temporary accommodation arranged by local authorities is now in self-contained housing, with 93 per cent of families with children in temporary accommodation in self-contained settings at end December 2007199. The Government separately reports, within the P1E quarterly statistics, 6.6 households accepted as homeless who are staying in ‘homeless at home’ arrangements. This usually signifies temporary arrangements with parents, other relatives or friends, though some of those who are homeless at home are staying in their own accommodation which they are about to lose. In line with the general trends on temporary accommodation, the numbers in homeless at home arrangements started to decline from 2005 onwards, and there were 5,510 families with dependant children or an expectant mother accepted as 194 Holder, T., Curteis, S., Griffiths, S., Hunter, G. and James, K. (2002) Life on Hold: “I can’t even think about tomorrow”: The housing and support needs of families in temporary accommodation in Leeds. Leeds City Council (Unpublished report); Sawtell, M. (2002) Lives on Hold: Homeless families in temporary accommodation, London: The Maternity Alliance; Walters, S. and East, L. (2001) ‘The cycle of homelessness in the lives of young mothers: the diagnostic phase of an action research project’ Journal of Clinical Nursing, 10, 171-179. 195 London Research Centre (1991) Length of Stay in Temporary Accommodation: A Study of Homeless Households in London London: London Research Centre; Spatford, T. (2003) Developing Cross Borough Support for Homeless Children and Families London: Newham Children’s Fund Partnership Board and Kings Cross Homelessness Project; Homelessness Directorate (2003) Reducing B&B Use and Tackling Homelessness – What’s Working: A Good Practice Handbook London: Homelessness Directorate/ODPM.196 Bayswater Hotel Homelessness Project (1987) Speaking for Ourselves: Families in Bayswater B&B. London: Bayswater Hotel Homelessness Project; Murie, A. and Jeffers, S. (1987) Living in Bed and Breakfast: the Experience of Homelessness in London, Bristol: University of Bristol, School for Advanced Urban Studies; Niner, P. (1989) Homelessness in Nine Local Authorities: Case Studies of Policy and Practice, London: HMSO; Thomas, A. and Niner, P. (1989) Living in Temporary Accommodation: A Survey of Homeless People, London: HMSO; Crane, H. (1990) Speaking from Experience: Working with Homeless Families, London: Bayswater Hotel Homelessness Project; Carter, M. (1995) Out of Sight: London’s Continuing B&B Crisis, London: London Homelessness Forum.197 Communities and Local Government Statistical Release. Statutory Homelessness: 4th quarter 2007. England. Table 6.198 Homelessness (Suitability of Accommodation) (England) Order 2003 (SI 2003/3326) 199 The majority of this self-contained temporary accommodation is provided via arrangements with the private rented sector. In the first six months of 2005, on average, 60 per cent of all self-contained temporary accommodation in England was provided through HALS (Housing Association Leasing Schemes) or Private Sector Leasing (PSL) schemes, and another 12 per cent through other private sector stock. The remaining 28 per cent was provided by local authorities and housing associations from their own social rented stock. (Source: P1E returns for financial year 2005/6 provided to CHP by Communities and Local Government, for details on P1E see: http://www.communities.gov.uk/). (CHP analysis)120 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldshomeless living as ‘homeless at home’ at end December 2007 (this is in addition to the 59,990 families in temporary accommodation noted above). While the quarterly P1E statistics provide useful ‘snapshots’ on the use of 6.7 temporary accommodation for families with children, they provide only very basic data on families’ experiences of temporary accommodation200. Moreover, as noted in the introduction (see para 1.9), most of the existing research studies in this area have been qualitative and/or small scale. This chapter therefore draws on data from Survey 1 – a survey of families accepted as being owed the main homeless duty between 1st January 2005 and 30 June 2005 – to provide a detailed statistical account of families’ experiences of temporary accommodation. Topics covered include:• the housing status (temporary or settled) of families at point of survey;• time spent in temporary accommodation;• types of temporary accommodation experienced; • moves made between temporary accommodation addresses; and• a summary of temporary accommodation ‘pathways’.As noted above, the impacts of prolonged stays in temporary accommodation 6.8 are of particular interest. The last section of the chapter compares temporary accommodation experiences of Survey 1 families to those of families accepted as owed the main homelessness duty and in temporary accommodation for more than a year (Survey 4)201.As in other chapters, we present two types of statistical analysis here: 6.9 bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a range of other factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected202. 200 This (limited) P1E data is referred to as appropriate throughout this chapter and demonstrates that the regional and other patterns found in the present study follow the same broad pattern as these official statistics. However, readers should not expect the data provided in this study to match relevant P1E statistics for two reasons. First, within the P1E families who are staying in ‘homeless at home’ arrangements are treated as an entirely separate category, but within the current study those staying with family and friends, or in their own accommodation which they are about to lose, are considered alongside all other families in temporary accommodation. Second, the ‘sampling frames’ are different for P1E and Survey 1 in the present study: the former is a snapshot of families in temporary accommodation at a certain point in time, whereas the latter is representative of all families accepted as homeless over a six month ‘time window’ (see Appendix 1). This research design was chosen to counter the tendency in snapshot data to emphasise the position of those in temporary accommodation for extended periods, to the detriment of those whose experience is shorter-term.201 Survey 4 was required because the ‘time-window’ design for Survey 1, while delivering a representative sample of those accepted as homeless over a six month period, by definition excluded those with prolonged stays in temporary accommodation. See Chapter 1 for a full explanation. 202 See Chapter 1 and Appendix 2 for more detail. Chapter 6 Families’ experience of temporary accommodation 121This survey evidence demonstrates that the overwhelming factor driving 6.10 temporary accommodation experience in England was where a family was accepted as homeless. While demographic variables, such as household size and age of adult respondent, had a minor influence on temporary accommodation experience, generally such factors had little independent effect once region and other geographical variables were taken into account. Key points • By point of survey (on average 9 months after acceptance as homeless), 55 per cent of families were in settled housing, and 45 per cent were still living in temporary accommodation. However, in London only 18 per cent of families had moved on to settled housing, as compared with 76 per cent in the North and Midlands. • Around one fifth (21 per cent) of all families had moved directly into settled housing without a stay in temporary accommodation. Most of these families were in the North and Midlands, and very few were in London. • Of those in settled housing at point of survey, 91 per cent were in social rented housing.• Amongst those families still in temporary accommodation at point of survey, 78 per cent were in self-contained temporary accommodation; this form of provision was especially predominant in London and the South. Only 2 per cent of all families still in temporary accommodation were in B&B hotels.• ‘Pathways’ through temporary accommodation were strongly influenced by where a family was accepted as homeless. Those accepted in London, and to a lesser extent in the South, were likely to experience prolonged stays in temporary accommodation, and to spend much of their time in self-contained temporary accommodation. Families in the North and Midlands typically experienced a relatively short stay in temporary accommodation (very often temporary arrangements with family or friends) before being moved on to settled housing. • Multiple moves between temporary accommodation addresses were generally rare (8 per cent) but much more common amongst those families in temporary accommodation for more than a year (Survey 4 families) (43 per cent). • All families in temporary accommodation for more than one year were in self-contained temporary accommodation, and the vast majority were in London. Their average stay in temporary accommodation (at point of survey) was 2.9 years. Levels of frustration at the length of wait for settled housing were high amongst these families, as was dissatisfaction regarding the information provided by local authorities regarding progress made toward providing them with settled housing.122 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsHousing situation at point of surveyOn average, families were surveyed 9 months after acceptance as homeless. 6.11 By point of survey, just over half of families were in settled housing (55 per cent)203, and the remaining 45 per cent were still in temporary accommodation. In London, families were much more likely to still be in temporary 6.12 accommodation at point of survey (82 per cent) than they were elsewhere in England. Higher proportions of families were still in temporary accommodation in the South than in the North and Midlands (48 per cent and 24 per cent respectively) (Table 6.1). Table 6.1: Housing situation of families at point of survey, by percentage of all families in each regionRegion In settled housing at point of surveyIn TA at point of surveyBaseLondon 18% 82% 641South 52% 48% 734North and Midlands76% 24% 678All 55% 45% 2,053Source: Survey 1In total, 46 per cent of all families still in temporary accommodation at point 6.13 of survey were in London, 30 per cent were in the South, and 24 per cent were in the North and Midlands204.It might have been supposed that this regional variation in rates of moving 6.14 on to settled housing reflected the point at which fieldwork took place in relation to when families were accepted as homeless. However, the average time between acceptance and survey was 8.9 months in both the North and Midlands and in the South, while in London it was only very slightly higher at 9 months (the median in all these broad regions was 9 months). Thus it was regional disparities in rates of move on to settled housing, rather than the research timetable, that accounted for the patterns indicated in Table 6.1.203 One fifth (21 per cent) of all families had moved directly into settled housing on being accepted as homeless without a stay in temporary accommodation.204 This geographically-driven pattern is broadly consistent with the quarterly P1E data on families in temporary accommodation in England, which indicates that these families are heavily concentrated in London. For 2005/6, for example, on average 63 per cent of families in temporary accommodation were in London, 27 per cent were in the South, and only 10 per cent were in the North and Midlands. (Source: P1E returns for financial year 2005/6 provided to CHP by Communities and Local Government, CHP analysis). Our statistics and those in P1E do not match exactly for reasons given in footnote 200 above. Chapter 6 Families’ experience of temporary accommodation 123The 55 per cent of Survey 1 families who had been provided with settled 6.15 housing were overwhelmingly in social rented housing (91 per cent). A small number of those in settled housing reported that they had been accommodated in the private rented sector (8 per cent), and a tiny proportion (less than 1 per cent) were in owner occupation. As Table 6.2 demonstrates, the6.16 45 per cent of families who were in still in temporary accommodation at point of survey were mainly living within self-contained housing (flats or houses) that was being used on a temporary basis (78 per cent of all those still in temporary accommodation)205. Overall, just 2 per cent of families in temporary accommodation at point of survey were in a B&B hotel, while 6 per cent were living in a hostel or supported accommodation. Around one in seven (14 per cent) were living with friends and relatives (including 7 per cent who were staying with the parents of the adult respondent, and 7 per cent who were staying with friends or other relatives)206. Table 6.2: Current living situation of families in temporary accommodation at point of surveyType of accommodation Percentage Self-contained 78%Hostel or supported housing 6%B&B hotel 2%Parents 7%Friends or (other) relatives 7%Base 1,130Source: Survey 1There was also a strong regional pattern with regards to the type of 6.17 temporary accommodation these families were living in (Table 6.3). Thus 88 per cent of families in temporary accommodation in London were in self-contained settings, as compared with 52 per cent of those in temporary accommodation in the North and Midlands. Use of arrangements with parents (at 19 per cent), and with friends and (other) relatives (at 22 per cent), was much higher for the minority of families still in temporary 205 This figure included small numbers of owner occupiers and occupants of tied housing who were living in a home they were about to lose (1 per cent of families). Almost all adult respondents in self-contained temporary accommodation described themselves as being in “social rented” housing. However, this was known not to be the case as most of this temporary accommodation is provided through the private sector (see footnote 199 above). Participating local authorities were also not always able to provide consistent information on the way in which specific addresses had been made available as temporary accommodation.206 This breakdown of temporary accommodation types at point of survey broadly reflects the pattern within the P1E quarterly statistics (see para 6.5). It does not match exactly for the reasons noted above in footnote 200, and in particular the separate categorisation of those who are ‘homeless at home’ in the P1E statistics. 124 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsaccommodation in the North and Midlands than was the case in London and the South. Table 6.3: Current living situation of families in temporary accommodation at point of survey, by regionType of accommodationLondon South North and MidlandsAllSelf-contained 88% [72%] 85% [41%] 52% [13%] 79% [36%]Hostel or supported housing4% [3%] 10% [5%] 6% [2%] 6% [3%]B&B hotel 3% [3%] 0% [0%] 1% [<1%] 2% [1%]Parents 3% [2%] 4% [2%] 19% [5%] 7% [3%]Friends or (other) relatives2% [2%] 1% [1%] 22% [5%] 7% [3%]Total 100% [82%] 100% [48%] 100% [24%] 100% [45%]Base 544 [641] 418 [734] 168 [678] 1,130 [2,053]Source: Survey 1 Percentages in brackets indicate the proportion of all families accepted as homeless that the relevant group constitutes.Length of time in temporary accommodation As families in settled housing had, by time of survey, ‘completed’ their stay in 6.18 temporary accommodation, whereas those still in temporary accommodation had not, these two groups are considered separately in this section. The section does conclude, however, by considering the overall prevalence of stays of more than six months in temporary accommodation across all families accepted as homeless207.How quickly were families in settled housing provided with their new homes?Across England as a whole, families in settled housing reported having been 6.19 in temporary accommodation prior to entering that housing for an average of 4.7 months (median 3 months). The average length of stay in temporary accommodation for families in settled housing in the North and Midlands was 3.9 months (median 2 months); it was somewhat longer in the South (average 5.3 months, median 4 months); and longest in London (average 6.5 months, median 4 months). 207 Recall can be a particular problem with time-related questions, so all of the data on duration of stay in temporary accommodation should be treated as broadly indicative rather than precise. Chapter 6 Families’ experience of temporary accommodation 125As Table 6.4 demonstrates, amongst (the large proportion of) families in 6.20 the North and Midlands who were in settled housing by point of survey, 21 per cent had experienced a stay in temporary accommodation exceeding six months. This contrasts with 41 per cent of (the much smaller number of) families in settled housing in London who had spent at least six months in temporary accommodation. Table 6.4: Time spent in temporary accommodation by those families in settled housing at point of surveyNo temporary accom-modation stay Up to 3 monthsMore than 3 months up to 6 monthsOver 6 monthsTotal BaseLondon 33% [6%] 13% [2%] 13% [2%] 41% [7%] 100% [18%] 97* [641]South 33% [17%] 22% [11%] 16% [8%] 30% [16%] 100% [52%] 316 [734]North and Midlands39% [30%] 19% [15%] 20% [20%] 21% [16%] 100% [76%] 510 [678]All 37% [21%] 18% [11%] 18% [10%] 25% [13%] 100% [55%] 923 [2,053]Source: Survey 1 Percentages in brackets indicate the proportion of all families accepted as homeless that the relevant group (in settled housing) constitutes. * The base number of families who had been provided with settled accommodation in London is small, so these estimates must be treated with caution. Table 6.4 also demonstrates that around a third (37 per cent) of families 6.21 in settled housing, comprising 21 per cent of all families accepted as homeless, had been moved directly into settled housing without a temporary accommodation stay.Very low numbers of families accepted in London had moved straight into 6.22 settled housing: while this was reported by 33 per cent of those in settled housing in the capital, this small group accounted for only 6 per cent of all families accepted in London. By contrast, 39 per cent of families in settled housing in the North and Midlands (30 per cent of all families accepted in this broad region) had moved straight into settled housing.Families accepted in areas of higher housing stress6.23 208 were unlikely to have moved straight into settled housing. Only 11 per cent of all families accepted in these areas, as compared to 31 per cent of families in areas of less housing stress, had moved straight into settled housing.Adult respondents with an ethnic minority background were unlikely to have 6.24 moved directly into settled housing (12 per cent had done so as compared to 208 Local authorities with relatively lower and higher levels of housing stress were identified by employing the Wilcox affordability ratio. See Appendix 2 for a full explanation. 126 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds24 per cent of other adult respondents). A similar pattern existed in relation to former asylum seekers (11 per cent moved straight into settled housing, compared to 22 per cent of other adult respondents). However, regression analysis indicated that, when other factors were held constant209, neither ethnicity nor being a former asylum seeker had an independent effect on likelihood of having moved straight into settled housing. Instead, the independent effects were restricted to region (see para 6.22) and housing stress (see para 6.23). How long had those still in temporary accommodation been in temporary accommodation?Across England, those still in temporary accommodation had experienced an 6.25. average stay in temporary accommodation of 11 months (median 11 months) by point of survey. In London, this average was slightly higher at 11.6 months (median 12 months); in the South it was 10.6 months (median 11 months); and in the North and Midlands it was 10 months (median 10 months). As is shown in Table 6.5, the great majority (85 per cent) of families who 6.26. were still in temporary accommodation at point of survey had been there for more than six months, and this figure was relatively constant across the broad regions. Table 6.5: Time spent in temporary accommodation by those families in temporary accommodation at point of surveyUp to 3 monthsMore than 3 months up to 6 monthsOver 6 monthsTotal BaseLondon 3% [3%] 8% [7%] 89% [73%] 100% [82%] 544 [641]South 4% [2%] 14% [7%] 83% [40%] 100% [48%] 734 [418]North and Midlands6% [1%] 12% [3%] 82% [19%] 100% [24%] 168 [678]All 4% [2%] 11% [5%] 85% [39%] 100% [45%] 1,130 [2,053]Source: Survey 1 Percentages in brackets indicate the proportion of all families accepted as homeless that the relevant group (still in temporary accommodation) constitutes.It must be borne in mind, however, that the relative homogeneity of 6.27 these results arises from the ‘incomplete’ nature of these families’ stays in temporary accommodation: the time that they had spent in temporary 209 The factors controlled for in this regression analysis were demographic and geographical variables.Chapter 6 Families’ experience of temporary accommodation 127accommodation at point of survey was effectively only a record of the time that had elapsed between acceptance as homeless and when they were interviewed for the present research. The ultimate length of stay in temporary accommodation was likely to be much longer for those in London, and to a lesser extent for those in the South, than for families in the North and Midlands. The data presented in Table 6.5 is indicative of this insofar as it notes that families still in temporary accommodation, having already spent six months there, comprised 73 per cent of all families accepted as homeless in London, 40 per cent of those accepted in the South, but only 19 per cent of those in the North and Midlands (see also Figure 6.1 below) 210. We asked adult respondents still in temporary accommodation about 6.28 the information they had received regarding progress with finding them settled housing. Overall, 59 per cent of respondents still in temporary accommodation had not been told how much longer they would have to wait before being provided with settled housing. There was no relationship between likelihood of being told how much longer they would have to wait and any geographical or demographic variables. It might have been anticipated that larger households would face more uncertainty about how long they would have to wait than would be the case for smaller households, as larger social housing is in short supply, but no evidence was found of this.Overall, how prevalent are stays in temporary accommodation of more than 6 months?As is shown in Figure 6.1, 80 per cent of all families accepted in London had 6.29 experienced a temporary accommodation stay exceeding six months (at point of survey), compared to 55 per cent of families in the South, and 35 per cent in the North and Midlands211. 210 The P1E data also indicate that the ultimate stay in temporary accommodation for households accepted as homeless differs substantially depending on where they are accepted. For example, during 2005/6, 52 per cent of all households leaving temporary accommodation for settled housing had been in temporary accommodation for less than six months, but the figure for London was only 19 per cent. Across England, 13 per cent of households entering settled housing from temporary accommodation had been resident in temporary accommodation for more than two years, but this was true of 36 per cent of those accepted in London (CHP analysis). For information on the P1E data see: http://www.communities.gov.uk/ 211 This section considers the prevalence of prolonged stays in temporary accommodation across all Survey 1 families. Thus the proportions given for each region include both those still in temporary accommodation having stayed there for 6 months or more, and those in settled housing after a temporary accommodation stay of at least 6 months.128 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 6.1: Distribution of temporary accommodation stays across regionsNo stay in temporary accommodation6 months3-6 monthsUp to 3 months0%20%40%60%80%100%North and MidlandsSouthLondon80%9%6%5%55%15% 17%13%35%31%16% 16%Source: Survey 1; Base: 2,053 respondentsLondon clearly stands out on its own. While stays in temporary 6.30 accommodation in the South are longer than in the North and Midlands, these two broad regions have more in common with each other than with the capital. The extent of long stays in temporary accommodation in London is not unexpected, as this is consistent with the relevant data available from official P1E statistics212. Nevertheless, these findings highlight the stark reality of what are clearly markedly different experiences of statutory homelessness in different parts of England. Regression analysis confirmed that, other things being equal6.31 213, being accepted as homeless in London was the most important independent influence on whether a family would experience a stay of six months or longer in temporary accommodation. However, there was also an independent, albeit smaller, effect of being in a larger family group (four or more people) which increased the likelihood of stays over 6 months. Given that families still in temporary accommodation at point of survey had 6.32 not yet completed their stay in temporary accommodation, it was not possible to determine what the average length of stay in temporary accommodation would ultimately be for families accepted as homeless as a whole. However, at point of survey it was 7.3 months, with a median of 7 months. 212 See footnote 206 above.213 The factors controlled for in this regression analysis were demographic and geographical variables.Chapter 6 Families’ experience of temporary accommodation 129Types of temporary accommodation experienced Table 6.6 shows all the forms of temporary accommodation experienced by 6.33 families since they were accepted as homeless. As can be seen, self-contained temporary accommodation was the most 6.34 commonly experienced form of provision, reported by 59 per cent of those with a temporary accommodation stay (50 per cent of all families accepted as homeless)214. Living with parents, friends and (other) relatives, and living in a hostel, were each experienced by around a quarter of those with a temporary accommodation stay (and around one fifth of all families accepted as homeless). B&B hotels were the least commonly experienced form of temporary accommodation, reported by 15 per cent of those with a temporary accommodation stay (comprising 12 per cent of all families accepted as homeless)215. Table 6.6: Types of temporary accommodation experienced by families (families who had stayed in temporary accommodation)Type of accommodationLondon South North and MidlandsAllSelf-contained 84% [79%] 64% [56%] 38% [29%] 59% [50%]Hostel or supported housing17% [15%] 29% [24%] 25% [17%] 24% [19%]B&B hotel 22% [21%] 18% [15%] 8% [5%] 15% [12%]Parents 19% [18%] 25% [21%] 29% [20%] 25% [20%]Friends or (other) relatives21% [19%] 20% [17%] 36% [25%] 27% [21%]Base 608 [641] 643 [734] 464 [678] 1,715 [2,053]Source: Survey 1 Percentages in brackets are for all families accepted as homeless. Families could have experienced more than one form of temporary accommodation, so percentages do not sum to 100%.214 Readers will note that this suggests a less predominant role for self-contained temporary accommodation in families’ experiences than that indicated by the P1E quarterly statistics (84 per cent of families in self-contained temporary accommodation at end June 2005). This is a good example of the difference between the time window/experiential approach of the current study (which takes fuller account of shorter stays in temporary accommodation, for which shared forms of temporary accommodation are often employed) and the ‘snapshot’ nature of the P1E data (which tends to capture mainly the position of those experiencing longer stays in temporary accommodation, where self-contained provision is more dominant). See para 6.46 below on temporary accommodation ‘pathways’ for further discussion. 215 It should be noted that, as 45 per cent of all families had not ‘completed’ their temporary accommodation stay at point of interview, the final proportion experiencing various forms of temporary accommodation may be a little higher. However, the ‘pathways’ analysis presented below suggests that most moves between temporary accommodation types are made towards the beginning of a temporary accommodation episode, so it is unlikely that the proportion experiencing particular forms of temporary accommodation will rise by much (see para 6.46).130 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 6.6 also makes clear that self-contained temporary accommodation 6.35 is far more widely experienced in London (by 84 per cent of those with a temporary accommodation stay), than it is in the North and Midlands (by 38 per cent of those who had stayed in temporary accommodation). B&B hotels were also used more often in London (reported by 22 per cent of those with a temporary accommodation stay) than in the North and Midlands (reported by only 8 per cent of those with a temporary accommodation stay). Conversely, temporary arrangements with parents, and with friends and (other) relatives, were more commonly employed in the North and Midlands than in the capital (experienced by 29 per cent and 36 per cent respectively in the North and Midlands, as compared to 19 per cent and 21 per cent in London).Regression analysis confirmed that, other things being equal6.36 216, region had the most powerful independent effect on whether a family had experienced self-contained temporary accommodation (it was less likely for families in the North and Midlands). However, there were also (weaker) independent effects of the following: • household type: 70 per cent of couple households who had stayed in temporary accommodation, as compared with 55 per cent of lone women parent households, had experienced self-contained temporary accommodation. • rurality: 62 per cent of those in urban areas with a temporary accommodation stay, as compared with 50 per cent in rural areas, had experienced self-contained temporary accommodation (this may indicate that rural authorities face particular difficulties in obtaining private or social rented housing for use as temporary accommodation)217. • age of adult respondent: 64 per cent of adult respondents aged over 25 who had stayed in temporary accommodation, as compared with 52 per cent of those under this age, had experienced self-contained temporary accommodation. This was linked to the greater propensity of younger adult respondents to experience temporary arrangements with parents, friends and (other) relatives (see para 6.38 below). Additionally, there was an independent effect of time spent in temporary 6.37 accommodation: 72 per cent of families who had spent more than six months in temporary accommodation, as compared with only 43 per cent of those who had had a shorter stay in temporary accommodation, had experienced self-contained temporary accommodation (see ‘pathways’ discussion below, para 6.46).216 Factors controlled for in this regression analysis included demographic and geographical variables and length of time in temporary accommodation. 217 For the definitions of urban and rural used in this research see Appendix 2. Chapter 6 Families’ experience of temporary accommodation 131We also conducted regression analyses on each of the other forms of 6.38 temporary accommodation to determine what factors, if any, had an independent effect on the likelihood of them being experienced. When a range of variables were held constant218, the independent effects detected were as follows: • experiencing B&B hotels was associated with being accepted in an area of ‘higher housing stress’219 (this may reflect the particular difficulty that local authorities in these areas face in securing alternative forms of temporary accommodation). • experiencing hostels was associated with being accepted in a rural area220. • temporarily staying with parents, friends or (other) relatives was associated with adult respondents being aged under 25. Staying with friends and relatives was additionally associated with being accepted in the North and Midlands. Most families with a temporary accommodation stay had experienced at least 6.39 some time in shared forms of provision221, with only around a quarter (28 per cent) having only ever experienced self-contained temporary accommodation. Families accepted in London were more likely than those accepted elsewhere to have stayed only in self-contained temporary accommodation (38 per cent as compared to 24 per cent). Much smaller proportions of families had only ever experienced any other single form of temporary accommodation, including less than 1 per cent who had experienced only B&B hotels.Moves between temporary accommodation addresses Table 6.7 shows that 56 per cent of those with experience of temporary 6.40 accommodation had lived at only one temporary accommodation address. This group accounted for 43 per cent of all families accepted as homeless and, together with the 21 per cent of families who experienced no temporary accommodation at all, this meant that 64 per cent of all families accepted as homeless had experienced no moves between temporary accommodation addresses. One third (34 per cent) of families with experience of temporary 6.41 accommodation had stayed in two temporary accommodation addresses (27 per cent of all families accepted as homeless), and 10 per cent had lived 218 Factors controlled for in each of these regression analyses included demographic and geographical variables. 219 Local authorities with relatively lower and higher levels of housing stress were identified by employing the Wilcox affordability ratio. See Appendix 2 for a full explanation.220 This links with the point above about the possible lack of availability of self-contained temporary accommodation in rural areas, see para 6.36.221 That is, in temporary arrangements with parents, friends or (other) relatives, in B&B hotels, or in hostels. 132 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsat three or more temporary addresses (8 per cent of all families accepted as homeless). Thus it was clear that multiple moves between temporary accommodation addresses were quite rare222. Table 6.7: Number of different temporary accommodation addresses reported by families (families who had stayed in temporary accommodation) No TA stay1 TA address2 TA addresses3 or more TA addressesTotal BaseNorth and Midlands0% [32%]66% [44%]27% [19%]7% [5%]100% [100%]464 [678]South 0% [18%]50% [40%]37% [31%]13% [11%]100% [100%]643 [734]London 0% [7%]47% [44%]41% [38%]12% [11%]100% [100%]608 [641]All 0% [21%]56% [43%]34% [27%]10% [8%]100% [100%]1,715 [2,053]Source: Survey 1 Figures for all families accepted as homeless are shown in brackets. There were distinct regional patterns with regards to experience of moves 6.42 between temporary accommodation addresses. One half of the families who had experienced a temporary accommodation stay in London (53 per cent) and in the South (50 per cent) had made at least one move between temporary accommodation addresses. This compared to only one third (34 per cent) of families with a temporary accommodation stay in the North and Midlands (24 per cent of all families accepted in this broad region) (Table 6.7).One half (54 per cent) of families accepted in the areas of higher housing 6.43 stress, who had stayed in temporary accommodation, had made at least one move between temporary accommodation addresses (accounting for 50 per cent of all families accepted in these areas). This compared to only 35 per cent of families in areas of lower housing stress who had stayed in temporary accommodation moving at least once (25 per cent of all families accepted in these areas)223.Regression analysis confirmed that, other things being equal6.44 224, both broad region and housing stress had an independent influence on the likelihood 222 However, it should be noted that, as 45 per cent of all families had not ‘completed’ their temporary accommodation stay at point of interview, the final proportion experiencing multiple moves may be higher. That said, the ‘pathways’ analysis presented below suggests that most moves are made at beginning of a temporary accommodation episode, so it is unlikely that the proportion will rise by much (see para 6.46).223 Local authorities with relatively lower and higher levels of housing stress were identified by employing the Wilcox affordability ratio. See Appendix 2 for a full explanation. 224 Factors controlled for in this regression analysis were geographical and demographic variables, and length of time in temporary accommodation. Chapter 6 Families’ experience of temporary accommodation 133of making a move between temporary accommodation addresses. However, the most powerful independent effect detected was length of time in temporary accommodation: 58 per cent of families who had stayed in temporary accommodation for six months or more had made at least one move, compared to only 30 per cent of those who had stayed in temporary accommodation for a shorter period. As the ‘pathways’ analysis below demonstrates, the purpose of many of 6.45 these moves between temporary accommodation addresses seemed to be to relocate families from shared forms of provision to self-contained temporary accommodation, especially in London and the South, where stays in temporary accommodation were often lengthy. Pathways through temporary accommodation Figure 6.2 summarises families’ ‘pathways’ through temporary 6.46 accommodation, and confirms that they were strongly related to where they were accepted as homeless. The main points to note are as follows:• One fifth (21 per cent) had not stayed in temporary accommodation at all. The majority of these were in the North and Midlands (67 per cent).• Families in London and the South were much more likely to still be in temporary accommodation at point of survey (77 per cent of all families in temporary accommodation were in London and the South), and these families were likely to be in self-contained temporary accommodation (86 per cent). • Families in London and the South in temporary accommodation were likely to have been moved into self-contained temporary accommodation on acceptance (40 per cent), or to have been moved from shared forms of provision into self-contained temporary accommodation by point of survey (46 per cent). • The (relatively small number of) families still in temporary accommodation in the North and Midlands were more likely to be in shared forms of provision than those in temporary accommodation in London or the South (47 per cent as compared with 14 per cent). • Across all regions, families in settled housing at point of survey (after a spell in temporary accommodation) were likely to have been moved into this housing straight from shared forms of temporary accommodation (71 per cent in London and the South and 75 per cent in the North and Midlands). These families had generally spent less time in temporary accommodation than those still in temporary accommodation, and throughout England it appeared that self-contained temporary accommodation was less likely to be provided where it was the case that a family would soon be provided with settled housing. 134 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 6.2: Summary of temporary accommodation pathwaysLondon,the South77% of allfamilies in TA77% in TA for more than six months 72% in TA for more than six months 41% in TA for more than six months 27% in TA for more than six monthsNorth andMidlands23% of allfamilies in TALondon,the South37% of allfamilies in settled housingafter a TA stayNorth andMidlands63% of allfamilies insettled housingafter a TA stayLondon,the South23% of allfamilies insettled housingwithout a TAstayMoved straight into settledhousing with no temporaryaccommodation stay:21% of all familiesNorth andMidlands67% of allfamilies insettled housingwithout a TAstayMoved straight into self-contained TAafter application: 40%Moved into self-contained TA by point of surveyfrom parents, other relatives or friends, hostels,or B&B: 46%Parents, other relatives or friends, hostels,or B&B at point of survey: 14%Moved straight into self-contained TAafter application: 25%Moved into self-contained TA by point of surveyfrom parents, other relatives or friends, hostels,or B&B: 28%Parents, other relatives or friends, hostels,or B&B at point of survey: 47%Rehoused from self-contained TA: 29%Rehoused from parents, other relativesor friends, hostels, or B&B: 71%Rehoused from self-contained TA: 25%Rehoused from parents, other relativesor friends, hostels, or B&B: 75%In settled housing at point ofsurvey following temporaryaccommodation stay:34% of all familiesIn temporary accommodationat point of survey:45% of all familiesFamilies acceptedas homelessChapter 6 Families’ experience of temporary accommodation 135Families in temporary accommodation for more than one yearAs detailed in Chapter 2, the great majority (82 per cent) of families in 6.47 temporary accommodation for over one year (Survey 4 families) were accepted as homeless in London. All of the other Survey 4 families (18 per cent) were accepted in the South (see para 2.66). All of these Survey 4 families were in self-contained temporary 6.48 accommodation by the point of survey. As noted above, Survey 1 excluded, by definition, those with prolonged 6.49 stays in temporary accommodation, so Survey 4 was specifically designed to capture the experiences of this group. Thus, as would be expected, the average time spent in temporary accommodation by Survey 4 families at point of survey was longer, at 2.9 years, with a median of 2.6 years (as compared with an average of 7.3 months, and a median of 7 months, for Survey 1 families at point of survey225). Table 6.8 below presents the types of temporary accommodation families 6.50 in Survey 1 (with a temporary accommodation stay) and in Survey 4 had experienced. It is clear that experience of self-contained temporary accommodation and B&B hotels was more common, and temporary arrangements with parents were less common, amongst Survey 4 than Survey 1 families. This is in keeping with the regional analysis within Survey 1 above which showed that families in London, where most Survey 4 families were located, were more likely to experience self-contained temporary accommodation, and B&B hotels, than families elsewhere, and were less likely to experience temporary arrangements with parents. 225 It must always be borne in mind that for many of those families in Survey 1, as well as all of those in Survey 4, their stay in temporary accommodation was not yet complete at point of survey. 136 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 6.8: Types of temporary accommodation experienced by Survey 4 families and by Survey 1 families (with a temporary accommodation stay)Type of temporary accommodation Survey 4 Survey 1Self-contained 100% 63%Hostel, foyer or other supported housing24% 24%B&B hotel 36% 15%Parents 15% 25%Friends or (other) relatives 22% 27%Base 571 1,715Source: Survey 1 (with a temporary accommodation stay) and Survey 4. Respondents could report experience of more than one type of temporary accommodation so percentages do not sum to 100%. Figures are for point of survey, and percentages include, for both surveys, families still in temporary accommodation, who might go on to experience other forms of temporary accommodation. Survey 4 families were more likely to have experienced moves between 6.51 temporary accommodation addresses than were Survey 1 families (Table 6.9). Most notably, 43 per cent of Survey 4 families had lived in three or more temporary accommodation addresses, compared to only 10 per cent of those in Survey 1. This reinforces the importance of the relationship between length of time in temporary accommodation and moves noted above (see para 6.44). Table 6.9: Number of temporary accommodation addresses experienced by Survey 4 families and by Survey 1 families (with a temporary accommodation stay)Number of different temporary accommodation addressSurvey 4 Survey 1 Difference1 address 18% 56% -38%2 addresses 39% 34% 15%3 or more addresses 43% 10% 33%Total 100% 100% –Base 571 1,715 –Source: Survey 1 (with a temporary accommodation stay) and Survey 4. Figures are for point of survey, and percentages include, for both surveys, families still in temporary accommodation, who might go on to experience more moves between TA addresses. Chapter 6 Families’ experience of temporary accommodation 137However, it is important to note that almost all of these Survey 4 families 6.52 (95 per cent) had been in their current temporary accommodation for at least six months when they were surveyed. In fact, on average, they reported having been in this particular temporary accommodation for 31 months (2.5 years) (the median was 29 months, 2.4 years). So Survey 4 families were not necessarily experiencing current housing instability, even though they had experienced more moves than those in Survey 1.Table 6.10 summarises the pathways through temporary accommodation 6.53 taken by Survey 4 families. Half (50 per cent) of all of these families had moved from some form of shared provision into self-contained temporary accommodation (and the majority of this group had experienced at least two moves between temporary accommodation addresses). The other 50 per cent of families had only experienced self-contained temporary accommodation. This comprised 32 per cent of all Survey 4 families who had moved between self-contained temporary accommodation addresses (most had moved only once), and 18 per cent of all Survey 4 families who had stayed only at their current (self-contained) temporary accommodation address.Table 6.10: Pathways through temporary accommodation for Survey 4 familiesPathway through TA (at point of survey) Survey 4 families Moved from shared temporary accommodation into self-contained TA 50%Only experienced self-contained TA: had moved between TA addresses 32%Only experienced self-contained TA: no moves 18%Total 100%Base 571Source: Survey 4. Figures are for point of survey, and relate to families still in temporary accommodation, some of whom might make further moves between temporary accommodation addresses. The 50 per cent of Survey 4 families who had only experienced self-contained 6.54 temporary accommodation was higher than the 28 per cent of Survey 1 families (with a temporary accommodation stay) who had only experienced this form of provision. This effect was associated with London: the great majority of Survey 4 families were accepted in London, and families accepted in London in Survey 1 were more likely than those accepted elsewhere to have only stayed in self-contained temporary accommodation (see para 6.39).138 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsWe asked Survey 4 adult respondents whether they had had any choice over 6.55 their (often long-term) temporary accommodation226. Only around one tenth (11 per cent) of these families reported that they had been given a choice when allocated their current temporary accommodation. We also asked whether they had rejected any offers of settled housing since their acceptance as homeless. Only 7 per cent reported that they had, which is perhaps unsurprising since evidence indicates that local authorities often make only one ‘reasonable offer’ of settled housing as discharge of their statutory homelessness duty (see footnote 268).We also investigated Survey 4 adult respondents’ perceptions of their 6.56 extended stay in temporary accommodation and their views on the reasons for this. More than two thirds (69 per cent) of these adult respondents reported that they were ‘not informed at all’ about the progress that the council was making towards securing them settled accommodation; an additional 20 per cent considered themselves ‘not very well informed’; and only 11 per cent either ‘fairly’ or ‘very’ well informed. Over half (59 per cent) of Survey 4 adult respondents reported that they were ‘very frustrated’ about their length of wait for settled housing and 28 per cent that they were ‘a bit frustrated’. Figure 6.3 summarises the reasons why Survey 4 adult respondents thought it 6.57 was taking the council so long to find settled housing for them. One quarter (24 per cent) attributed the long wait to a (general) shortage of housing/accommodation, and 18 per cent to the large number of people on the waiting list. Smaller proportions blamed the prolonged wait on the grading/priority system (8 per cent), to the lack of large properties (7 per cent), or to the bidding system (in choice-based lettings schemes) (6 per cent). Only very small numbers mentioned other possible factors. One quarter (27 per cent) said that they did not know why they had been in temporary accommodation for so long.226 See also Chapter 8 (para 8.12) for a discussion of choice over settled housing.Chapter 6 Families’ experience of temporary accommodation 139Figure 6.3: Reasons why survey 4 respondents thought they had had to wait so long in temporary accommodation27%24%18%8%7%6%3%2%2%2%2%Have rent arrearsPrevious offer was unsuitableBecause want to stay in same areaImmigrationHave to wait for many yearsHave to use a bidding systemA lack of large propertiesBecause of gradingh/priority systemA lot of people on waiting listThere is a shortage of housingDon't know why0% 5% 10% 15% 20% 25% 30%Source: Survey 1; Base: 571 respondents. Multiple responses were possible. Conclusions This chapter has reviewed families’ experiences of temporary accommodation. 6.58 It has demonstrated that by far the most important determinant of these experiences is where a family is accepted as homeless. Families accepted in London, and to a lesser extent those accepted in the 6.59 South, often experienced long stays in temporary accommodation, and around half had made at least one move between temporary accommodation addresses. These families spent much of their time in self-contained temporary accommodation. Use of temporary arrangements with parents, friends and (other) relatives was 6.60 much commoner in the North and Midlands, with families (typically) staying in such arrangements for a relatively short period before moving on to settled housing. Most of those who did not experience temporary accommodation at all, because they were accommodated directly into settled housing, were in the North and Midlands. 140 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsIn the main, demographic characteristics had no independent influence on 6.61 temporary accommodation experiences once region and other geographical variables (such as rurality and housing affordability) were taken into account. However, there was a (modest) independent effect of household size on the likelihood of spending over 6 months in temporary accommodation (likelier amongst families with four or more people). In addition, lone women parents were less likely than couple households to experience self-contained temporary accommodation, and this was also the case for adult respondents under 25 as compared to those over this age (younger adult respondents were likeliest to experience arrangements with parents, friends or other relatives). The next chapter moves on to consider physical and other conditions in the 6.62 temporary accommodation experienced by these families. Chapter 7 Conditions in temporary accommodation 141Chapter 7:Conditions in temporary accommodationIntroductionThe overall objective of this study was to understand the causes, experiences 7.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data was collected in five separate surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys).The physical and other conditions in temporary accommodation provided to 7.2 families accepted as homeless have long been a matter of policy concern. As noted in Chapter 6, numerous reports of the poor conditions experienced by families in B&B hotels227 resulted in a Government commitment to end the long-term use of this form of temporary accommodation for families under the homelessness legislation228. However, there has been little investigation of the conditions pertaining in other forms of temporary accommodation, in particular self-contained temporary accommodation and temporary arrangements with family and friends. This chapter therefore draws on data from Survey 1 – a survey of families 7.3 accepted as being owed the main homeless duty between 1st January 2005 and 30 June 2005 – to explore families’ experiences of temporary accommodation conditions, and in particular to compare reported conditions across different types of provision. Please note that this chapter presents adult respondents’ views of conditions in temporary accommodation, rather than drawing on independent or official inspections of such accommodation229. 227 Bayswater Hotel Homelessness Project (1987) Speaking for Ourselves: Families in Bayswater B&B, London: Bayswater Hotel Homelessness Project; Murie, A. and Jeffers, S. (1987) Living in Bed and Breakfast: The experience of homelessness in London, Bristol: University of Bristol, School for Advanced Urban Studies; Crane, H. (1990) Speaking from Experience: Working with homeless families, London: Bayswater Hotel Homelessness Project; Niner, P. (1989) Homelessness in Nine Local Authorities: Case studies of policy and practice, London: HMSO; Thomas, A. and Niner, P. (1989) Living in Temporary Accommodation: A survey of homeless people, London: HMSO; Carter, M. (1995) Out of Sight: London’s continuing B&B crisis, London: London Homelessness Forum; Carter, M. (1997) The Last Resort: Living in Bed and Breakfast in the 1990s London: Shelter. 228 Since April 2004 it has been prohibited for local authorities to accommodate families accepted as homeless in B&B hotels for more than six weeks (Homelessness (Suitability of Accommodation) (England) Order 2003 (SI 2003/3326)) 229 This is important to bear in mind where comparisons are made to findings from the English House Conditions Survey (EHCS) where surveyors are responsible for assessing the state of repair of properties, and local liveability factors, such as poor quality environments. 142 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThe chapter draws on data about the two temporary accommodation settings 7.4 about which detailed information was sought:• the current temporary accommodation occupied by families still in temporary accommodation; and• the last temporary accommodation occupied by families provided with settled housing (where they had had experience of temporary accommodation).This technique, of drawing on both current and most recent experience of 7.5 temporary accommodation, has been adopted to maximise the coverage of temporary accommodation conditions in this analysis. Asking only about current temporary accommodation at point of survey would have meant that under half of Survey 1 adult respondents would have been able to report on temporary accommodation conditions (only 45 per cent of families were still in temporary accommodation when interviewed, see para 6.11), whereas eight out of ten were able to do so using this approach230. The chapter begins by describing the type of current or most recent 7.6 temporary accommodation reported on by families accepted as homeless. This combined dataset is then employed to explore adult respondents’ views and experiences with respect to the following aspects of temporary accommodation:• space, rooms and sharing;• satisfaction with facilities;• access to household items and amenities;• physical conditions; • sense of safety; and• overall satisfaction. As noted in earlier chapters, the experience of those with prolonged stays 7.7 in temporary accommodation is of particular interest. The last section of this chapter considers the temporary accommodation conditions reported by adult respondents in families accepted as being owed the main homelessness duty and in temporary accommodation for more than a year (Survey 4 adult respondents)231.230 As noted in Chapter 6, around one fifth (21 per cent) of all families accepted as homeless were accommodated directly into settled housing without a stay in temporary accommodation, and so are not referred to in this chapter. When the term ‘all adult respondents’ is used in this chapter, it refers to adult respondents in this combined dataset only, i.e. those able to report on conditions in their current or last temporary accommodation. 231 Survey 4 was required because the ‘time-window’ design for Survey 1, while delivering a representative sample of those accepted as homeless over a six month period, by definition excluded those with prolonged stays in temporary accommodation. See Chapter 1 and Appendix 1 for a full explanation. Chapter 7 Conditions in temporary accommodation 143As elsewhere in this report, we present two types of statistical analysis in 7.8 this chapter: bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a range of factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected by regression analysis232.This evidence reveals that overall satisfaction levels varied little between 7.9 different types of temporary accommodation. However, there was often a relationship between the perceived attributes of temporary accommodation and its type. Thus, self-contained temporary accommodation performed best with regards to the living space afforded to families, but in some respects it was described as offering the worst physical conditions. Arrangements with friends and relatives afforded families access to the broadest range of household items and amenities and the best physical conditions, but also presented the greatest problems with respect to space standards. Access to household items and amenities was often more restricted in B&B hotels and in hostels than in other forms of temporary provision. Survey 4 adult respondents were more likely to express dissatisfaction with cooking, bathroom and sleeping arrangements, and with overall living space, than Survey 1 adult respondents. This was associated with their larger average household size. Key points • Self-contained temporary accommodation was perceived to offer better space standards than other forms of temporary accommodation, and was rated most highly with regards to cooking, sleeping, bathroom and other facilities. However, the worst physical conditions were often reported in this type of provision, particularly with respect to damp, décor and state of repair.• Temporary arrangements with friends and relatives offered families the best physical conditions and access to the widest range of household items and amenities (such as washing machines, tumble dryers, showers, and gardens/play areas). Families also felt safest when in this form of temporary accommodation. However, concerns about lack of space and privacy were at their most acute in these arrangements.232 See Chapter 1 and Appendix 2 for more detail.144 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds• Access to household items and amenities (including kitchens and living rooms) was often more restricted in hostels and B&B hotels than in other forms of temporary accommodation. However, the worst physical conditions and space standards were not generally reported in these forms of temporary accommodation.• Overall satisfaction levels differed little between temporary accommodation types, but rather were influenced by specific attributes of temporary accommodation – such as living space, perceptions of safety, sharing rooms with other households, and physical standards – independent of type.• Survey 4 adult respondents were far less satisfied with cooking, bathroom and sleeping arrangements in their temporary accommodation than were Survey 1 respondents reporting on self-contained temporary accommodation. This was explained in part by the stronger representation of larger households (five or more people) amongst Survey 4 families, who tended in both surveys to report lower levels of satisfaction with these arrangements, and also with their overall living space.The current or most recent temporary accommodation occupied by families Table 7.1 shows the types of current/last temporary accommodation reported 7.10 on in the combined data set used in this chapter. Table 7.1: Type of temporary accommodation (current or last occupied temporary accommodation) Accommodation typeCurrent temporary accommodation (families in temporary accommodation)Last temporary accommodation (families in settled housing)AllSelf-contained 79% 27% 57%Hostel 8% 31% 18%B&B hotel 5% 4% 5%Parents 1% 15% 7%Friends or (other) relatives 7% 23% 14%Total 100% 100% 100%Base* 1,096 582 1,678Source: Survey 1 *Base is 1,678 families for whom data were available reporting on either their current temporary accommodation or the last temporary accommodation they had occupied prior to settled housing. Chapter 7 Conditions in temporary accommodation 145The notable distinction in the accommodation profile between those currently 7.11 in temporary accommodation (overwhelmingly in self-contained temporary accommodation), and the last temporary accommodation of those now in settled housing (much stronger representation of hostels and arrangements with parents, friends or (other) relatives), relates to the regionally-driven ‘pathways’ through homelessness reported in Chapter 6. In other words, families still in temporary accommodation tended to be in London or the South, and to have been provided with self-contained temporary accommodation in a context where stays in temporary accommodation were relatively prolonged. In contrast, families in settled housing, who were mainly in the North and Midlands, were likely to have stayed for a relatively short period in various forms of shared temporary accommodation before being moved into settled housing. Table 7.2 confirms this broad regional pattern within the combined dataset. 7.12 Thus, current/last temporary accommodation amongst families accepted in London was overwhelmingly in self-contained housing (82 per cent); within the South this figure was 62 per cent. In the North and Midlands the proportion of self-contained temporary accommodation was much lower at 35 per cent, with greater use of arrangements with parents, friends and (other) relatives. Table 7.2: Type of temporary accommodation reported on by families by broad region (current or last occupied temporary accommodation)Self-contained Hostel B&B hotelParents Friends or (other) relatives Total BaseLondon 82% 10% 1% 1% 6% 100% 601South 62% 21% 5% 7% 5% 100% 623North and Midlands35% 21% 7% 12% 25% 100% 454All 57% 18% 5% 7% 14% 100% 1,678Source: Survey 1These specific types of temporary accommodation are combined into three 7.13 broad categories for the remainder of this chapter, to aid statistical reliability:• ‘self-contained’ (57 per cent of all current/last temporary accommodation); • ‘hostels and B&B hotels’ (23 per cent of all current/last temporary accommodation); and146 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds• ‘friends and relatives’ (comprising parents, other relatives, and friends) (21 per cent of all current/last temporary accommodation). Space, rooms and sharing in temporary accommodationThis section examines a range of aspects of space, rooms and sharing in 7.14 temporary accommodation, focusing on: • bedrooms and sleeping arrangements; • kitchens, bathrooms and living rooms; and • overall satisfaction with living space. Bedrooms and sleeping arrangementsFamilies staying in self-contained temporary accommodation always slept in 7.15 dedicated bedrooms. However, one fifth (22 per cent) of families reporting on hostels and B&B hotels did not have access to a separate bedroom or bedrooms: for these families, the sleeping area was combined with other functions, e.g. ‘studio’ designs with one living area and/or bed-sitting-rooms. In addition, 13 per cent of adult respondents reporting on arrangements with friends or relatives said that their family was not sleeping in dedicated bedrooms in this accommodation, but rather in a room or rooms with another purpose, such as a living room. A very small number (1 per cent) of adult respondents reporting on 7.16 arrangements with friends or relatives reported that at least one member of the family had to share a bedroom with a person/people from outside their immediate family233. This was not reported in any other type of temporary accommodation234. Sharing of bedrooms 7.17 within the immediate family was far commoner. In total, 58 per cent of adult respondents reported sharing of bedrooms between immediate family members in their current/last temporary accommodation. Bedroom sharing within the immediate family was most frequently reported 7.18 in arrangements with friends or relatives (experienced by 75 per cent of all families who were/had been living in these arrangements), but was also widespread in hostels and B&B hotels (64 per cent of those reporting on this 233 An adult respondent’s ‘immediate family’ was defined as their partner, child(ren) and any other members of their household with whom they intended to live in their ‘settled’ accommodation. 234 Findings in relation to sharing other rooms, such as kitchens (see below) suggest that where a family was living on a temporary basis with relatives, adult respondents would not always classify these relatives as being outside their “immediate family”. Thus this may be an underestimate of cases of sharing bedrooms with relatives who were not part of the immediate family. Chapter 7 Conditions in temporary accommodation 147form of temporary accommodation). It was less common in self-contained temporary accommodation (49 per cent). As Table 7.3 indicates, 53 per cent of those who reported sharing of 7.19 bedrooms within their immediate family said that this was problematic (accounting for 30 per cent of all adult respondents). Bedroom sharing was most likely to be considered a problem in arrangements 7.20 with friends or relatives (70 per cent of adult respondents whose families were/had been sharing bedrooms in these arrangements thought it problematic). However, it was also quite commonly reported as a problem in hostels or B&B hotels (by 58 per cent of those whose families were/had been sharing bedrooms in hostels or B&B hotels). By contrast, bedroom sharing seemed to be less problematic in self-contained temporary accommodation (reported as a problem by 42 per cent of adult respondents whose families were/had been sharing bedrooms in this form of provision). Table 7.3: Extent of problematic bedroom sharing among families sharing bedrooms, by type of current/ last temporary accommodationAccommodation type Sharing a problemSharing not a problemTotal BaseSelf-contained 42% 58% 100% 534Hostels or B&B hotels 58% 42% 100% 217Friends or relatives 70% 30% 100% 190All 53% 47% 100% 941Source: Survey 1 Base: all families where sharing bedrooms within the immediate family was reported in current or last temporary accommodation.There were no regional variations in the extent to which bedroom sharing was 7.21 thought to be problematic. While larger families (with four or more members) were more likely to report having shared bedrooms within their current/last temporary accommodation (80 per cent compared to 49 per cent of other households), they were no more likely than smaller families to report this sharing as problematic. However, lone parents who reported bedroom sharing within their immediate family were more likely to view this as problematic (58 per cent) than couples with children (44 per cent), presumably because in some of the latter cases this was restricted to an adult couple sharing a bedroom.148 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAmong those who said that bedroom sharing was problematic, the most 7.22 commonly reported difficulties were lack of privacy (noted by 72 per cent of those who reported problematic bedroom sharing) and a lack of space (68 per cent). One third (37 per cent) of the adult respondents who reported problematic bedroom sharing in their current/last temporary accommodation said that children of different genders were sharing inappropriately. See Figure 7.1. Figure 7.1: Problems associated with sharing bedrooms in current/last temporary accommodation72%0%20%40%60%80%100%Inappropriate for childrento be sharing bedroomLack of spacePercentage of families reporting problematicbedroom sharing in current or lasttemporary accommodationLack of privacy68%37%Source: Survey 1 Base: 501 adult respondents who reported that sharing bedrooms was problematic in their current or last temporary accommodation. Multiple responses were possible.Adult respondents were asked how satisfied they were with sleeping 7.23 arrangements in their temporary accommodation. Overall, 78 per cent said that they were very or fairly satisfied. However, satisfaction with sleeping arrangements was strongly related to type of temporary accommodation. When reporting on self-contained temporary accommodation, 94 per cent of adult respondents said that they were very or fairly satisfied with the sleeping arrangements. This figure dropped to 64 per cent for hostel and B&B hotels, and to 45 per cent for arrangements with friends or relatives. This follows a similar pattern to the findings on ‘problematic’ bedroom sharing, which, as just noted, was reported most often in arrangements with relatives or friends, next most often in hostels and B&B hotels, and least often in self-contained temporary accommodation. Chapter 7 Conditions in temporary accommodation 149Bedroom StandardThe ‘Bedroom Standard’ is a normative measure of occupation density, based 7.24 on the ages and composition of the family, developed by the Government Social Survey in the 1960s for use in social surveys235. In most instances (88 per cent) the current/last temporary accommodation 7.25 occupied by families accepted as homeless met the Bedroom Standard, as compared with 95 per cent for all rented dwellings in England236. This was true for 89 per cent of self-contained temporary accommodation and 84 per cent of temporary accommodation with friends or relatives. It was also the case that the bulk of hostel and B&B hotel accommodation met the Bedroom Standard (89 per cent), although it must be remembered that the Bedroom Standard defines bed-sitting-rooms as equivalent to bedrooms237. There were no regional variations with respect to the proportion of temporary accommodation that failed to meet the Bedroom Standard. However, it was clear that the Bedroom Standard was not a reliable indicator 7.26 of satisfaction with bedroom sharing arrangements. In fact, 91 per cent of all adult respondents who reported problematic bedroom sharing (see Table 7.3 above) were in temporary accommodation that met the Bedroom Standard. Kitchens, bathrooms and living rooms As Table 7.4 demonstrates, all families whose current/last temporary 7.27 accommodation was self-contained housing had access to a kitchen and bathroom in that accommodation. Almost all (98 per cent) of these families also had access to a living room used only for that purpose (e.g. it was not also used as a bedroom)238. In contrast, 10 per cent of those whose current/last temporary 7.28 accommodation was with friends and relatives reported that they had no access to a kitchen239; and a further 47 per cent said that they had to share their kitchen with members of the host household240. Moreover, 4 per cent 235 See: http://www.communities.gov.uk/index.asp?id=1152919 4.2 As originally modeled, the Bedroom Standard assesses the notional number of bedrooms allocated for each household in accordance with its composition by age, gender, marital status and relationships of family members. Adults identified as partners were assumed to be sharing a bedroom. Any third adult aged over 21 is also assumed to need a separate bedroom. Within the Bedroom Standard, any two children (under 10) and two adolescents aged 10-20 of the same gender are assumed as being able to share a bedroom, a younger child (under 10) can also share a bedroom with an adolescent of the same gender. Unpaired children or adolescents of different genders aged over 10 are assumed to require separate bedrooms. This standard is then compared with the actual number of bedrooms (note that this includes bed-sitting rooms) available for the “sole use of the household”.236 Including 95 per cent of private rented dwellings and 94 per cent of social rented dwellings. (DCLG (2006) Housing in England 2004/5: A Report principally from the 2004/05 Survey of English Housing.)237 See footnote 235 above.238 The 2 per cent reporting on self-contained temporary accommodation who did not have access to a dedicated living room did have access to a living room, but it was also being used as a bedroom. 239 It is important to note here that families were very unlikely to lack access to a cooker or fridge in their current or last temporary accommodation (see Table 7.7 below). However, access to a kitchen was not universal.240 We suspect that the percentage of those staying with friends and relatives who are sharing kitchens with members of the host household is likely to be higher as some adult respondents may have misinterpreted this question and categorised their parents or 150 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldssaid they had no access to a bathroom in this accommodation (this did not mean no access to a toilet or sink, but it did mean no access to a shower or bath), and another 53 per cent reported that they had to share their bathroom with members of the host household241. Finally, 13 per cent said that they had no access to a living room (used only for that purpose) in their temporary accommodation with friends and relatives, and again almost half (45 per cent) said that they had to share their living room with members of the host household242.Those reporting on hostels and B&B hotels were the group most likely to say 7.29 that they had no access to a kitchen. Worryingly, lack of access to a kitchen was reported by one in five (22 per cent) of these families. However, families reporting on hostels and B&B hotels were less likely to have to share a kitchen than families staying in temporary arrangements with friends and relatives (29 per cent as compared with 47 per cent). This meant that there was little difference between the two temporary accommodation types with regards to exclusive access to a kitchen. At 3 per cent, the proportion of families reporting on hostels and B&B hotels 7.30 who said that they had no access to a bathroom was very similar to that for those reporting on arrangements with friends and relatives. But again a smaller percentage of families in hostels and B&B hotels had a shared bathroom (42 per cent). This meant that they were more likely than those reporting on arrangements with friends and relatives to have exclusive access to a bathroom (55 per cent) (see Table 7.4). Families whose current/last temporary accommodation was a hostel or B&B 7.31 hotel were less likely than other families to have access to a living room used solely for that purpose – one third (35 per cent) had no such a room in this accommodation. But again, where a living room was available to these families, it was less likely to be shared with members of another household than was the case in temporary arrangements with family and friends, meaning that the proportion with exclusive access to living rooms was virtually identical across these two temporary accommodation types. other relatives as part of their “immediate family”, and thus not a “different household” with whom they had to share. 241 See preceding footnote – the same misconception might have arisen with regards to sharing of bathrooms. 242 See preceding footnote – the same misconception might have arisen with regards to sharing of living rooms. If we are correct in thinking that the proportion amongst those staying with friends and relatives who are sharing these three types of rooms is likely to be an underestimate, then the marginal advantage that those in hostels and B&B have with regards to exclusive access to such rooms (see Table 7.4 below) is likely to be greater in reality. Chapter 7 Conditions in temporary accommodation 151Table 7.4: Access to kitchens, bathrooms, living rooms (current or last occupied temporary accommodation)Accommodation type No access to kitchen*Shared kitchenOwn kitchenTotal BaseSelf-contained housing 0% 0% 100% 100% 1,073Friends or relatives 10% 47% 43% 100% 257Hostel and B&B hotels 22% 29% 49% 100% 348All 10% 18% 72% 100% 1,678Accommodation typeNo access to bathroom**Shared BathroomOwn bathroomTotal BaseSelf-contained housing 0% 0% 100% 100% 1,073Friends or relatives 4% 53% 43% 100% 257Hostel and B&B hotels 3% 42% 55% 100% 348All 4% 22% 74% 100% 1,678Accommodation type No access to living room***Shared living roomOwn living room Total BaseSelf-contained housing 2% 0% 98% 100% 1,073Friends or relatives 13% 45% 42% 100% 257Hostel and B&B hotels 35% 22% 43% 100% 348All 14% 15% 71% 100% 1,678Source : Survey 1 * No access to a kitchen did not mean no access to a cooker or fridge. ** No access to a bathroom did not mean no access to a toilet or sink, but it did mean a family lacked access to a shower and/or bath. *** A living room was only counted as available to a family if it had no other use (e.g. it was not being used as a bedroom). To summarise, 85 per cent of all adult respondents reported that their 7.32 families had access to all three of these rooms in their current/last temporary accommodation (72 per cent had exclusive access to all three rooms). However, there were sharp variations by temporary accommodation type. Virtually all (98 per cent) of those reporting on self-contained temporary accommodation said that their families had access to all three rooms (none of whom shared any of these rooms). This figure dropped to 81 per cent amongst those reporting on arrangements with friends and relatives (the proportion of this group with exclusive access to all three rooms was 40 per cent). Amongst those reporting on hostels and B&B hotels, 56 per cent had access to all three rooms (the proportion with exclusive access to all three rooms was 33 per cent)243. 243 Families in hostels and B&B hotels or in arrangements with friends and relatives who had exclusive access to all three of these rooms, and to bedrooms used only by the immediate family, are not described in this report as having ‘self-contained’ accommodation because it is of a managed nature (hostels and B&B hotels) or they are having to share a house/flat with a host household (friends and relatives). In other words, the term ‘self-contained’ is used to denote not only exclusive access to all of the conventional rooms that one would expect in a home, but also for these rooms to be located in an ordinary house or flat used only by the immediate family. 152 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsBearing in mind that not all families had access to all of these rooms in their 7.33 current/last temporary accommodation, 15 per cent of adult respondents said that their family had experience of sharing at least one them with members of another household. This included one quarter (23 per cent) of those reporting on hostels and B&B hotels, and 41 per cent of those reporting on temporary arrangements with friends or relatives. Figure 7.2 illustrates the nature of the problems identified with sharing these 7.34 rooms with members of other households. The most commonly identified issue was a lack of privacy (mentioned by 39 per cent of adult respondents who reported sharing one or more of these rooms). Other problems reported included “too many people sharing” (26 per cent), “having to queue” (16 per cent), and the “poor hygiene of others” (14 per cent). In all, 52 per cent of those sharing rooms in their current/last temporary accommodation with members of another household identified at least one problem (accounting for 9 per cent of all adult respondents).Figure 7.2: Problems reported with sharing one or more rooms with other households in current or last temporary accommodation (any sharing of kitchen, living room, bathroom)0% 10% 20% 30% 40%Poor hygieneof othersHave to queueToo manypeople sharingLack of privacy 39%26%16%14%Source: Survey 1 Base: 369 adult respondents whose families had shared rooms with other households in their current or last temporary accommodation.Satisfaction with overall living spaceJust over half (54 per cent) of adult respondents felt that, overall, there was 7.35 enough living space in their current/last temporary accommodation. However, satisfaction with space was strongly associated with accommodation type. Thus, while 69 per cent of adult respondents reporting on self-contained temporary accommodation were satisfied with their living space, this was Chapter 7 Conditions in temporary accommodation 153true of only 41 per cent of those reporting on hostels and B&B hotels, and even fewer (22 per cent) of those reporting on arrangements with friends and relatives (Table 7.5). Table 7.5: Satisfaction with living space by temporary accommodation type (current or last occupied temporary accommodation)Accommodation typeSatisfied with living spaceNot satisfied with living spaceTotal BaseSelf-contained 69% 31% 100% 1,073Friends or relatives 22% 78% 100% 257Hostel and B&B hotels 41% 59% 100% 348All 54% 46% 100% 1,678Source: Survey 1. Figure 7.3 shows the types of problems reported by the 46 per cent of adult 7.36 respondents who said that they were not satisfied with the space available to their families in their current/last temporary accommodation. As can be seen, lack of privacy for adults was the biggest concern reported by this group (64 per cent), followed by a lack of privacy for children (51 per cent), and a lack of space for children to play (50 per cent).Figure 7.3: Problems reported with lack of space (current/last temporary accommodation)0% 10% 20% 30% 40% 50% 60% 70%Nowhere suitablefor homeworkCauses argumentsClaustrophobia or feelingsuffocatedInsufficient room forsmall itemsInsufficient room forlarge itemsInsufficient space for playLack of privacy for childrenLack of privacy for adults 64%51%50%39%34%33%33%17%Source: Survey 1 Base: 742 respondents who reported a ‘lack of space’ in their current or last temporary accommodation. Multiple responses were possible154 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAround four in ten (39 per cent) of those who were dissatisfied with the 7.37 space in their temporary accommodation said that there was insufficient room for large items (defined in the survey through examples including sofas and wardrobes), and one third (34 per cent) reported insufficient storage for small items (defined in the survey as including children’s toys and the families’ clothes). Likewise, a third (33 per cent) reported feelings of claustrophobia, and a third (33 per cent) said that the lack of space was ‘causing arguments’. A lack of suitable space for homework was not frequently identified (17 per cent), but this may be explained by many of the children in the families being too young to be set homework (see para 2.18). Adult respondents in London were more likely than those elsewhere to report 7.38 that they had sufficient living space in their temporary accommodation (64 per cent did so). This reflected the higher use of self-contained temporary accommodation in the capital (see Table 7.2).Satisfaction with facilities in temporary accommodationWe asked adult respondents how satisfied they were with the living room, 7.39 cooking, laundry and bathroom facilities in their current/last temporary accommodation. As can be seen in Table 7.6, satisfaction levels were generally high, though 7.40 self-contained temporary accommodation was consistently rated most favourably, and hostels and B&B hotels least favourably, with respect to these facilities.Chapter 7 Conditions in temporary accommodation 155Table 7.6: Satisfaction with facilities (current or last occupied temporary accommodation)Cooking facilities Satisfied Neutral Dissatisfied Total BaseSelf-contained 95% 1% 4% 100% 1,073Hostel or B&B hotel 77% 6% 17% 100% 257Friends or relatives 85% 8% 8% 100% 348All 89% 3% 8% 100% 1,678Laundry facilities Satisfied Neutral Dissatisfied Total BaseSelf-contained 91% 1% 7% 100% 1,073Hostel or B&B hotel 65% 7% 28% 100% 257Friends or relatives 84% 7% 9% 100% 348All 84% 4% 12% 100% 1,678Bathroom facilities Satisfied Neutral Dissatisfied Total BaseSelf-contained 92% 1% 7% 100% 1,073Hostel or B&B hotel 72% 8% 20% 100% 257Friends or relatives 77% 9% 14% 100% 348All 85% 4% 11% 100% 1,678Living room facilities Satisfied Neutral Dissatisfied Total BaseSelf-contained housing 97% 1% 3% 100% 1,073Hostel or B&B hotel 83% 6% 11% 100% 257Friends or relatives 83% 8% 9% 100% 348All 91% 3% 5% 100% 1,678Source: Survey 1Satisfaction with cooking facilities in current/last temporary accommodation 7.41 was strongly associated with the availability of a kitchen. One third (36 per cent) of families without access to a kitchen (see Table 7.4) reported that they were dissatisfied with the cooking facilities, compared to only 6 per cent of those with access to a kitchen. A similar pattern existed in relation to laundry facilities, where 36 per cent of families without access to a kitchen expressed dissatisfaction, compared to 11 per cent of those with a kitchen in their current/last temporary accommodation. However, there was found to be no association between access to a dedicated living room and satisfaction with living room facilities. While the number of families lacking a bathroom altogether was too small 7.42 to test for statistical associations, analysis did reveal that sharing a bathroom with other households led to a lower satisfaction rate with these facilities. Thus, among those who shared bathrooms in their current/last temporary 156 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsaccommodation (see Table 7.4), 23 per cent were dissatisfied with these facilities, compared to 8 per cent of those who did not share their bathroom. Similarly, sharing of living rooms tended to produce a somewhat lower satisfaction rate with those facilities. Among those who shared living rooms (again see Table 7.4), 15 per cent were dissatisfied, compared to 4 per cent of those who did not share.Access to household items and amenities in temporary accommodationTable 7.77.43 shows the specific household items and amenities which families had access to in their current/last temporary accommodation. Adult respondents were presented with a predefined list of items, and this list referred to household items and amenities available within their accommodation, not to facilities nearby (such as a park or laundrette). Access to some items was very similar across all types of temporary 7.44 accommodation. Thus, regardless of whether they were reporting on self-contained temporary accommodation, hostels and B&B hotels, or arrangements with friends or relatives, almost all adult respondents said that they had access to a cooker and fridge, a bath and a television (Table 7.7). However, some household items and amenities widely available in friends’ 7.45 and relatives’ houses, such as washing machines (94 per cent) and tumble dryers (85 per cent), were less common in self-contained temporary accommodation (84 per cent and 61 per cent respectively), or in hostels and B&B hotels (74 per cent and 67 per cent). The same was also true for showers, found in 72 per cent of arrangements with friends and relatives, but only 51 per cent of self-contained temporary accommodation and 55 per cent of hostels and B&B hotels. Adult respondents reporting on arrangements with friends and relatives were likewise more likely to have access to a BT landline (or equivalent fixed telephone line) (78 per cent), than those reporting on self-contained temporary accommodation (57 per cent) or hostels and B&B hotels (52 per cent).Perhaps most importantly, a garden or other suitable outside play area for 7.46 children was much more commonly found in arrangements with friends or relatives (72 per cent), than in hostels or B&B hotels (47 per cent), or in self-contained temporary accommodation (34 per cent). Indeed, of all the household items and amenities asked about, this was the one most likely to be lacking in self-contained temporary accommodation244. This result is related in part to the concentration of self-contained temporary 244 Suitable outside play areas for children are now a policy priority in housing development, see Communities and Local Government (2007) Homes for the Future: More affordable, more sustainable, London: Communities and Local Government (para 6.2). Chapter 7 Conditions in temporary accommodation 157accommodation in London, where access to gardens is generally lower than elsewhere in the country, however the proportion with access to a garden in self-contained temporary accommodation is low even for London245. Table 7.7: Access to household items and amenities (percentage of current or last occupied temporary accommodation that had facilities) Self-contained Friends or relativesHostels and B&B hotelsAllFridge 99% 97% 94% 96%Cooker 98% 96% 95% 96%TV 97% 97% 84% 94%Bath 97% 94% 90% 94%Freezer 91% 96% 77% 88%Washing machine 84% 94% 74% 83%Food storage 80% 84% 72% 78%Dinner table 71% 79% 69% 71%Tumble dryer 61% 85% 67% 66%BT landline 57% 78% 52% 60%Shower 51% 72% 55% 56%Garden or suitable play area 34% 72% 47% 44%Computer 39% 45% 19% 36%Base 1,073 257 348 1,678Source: Survey 1Computers were available to only a minority of families in all forms of 7.47 temporary accommodation, dropping to one in five (19 per cent) of those reporting on hostels or B&B hotels. There were no regional distinctions in the household items and amenities 7.48 available to families, other than those attributable to geographical patterns in the types of temporary accommodation used. To summarise, families reporting on arrangements with friends or relatives 7.49 tended to have access to the widest set of household items and amenities, and those reporting on hostels and B&B hotels to the narrowest. 245 The Survey of English Housing (SEH) 2005/6 indicated that, across all tenures, having a garden was less likely in London than outside the capital. In London, 49 per cent of households in the social rented sector, and 54 per cent of those in the private rented sector, had a garden (DCLG Survey of English Housing (SEH) 2005/06). 158 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsPhysical conditions within temporary accommodationSpecific problems with physical conditions in temporary accommodation Table 7.8 summarises the physical conditions problems reported in current/last 7.50 temporary accommodation. Adult respondents were prompted with a pre-defined list of potential problems. As can be seen, the most commonly reported problem related to accessibility 7.51 (44 per cent of all temporary accommodation was said to be difficult to access with a pram or buggy), although this was less of an issue in hostels and B&B hotels than in other forms of temporary accommodation246. Other relatively common problems included temporary accommodation being dirty when families first arrived (31 per cent of all temporary accommodation), not being in reasonable decorative order when families first arrived (30 per cent), and a lack of control over heating (29 per cent). Dampness was also quite a widespread problem (identified in 29 per cent 7.52 of all temporary accommodation), but this finding must be seen in the context of the high rates of dampness found in rented housing in England generally247. Likewise, while 24 per cent of all adult respondents said that their temporary accommodation was not in a reasonable state of repair when they first arrived, 18 per cent of all heads of household in rented accommodation in England have rated its current state of repair as fairly or very poor248. In most respects self-contained temporary accommodation was described as 7.53 having the worst physical conditions (this was especially notable on damp, décor and repair), and the only dimension on which it was rated most highly was control over heating. Across the majority of measures, arrangements with friends and relatives were perceived to offer by far the best physical conditions. 246 Family-sized housing that is accessible for baby buggies is now a policy priority in housing development, see Communities and Local Government (2007) Homes for the Future: More affordable, more sustainable, London: Communities and Local Government. (para 6.2).247 Dampness was found in 38 per cent of local authority stock, 33 per cent of housing association stock, and 42 per cent of private rented stock (DCLG, 2005, English House Condition Survey).248 This includes 20 per cent of heads of household in local authority housing, 14 per cent of those in housing association accommodation, and 15 per cent of those in the private rented sector (DCLG, 2005, English House Condition Survey.)Chapter 7 Conditions in temporary accommodation 159Table 7.8: Problems with conditions in temporary accommodation, by temporary accommodation type (current or last occupied temporary accommodation) Physical condition problemSelf-contained Friends or relativesHostel orB&B hotelAllDifficult to access with pram or buggy 45% 48% 31% 44%Dirty when arrived 39% 7% 30% 31%Not reasonably decorated when arrived38% 7% 26% 30%Damp 40% 11% 18% 29%Lack of control over heating18% 40% 46% 29%Not in reasonable repair when arrived32% 9% 19% 24%Infestation* 19% 5% 15% 15%Conditions pose a risk to children’s safety**20% 3% 11% 15%Base 1,073 257 348 1,678Source: Survey 1 *Mice, rats, fleas, bedbugs or cockroaches. **Adult respondent answered yes when asked whether or not “this place is in such a bad condition that you worry about your child/children’s safety”.Table 7.9 shows these data by region. As can be seen, problems with physical 7.54 conditions tended to be reported at a higher rate in London and in the South, than in the North and Midlands. This was partly, but not entirely, accounted for by self-contained temporary accommodation being more commonly used in the first two of these broad regions, and arrangements with friends and relatives being more prevalent in the North and Midlands (see para 7.59). One particular point to note is that infestation was more frequently reported 7.55 in temporary accommodation in London than elsewhere. Again, this finding was partly but not entirely accounted for by the heavy use of self-contained temporary accommodation in the capital. While no directly comparable national data is available, English House Conditions Survey (EHCS) data on mouse infestations indicates that these are more prevalent in London than in other regions249. 249 DEFRA (2005) Rodent Infestations in Domestic Properties in England, 2001: A Report arising from the 2001 EHCS. 160 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 7.9: Problems with conditions in temporary accommodation, by broad region (current or last occupied temporary accommodation)Physical condition problemLondon South North and MidlandsAllDifficult to access with pram or buggy45% 53% 37% 44%Dirty when arrived 35% 41% 20% 31%Not reasonably decorated when arrived34% 38% 20% 30%Damp 38% 34% 17% 29%Lack of control over heating20% 34% 31% 29%Not in reasonable repair when arrived30% 27% 17% 24%Infestation 27% 13% 7% 15%Conditions pose a risk to children’s safety17% 19% 9% 15%Base 601 623 454 1,678Source: Survey 1Distribution of multiple physical problems within temporary accommodation Table 7.10 shows the distribution of multiple problems with the physical 7.56 condition of temporary accommodation, as reported by adult respondents. One fifth of temporary accommodation (21 per cent) was reported not to have any of the physical conditions problems portrayed in Table 7.8; 44 per cent was described as having one or two of these problems; and 35 per cent as having three or more of the relevant problems. At 45 per cent, self-contained temporary accommodation was the form of 7.57 provision most often described as having three or more physical conditions problems. One third (33 per cent) of hostels and B&B hotels were said to have at least three physical problems, but this was true of only 11 per cent of arrangements with friends or relatives. Chapter 7 Conditions in temporary accommodation 161Table 7.10: Frequency with which problems with physical conditions tended to be reported in temporary accommodation (current or last occupied temporary accommodation)Type of TA No problems reported1 or 2 problems reported3 or more problems reportedTotal BaseSelf-contained 17% 38% 45% 100% 1,073Friends or relatives 33% 57% 11% 100% 257Hostel or B&B hotel19% 49% 33% 100%348All 21% 44% 35% 100% 1,678Source: Survey 1 There was also an association between broad region and the prevalence 7.58 of multiple physical problems in temporary accommodation. Temporary accommodation in London was most likely to be described as having three or more physical problems (46 per cent), with the South close behind (43 per cent), and a far lower figure (22 per cent) reporting this number of problems in the North and Midlands (Table 7.11).Table 7.11: Frequency with which problems with physical conditions were reported in temporary accommodation, by broad region (current or last occupied temporary accommodation)Broad region No problems reported1 or 2 problems reported3 or more problems reportedTotal BaseLondon 15% 39% 46% 100% 601South 14% 43% 43% 100% 623North and Midlands30% 49% 22% 100%454All 21% 44% 35% 100% 1,678Source: Survey 1 We investigated the combined effect of broad region and temporary 7.59 accommodation type on the prevalence of multiple physical conditions problems in temporary accommodation. We found that, while in the North and Midlands, and in the South, other forms of temporary accommodation were reported to have fewer physical problems than self-contained temporary accommodation, this was not the case in London. Within London, other forms of temporary accommodation were approximately as likely as self-162 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldscontained temporary accommodation to be described by adult respondents as having multiple problems with physical conditions (Table 7.12). Table 7.12: Frequency with which three or more physical problems with temporary accommodation were reported, by temporary accommodation type and broad region (current or last occupied temporary accommodation)Broad region 3 or more problems reported in self-contained TA Base 3 or more problems reported in other TABaseLondon 46% 498 44% 110South 47% 413 35% 230North and Midlands 40% 162 12% 302All 45% 1,073 20% 642Source: Survey 1 Regression analysis confirmed that, other things being equal7.60 250, the independent influences on likelihood of reporting three or more problems with physical conditions were region (less likely in the North and Midlands), and temporary accommodation type (most likely in self-contained temporary accommodation). Safety in temporary accommodationThere were two key aspects of safety. The first was how safe families felt 7.61 inside their temporary accommodation, and the second was whether they felt safe in the area in which the temporary accommodation was located. Safety inside temporary accommodationMost (79 per cent) of adult respondents reported that their families felt safe 7.62 inside their temporary accommodation. However, as is shown in Table 7.13, there was some variation between temporary accommodation types. Families were likeliest to feel safe when living in temporary arrangements with friends or relatives (only 7 per cent did not). On the other hand, a quarter (25 per cent) of both those reporting on self-contained temporary accommodation, and on hostels and B&B hotels, said that their families did not feel safe when inside this accommodation. 250 Factors controlled for in this regression analysis included demographic and geographical variables, and temporary accommodation type. Chapter 7 Conditions in temporary accommodation 163Table 7.13: Safety within temporary accommodation (current or last occupied temporary accommodation)Accommodation typeFelt safe inside temporary accommodation Did not feel safe inside temporary accommodation Total BaseSelf-contained 75% 25% 100% 1,073Friends or relatives 93% 7% 100% 257Hostel and B&B hotels75% 25% 100% 348All 79% 21% 100% 1,678Source: Survey 1Regression analysis indicated that, when a range of variables were held 7.63 constant251, the only factor which had an independent influence on perceptions of safety inside temporary accommodation was whether or not the adult respondent perceived the surrounding area to be safe. Thus the high proportion of adult respondents who reported feeling safe within arrangements with friends or relatives was explained by this form of temporary accommodation being likely to be viewed as located within a safe area (see below, para 7.69). Neither demographic factors (such as ethnicity), nor region or other geographical variables, had an independent influence on feelings of safety inside temporary accommodation. Likewise, violent relationship breakdown as reason for applying as homeless exerted no independent effect252. One might anticipate that sharing rooms with members of other households 7.64 in a hostel or B&B hotel would be associated with feeling unsafe inside this temporary accommodation, but this was not found to be the case. Only 6 per cent of those reporting on hostels and B&B hotels said that they felt unsafe because of the bad behaviour of other households in their accommodation253.251 This regression analysis controlled for demographic variables; geographical variables; type of temporary accommodation; whether the lounge/living room, kitchen or bathroom were shared with other households; whether the area in which temporary accommodation was situated was reported as unsafe; and violent relationship breakdown as a reason for applying as homeless. 252 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.253 Though a variety of other problems with sharing rooms with other households were reported, see Figure 7.2.164 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsSafety in the area in which temporary accommodation was locatedOne quarter (25 per cent) of adult respondents reported that they or their 7.65 family felt unsafe in the area in which their temporary accommodation was located and, as indicated above, this was strongly associated with their feeling unsafe when inside their temporary accommodation too (see 7.63)254. The reasons why some families reported feeling unsafe in the area where their 7.66 temporary accommodation was situated are portrayed in Figure 7.4. Figure 7.4 : Reasons why families felt surrounding area was unsafe (current or last temporary accommodation)41%40%39%30%9%8%7%3%0% 10% 20% 30% 40% 50%Threat from otherfamily membersRacists in areaThreat from former partnerLocal children bullyrespondent's childDrugs problem in areaAnti-social behaviour fromneighboursCrime problem in areaAnti-social behaviour fromlocal childrenSource: Survey 1 Base: 331 adult respondents who perceived that their current/most recent temporary accommodation was in an ‘unsafe’ area. Multiple responses were possible.Anti-social behaviour and crime were the main reasons why adult respondents 7.67 reported feeling that the surrounding area was unsafe. Adult respondents were equally likely to report anti-social behaviour from children (41 per cent) as from adults (39 per cent). “Drug problems in area” was reported by around one third (30 per cent) of those who reported that they felt it was unsafe. A small number of adult respondents reported a threat from a former partner 7.68 (8 per cent of those who felt unsafe in the area; 2 per cent of all adult respondents). Racists in area was reported by 7 per cent of adult respondents who felt unsafe (1 per cent of all adult respondents).254 SEH data is not directly comparable, but it does indicate that social renters are more likely than other heads of household in England to report crime as a problem in their area: 20 per cent of social renters said it was a serious problem; and 33 per cent a problem but not serious. Amongst private renters, 10 per cent said it was a serious problem, and 32 per cent a problem but not serious. SEH data also indicates that Londoners, lone parents, and people living in deprived areas are all more likely to report crime as a problem in their area. These are national patterns that could impact on this aspect of our findings (DCLG (2006) Housing in England 2004/5. A Report Principally from the 2004/5 Survey of English Housing). Chapter 7 Conditions in temporary accommodation 165As noted above, there was an association between type of temporary 7.69 accommodation and feeling safe in the surrounding area, with 89 per cent of those reporting on arrangements with friends and relatives feeling that this accommodation was in a safe area, compared to 70 per cent in hostels and B&B hotels, and 71 per cent in self-contained temporary accommodation. Regression analysis confirmed that, other things being equal7.70 255, the only independent influence on perceptions of area safety was temporary accommodation type (more likely to feel safe in the area if staying with friends or relatives). There was no independent effect exerted by demographic variables (including ethnicity) or by geographical variables. Overall satisfaction with temporary accommodationRespondents were asked to rank overall satisfaction with their current/last 7.71 temporary accommodation using a score of between 1 and 10 (where ten was ‘excellent’). In overall terms, average satisfaction scores were close to six out of ten 7.72 (an average of 5.8)256. The average scores for self-contained temporary accommodation (5.8) and for hostels and B&B hotels (5.5) were slightly lower than those for arrangements with friends or relatives (6.2) (the median in all instances was 6). For the purposes of the regression analysis, adult respondents were divided 7.73 into those who were less satisfied (a score of 5 or less), and those who were more satisfied (a score or 6 and above), and they formed two roughly equal groups (48 per cent gave a score of 5 or less, 52 per cent a score of 6 or more). There was no difference between broad regions or temporary accommodation 7.74 types on this measure. However, regression analysis showed that specific attributes of temporary accommodation did exert an independent influence on relative satisfaction when other variables were held constant257. Adult respondents who reported feeling unsafe were far less likely than other adult respondents to report satisfaction levels of 6 or more out of 10. Adult respondents who reported problems due to sharing with other households, those who said that their families lacked sufficient living space, and those who reported three or more problems with the physical conditions in their 255 This regression analysis controlled for demographic characteristics; geographical variables; type of temporary accommodation; and violent relationship breakdown as a reason for applying as homeless. 256 However, there was a standard deviation of 2.64, indicating that substantial variation from the average was occurring.257 Factors controlled for in this regression analysis included: demographic variables; geographical variables; type of temporary accommodation; and temporary accommodation conditions. 166 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldstemporary accommodation, were also less likely to report satisfaction levels of 6 or more out of 10. These relationships are illustrated in Table 7.14. Table 7.14: Factors influencing satisfaction with temporary accommodation (current or last occupied temporary accommodation)Factor Satisfaction score five or less out of 10Satisfaction score six or more out of 10Total BaseFeeling unsafe in temporary accommodation 80% 20% 100% 395Problems sharing with other households68% 32% 100% 196Three or more physical problems with temporary accommodation 66% 34% 100% 673Not enough living space 61% 39% 100% 763All 48% 52% 100% 1,678Source: Survey 1Families in temporary accommodation for over one year As noted in Chapter 1 (para 1.20), Survey 4 was required because, by 7.75 definition, Survey 1 did not include those with prolonged stays in temporary accommodation. As all the Survey 4 families were in self-contained temporary accommodation at point of interview (see Chapter 6), we confine the comparisons here to Survey 1 families reporting on self-contained temporary accommodation. Survey 4 families were less likely than Survey 1 families (reporting on self-7.76 contained temporary accommodation) to be satisfied with their living space (58 per cent as compared to 69 per cent). This difference was explained in part by the higher number of larger households among the Survey 4 families (51 per cent of Survey 4 families, compared to 30 per cent of Survey 1 families, had four or more members). Adult respondents in larger households in both Survey 1 and Survey 4 were less likely to report satisfaction with their living space than other households258.258 In Survey 1, only 63 per cent of adult respondents in households with four or more members reported satisfaction with their living space, compared to 82 per cent of adult respondents in smaller households. In Survey 4, 53 per cent of these households reported satisfaction with their living space, compared to 65 per cent in smaller households.Chapter 7 Conditions in temporary accommodation 167Almost all (93 per cent) of Survey 4 families were in temporary 7.77 accommodation that met the Bedroom Standard. This is very similar to the 89 per cent compliance found in Survey 1 self-contained temporary accommodation. In total, 54 per cent of all Survey 4 families shared bedrooms within the 7.78 immediate family (close to the 58 per cent for Survey 1 families reporting on self-contained temporary accommodation). This bedroom sharing was considered problematic more frequently by Survey 4 adult respondents (by 56 per cent of those sharing bedrooms) than by Survey 1 adult respondents reporting on self-contained temporary accommodation (42 per cent of those sharing bedrooms). As with satisfaction with living space, this finding appeared to be related to the greater representation of larger households in Survey 4 than Survey 1259. All Survey 4 families had exclusive access to a living room, kitchen and 7.79 bathroom in their temporary accommodation. Table 7.15 portrays levels of satisfaction with facilities amongst Survey 4 7.80 adult respondents in their temporary accommodation. It is clear that general levels of satisfaction were much lower amongst Survey 4 than Survey 1 adult respondents (reporting on self-contained temporary accommodation), and that this was particularly the case with respect to bathroom, cooking and sleeping arrangements. 259 In Survey 1, very large households, defined as having five or more members, were quite rare (12 per cent). However, in Survey 4, they were much more common (37 per cent). These very large households were more likely to report problematic bedroom sharing across both surveys. In Survey 1, 70 per cent of very large households who shared bedrooms in self-contained temporary accommodation said that this was problematic, as compared to 49 per cent of smaller households. In Survey 4, 62 per cent of very large households sharing bedrooms reported that this was problematic, as compared to 52 per cent of other households.168 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 7.15: Satisfaction with facilities (Survey 4 adult respondents and Survey 1 adult respondents reporting on self-contained temporary accommodation)Cooking facilities Satisfied Neutral Dissatisfied Total BaseSurvey 1 95% 1% 4% 100% 1,073Survey 4 69% 8% 23% 100% 571Difference -26% +7% +19% – –Laundry facilities Satisfied Neutral Dissatisfied Total BaseSurvey 1 91% 1% 7% 100% 1,073Survey 4 72% 9% 19% 100% 571Difference -19% +8% +12% – –Bathroom facilities Satisfied Neutral Dissatisfied Total BaseSurvey 1 92% 1% 7% 100% 1,073Survey 4 56% 9% 35% 100% 571Difference -40% +8% +28% – –Living room facilities Satisfied Neutral Dissatisfied Total BaseSurvey 1 97% 1% 3% 100% 1,073Survey 4 84% 4% 12% 100% 571Difference -13% +3% +9% – –Sleeping arrangements Satisfied Neutral Dissatisfied Total BaseSurvey 1 94% 1% 5% 100% 1,073Survey 4 58% 6% 35% 100% 571Difference -36% +5% +30% – –Source: Survey 4 and Survey 1 (adult respondents reporting on self-contained temporary accommodation) Again, a key factor that appeared to influence the views of Survey 4 adult 7.81 respondents was household size. Those in the largest households (with five or more members) were more likely to express dissatisfaction with one or more of bathroom, cooking and/or sleeping arrangements than adult respondents in smaller households (67 per cent and 51 per cent respectively). The household items and amenities available to Survey 4 families mirrored 7.82 those among Survey 1 families reporting on self-contained temporary accommodation almost exactly. There were only two differences. Families in Survey 4 were more likely to have access to a landline telephone (74 per cent compared to 57 per cent of Survey 1 families whose current/last temporary accommodation was self-contained housing), alongside access to a computer (51 per cent of Survey 4 families, compared to 39 per cent of Survey 1 Chapter 7 Conditions in temporary accommodation 169families whose current/last temporary accommodation was self-contained housing).Survey 4 families were as likely to feel safe inside their temporary 7.83 accommodation as Survey 1 families reporting on self-contained temporary accommodation (78 per cent and 75 per cent respectively thought it safe). Likewise, there was effectively no difference between Survey 4 and Survey 1 adult respondents with respect to whether they felt unsafe in the area in which their self-contained temporary accommodation was located (29 per cent and 27 per cent respectively thought it unsafe). As was the case with Survey 1, regression analysis indicated that, other things being equal260, the main independent effect on feeling safe inside temporary accommodation was feeling safe in the surrounding area. However, in the case of Survey 4, a (weaker) independent effect of household type was also detected (women lone parents were more likely to feel unsafe). As Table 7.16 demonstrates, Survey 4 adult respondents described very 7.84 similar physical conditions in their temporary accommodation as Survey 1 adult respondents reporting on self-contained temporary accommodation. The only substantial difference related to infestation, with Survey 4 adult respondents more likely to identify this as a problem. This effect was found to be caused by the very strong representation of families accepted in London in Survey 4 (82 per cent of the total, see para 2.66 above), as infestation was more commonly reported in the capital than elsewhere in both Survey 1 and Survey 4 261. 260 The factors controlled for in this regression analysis were: demographic characteristics; geographical variables; whether the area in which temporary accommodation was situated was reported as unsafe; and violent relationship breakdown as a cause of homelessness.261 One quarter (26 per cent) of Survey 1 adult respondents accepted in London reporting on self-contained temporary accommodation said that it was infested; as compared to only 14 per cent of those reporting on self-contained temporary accommodation elsewhere. In Survey 4, 33 per cent of families accepted in London reported infestations and, although the Survey 4 adult respondents accepted outside of London were a small minority, they were much less likely to report infestations in their temporary accommodation (the margin of error is too great to state the actual percentage). As reported in para 7.55 above, limited national data from the 2001 EHCS is consistent with this pattern, indicating that that mouse infestations are more common in London than in other regions in England. 170 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 7.16: Physical problems with temporary accommodation (Survey 4 adult respondents and Survey 1 adult respondents reporting on self-contained temporary accommodation )Accommodation type Survey 4 Survey 1 DifferenceNot easy to access with pram or buggy 45% 45% 0%Dirty when arrived 34% 39% -5%Not reasonably decorated when arrived39% 38% +1%Damp 43% 40% +3%Lack of control over heating 18% 18% 0%Not in reasonable repair when arrived33% 32% +1%Infestation 29% 19% + 10%Conditions pose a risk to children’s safety18% 20% -2%Base 571 1,073 –Source: Survey 4 and Survey 1 (adult respondents reporting on self-contained temporary accommodation)Survey 4 respondents were also asked to rank overall satisfaction with 7.85 their current temporary accommodation using a score of between one and ten (again where ten was ‘excellent’). On average, Survey 4 respondents ranked their current temporary accommodation with a satisfaction score of 5.8 (median 6)262. These results were identical to those for Survey 1 adult respondents reporting on self-contained temporary accommodation. As in Survey 1, when adult respondents were divided into those who were 7.86 less satisfied (a score of 5 or less), and those who were more satisfied (a score or 6 or above), this again formed two roughly equal groups (47 per cent gave a score of five or less, and 53 per cent a score of 6 or more). Comparison of satisfaction levels between regions and between temporary accommodation types was not possible for Survey 4 (most were in London and all were in self-contained temporary accommodation)263. However, three of the same specific attributes that influenced levels of satisfaction with temporary accommodation amongst Survey 1 adult respondents were also associated with satisfaction levels amongst Survey 4 adult respondents. For both sets of respondents, feeling unsafe within temporary accommodation was the most important factor associated with dissatisfaction, though insufficient living 262 Though the standard deviation of 2.4 showed there was some variation. 263 In any case, these factors did not affect satisfaction in Survey 1, where they could be investigated.Chapter 7 Conditions in temporary accommodation 171space and multiple problems with physical conditions, were also important factors (see Table 7.17)264. Table 7.17: Factors associated with satisfaction with temporary accommodation (Survey 4 adult respondents and Survey 1 respondents reporting on self-contained temporary accommodation)Factor Satisfaction score five or less out of 10Satisfaction score six or more out of 10Total BaseThree or more problems with physical conditions in temporary accommodation 60% [67%] 40% [33%] 100% [100%]205 [415]Not feeling safe within temporary accommodation 79% [80%] 21% [20%] 100% [100%]129 [260]Insufficient living space61% [66%] 39% [34%] 100% [100%]236 [330]All47% [48%] 53% [52%] 100% [100%]571 [1,073]Source: Survey 4 and Survey 1 (adult respondents reporting on self-contained temporary accommodation) Figures for Survey 1 are shown in italicsConclusions This chapter has reviewed conditions in temporary accommodation for 7.87 families accepted as homeless, as reported by adult respondents. Overall satisfaction levels varied little between different types of temporary accommodation, but there was often a relationship between specific attributes of temporary accommodation and its type.The most common form of temporary accommodation – self-contained 7.88 housing – was perceived to afford families the best space standards, and satisfaction with cooking, sleeping, bathroom and other facilities was also highest in this form of temporary provision. However, the worst physical conditions were often reported in self-contained temporary accommodation, particularly with respect to damp, décor and state of repair265.264 The other factor that exerted an independent effect on Survey 1 adult respondents’ satisfaction with temporary accommodation – problems with sharing rooms with other households – did not affect any Survey 4 families as all were in self-contained temporary accommodation.265 These mixed findings on conditions within self-contained temporary accommodation should be viewed alongside the evidence presented in Chapter 13 which indicate a consistent pattern whereby this form of temporary provision is associated with a better quality of life for both adults and children than other types of temporary accommodation.172 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTemporary arrangements with friends and relatives offered families the 7.89 best physical conditions and access to the widest range of household items and amenities. Families also felt safest when in this form of temporary accommodation. However, concerns about space standards were at their most acute in these arrangements. Access to household items and amenities (including kitchens and dedicated 7.90 living rooms) was often more restricted in hostels and B&B hotels than in other forms of temporary provision, but the worst physical conditions and space standards were not generally reported in these forms of temporary accommodation.Survey 4 adult respondents were far less satisfied with cooking, bathroom 7.91 and sleeping arrangements in their temporary accommodation than were Survey 1 respondents reporting on self-contained temporary accommodation. This was explained in part by the stronger representation of larger households amongst Survey 4 families, who tended in both surveys to report lower levels of satisfaction with these arrangements, and also with their overall living space. The next chapter moves on to consider conditions in the settled housing 7.92 allocated to families accepted as homeless. Chapter 8 Conditions in settled housing 173Chapter 8:Conditions in settled housing IntroductionThe overall objective of this study was to understand the causes, experiences 8.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data was collected in five separate surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys).The nature and quality of the settled housing provided to families accepted 8.2 as being owed the main homelessness duty is important in the context of this report because a principal aim of the statutory homelessness framework is to meet families’ housing needs in a sustainable way266. One might anticipate that such settled housing should be of better quality and/or more suitable for their needs than the temporary accommodation provided to these families, but that has never been the subject of comparative analysis. Previous research has indicated that families accepted as homeless may accept inferior offers of social housing as compared with other housing applicants267, but this has largely been based on indirect evidence (particularly the limited number of offers of settled housing received by these families268), rather than on direct evidence regarding the relative quality of the housing which they obtain. This chapter therefore draws on data from Survey 1 – a survey of families 8.3 accepted as being owed the main homeless duty between 1st January 2005 and 30 June 2005 – to consider the physical conditions and other features of settled housing provided to families accepted as homeless. After summarising the key characteristics of families in settled housing by point of survey, topics covered include:• tenure and choice in settled housing; • living space;266 Bramley, G., Fitzpatrick, S., Karley, N.K., Monk, S. and Pleace, N. (2005) Evaluation of English Housing Policy 1975-2000. Theme 1: Supply, Need and Access, London: ODPM.267 Fitzpatrick, S. and Stephens, M. (1999) ‘Homelessness, need and desert in the allocation of council housing’, Housing Studies, 14 (4) 413-431.268 The proportion of local authorities allowing only a single ‘reasonable’ offer of accommodation to households accepted as homeless and owed the main homelessness duty rose from 25 per cent in 1991 to 75 per cent by 2000 (Pawson, H., Levinson, D., Lawton, G., Parker, J. and Third, H. (2001) Local Authority Policy and Practice on Allocations, Transfers and Homelessness, London: DTLR.)174 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds• satisfaction with facilities;• access to households items and amenities;• physical conditions;• sense of safety; and • overall satisfaction.Much of the discussion below involves comparisons between conditions in 8.4 settled housing and those in temporary accommodation269. As the settled housing provided to families was always self-contained (see para 8.11 below), these comparisons are confined to self-contained temporary accommodation, in order to compare ‘like with like’270. We also provide some comparisons with general conditions in the social rented sector, as this was where almost all of these families were accommodated. Please note that, as with Chapter 7, all findings on accommodation conditions are based on adult respondents’ views rather than on an independent inspection271.As elsewhere in this report, we present two types of statistical analysis in 8.5 this chapter: bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a range of factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected by regression analysis272. This survey evidence indicates that conditions in settled housing, as reported 8.6 by adult respondents, were in some respects better than those found in self-contained temporary accommodation (e.g. space standards and some aspects of physical conditions), but on some dimensions were rather worse (e.g. satisfaction with bathroom, cooking and laundry facilities). Nevertheless, overall levels of satisfaction were markedly higher for settled housing than for self-contained temporary accommodation. 269 This draws on the combined Survey 1 dataset, as presented in Chapter 7, which incorporates evidence on both conditions in current temporary accommodation (for those still in temporary accommodation) and most recent temporary accommodation (for those in settled housing). The adult respondents whose responses are being compared are not entirely separate groups as some of those who provided data on their ‘most recent temporary accommodation’ (where it was self-contained) will also have provided data on their (current) settled housing.270 Adult respondents assessed self-contained temporary accommodation more favourably than other forms of temporary accommodation in some respects (e.g. space standards) and less favourably on others (e.g. physical conditions) (see Chapter 7). Thus, it is not the case that using self-contained temporary accommodation as the comparison point systematically provided a ‘higher benchmark’ for comparison with settled housing than would have been the case using all temporary accommodation.271 This is important to bear in mind where comparisons are made to the EHCS where surveyors are responsible for assessing the state of repair of properties and local liveability factors, such as poor quality environments.272 See Chapter 1 and Appendix 2 for more detail.Chapter 8 Conditions in settled housing 175Key points • At point of survey, 55 per cent of all families accepted as homeless were in settled housing. This included 76 per cent of families accepted in the North and Midlands, 52 per cent of those accepted in the South, and only 18 per cent of those accepted in London. • Almost all (91 per cent) of these families had been provided with social rented housing. • Only 25 per cent of all families in settled housing reported being given any choice over this housing.• Families in settled housing were less likely to share bedrooms (within the immediate family) than families reporting on self-contained temporary accommodation, and where they did share bedrooms they were less likely to consider it problematic. • Overall satisfaction with living space was greater amongst adult respondents in settled housing than amongst those reporting on self-contained temporary accommodation.• Satisfaction with cooking, laundry and, especially, bathroom facilities was much lower amongst adult respondents in settled housing than amongst those reporting on self-contained temporary accommodation. • Access to specific household items and amenities was very similar across the two groups, though families in settled housing were more likely to have access to a garden/outdoor play space, and less likely to have access to a shower, than those reporting on self-contained temporary accommodation. • Several problems with physical conditions were less commonly reported in settled housing than in self-contained temporary accommodation (e.g. damp, infestation, conditions which pose a risk to children), though settled housing was more likely to be reported as in poor decorative order when families first arrived.• Overall satisfaction with settled housing was markedly higher than for self-contained temporary accommodation.The families in settled housing The key contextual data on families in settled housing has already been 8.7 provided in Chapter 6. As noted there, 55 per cent of families accepted as homeless had been provided with settled housing by the time of survey, but there was a strong regional dimension to this: only 18 per cent of families accepted in London were in settled housing by point of survey, as compared 176 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldswith 52 per cent in the South, and 76 per cent in the North and Midlands (see Chapter 6)273. In total, 63 per cent8.8 of all families in settled housing at point of survey had been accepted as homeless in the North and Midlands, 28 per cent in the South, and only 8 per cent in London (see Table 8.1)274. Table 8.1: Location of families in settled housing at point of surveyRegion Percentage BaseLondon 8% 97South 28% 316North and Midlands 63% 510Total 100% 100%Source: Survey 1In most respects, families in settled housing differed little in their 8.9 characteristics from those still in temporary accommodation. In particular, larger family groups (with four or more members), who might be expected to be less likely to be found settled housing quickly, were not found to be any more or less likely to be in settled housing at point of survey (see para 2.68 on the increased likelihood of being in temporary accommodation for over one year). Regression analysis confirmed that, other things being equal8.10 275, the only independent influence on whether or not a household had entered settled housing by point of survey was broad region (being accepted in the North and Midlands made it more likely, being accepted in London less likely). Demographic characteristics, including ethnicity and being a former asylum seeker, had no independent effect once this was taken into account. Tenure and choice in settled housing As noted above, the settled housing that families had been provided with 8.11 was always self-contained. In the great majority of cases this was described as social rented housing (91 per cent), with a further 8 per cent of families 273 These regional disparities were not due to differences in the time elapsed between acceptance and interview: in the North and Midlands the average time between acceptance and survey for families was 8.9 months, while in London it was 9 months, and in the South 8.9 months (the median in all cases was 9 months).274 Given the heavy concentration of self-contained temporary accommodation in London (see Chapter 6), this chapter is, to a large extent, comparing settled housing in the North and Midlands to self-contained temporary accommodation in the capital. However, as self-contained temporary accommodation was reported as having fairly similar characteristics across all regions (see Chapter 7) this is not a major limitation for most purposes of the analysis. Where it is a concern, this is highlighted.275 The factors controlled for in this regression analysis were geographical and demographic variables.Chapter 8 Conditions in settled housing 177reporting that they had moved on to the private rented sector, and a very small number (under 1 per cent) that they were now in owner occupied housing.Families in settled housing were asked if they had been given any choice 8.12 when they were allocated their home. Only 25 per cent reported that they had had a choice over their settled accommodation276. Those adult respondents accepted in areas of less housing stress were more likely to report that they had been given a choice over their settled housing (38 per cent) than those in areas of higher housing stress (24 per cent)277. Regression analysis confirmed that, other things being equal278, housing stress within different areas was the only factor which had an independent influence on the likelihood of having been given a choice279.Living space in settled housingOne third of8.13 adult respondents in settled housing (36 per cent) reported that members of their family were sharing bedrooms within the immediate family (this compares to 49 per cent of adult respondents reporting on self-contained temporary accommodation). Of those sharing bedrooms, 25 per cent reported that this was a problem (as compared to 42 per cent of those sharing bedrooms in self-contained temporary accommodation reporting this as problematic). Therefore both the prevalence of bedroom sharing and problems with bedroom sharing were less common in settled housing than in self-contained temporary accommodation. Where bedroom sharing was reported to be problematic, the reasons given 8.14 matched those cited in relation to self-contained temporary accommodation: inappropriate for children of different genders to share; insufficient space; and, particularly, a lack of privacy (see Figure 7.1). Most settled housing met the Bedroom Standard (90 per cent)8.15 280. This was true of an effectively identical proportion of self-contained temporary accommodation (89 per cent). 276 This is broadly in line with the findings of Pawson et al (2001) (see footnote 265 above). One might have expected the emphasis on choice based lettings schemes in the period since that report was published to mean that families accepted as homeless would now report a higher level of choice, but this does not appear to be the case.277 Local authorities with relatively lower and higher levels of housing stress were identified by employing the Wilcox affordability ratio. See Appendix 2 for a full explanation.278 The factors controlled for in this regression were geographical and demographic variables (but being a former asylum seeker was not included in any of the regression models restricted to those in settled housing as the numbers in settled housing were very small.)279 This is consistent with data from the SEH which indicates that less choice is offered to new social rented tenants, and those wishing to transfer within the sector, in areas of housing pressure (DCLG (2006) Housing in England 2004/5. A Report Principally from the 2004/5 Survey of English Housing).280 See footnote 235 for a definition and explanation of the Bedroom Standard. Data from the SEH indicates that 94 per cent of all households in the social rented sector are in housing that meets the Bedroom Standard (DCLG (2006) Housing in England 2004/5. A Report Principally from the 2004/5 Survey of English Housing).178 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAll families in settled housing had access to kitchens, living rooms used only 8.16 for this purpose (i.e. not as a bedroom), and bathrooms. As one would expect in self-contained accommodation, this access was always exclusive (i.e. none of these rooms were shared with members of any other households.) This matched the position in self-contained temporary accommodation281.Overall space standards tended to be viewed as better in settled housing 8.17 than in self-contained temporary accommodation. Thus 86 per cent of adult respondents in settled housing reported that they had sufficient living space for their family, as compared to 69 per cent in self-contained temporary accommodation282. Among those reporting insufficient living space in their settled housing, the 8.18 main complaints were lack of privacy and lack of storage. Satisfaction with facilities in settled housingTable 8.2 summarises adult respondents’ satisfaction with a range of facilities 8.19 in settled housing, and compares these data with satisfaction levels reported by adult respondents in self-contained temporary accommodation. As can be seen, while the majority were satisfied with respect to all of these 8.20 types of facilities, satisfaction was generally lower in settled housing than in self-contained temporary accommodation. Bathroom facilities in particular were much more poorly rated in settled housing than in self-contained temporary accommodation (only 72 per cent reported satisfaction as compared with 92 per cent in self-contained temporary accommodation)283. There were smaller differences in respect of laundry and cooking facilities, but again settled housing comes out worse than self-contained temporary accommodation. Satisfaction with living rooms and sleeping arrangements was generally high in both settings.281 With the minor difference that 2 per cent of families in Survey 1 reporting on self-contained temporary accommodation did not have access to a dedicated living room, because that room was also being used as a bedroom.282 Data from the SEH indicates that 3 per cent of all social rented tenants were unhappy with the size of their property (DCLG (2006) Housing in England 2004/5. A Report Principally from the 2004/5 Survey of English Housing).283 This seems likely to be related to the lower availability of showers in settled housing than in self-contained temporary accommodation (see Table 8.3).Chapter 8 Conditions in settled housing 179Table 8.2: Satisfaction with facilities in settled housing, compared to current or last temporary accommodation that was self-contained housingCooking facilities Satisfied Neutral Dissatisfied Total BaseSettled housing 85% 3% 12% 100% 923Self-contained temporary accommodation (current or last)95% 1% 4% 100% 1,073Difference -10% +2% +8% – –Laundry facilities Satisfied Neutral Dissatisfied Total BaseSettled housing 81% 5% 13% 100% 923Self-contained temporary accommodation (current or last)91% 1% 7% 100% 1,073Difference -10% +4% +6% – –Bathroom facilities Satisfied Neutral Dissatisfied Total BaseSettled housing 72% 7% 20% 100% 923Self-contained temporary accommodation (current or last)92% 1% 7% 100% 1,073Difference -20% +6% +13% – –Living rooms Satisfied Neutral Dissatisfied Total BaseSettled housing 95% 1% 4% 100% 923Self-contained temporary accommodation (current or last)97% 1% 3% 100% 1,073Difference -2% 0% +1% – –Sleeping arrangements Satisfied Neutral Dissatisfied Total BaseSettled housing 89% 2% 9% 100% 923Self-contained temporary accommodation (current or last)94% 1% 5% 100% 1,073Difference -5% +1% +4% – –Source: Survey 1 Bases: 1,073 adult respondents for whom data were available reporting on their current or last temporary accommodation (where that temporary accommodation was self-contained housing), and 923 respondents who were living in settled housing at point of survey. 180 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAccess to household items and amenities in settled housingTable 8.3 shows the access to specific household items and amenities 8.21 reported by adult respondents in settled housing, and, again, these data are compared to the results for self-contained temporary accommodation. 284Table 8.3: Access to household items and facilities in settled housing, compared to current or last temporary accommodation that was self-contained housingFacilities Settled housing Self-contained temporary accommodation (current or last)DifferenceBath 100% 97% +3%Fridge 99% 99% 0%Cooker 99% 98% +1%TV 99% 97% +2%Freezer 95% 91% +4%Washing machine 94% 84% +10%Food storage 88% 80% +8%Tumble dryer 72% 61% +11%Garden or suitable play area 67% 34% +33%Dinner table 66% 71% -5%BT landline 59% 57% +2%Computer 36% 39% -3%Shower 31% 51% -20%Base 923 1,073 –Source: Survey 1As can be seen, there was often little difference between self-contained 8.22 temporary accommodation and settled housing with respect to the amenities that families had access to, although washing machines and tumble dryers were more common in the latter. For reasons that are not clear from the survey responses, showers were less often a feature of the bathrooms in settled housing than was the case in self-contained temporary accommodation. 284 This table is arranged in descending order of access to the relevant facilities in settled housing. As this differed slightly from rates of access in temporary accommodation, the order is slightly different from that in Table 7.7.Chapter 8 Conditions in settled housing 181Access to suitable gardens or play areas was much more common for 8.23 families in settled housing than it was for families reporting on self-contained temporary accommodation (67 per cent compared to 39 per cent). This was the most pronounced difference between settled housing and self-contained temporary accommodation. This is in part a regional effect: self-contained temporary accommodation was mainly in London, where access to gardens is generally more restricted (see also para 7.46)285. Physical conditions in settled housingTable 8.4 shows the physical conditions problems identified by adult 8.24 respondents in settled housing, and compares these data with the problems identified in self-contained temporary accommodation. Settled housing tended to be more accessible than self-contained temporary 8.25 accommodation (32 per cent of adult respondents said that it was difficult to access their settled housing with a pram or buggy, as compared with 45 per cent who reported this problem in self-contained temporary accommodation). Settled housing was also less likely than self-contained temporary accommodation to be reported as damp (27 per cent286 as compared to 40 per cent) or as suffering from an infestation (9 per cent as compared to 19 per cent). Physical conditions were not as likely to be viewed as a risk to the safety of children by adult respondents in settled housing (5 per cent) as by those reporting on self-contained temporary accommodation (20 per cent).285 However, even for London, access to gardens within self-contained temporary accommodation was low (see para 7.46). Data from the SEH indicates that 69 per cent of all social rented tenants in England have access to garden, so the proportion of those in settled housing in Survey 1 who had access to a garden is comparable to the national position (DCLG (2006) Housing in England 2004/5. A Report Principally from the 2004/5 Survey of English Housing).286 This appears to compare well with the 38 per cent of all local authority stock, and 33 per cent of housing association properties, that were found to have problems with condensation or dampness in the EHCS (DCLG, 2005 English House Condition Survey). However, it must be borne in mind that EHCS data reflects the results of a professional inspection and so is not directly comparable with reports from adult respondents.182 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 8.4: Problems with conditions in settled housing, compared to current or last occupied temporary accommodation that was self-contained housingProblem Settled housing Self-contained temporary accommodation (current or last)DifferenceDifficult to access with pram or buggy 32% 45% -13%Dirty when arrived 45% 39% +6%Not reasonably decorated when arrived66% 38% +28%Damp 27% 40% -13%Lack of control over heating 8% 18% -10%Not in reasonable repair when arrived34% 32% +2%Infestation* 9% 19% - 10%Conditions which pose a risk to children’s safety**5% 20% -15%Base 923 1,073 –Source: Survey 1 *Mice, rats, fleas, bedbugs or cockroaches. ** Adult respondent answered yes when asked whether or not “this place is in such a bad condition that you worry about your child/children’s safety”. On the other hand, settled housing was more likely to be described as in 8.26 poor decorative order when respondents first arrived than was the case for self-contained temporary accommodation (66 per cent compared to 38 per cent). There was no substantial difference between self-contained temporary accommodation and settled housing with regards to whether they were in a reasonable state of repair287, or dirty, when families first arrived. The distinct regional distribution of settled housing, mainly located in the 8.27 North and Midlands, and self-contained temporary accommodation, most of which was in London, needs to be noted as an explanation for some of these patterns. Damp and infestation, for example, was generally more common in temporary accommodation in London than elsewhere (see Table 7.9).287 The figure for settled housing is, however, somewhat higher than the proportion of all social rented stock which has been assessed as in fairly or very poor repair (20 per cent in local authority housing and 14 per cent in the housing association sector) (DCLG, 2005, English House Condition Survey).Chapter 8 Conditions in settled housing 183The frequency with which multiple problems occurred in settled housing 8.28 was found to be effectively identical to the levels reported in self-contained temporary accommodation (Table 8.5). Thus, 45 per cent of both adult respondents in settled housing and those reporting on self-contained temporary accommodation described their accommodation as having three or more of the problems represented in Table 8.4.Table 8.5: Frequency with which problems with physical conditions were reported in settled housing, compared to current or last occupied temporary accommodation that was self-contained housingType accommodation No problems reported1 or 2 problems reported3 or more problems reportedTotal BaseSettled housing 16% 39% 45% 100% 923Self-contained temporary accommodation (current or last)17% 38% 45% 100% 1,073Source: Survey 1Safety in settled housingAdult respondents generally reported that they and their immediate family 8.29 felt safe in their settled housing (87 per cent). This was higher then those who reported feeling safe within self-contained temporary accommodation (75 per cent). One fifth of adult respondents (19 per cent) said that they or their family felt 8.30 unsafe in the area within which their settled accommodation was located. This figure was similar to that for self-contained temporary accommodation (25 per cent)288. The main reasons given by the 19 per cent of respondents who felt that the 8.31 area surrounding their settled housing was unsafe were anti-social behaviour from local children and young people (52 per cent); anti-social behaviour from adults (41 per cent); crime in the area (39 per cent); and drug dealers in the area (31 per cent). As can be seen in Table 8.6, these were very similar to the reasons given by those adult respondents who thought that their self-contained temporary accommodation was located in an unsafe area. 288 SEH data is not directly comparable, but it does indicate that social renters are more likely than other heads of household in England to report crime as a problem in their area: 20 per cent said it was a serious problem; and 33 per cent a problem but not serious (DCLG (2006) Housing in England 2004/5. A Report principally from the 2004/5 Survey of English Housing).184 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 8.6: Why respondents felt unsafe in the area surrounding their settled housing, compared to current or last occupied temporary accommodation that was self-contained housingReason for feeling unsafe Settled housing Self-contained temporary accommodation (current or last)DifferenceAnti-social behaviour from children or young people 51% 49% +2%Anti-social behaviour from adults 41% 47% -6%Crime in area 40% 43% -3%Drug dealers in area 31% 33% -2%Base 169 297 –Source: Survey 1 Bases: 297 adult respondents who reported that they or their family felt unsafe in the area in which their current/last temporary accommodation was located (where that temporary accommodation was self-contained housing), and 169 adult respondents who were living in settled housing at point of survey who reported that they felt that the surrounding area was unsafe As was noted with regards to temporary accommodation (see Chapter 7, 8.32 para 7.63), the only independent effect on whether an adult respondent and their family felt safe inside their settled housing was whether they felt the area in which it was situated was safe289. Neither demographic factors (such as ethnicity), nor region or other geographical variables, had an independent influence on feelings of safety inside settled housing. Likewise, violent relationship breakdown as a reason for applying as homeless exerted no independent effect290.Overall satisfaction with settled housing Adult respondents were asked to rate their settled housing out of ten 8.33 (where ten was ‘excellent’). The average score was 7, and the median was a score of 8. These scores were higher than for self-contained temporary accommodation, which had an average of 5.8 and a median of 6.Three quarters (75 per cent) of adult respondents rated their overall 8.34 satisfaction with their settled housing as six or more out of ten. In contrast, only 53 per cent of adult respondents reporting on self-contained temporary accommodation gave this accommodation the same rating. Levels of satisfaction with settled housing were therefore markedly higher than was 289 This regression analysis controlled for: demographic variables; geographical variables; violent relationship breakdown as a reason for applying as homeless; whether area perceived to be unsafe.290 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.Chapter 8 Conditions in settled housing 185the case for self-contained temporary accommodation (and indeed for other forms of temporary accommodation). Regression analysis indicated a very similar pattern of influences on overall 8.35 satisfaction with settled housing as were identified in relation to temporary accommodation (see Table 7.14). Thus, other things being equal291, the only independent effects related to the following three attributes of settled housing: • sense of safety: among those who did not feel safe within their settled housing, 68 per cent reported a satisfaction score of five or less for their settled housing, compared to only 19 per cent of those who reported feeling safe in their housing.• physical conditions: adult respondents who reported three or more physical conditions problems were more likely (37 per cent) than those who reported fewer problems (16 per cent) to give their settled housing an overall satisfaction score of less than 5. • living space: those who expressed dissatisfaction with the level of living space in their settled housing were also more likely (44 per cent) than other adult respondents in settled housing (23 per cent) to give their homes a satisfaction score of five or less.ConclusionsThis chapter has reviewed conditions in the settled housing provided to 8.36 families accepted as homeless, as described by adult respondents in these families. It found that, in some respects, conditions were better in settled housing 8.37 than in self-contained temporary accommodation. This was true with regards to living space, access to gardens, and some aspects of physical conditions. However, satisfaction with cooking, laundry, and, especially, bathroom facilities was considerably lower in settled housing than in self-contained temporary accommodation. Despite these mixed results on accommodation conditions, overall satisfaction was markedly higher with settled housing than with self-contained temporary accommodation. This suggests that the security offered by settled (usually social) housing may be an important factor for families accepted as homeless.The next chapter moves on to consider the health and support needs of 8.38 adults within families accepted as homeless.291 This regression analysis controlled for: demographic variables; geographical variables; and accommodation conditions.186 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsChapter 9:Health and social support IntroductionThe overall objective of this study was to understand the causes, experiences 9.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data were collected in five separate surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys).A range of research studies have suggested poor health outcomes and poor 9.2 access to NHS services amongst families accepted as homeless. There is some evidence of children and parents in these families experiencing mental health problems at disproportionately high rates292. It has also been suggested that families accepted as homeless may suffer from poor continuity of health care, and restricted access to permanent GP registration, or difficulties in gaining access to GPs with whom they are registered, because of moves between temporary accommodation addresses293. A potential problem with diet, due to the difficulties associated with preparing food in B&B hotels and other managed forms of temporary accommodation with poor cooking facilities, has also been identified294. As noted in earlier chapters (see para 1.10), much of the relevant research has focussed on B&B hotels, to the neglect of other forms of temporary accommodation. Some families accepted as homeless may require low intensity or practical 9.3 support to help them set up home or to sustain independent living, such as the resettlement and tenancy sustainment services funded through the Supporting People programme295. It is also possible that families may require social services assistance with respect to parenting, or with complex issues such as mental health, drug or alcohol problems296.292 Vostanis P, Tischler V, Cumella S, Bellerby T. (2001) ‘Mental health problems and social supports among homeless mothers and children victims of domestic and community violence’ International Journal of Social Psychiatry 47, (4) 30-40.293 Quilgars, D. and Pleace, N. (2003) Delivering Health Care to Homeless People: An effectiveness review, Edinburgh : NHS Scotland http://www.healthscotland.com/documents/browse/425/1136.aspx294 Pleace, N and Quilgars, D (1996) Health and Homelessness in London: A review, London: The King’s Fund.295 ODPM (2005) Supporting People Working Paper 7: Homeless Families, London: ODPM.296 Jones, A., Pleace, N. and Quilgars, D. (2002) Firm Foundations: an Evaluation of the Shelter Homeless to Home Service, London: Shelter.Chapter 9 Health and social support 187An experience of homelessness can also mean that a family is cut off from 9.4 its existing social networks. This includes, but is not limited to, circumstances in which a woman and her children are escaping violence and have to travel some distance to escape the threat from a former partner297. A lack of informal social support from family and friends, or the loss of such support, has been linked to recurrent homelessness298 and, within the general population, to poorer general and mental health299. This chapter draws on data from Survey 1 – a survey of families accepted 9.5 as being owed the main homeless duty between 1st January 2005 and 30 June 2005 – to explore adult respondents’ health and social support needs, and any changes in these that may be attributable to the experience of homelessness and temporary accommodation300. The following topics are covered: • general health and longstanding illness/disability;• changes in diet;• mental health problems;• drug and alcohol problems; • support needs and service use; and• access to social support. The experiences and circumstances of those with prolonged stays in 9.6 temporary accommodation are of particular interest. The last section of the chapter compares the health and well-being of Survey 1 adult respondents to that of adult respondents in families accepted as owed the main homelessness duty and in temporary accommodation for more than a year (Survey 4)301.It should be noted that self-completion questions were used to gather the 9.7 most sensitive material reported in this chapter, including that on drugs, alcohol and mental health problems.297 Holder, T; Curteis. S; Griffths, S; Hunter, G. and James, K. (2002) Life on Hold: “I can’t even think about tomorrow”: The housing and support needs of families in temporary accommodation in Leeds, Leeds City Council (unpublished report); Jones, A; Pleace, N. and Quilgars, D (2002) Firm Foundations: An evaluation of the Shelter Homeless to Home service, London: Shelter.298 Jones, A; Pleace, N. and Quilgars, D (2002) Firm Foundations: An evaluation of the Shelter Homeless to Home service, London: Shelter.299 Cohen, S. and Wills, T. (1985) ‘Stress, social support and the buffering hypothesis’, Psychological Bulletin, 98: 310-357.300 The needs and experiences of children within these families are explored in Chapter 11.301 Survey 4 was required because the ‘time-window’ design for Survey 1, while delivering a representative sample of those accepted as homeless over a six month period, by definition excluded those with prolonged stays in temporary accommodation. See Chapter 1 and Appendix 1 for a full explanation.188 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAs elsewhere in this report, we present two types of statistical analysis in 9.8 this chapter: bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a range of other factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected302.This survey evidence suggests that, overall, adult respondents appeared to be 9.9 a relatively disadvantaged group with regards to health and social support, but were not, in the main, extremely vulnerable (see Chapter 3 on these adult respondents’ long-term history). Moreover, the impacts of homelessness and temporary accommodation on the health and social support circumstances of adult respondents seemed negligible, or marginally positive. Key points • Adult respondents generally had poorer self-reported general health than the general population of the same age. • For most adult respondents (66 per cent) there was no change in self-reported health status since leaving their last settled accommodation, and where it had changed, health status was more likely to have improved than deteriorated.• There was a strong association between improvements in general health status and violent relationship breakdown as a cause of homelessness303.• Diet was also reported to be more likely to have improved than deteriorated since families left their last settled accommodation, although again no change was reported for most families (65 per cent). Deteriorations in diet were associated with current financial difficulties.• One quarter (27 per cent) of adult respondents reported current anxiety, depression or other mental health problems. This was a higher rate than is found in the general population (18 per cent); a difference that is partly accounted for by the preponderance of women amongst adult respondents. Mental health problems were less commonly reported by ethnic minority than White adult respondents. • Very few adult respondents (3 per cent) self-reported current drug and/or alcohol problems.303302 See Appendix 2 for more detail on analysis.303 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.Chapter 9 Health and social support 189• One third (35 per cent) of adult respondents reported at least one unmet need for ‘practical support’ (such as with acquiring furniture or managing money), but only 4 per cent reported an unmet need for ‘personal support’ (for example, with mental health problems or parenting skills).• Families in temporary accommodation were more likely than those in settled housing to report seeing key workers or housing support workers. There was otherwise little distinction in the use of several key NHS, care and support services between those in temporary and settled housing. • GP registration was near universal (98 per cent) amongst adult respondents, and most reported that these GP services were within easy reach of where they were staying. • The overwhelming majority (87 per cent) of adult respondents had access to some level of emotional and/or instrumental forms of social support, but nonetheless this was a lower rate of access than that found in the general population. Ethnic minority adult respondents tended to have less access to instrumental support than other adult respondents. Access to these forms of support had seldom changed since adult respondents had left their last settled accommodation. • An overall net reduction in contact with friends was reported by adult respondents since they left their last settled accommodation, but there was little net change in contact with family Only very small numbers of adult respondents had no contact at all with family and/or friends at point of interview (though this was somewhat more common amongst former asylum seekers). • Survey 4 adult respondents had a similar health profile to that of Survey 1 adult respondents, but were less likely to report mental health problems (this was linked to the high proportion of ethnic minorities in Survey 4). Survey 4 adult respondents were more likely to report unmet practical support needs, but had a similarly low level of self-reported personal support needs as those in Survey 1. Survey 4 respondents had similar levels of access to emotional support than Survey 1 adult respondents, but less access to instrumental support (again this was attributable to the high proportion of ethnic minorities in the latter Survey).General healthThis section considers adult respondents’ general health status at point 9.10 of interview, and investigates whether there were any changes in this status which may be associated with homelessness and living in temporary accommodation. It also considers the prevalence of longstanding illness and disability amongst this group. 190 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsCurrent general health at point of interviewAdult respondents were asked to self-assess their general health according to 9.11 the same scale used by the Health Survey for England (HSE) 2005 (Table 9.1). They were most likely to rate their general health as “good” (40 per cent) or “very good” (29 per cent). A further 22 per cent rated it as “fair”. Only 10 per cent of respondents rated their health as “bad” (8 per cent) or “very bad” (2 per cent). There was no distinction between the self-reported general health status 9.12 for adult respondents still in temporary accommodation and those in settled housing at point of survey.However, as Table 9.1 demonstrates, self-reported general health status was 9.13 generally poorer among adult respondents than among the same age group in the general population. These effects generally increased with age. Overall, 14 per cent fewer adult respondents aged 16-54 reported ‘very good’ or ‘good’ health compared to the same age range in the general population.Chapter 9 Health and social support 191Table 9.1: Self-assessed general health of Survey 1 adult respondents at point of survey by age, compared with the general population (aged 16-54)Age groupVery goodGood Fair Bad Very badTotal Base16-24 Survey 136% 44% 16% 3% 1% 100% 78716-24 HSE 44% 44% 10% 2% <1% 100% 902Difference -8% 0% +6% +1% 1% – –25-34 Survey 128% 38% 25% 8% 1% 100% 68825-34 HSE 41% 45% 11% 2% <1% 100% 1,162Difference -13% -7% +14% +6% +1% – –35-44 Survey 121% 40% 26% 11% 3% 100% 44535-44 HSE 37% 46% 13% 3% 1% 100% 1,367Difference -16% -6% +13% +8% +2% – –45-54 Survey 118% 24% 29% 25% 4% 100% 16945-54 HSE 31% 45% 17% 6% 1% 100% 1,335Difference -13% -21% +12% +19% +3% 100% –AllSurvey 129% 40% 22% 7% 2% 100% 2,037All HSE (16-54)38% 45% 13% 4% 1% 100% 4,766Difference -9% -5% +9% +3% +1% – –Source: Survey 1 and HSE (2005) (16-54) CHP analysis. Only a few Survey 1 adult respondents (less than 1%) were aged over 54 which meant numbers were too small to allow robust comparison with the general population of England. Changes in general health since last settled accommodationAdult respondents were also asked about their general health in their ‘last 9.14 settled accommodation’ prior to acceptance as homeless, as a means of investigating whether there was evidence of changes in health status that could be associated with the experience of homelessness and temporary accommodation304. 304 In all, 71 per cent of all families in Survey 1 had a ‘last settled accommodation’ that fulfilled relevant criteria for this to be deemed a ‘valid’ comparison point for the purposes of this research, and 29 per cent did not. See Appendix 1 for a full explanation. All references to ‘changes since last settled accommodation’ in this and subsequent chapters relate to those adult respondents/families who reported a ‘last settled accommodation’ that was such a valid comparison point.192 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFor the majority (66 per cent) self-reported health status had remained the 9.15 same between last settled accommodation and point of survey. In 22 per cent of cases it had got better, and in 12 per cent it had got worse. Thus adult respondents’ health status was approximately twice as likely to have improved as deteriorated since leaving their last settled accommodation, although for most there was no change. Changes in health status were not associated with whether adult respondents 9.16 were in temporary or settled accommodation at point of survey, nor was any relationship detected with particular temporary accommodation experiences. However, regression analysis indicated that, other things being equal9.17 305, there was a strong independent relationship between violent relationship breakdown306 as a cause of homelessness and improvements in general health. Almost half (44 per cent) of adult respondents who reported violent relationship breakdown said that they had experienced an improvement in their general health since their last settled accommodation, compared to just 17 per cent of other adult respondents. Regression analysis also indicated that, other things being equal9.18 307, there were two independent effects associated with deteriorations in general health between last settled accommodation and point of survey:• adult respondents who reported three or more physical problems with their current accommodation reported deteriorations in their health at a higher rate (20 per cent), than those who did not report three or more problems (10 per cent).• those experiencing a deterioration in their financial situation since their last settled home were also more likely to report a deterioration in health (21 per cent), than those whose financial situation had not deteriorated (9 per cent).Longstanding illness and disabilityOne quarter (25 per cent) of all adult respondents aged 16-54 reported a 9.19 longstanding illness or disability308. This is slightly lower than the rate (33 per cent) reported for the general population aged 16-54 in the HSE (2005). This 305 This regression analysis controlled for: demographic characteristics; geographical variables; current accommodation type; temporary accommodation experiences; accommodation conditions; causes of homelessness; reduced or increased contact with family; how managed financially and whether this has changed since last settled accommodation.306 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.307 This regression analysis controlled for: demographic characteristics; geographical variables; current accommodation type; temporary accommodation experiences; accommodation conditions; household became workless since last settled accommodation; causes of homelessness; how managed financially and whether this has changed since last settled accommodation.308 Adult respondents were asked the following question harmonised with the HSE (2005): ‘Do you have any longstanding illness, disability or infirmity? By longstanding I mean anything that has troubled you over a period of time, or that is likely to affect you over a period of time.’ Analysis was restricted to age 16-54 to allow for meaningful comparison with the HSE (2005) as very few adult respondents to Survey 1 were aged over 54.Chapter 9 Health and social support 193difference is explained in part by the younger age profile of adult respondents to Survey 1 (even when analysis is restricted to the 16-54 age range). However, the rate of ‘limiting’ longstanding illness or disability9.20 309 was effectively identical amongst adult respondents in families accepted as homeless (18 per cent) to that within the general population aged 16-54 (17 per cent) despite adult respondents’ lower average age. A small number (5 per cent) of adult respondents reported having dyslexia. 9.21 Estimates of dyslexia in the general population vary between 2 and 15 per cent, depending on whether mild or severe dyslexia is being recorded310. A small proportion of adult respondents (2 per cent) reported having another learning difficulty of some kind, and there is evidence of a similar level within the general population311.Changes in diet since last settled accommodation Despite the concerns raised by previous research9.22 312, most adult respondents did not report that their own and their children’s diet had deteriorated since leaving their last settled accommodation. In fact, 65 per cent reported that it had remained the same, 24 per cent reported that it had improved, and only 11 per cent said that it had got worse. Thus adult respondents were approximately twice as likely to report an improvement as a deterioration in their families’ diet, but most reported no change. There was a very small net improvement in diet (amounting to 4 percentage 9.23 points) amongst those still in temporary accommodation and a more substantial net improvement (amounting to 19 percentage points) amongst those in settled housing. No patterns were detected with regards to the type of temporary accommodation that a family was living in and changes in quality of diet. Improvements or deteriorations in diet where not found to be associated with satisfaction with cooking facilities in either temporary or settled accommodation313. An association was identified between current self-reported financial hardship 9.24 and deterioration in diet since last settled accommodation. Those adult respondents who reported a poor current financial situation (ranging from 309 Defined as a long-term illness or disability that ‘limits your activities in any way’.310 Source: Parliamentary Office of Science and Technology (2004) Postnote: Dyslexia and Dyscalculia July 2004 Number 226 available at: http://www.parliament.uk/documents/upload/POSTpn226.pdf311 Source: British Institute of Learning Difficulties (undated) Factsheet – learning disabilities http://www.bild.org.uk/docs/05faqs/Factsheet%20Learning%20Disabilities.pdf312 See paragraph 9.2313 See Table 7.6 for an account of satisfaction with cooking facilities in various types of temporary accommodation.194 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds‘not managing very well’ through to being in ‘deep financial trouble’) were more likely (20 per cent) to report a deterioration in diet than those who reported a better current financial situation (7 per cent)314. Mental health problemsCurrent “anxiety, depression or other mental health problems” were reported 9.25 by 27 per cent of all adult respondents (see Table 9.2). As with the general population, women adult respondents reported current 9.26 mental health problems at a higher rate than male adult respondents (28 per cent compared to 20 per cent). Partly as a result of the preponderance of women amongst adult respondents in families accepted as homeless, the overall proportion of adult respondents reporting anxiety, depression or other mental health problems was greater than the proportion of adults in the general population reporting similar concerns in the HSE (2005)315. Table 9.2: Approximate comparison of self-reported current mental health problems among adult respondents with the general population, by gender (aged 16-54)Gender Survey 1 adult respondents **Base HSE (2005)*Base DifferenceFemale 28% 1,736 21% 2,438 +7%Male 20% 307 15% 1,923 +5%All 27% 2,043 18% 4,361 +9%Source: Survey 1 and HSE 2005 (aged 16-54 years). CHP analysis. *Based on “Is respondent not anxious or depressed, anxious or depressed, or very anxious or very depressed” (the proportion reporting any level of depression or anxiety are reported here). **Adult respondents to Survey 1 were asked whether they still had depression, anxiety or other mental health problems if they reported any lifetime experience of these problems.Adult respondents in settled housing reported current mental health problems 9.27 at a near identical rate to those still in temporary accommodation at point of survey (27 per cent compared to 26 per cent). Likewise, there were no associations detected between temporary accommodation experiences and self-reported current mental health problems.314 See Chapter 10 for a discussion of overall financial circumstances.315 Note that this is an approximate comparison. Adult respondents to Survey 1 were asked if they currently had depression, anxiety or other mental health problems, whereas the HSE asked about “anxiety and depression”. Please also note that is a broad definition of mental health problems and does not necessarily indicate serious mental health problems. As with general health, the analysis is limited to the 16-54 age group to allow for meaningful comparisons with HSE (2005).Chapter 9 Health and social support 195However, regression analysis indicated that, other things being equal9.28 316, the following factors had an independent effect on self-reported current mental health problems:• experience of sexual assault as an adult: those who reported this experience were more likely to report current mental health problems (59 per cent) than other adult respondents (24 per cent).• violent relationship breakdown as a cause of homelessness: 44 per cent of those who reported violent relationship breakdown as a reason for applying as homeless reported current mental health problems, as compared to 23 per cent of other adult respondents317. • perceptions of safety in current accommodation (both temporary and settled): 36 per cent of those who felt unsafe in their current accommodation reported current mental health problems, as compared with 25 per cent of other adult respondents318.• current financial difficulties: 35 per cent of those with current financial difficulties also reported current mental health problems, compared to 23 per cent of other adult respondents319. • experience of drug and alcohol problems: one half (50 per cent) of those adult respondents reporting any history of drug and/or alcohol problems reported current mental health problems, compared to 24 per cent of other respondents.Regression analysis further indicated that, other things being equal9.29 320, two groups of adult respondents – young people and those with an ethnic minority background – were less likely than other adult respondents to self-report current mental health problems. Around one fifth (21 per cent) of adult respondents aged under 25 reported current mental health problems, as compared to 30 per cent of older respondents. Among ethnic minority adult respondents, current mental health problems were reported at about half the rate found among White adult respondents (16 per cent compared to 30 per cent).316 Factors controlled for in this regression analysis included: demographic characteristics; geographical variables; current accommodation type; temporary accommodation experiences; accommodation conditions; causes of homelessness; any experience of domestic violence; whether experienced sexual abuse as a child; whether experienced sexual assault as an adult; whether had ever had a problem with drugs and/or alcohol; current financial difficulties; and household ‘workless’.317 Violent relationship breakdown with a partner was reported by 13 per cent of all adult respondents as a reason for applying as homeless, and another 3 per cent of adult respondents reported some other form of violent relationship breakdown as a reason for their homelessness.318 See Chapters 7 and 8 for a full discussion of adult respondents’ perceptions of safety in temporary and settled accommodation respectively.319 Current financial situation ranging from ‘not managing very well’ through to being in ‘deep financial trouble’ (see Chapter 10).320 See footnote 316 for factors controlled for in this regression.196 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsWe asked adult respondents who reported current mental health problems 9.30 whether these had changed as compared with when they were living in their last settled accommodation. Amongst those with current mental health problems, 47 per cent reported that these had got worse since they left their last settled accommodation, and 23 per cent reported that they had improved, while 26 per cent reported that they had stayed the same321. However, it should be borne in mind that these figures do not take into account those adult respondents who may have had mental health problems in their last settled accommodation but who no longer had them at all. As was reported in Chapter 3 (see para 3.22), one half of all adult respondents (52 per cent) reported having ever experienced mental health problems; clearly a much higher figure than those reporting current mental health problems (27 per cent). It is not known whether these past mental health problems were present when the relevant adult respondents were living in their last settled accommodation.Drug and alcohol problemsSelf-reported problematic drug or alcohol use was not widespread among 9.31 the adult respondents to Survey 1322. As was reported in Chapter 3 (see para 3.26), around one in ten (11 per cent) said that they had ever had a problem with drinking and/or drug use (including solvents), and 3 per cent self-reported current problems with alcohol and/or drugs (see Table 9.3). Table 9.3: Self-reported drug and alcohol problems Drug problem (including solvents)Alcohol problemAlcohol and/or drug problemBaseCurrent problem 2% 1% 3% 2,053Ever had problem 9% 6% 11% 2,053Source: Survey 1 The limitations of self-reporting of problematic drug and/or alcohol use 9.32 must be acknowledged323, as people may find it difficult to admit (even to themselves) that they have a problem with substance misuse, especially in a culture in which excessive alcohol consumption is arguably a social norm324. 321 3% of respondents reported they were unsure whether their current mental health problems were better or worse.322 Adult respondents were asked “have you ever had problems with drugs or solvents?” and “have you ever had alcohol problems?”, and if they responded “yes” to either, whether they still had these problems. Questions that explore whether or not an individual thinks their drug or alcohol use is problematic, such as those employed in Survey 1, are unusual in national surveys which tend to focus on consumption patterns and to employ long suites of questions about such patterns, for which there was insufficient space in this wide-ranging questionnaire. Direct comparison with the general population is not therefore possible323 It should be noted that this part of the questionnaire was self-completion to encourage candidness as far as possible.324 Institute of Alcohol Studies (2005) Excessive and Problem Drinking in England and Wales: IAS Factsheet http://www.ias.org.uk/Chapter 9 Health and social support 197This may be a particular concern for parents with children who may fear social services involvement if they admit to a drug or alcohol problem. Nevertheless, there are a number of grounds for thinking that those with 9.33 problematic drug and/or alcohol use were genuinely a very small subgroup of all adult respondents. First, many adult respondents were willing to report other deeply personal problems, such as depression or experiencing domestic violence. Second, findings from across the survey are consistent on this point of low levels of drug/alcohol problems: such problems were almost never given as a contributory cause of homelessness (see Figure 5.1), and very few adult respondents reported receiving professional help with such problems (see Table 9.11 below). Third, the same survey techniques employed in this study found far higher levels of drug/alcohol problems being reported by 16-17 year olds325. Fourth, while not a directly comparable group, it is relevant to note that relatively low levels of drug/alcohol problems were also found amongst families accepted as homeless assisted under the Supporting People programme326.Support needs and service useAdult respondents were asked whether they or their families were getting, 9.34 or needed, help with a range of 14 potential types of support. These fell into two broad categories – ‘practical support’ and ‘personal support’. Practical supportTable 9.4 summarises the types of practical support currently received 9.35 by families, and also notes any unmet practical support needs that were reported. Overall, 63 per cent of adult respondents reported being in receipt of 9.36 one or more forms of practical support since applying as homeless. The types of practical support reported most frequently related specifically to accommodation – with 29 per cent of families receiving help with repairs, and 28 per cent receiving practical or financial help getting furniture or other household equipment. Around one fifth (21 per cent) reported receiving help with filling in forms or applying for benefits. All other forms of practical support were reported by smaller numbers.325 As is reported in Chapter 13, 37 per cent of young people accepted as homeless 16-17 year olds self-reported that they had ever had problems due to substance misuse, and 16 per cent reported that they still did.326 For information on the Supporting People client database see http://www.spclientrecord.org.uk.198 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAdult respondents who were in settled housing were more likely (68 per 9.37 cent) to have received one or more practical support services than those in temporary accommodation at point of survey (57 per cent). However, there was little difference on most specific types of practical support received, except help with repairs (received by 36 per cent of those in settled housing, as compared to 21 per cent of those in temporary accommodation), and help getting furniture (33 per cent of those in settled housing and 23 per cent of those still in temporary accommodation). Both of these differences were probably explained by those in settled housing getting help to set up in their new home.Regression analysis indicated that, other things being equal9.38 327, the following factors had an independent effect on the likelihood of having received practical support:• being in settled housing (see para 9.37). • having personal support needs: 87 per cent of adult respondents with at least one ‘personal support need’328 were receiving practical support, as compared to 58 per cent of other adult respondents.• having any experience of hostel or B&B hotels as temporary accommodation: 70 per cent of those who had experienced these forms of temporary accommodation were had received practical support, compared to 60 per cent of other adult respondents.• being in a workless household: 68 per cent of workless households had received practical support, compared to 52 per cent of other respondents. • being accepted in London: receipt of practical support was reported at a lower rate in London (51 per cent) than elsewhere (66 per cent).327 Factors controlled for in this regression included: demographic characteristics; geographical variables; current accommodation type; current mental health problems; experience of drug or alcohol problems; causes of homelessness; temporary accommodation experiences; accommodation conditions; workless household; difficulty managing financially; personal support needs.328 See para 9.42 below.Chapter 9 Health and social support 199Table 9.4: Families’ self-reported receipt and need for practical supportReceived helpUnmet need for helpDo not need helpTotalAdvice or help with repairs to your accommodation29% 8% 63% 100%Practical or financial help getting furniture or other household equipment28% 12% 60% 100%Someone to help you fill in official forms or apply for benefits21% 6% 73% 100%Someone to speak for you to official people16% 7% 77% 100%Advice or help finding a job 11% 7% 82% 100%Advice or help getting into education or training10% 8% 82% 100%Advice or help with managing money, budgeting or dealing with debts9% 10% 82% 100%Advice or help getting to see a doctor or accessing other health services9% 2% 89% 100%Advice or help getting childcare 7% 6% 87% 100%Advice or help getting your children into school6% 2% 92% 100%Source: Survey 1 Base 2,053One third (35 per cent) of all adult respondents reported at least one unmet 9.39 practical support need, but the proportion self-reporting any particular unmet need for practical support was much smaller. The most frequently identified unmet need – help with securing furniture or household equipment – was reported by 12 per cent of all adult respondents (Table 9.4). Adult respondents in temporary accommodation reported they had one or more unmet practical support need at a higher rate than adult respondents in settled housing (42 per cent compared to 30 per cent).200 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsRegression analysis indicated that, other things being equal9.40 329, the following factors had an independent effect on the likelihood of self-reporting unmet practical support needs:• being in a workless household: 40 per cent of adult respondents in workless households reported an unmet practical support need, compared to 28 per cent of other respondents.• being in financial difficulty: 49 per cent of those who reported current financial difficulties also reported unmet practical support needs, compared to 29 per cent of other adult respondents.• being in accommodation with physical conditions problems: 47 per cent of those who reported three or more physical problems with their accommodation reported unmet practical support needs, compared to 34 per cent of other adult respondents.Once these and other factors were taken into account, there was no 9.41 independent effect of being in settled or temporary accommodation. Personal support As demonstrated in Table 9.5, the most common form of personal support 9.42 received by adult respondents was help with mental health problems (12 per cent)330. Only small numbers reported receipt of any other form of personal support331. Very few self-reported any unmet personal support needs (4 per cent of all adult respondents reported one or more unmet personal support needs, as defined in Table 9.5). 329 Factors controlled for in this regression included: demographic characteristics; geographical variables; current accommodation type; current mental health problems; experience of drug or alcohol problems; causes of homelessness; temporary accommodation experiences; accommodation conditions; workless household; difficulties managing financially; personal support needs.330 This is considerably lower than the 27 per cent who reported current mental health problems. Nonetheless, only 2 per cent self-identified an unmet need for this type of personal support.331 The small proportions reporting receipt of, or need for, personal support with drug or alcohol problems are consistent with the low numbers reporting these sorts of problems (see Table 9.3).Chapter 9 Health and social support 201Table 9.5: Families’ self-reported need for personal supportGetting helpUnmet need for helpDo not need helpTotalAdvice or help dealing with mental health problems, including depression and anxiety12% 2% 86% 100%Advice or help with parenting6% 2% 91% 100%Advice or help dealing with drug problems1% 1% 98% 100%Advice or help dealing with alcohol problems1% 0% 99% 100%Source: Survey 1 Base 2,053Overall, only 16 per cent of adult respondents to Survey 1 said that they 9.43 had received help with any personal support needs or had unmet needs of this type. Regression analysis indicated that, other things being equal332, the following factors had an independent effect on the likelihood of self-reporting personal support needs (met or unmet)333:• current mental health problems: 36 per cent of adult respondents who reported a current mental health problem, as compared to just 8 per cent of other adult respondents, reported one or more personal support needs.• violent relationship breakdown as a cause of homelessness: 25 per cent of those who had experienced violent relationship breakdown reported personal support needs, as compared to 14 per cent of other adult respondents334.• practical support needs: adult respondents with one or more practical support needs reported personal support needs at a higher rate (22 per cent) than other adult respondents (6 per cent).There was no association between the presence of personal support 9.44 needs and whether an adult respondent was living in settled or temporary accommodation.332 Factors controlled for in this regression included: demographic characteristics; geographical variables; current accommodation type; current mental health problems; experience of drug or alcohol problems; causes of homelessness; temporary accommodation experiences; accommodation conditions; workless household; current financial difficulties practical support needs.333 It should be noted that an assumption was made here for the purposes of analysis that all those who receive such personal support do in fact need it, and thus can be combined with those identifying an unmet need for such help in order to identify all those with ‘personal support needs’.334 Violent relationship breakdown with a partner was reported by 13 per cent of all adult respondents as a reason for applying as homeless, and another 3 per cent of adult respondents reported some other form of violent relationship breakdown as a reason for their homelessness.202 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsUse of the NHS and care and support servicesThis section reviews the use of a range of NHS and other care and support 9.45 services by adult respondents and their families. It gives specific attention to the issue of GP registration amongst families accepted as homeless. Table 9.6 shows the extent to which adult respondents had ever had contact 9.46 with a range of professional services. Perhaps the most notable points are that three in five (60 per cent) of adult respondents had, at some point, had contact with housing support workers, and one fifth (22 per cent) had seen a social worker. Much smaller numbers had ever seen mental health specialists, refuge workers or drug and alcohol workers. Table 9.6 also shows the proportion of adult respondents who had had 9.47 any contact with the range of specified services when living in their current (temporary or settled) accommodation. Table 9.6: Use of NHS, care and support services and other formal support services in current accommodation, by temporary and settled accommodationEver seenSeen in current accommodationTA Settled housing AllGPs 85% 82% 76% 79%Health visitor 62% 38% 35% 37%Key worker or housing support worker60% 35% 18% 25%Social worker 22% 11% 8% 10%Community psychiatric nurse/counsellor 8% 3% 4% 3%Psychologist or psychiatrist 8% 4% 2% 3%Women’s Aid or refuge worker8% 2% 2% 2%Drug or alcohol worker 3% 2% <1% 1%Base 2,053 1,130 923 2,053Source: Survey 1Chapter 9 Health and social support 203Use of GP services in current accommodation was widespread (79 per cent 9.48 overall), as would be expected of households that often contained young children335. There was also quite a sizeable proportion (37 per cent) who had made use of health visitor services in their current accommodation, which again would be expected given the presence of babies and toddlers in many of these families336. One in ten families (11 per cent) had seen a social worker while living in their current accommodation. Current contact with community mental health services was unusual, which 9.49 may seem surprising given the proportion (27 per cent) who reported current mental health problems (see para 9.25). However, this probably reflects the strategic emphasis within the NHS on GP rather than Community Mental Health Services for dealing with less severe forms of depression and other mental health problems337. The low levels of contact with drug and alcohol services is consistent with 9.50 the very low levels of current drug and alcohol problems reported by adult respondents (see Table 9.3).Table 9.6 indicates that adult respondents in temporary accommodation 9.51 were more likely than those in settled housing to report seeing key workers or housing support workers in their current accommodation. There was otherwise little distinction between service use in temporary and settled housing. As noted in the introduction9.52 338, there has been a particular concern about GP registration and accessibility amongst homeless groups. However, almost all (98 per cent) of families accepted as homeless were registered with a GP. Adult respondents were also asked if their GP was in easy reach of where they were living: 84 per cent responded ‘yes’ and 16 per cent replied ‘no’. This proportion did not differ between those still in temporary accommodation and those in settled housing. There was very little evidence that families accepted as homeless were making 9.53 inappropriate use of hospital A&E services because of lack of GP registration. Less than 1 per cent of all adult respondents stated that they (or another family member) had, since they applied as homeless, gone to a hospital because they were not registered with a GP. 335 68 per cent of all families accepted as homeless contained a child or children under five years old.336 Half (46 per cent) of all families accepted as homeless contained a child or children aged under one year old.337 Department of Health (1999) National Service Framework for Mental Health: Modern standards and service models http://www.doh.gov.uk/338 See para 9.2.204 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAccess to social supportThis section outlines the availability of social support to adult respondents 9.54 in families accepted as homeless, and investigates any changes in such support since families left their last settled accommodation. It first examines their access to ‘emotional’ and ‘instrumental’ forms of social support, and then explores any change in their contact with friends or relatives since last settled accommodation, as this may also be indicative of their access to social support. Emotional and instrumental supportFigure 9.1 shows the percentage of survey respondents reporting that they 9.55 had someone who they could count on to listen if they “needed to talk” (i.e. emotional support) and/or to “help out in a crisis” (i.e. ‘instrumental’ support)339. The graphic covers adult respondents’ current accommodation at point of survey (whether temporary or settled), and their last settled accommodation prior to being accepted as homeless, alongside a comparator from the British Household Panel Survey (BHPS) 2004340.Figure 9.1: Access to practical and emotional support for Survey 1 adult respondents at point of survey by age, compared with responses to British Household Panel Survey, 2004Last settled accommodationCurrent accommodation81% 79%96%78% 77%97%0%20%40%60%80%100%Have someone to count tohelp out in a crisisHave someone to count on tolisten when need to talkBHPS 2004 EnglandComparatorSource: Survey 1 and BHPS 2004 (wave 13, England only, 16+ yrs, CHP analysis). Base: Survey 1, 2,053 respondents for current accommodation, 1,344 adult respondents defined as having a ‘last settled accommodation’ valid for comparison purposes; BHPS, 5,500 households in England 339 Cohen, S. and Wills, T. (1985) ‘Stress, social support and the buffering hypothesis’ Psychological Bulletin, 98, 310-357.340 The questions used in Survey 1 to assess emotional and instrumental support were not harmonised completely with those used in the BHPS 2004 (as we combined the ‘yes, one person’ and ‘yes, more than one person’ options into a single response). Nevertheless, the comparison with the BHPS indicates important differences between the adult respondents and the population in England as a whole.Chapter 9 Health and social support 205As Figure 9.1 demonstrates, access to these forms of social support was, 9.56 while high, still lower for adult respondents than amongst the general population of England. A large majority of adult respondents had at least some access to emotional support (81 per cent) and instrumental support (78 per cent) at point of survey. These figures were very close to the levels of emotional and instrumental support available to adult respondents in their last settled accommodation, indicating that overall access to these forms of social support were little affected by the experience of homelessness and living in temporary accommodation. Access to one or both of these forms of social support was similar among 9.57 adult respondents in temporary accommodation at point of survey (83 per cent) and those who were in settled housing (89 per cent) (the overall figure was 87 per cent). Regression analysis indicated that, other things being equal9.58 341, being from an ethnic minority background exerted an independent effect which lowered the likelihood of having access to emotional and/or instrumental support: only 76 per cent of adult respondents with an ethnic minority background, as compared to 90 per cent of White adult respondents, had current access to at least one of these forms of social support 342. Likewise, other things being equal, adult respondents who had current mental health problems were less likely to have access to instrumental and emotional support than other respondents (81 per cent compared to 89 per cent). We also investigated the sources of emotional and/or instrumental 9.59 support that adult respondents had access to. As can be seen in Figure 9.2, parents were the most common source of support (parents were a particularly important source of instrumental support, especially for younger respondents), with friends the next most important source of support. Partners were mentioned as a source of support by a smaller proportion of adult respondents, as one might expect given the high proportion (65 per cent) of lone women parents amongst the adult respondents to Survey 1 (see Chapter 2). Children and workers were not often mentioned as sources of support.341 Factors controlled for in this regression analysis included: demographic characteristics; geographical variables; current accommodation type; temporary accommodation experiences; accommodation conditions; causes of homelessness; whether have current mental health problems; experience of drugs or alcohol problems; workless household; and current financial difficulties.342 It should be noted that being a former asylum seeker was also controlled for in this regression analysis, but did not have an independent effect, and so does not explain the finding on ethnic minorities.206 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 9.2: Sources of emotional and instrumental support50%44%36%27%9%3%2%1%0% 10% 20% 30% 40% 50% 60%OthersCommunity groupsChildrenWorkersPartnerOther familyFriendsParentsSource: Survey 1 Base: 1,768 adult respondents who identified sources of emotional and instrumental social support. ‘Workers’ includes social workers and housing support or key workers. Changes in social networksAdult respondents were asked whether there had been any change in the 9.60 level of contact they had with family343 and with friends between their last settled accommodation and point of survey. As is shown in Table 9.7, over half of adult respondents had experienced 9.61 some changes in their level of contact with family and/or friends. Those reporting more contact with family (28 per cent) balanced out those reporting less (27 per cent). However, almost double the proportion reported less contact with friends (36 per cent) than reported more contact (20 per cent). As can be seen, only very small proportions had no contact at all with family or friends at either point (but these figures were much higher for former asylum seekers: 16 per cent had no contact with family in either setting, and 10 per cent had no contact with friends in either setting).343 The phrase used was ‘contact with family members you like’ on the basis that contact with some or all of adult respondent’s family may not necessarily have been desired by the respondent.Chapter 9 Health and social support 207Table 9.7: Changes in contact with family between last settled housing and current accommodationMore contactSame contactLess contactNo contact at either pointTotal BaseContact with family28% 42% 27% 2% 100% 1,334Contact with friends 20% 40% 36% 3% 100% 1,333Source: Survey 1 Base: all adult respondents with a ‘last settled accommodation’ that was valid for comparison purposes.It is also worth noting that less than 1 per cent of those adult respondents 9.62 who had had contact with family in their last settled accommodation had completely ceased contact by point of interview, and with regards to friends this was true of just 4 per cent. Thus complete loss of pre-existing social support was very rare.No association was identified between changes in contact with family or 9.63 friends and whether an adult respondent was living in settled housing or in temporary accommodation at point of interview. Regression analysis indicated that, other things being equal9.64 344, relationship breakdown as a cause of homelessness had an independent (positive) effect on the likelihood of increased contact with family and/or friends. Thus, 35 per cent of those who had experienced relationship breakdown reported increased contact with family and/or friends, as compared with only 17 per cent of other adult respondents. Regression analysis also indicated that, other things being equal9.65 345, reduced contact with family and/or friends was associated with two factors:• being accepted as homeless in a rural area: 37 per cent of those accepted in a rural areas reported diminished contact with friends and/or family, as compared with 28 per cent of other adult respondents. • being under 25: 35 per cent of younger adult respondents reported less contact with family and/or friends, as compared with 26 per cent of older adult respondents. 344 Demographic characteristics; geographical variables; causes of homelessness; current accommodation type; temporary accommodation experiences; current mental health problems; history of drug/alcohol problems; where applied from as homeless; and current financial status.345 Demographic characteristics; geographical variables; causes of homelessness; current accommodation type; temporary accommodation experiences; current mental health problems; history of drug/alcohol problems; where applied from as homeless; and current financial status.208 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAdult respondents in temporary accommodation for more than one year This section reviews the health and support needs of adult respondents 9.66 who had stayed in temporary accommodation for over one year (Survey 4 adult respondents), and compares this to the findings on Survey 1 adult respondents who had generally spent a shorter period in temporary accommodation. Health amongst Survey 4 adult respondentsTable 9.8 compares the self-assessed general health of Survey 4 adult 9.67 respondents with that of Survey 1 adult respondents and the general population aged 16-54. As can be seen, there were only minor differences between the self-assessed general health of Survey 4 and Survey 1 adult respondents. Survey 4 respondents tended to be less likely to report very good or good health than the general population, something that was also true for Survey 1 respondents.Table 9.8: Self-assessed general health of Survey 4 adult respondents at point of survey, compared with Survey 1 adult respondents and general population (aged 16-54)Age group Very GoodGood Fair Bad Very BadTotal BaseSurvey 4 (16-54)24% 40% 26% 9% 1% 100% 571Survey 1 (16-54)29% 40% 22% 7% 2% 100% 2,037HSE (16-54) 38% 45% 13% 4% 1% 100% 4,766Source: Survey 4, Survey 1 and HSE (2005). (16-54) CHP analysis. Only a few Survey 4 respondents (1%) were aged over 54 which meant numbers were too small to allow robust comparison with the general population of England.Longstanding illness and disability was reported by 25 per cent of Survey 9.68 4 adult respondents, the same rate as was found among Survey 1 adult respondents, which was lower than that reported by the general population (33 per cent). Overall, 20 per cent of Survey 4 adult respondents reported a longstanding illness or disability that limited their activities, again this figure was very similar to that among Survey 1 respondents (18 per cent), and also in this case to the general population (17 per cent). Current anxiety, depression or other mental health problems were self-9.69 reported by 20 per cent of Survey 4 adult respondents. This was a somewhat lower rate than that found among Survey 1 adult respondents Chapter 9 Health and social support 209(27 per cent, see Table 9.2). This discrepancy was explained by the higher proportion of ethnic minority adult respondents in Survey 4 than in Survey 1: these respondents were less likely in both surveys to report mental health problems346. Regression analysis on Survey 4 indicated that, other things being equal9.70 347, the following factors had an independent effect on the likelihood of adult respondents reporting current mental health problems:• being from an ethnic minority background: only 14 per cent of those from an ethnic minority background, as compared with 30 per cent of White adult respondents, self-reported current mental health problems.• being a former asylum seeker: only 10 per cent of former asylum seekers self-reported current mental health problems, as compared to 25 per cent of other Survey 4 adult respondents348. • feeling unsafe in the area in which temporary accommodation was located: 31 per cent of those who felt unsafe in the local area reported current mental health problems, as compared to 15 per cent of other Survey 4 adult respondents349. • current financial difficulties: 26 per cent of those with current financial difficulties reported current mental health problems, as compared to 15 per cent of other Survey 4 adult respondents350.As with Survey 1 (see Table 9.3), self-reported drug and/or alcohol problems 9.71 were very unusual among the Survey 4 adult respondents. Only 6 per cent reported ever having had these problems and only 1 per cent reported that they currently had a problem with drugs and/or alcohol. Support needs and service use amongst Survey 4 adult respondentsAs Table 9.9 demonstrates, Survey 4 adult respondents tended to report 9.72 having received most forms of practical support at a lower rate than did Survey 1 adult respondents. This is explained to some extent by the fact that around half of Survey 1 adult respondents were in settled housing by 346 We conducted a regression analysis which indicated that, once ethnic minority backgrounds were controlled for, there was no distinction between Survey 1 and Survey 4 adult respondents with respect to likelihood of self-reporting current mental health problems. Factors controlled for in this regression were: ethnic minority background; being in Survey 1 or Survey 4. See para 9.29 for data on ethnic minorities and mental health in Survey 1. See para 9.70 for data on ethnic minorities and mental health in Survey 4.347 Factors controlled for in this regression included: demographic variables; relationship breakdown; and violent relationship breakdown, as a cause of homelessness; temporary accommodation conditions; current financial status.348 This result differs from Survey 1 where no particular link was detected between a history of seeking asylum and lower levels of self-reported mental health problems. This is most likely because the number of former asylum seekers was too small to detect the effect at the required degree of statistical significance in Survey 1, but their larger numbers in Survey 4 allowed this to be picked up.349 There was a similar finding on links between perceptions of safety and mental health problems in Survey 1, though in this case a higher incidence of mental health problems was found amongst those who felt unsafe inside their (temporary or settled) accommodation rather than in the local area (see para 9.28).350 This matches the finding for Survey 1, (see para 9.28).210 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldstime of interview: this group were more likely than those still in temporary accommodation to report having received practical support (see para 9.37). However, Survey 4 adult respondents were more likely to report unmet practical support needs than were Survey 1 adult respondents, particularly with regards to help with filling in official forms and someone to speak for them to official people.Table 9.9: Survey 4 families’ self -reported need for practical supportReceiving helpUnmet need for helpDo not need helpAdvice or help with repairs to your accommodation15% [29%] 13% [8%] 72% [63%]Practical or financial help getting furniture or other household equipment13% [28%] 17% [12%] 69% [60%]Someone to help you fill in official forms or apply for benefits13% [21%] 16% [6%] 71% [73%]Someone to speak for you to official people11% [16%] 17% [7%] 73% [77%]Advice or help finding a job 7% [11%] 5% [7%] 87% [82%]Advice or help getting into education or training8% [10%] 8% [8%] 84% [82%]Advice or help with managing money, budgeting or dealing with debts5% [9%] 7% [10%] 88% [82%]Advice or help getting to see a doctor or accessing other health services7% [9%] 5% [2%] 88% [89%]Advice or help getting childcare5% [7%] 4% [6%] 91% [87%]Advice or help getting your children into school4% [6%] 3% [2%] 92% [92%]Source: Survey 4 and Survey 1 Base 571 [2,053] Figures in brackets refer to Survey 1. Each row totals to 100%Table 9.10 shows the met and unmet personal support needs reported 9.73 by Survey 4 adult respondents. As with Survey 1, both receipt of personal support and identification of unmet needs for this type of support was low.Chapter 9 Health and social support 211Table 9.10: Families’ self-reported need for personal supportGetting helpUnmet need for helpDo not need helpAdvice or help dealing with mental health problems, including depression and anxiety8% [12%] 3% [2%] 91% [86%]Advice or help with parenting 3% [6%] 1% [2%] 96% [91%]Advice or help dealing with drug problems1% [1%] <1% [1%] 99% [98%]Advice or help dealing with alcohol problems1% [1%] 0% [0%] 99% [99%]Source: Survey 4 and Survey 1 Base 571 [2,053] Figures in brackets refer to Survey 1. Each row totals to 100%Table 9.11 compares use of a range of services in current accommodation 9.74 by Survey 4 and Survey 1 adult respondents. As can be seen, Survey 4 adult respondents were more likely than Survey 1 adult respondents to have made use of GPs, health visitors, key workers/housing support workers and social workers in their current accommodation. Some of these differences reflected the typically longer duration for which Survey 4 families had been in their current (temporary) accommodation compared to Survey 1 families in their current (temporary or settled) accommodation (see para 6.52). In the case of key workers specifically, this difference will also reflect the fact that half of Survey 1 adult respondents were in settled housing by point of survey, where they were less likely to see these workers than in temporary accommodation (see Table 9.6). 212 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 9.11: Use of NHS and other care and support services, for Survey 4 and Survey 1Survey 4 adult respondentsSurvey 1 adult respondents Difference GPs 95% 79% +16%Health Visitor 48% 37% +11%Key worker or housing support worker57% 25% +32%Social worker 18% 10% +8%Community psychiatric nurse/counsellor 6% 3% +3%Psychologist or psychiatrist 8% 3% +5%Drug or alcohol worker 1% 1% 0%Women’s Aid or refuge worker2% 2% 0%Base 571 2,053 –Source: Survey 1 and Survey 4GP registration among Survey 4 adult respondents was effectively universal 9.75 (over 99 per cent). Most respondents to Survey 4 reported that their GP was within easy reach (84 per cent). These figures matched closely those for Survey 1.Access to social support amongst Survey 4 adult respondentsSurvey 4 adult respondents were likely to report they had someone to listen 9.76 to them when they needed to talk (76 per cent), which was a similar level to that reported by Survey 1 adult respondents (81 per cent). However, Survey 4 adult respondents were less likely to report that they had someone who could help them out in a crisis (63 per cent) than was the case for Survey 1 adult respondents (78 per cent). The stronger presence of people with an ethnic minority background among Survey 4 respondents largely accounted for the lower rate at which these respondents reported access to instrumental support351. 351 Overall, 53 per cent of Survey 4 adult respondents with an ethnic minority background reported access to instrumental support, compared to 74 per cent of White adult respondents. The figures for Survey 1 were similar; with 61 per cent of adult respondents to Survey 1 with an ethnic minority background reporting access to instrumental support, compared to 81 per cent of White respondents.Chapter 9 Health and social support 213The main sources of instrumental and/or emotional support reported by 9.77 adult respondents to Survey 4 are shown in Figure 9.3. Parents played a less prominent role for Survey 4 adult respondents than for Survey 1 adult respondents (25 per cent and 50 per cent identified them as a source of support respectively), which probably reflects the typically higher ages of Survey 4 adult respondents (those who received help from parents in Survey 1 were mainly younger people). However, friends and other relatives, the other two main sources of social supports reported by Survey 1 adult respondents, were mentioned at similar rates by Survey 4 adult respondents (Figures 9.3 and 9.2).Figure 9.3: Sources of emotional and instrumental support47%34%34%25%10%6%4%1%0% 10% 20% 30% 40% 50%OthersCommunity groupsWorkersChildrensParentsPartnerOther familyFriendsSource: Survey 4. Base: 480 adults Survey 4 adult respondents who identified sources of emotional and instrumental social support ConclusionsThis chapter has reviewed the health and support needs of adult respondents 9.78 in families accepted as homeless. It found that the self-reported general health status of these adult respondents was often worse than that of the general population, but did not appear to be adversely affected by homelessness and stays in temporary accommodation. Likewise, access to social support (both emotional and instrumental) was somewhat lower than for the general population, but did not seem to have diminished since leaving last settled accommodation.214 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsDespite concerns expressed in previous research, most families’ diet was not 9.79 reported to have deteriorated since leaving their last settled accommodation, and GP registration was near universal amongst adult respondents in these families (with satisfactory ease of access to their GP also reported by most adult respondents).Rates of self-reported anxiety, depression and other mental health problems 9.80 were somewhat higher than that of the general population; this difference was partly accounted for by the high proportion of women amongst adult respondents. Self-reported current drug or alcohol problems were unusual amongst adult respondents in families accepted as homeless352, and the reported need for personal support was generally low. In general, it seemed that length of time in temporary accommodation had 9.81 little impact on health and social support needs. The findings on Survey 4 adult respondents who had been in temporary accommodation for over one year were in the main very similar to those for Survey 1 adult respondents, except that Survey 4 adult respondents were less likely to self-report mental health problems and had lower access to instrumental forms of social support. Both of these discrepancies were linked to the high proportion of ethnic minority adult respondents in Survey 4 (see para 9.29 and 9.76). It was also found that Survey 4 adult respondents were more likely than those in Survey 1 to self-report unmet needs for practical support, but they had similarly low levels of self-reported unmet needs for personal support. To summarise, adult respondents in families accepted as homeless appeared 9.82 to be a relatively disadvantaged group with regard to their health and access to social support but were not in the main extremely vulnerable (see also Chapter 3 on these adult respondents’ long-term histories). Moreover, the impacts of homelessness and temporary accommodation on adult respondents’ health and social support circumstances seemed marginal. 352 Though bear in mind the caveats expressed in para 9.32. Also, many of these families have experienced relationship breakdown, and it is possible that the partner (usually male) who has left the family may have had drug or alcohol problems that contributed to the family’s homelessness.Chapter 10 Employment, income and expenditure 215Chapter 10:Employment, income and expenditureIntroductionThe overall objective of this study was to understand the causes, experiences 10.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data was collected in five separate surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys).There is some limited evidence which indicates that families accepted as 10.2 homeless are likely to be living on low incomes, and dependent on means-tested benefits, but this research is now rather dated353. It has further been argued that these families, already economically disadvantaged, may find their financial circumstances further undermined by the disruption associated with homelessness because, for example, of the difficulties of taking up work or training opportunities while living in temporary accommodation, particularly in hostels and B&B hotels where there may be perceived financial disincentives associated with high rent levels354. There have been particular concerns about the direct costs to families accepted as homeless associated with the upheaval they experience, such as having to buy new school uniforms necessitated by homelessness-related school moves, having to replace belongings which are lost or have to be abandoned when they move into temporary accommodation, and so on. However, the economic status and financial circumstances of families 10.3 accepted as homeless has more often been assumed than demonstrated. The research which is available has tended to be qualitative and/or small scale in nature, and it is not possible to assume that the findings reflect the circumstances of all families accepted as homeless355. This chapter therefore 353 Prescott-Clarke, P., Clemens, S. and Park, A. (1994) Routes into Local Authority Housing: A study of local authority waiting lists and new tenancies, London: HMSO.354 Thomas, A. and Niner, P. (1989) Living in Temporary Accommodation: A survey of homeless people, London: HMSO; Hall, S., Powney, J. and Davidson, P. (2000) The Impact of Homelessness on Families, Edinburgh: Scottish Council for Research in Education; Sawtell, M. (2002) Lives on Hold: Homeless families in temporary accommodation, London: The Maternity Alliance.355 Jones, A; Pleace, N. and Quilgars, D (2002) Firm Foundations: An evaluation of the Shelter Homeless to Home service London: Shelter; Randall, G. and Brown, S. (2003) The Support Needs of Homeless People and their Families, London: ODPM; Fitzpatrick, S., Pleace, N. and Jones, A. (2006) The Support Needs of Homeless Families: An audit of provision for families affected by homelessness in Scotland, Edinburgh: NHS Health Scotland.216 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsdraws on data from Survey 1 – a survey of families accepted as being owed the main homeless duty between 1st January 2005 and 30 June 2005 – to provide a detailed account of:• economic status, worklessness and barriers to work amongst families accepted as homeless;• the educational qualifications and literacy levels of adult respondents;• income, debt, and major expenditure;• financial exclusion; and• overall financial circumstances.The experiences of those with prolonged stays in temporary accommodation 10.4 are of particular interest. The last section of the chapter compares the economic and financial circumstances of Survey 1 families to that of families accepted as owed the main homelessness duty and in temporary accommodation for more than a year (Survey 4)356.As in other chapters, we present two types of statistical analysis here: 10.5 bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a range of other factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected357. This survey evidence demonstrates that families accepted as homeless were 10.6 living on very low incomes, and levels of worklessness amongst these families far outstripped those in the general population. Moreover, families accepted as homeless often felt that they were struggling more financially than they had been in their ‘last settled accommodation’358 (although expenses directly associated homelessness seemed to be relatively minor problems in the context of the overall weak economic position of these families). Families in temporary accommodation for over one year were more likely than other families accepted as homeless to report financial problems, but this seemed related to the type of temporary accommodation they occupied (self-contained), and to their concentration in London, rather than to their length of time in temporary accommodation. 356 Survey 4 was required because the ‘time-window’ design for Survey 1, while delivering a representative sample of those accepted as homeless over a six month period, by definition excluded those with prolonged stays in temporary accommodation. See Chapter 1 and Appendix 1 for a full explanation.357 See Chapter 1 and Appendix 2 for more detail.358 See Chapter 1 and footnote 366 below for an explanation of how ‘last settled accommodation’ was defined in this research.Chapter 10 Employment, income and expenditure 217Key points• Almost one third (29 per cent) of all adults in families accepted as homeless were in paid work at the time of the survey. • Around two thirds (64 per cent) of all families accepted as homeless were ‘workless’ (contained no adults in paid work), compared with 14 per cent of all families with children in the general population. This disparity was only partly accounted for by the large proportion of lone woman parents with young children amongst families accepted as homeless. • There was a net increase in worklessness amongst families accepted as homeless since they left their last settled accommodation. However, ‘homelessness-specific’ barriers to work – such as ‘living in temporary accommodation’ or ‘the disruption caused by homelessness’ – were very seldom cited by adult respondents. • Over one third (36 per cent) of adult respondents in families accepted as homeless had no academic or vocational qualifications, compared to 10 per cent of adults in families in the general population.• Families accepted as homeless were generally living on very low incomes, with the average incomes of couple-headed households in particular lagging very far behind their counterparts in the general population. All families accepted as homeless were very likely to be in receipt of means-tested benefits. • Two in five families (41 per cent) were behind with at least one regular household bill or loan repayment.• Major expenditure (over £200) on items such as furniture, white goods and carpets was more common amongst families in settled housing at point of interview than amongst those still in temporary accommodation. Many families in settled housing had received a Community Care Grant, most probably to assist with the cost of setting up their new home.• Families accepted as homeless were more likely to self-report difficulties in managing financially than families with children in the general population. They were much more likely to report a deterioration (47 per cent) than an improvement (18 per cent) in their financial circumstances since leaving their last settled accommodation. • Survey 4 families differed little in their employment, educational and financial circumstances from Survey 1 families, but were more likely to report that, overall, they were struggling financially. This seemed related to the form of accommodation in which they were living (self-contained temporary accommodation), and to their concentration in London, rather than to the length of time they had stayed in temporary accommodation.218 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsEconomic status, worklessness and barriers to workThis section considers the employment circumstances of adults in families 10.7 accepted as homeless. It considers:• the economic status of all adults in families accepted as homeless; • overall levels of worklessness amongst these families; • any changes in levels of worklessness since last settled accommodation; and• barriers to work reported by adult respondents in families accepted as homeless. Current economic status of all adults in families accepted as homelessTable 10.1 shows the economic status of all adults (over 16) within families 10.8 accepted as homeless at the point of survey, and compares this with that of adults of working age in households containing children in the general population. Table 10.1: Economic status of all adults in families accepted as homeless, compared with all adults living in households containing children in general populationStatus Survey 1 Adults in families with children in England DifferenceIn paid work (including self-employed) 29% 70% -41%Looking after the home or family 38% 12% +26%In education or training 8% 9% -1%Unemployed and seeking work 12% 3% +9%Sick or disabled 7% 3% +4%Other (not in employment) 6% 3% +3%Total 100% 100% –Base 3,015 6,500 –Source: Survey 1 and (FACS) (2005) (England only). CHP analysis. Base: for Survey 1 is the data on the economic status of 3,015 adults across 2,053 households, this included all working age (18+) adults in the householdsIt is immediately clear that adults in families accepted as homeless were 10.9 far less likely to be in paid work than adults in families with children in the general population. However, at 29 per cent, levels of employment amongst adults in these families were perhaps higher than might have been expected. Chapter 10 Employment, income and expenditure 219The patterns shown in Table 10.1 (particularly the high rate of ‘looking after 10.10 home or family’) are linked in part to differences in household composition compared to the general population of families, primarily the much higher numbers of women lone parent families with young children amongst families accepted as homeless (see Chapter 2). These differences are explored further below. Worklessness amongst families accepted as homelessApproximately one third of all families accepted as homeless (30 per cent) 10.11 contained one adult in paid work, 5 per cent contained two adults in paid work, and 1 per cent contained three or more adults in paid work. Overall, therefore, 64 per cent of families accepted as homeless were ‘workless’ (contained no adults in paid work). During 2005, only 14 per cent of all families with children in the general population of England were workless359.However, as just noted, there were far more women lone parents amongst 10.12 families accepted as homeless than in the general population of England (65 per cent of all families accepted as homeless, compared to 20 per cent of all families with children in England360). These women lone parents were also far more likely to be in households containing children aged under five, ie of pre-school age, (67 per cent of women lone parents accepted as homeless361, compared to 36 per cent of women lone parents in the general population)362. Across England, families that are headed by a female lone parent are far more likely to be ‘workless’ (40 per cent) than are families headed by couples (4 per cent)363. Particularly low rates of paid work are found among women heading lone parent households who have pre-school children (55 per cent are workless). However, even once these household characteristics are taken into account, 10.13 families accepted as homeless were still more likely to be workless than their equivalents in the general population: • a large majority of women lone parents with children under five accepted as homeless were not in paid work (78 per cent), compared to 55 per cent of the same group in the general population.• women lone parents with no children aged under five were also less likely to be in paid work than their peers in the general population (61 per cent were not in paid work, compared to 32 per cent of this group in the general population).359 FACS (2005) (England only) CHP analysis.360 FACS (2005) (England only)361 Accounting for 44 per cent of all families accepted as homeless.362 Accounting for 10 per cent of all families containing children in England (FACS, 2005).363 Families headed by a lone male parent are unusual in the general population (under 1 per cent of families containing children), though they were more common among families accepted as homeless (4 per cent).220 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds• couples with children were far more likely to be workless than couple-headed families in the general population of England (47 per cent compared to 4 per cent). Regression analysis confirmed that, other things being equal10.14 364, a strong influence on whether a family accepted as homeless was workless at point of survey was household type (couple households were more likely to contain an adult in work (53 per cent) than were women lone parent households (28 per cent)). The other factors found to exert an independent effect were as follows:• age of adult respondent: families in which the adult respondent was aged under 25 were more likely to be workless (73 per cent) than families where the adult respondent was older (59 per cent). • the setting from which a homelessness application was made: families that had applied as homeless from managed forms of accommodation (e.g. hostels and B&B hostels) were more likely to be workless (82 per cent) than other families (62 per cent)365.• experience of temporary accommodation: families that had experienced a stay in temporary accommodation (of any length) were more likely (67 per cent) to be workless at point of survey than families who had not stayed in temporary accommodation at all (53 per cent). • long-term limiting illness: families in which the adult respondent had a long-term limiting illness or disability were more likely to be workless (73 per cent) than other families (61 per cent). Notably, there was no difference in levels of worklessness between those in 10.15 temporary or settled accommodation at point of survey. Changes in the working status of families since last settled accommodation Figure 10.1 summarises changes in the ‘working’ status of families accepted 10.16 as homeless since leaving their ‘last settled accommodation’366. Most families (74 per cent) had not seen any changes in their working status. 10.17 This group was composed of those families which contained no adults in paid work either at point of interview or in their last settled accommodation 364 Factors controlled for in this regression included: demographic characteristics; geographical variables; where applied from as homeless; current accommodation type; causes of homelessness; temporary accommodation experiences; current mental health problems; ever having had a drug or alcohol problem; long-term limiting illness or disability; educational qualifications.365 See Chapter 4.366 In all, 71 per cent of all families in Survey 1 had a last settled accommodation that fulfilled relevant criteria for this to be deemed a ‘valid’ comparison point for the purposes of this research, and 29 per cent did not. See Chapter 1 and Appendix 1 for a full explanation. All references to ‘changes since last settled accommodation’ in this and other chapters relate to those adult respondents/families who reported a ‘last settled accommodation’ that was such a valid comparison point.Chapter 10 Employment, income and expenditure 221(45 per cent), and those families that contained someone in paid work in both their last settled accommodation and at point of survey (29 per cent). However, 21 per cent of families had moved from a working to workless 10.18 status since leaving their last settled accommodation (i.e. they had an adult in paid work in their last settled accommodation but this was no longer the case by point of survey) (Figure 10.1). This was offset to only a small degree by 6 per cent of families having experienced the reverse. There was therefore a net 15 percentage point increase in the proportion of workless households amongst families accepted as homeless since they left their last settled accommodation.Figure 10.1: Changes in families’ working status since their last settled accommodation45%Adult(s) working inlast settledaccommodationbut not atcurrent addressAdult(s) working inlast settledaccommodationbut working atcurrent addressAdult(s) working inlast settledaccommodationand current addressWorkless in lastsettled accommodationand current address0%10%20%30%40%50%29%6%21%Source: Survey 1 Base: 1,344 families who had ‘last settled accommodation’ that was valid for comparison purposes. This change in the working profile of families accepted as homeless did not 10.19 appear to be explained by the departure of working adults from the relevant households (for example, as a result of relationship breakdown), as there was no association between changes in families’ working status and alterations in household composition.Likewise, there was no distinction between those families in temporary and 10.20 settled accommodation with regards to their likelihood of having moved from a working to workless status since leaving their last settled accommodation. 222 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsRegression analysis, which controlled for a wide range of variables10.21 367, detected no independent associations between any specific geographical, demographic, temporary accommodation experience or any other factors and the likelihood of families’ having become workless since leaving their last settled accommodation. This may suggest that it is the initial disruption of becoming homeless (experienced by all of these families) that leads to the rise in worklessness, rather than the specific nature of their temporary accommodation or other subsequent experiences.Barriers to work Data were collected on barriers to finding or seeking paid work among 10.22 the adults respondents in families accepted as homeless who were not in work. The reported barriers were compared to those reported by adults not in employment in families in the general population. As noted above, the difference between these two populations (i.e. the much higher number of women lone parents among the adults in families accepted as homeless) needs to be borne in mind. As Table 10.2 demonstrates, most barriers to work were reported at a higher 10.23 rate by adult respondents in families accepted as homeless than by adults in families in the general population, but these differences were usually quite marginal. Adult respondents in families accepted as homeless were less likely than adults in families in the general population to identify ‘not wanting to spend more time apart from children’ as a barrier to work368. Amongst families accepted as homeless, citing of this barrier was strongly influenced by having children aged under 5 (34 per cent of adult respondents who had children under 5 reported this as barrier, compared to 10 per cent who only had older children).367 Factors controlled for in this regression analysis included: geographical variables; demographic characteristics; causes of homelessness; where applied from as homeless; current mental health problems; ever had drug or alcohol problems; long-term limiting illness; current accommodation type; temporary accommodation experiences; changes in household composition (an adult member has left or joined household since last settled accommodation).368 This may reflect the stronger economic position of most families in the general population, which makes it financially easier to have one partner (usually the mother) at home with the children full time.Chapter 10 Employment, income and expenditure 223Table 10.2: Barriers to employment reported by adult respondents in families accepted as homeless and by adults in families containing children in general populationReasons Survey 1 adult respondentsAdults in families in England DifferenceNone, already looking 11% 5% +6%No child care available 11% 3% +8%Cannot afford child care 13% 8% +5%Don’t have the skills/ qualifications 7% 2% +3%Do not want to spend more time apart from children25% 39% -14%No work available 2% 1% +1%Would not be able to pay rent or mortgage3% 1% +2%Better off not working 5% 3% +2%Studying/ on a training course 8% 4% +4%Own illness/ disability 13% 10% +3%Transport problems 1% 1% 0%Child’s illness/ disability 3% 4% -1%Maternity leave/pregnancy 6% 2% +4%Other 2% 5% -3%Base 1,323 4,273 –Source: Survey 1 (all adult respondents not in work) and (FACS) (2005) (England only) (all adults not in work). Multiple responses were possible.We also asked a series of ‘homelessness specific’ questions on barriers to 10.24 seeking paid employment. Adult respondents not in work were very unlikely (less than 1 per cent) to report that ‘living in temporary accommodation’ or the ‘disruption caused by homelessness’ formed a barrier to seeking work. Related to this, it should also be noted from Table 10.2 that, despite the concerns articulated in previous research (see para 10.2), they were not notably more likely to report financial disincentives (‘better off not working’) than were adults not in work in the general population of families. 369369 Hoxhallari, L; Conolly, A. and Lyon, N. (2007) Families with Children in Britain: Findings from the 2005 Families and Children Study (FACS), DWP research report 424. 224 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThere was no distinction in the barriers to paid work reported by those adult 10.25 respondents still in temporary accommodation and those in settled housing at point of survey. Qualifications and literacy We asked adult respondents about their academic and other qualifications, and 10.26 compared this to the qualifications held by adults in families across England.Table 10.3 shows the highest academic qualification held by adult 10.27 respondents in families accepted as homeless compared to the highest qualification held by adults in families in the general population.370Table 10.3: Academic qualifications held by adult respondents in families accepted as homeless in comparison with adults in families in the general populationHighest academic qualificationAdult respondents (Survey 1)Adults in families in EnglandDifferenceGCSE(s) grade A-C or equivalent 35% 35% 0%A-levels or equivalent 6% 14% -8%Degree and/or higher degree 4% 27% -23%Base 2,053 6,500 –Source: Survey 1 and (FACS) (2005) (England only)As can be seen, qualification levels were much poorer amongst adult 10.28 respondents in families accepted as homeless than amongst the adults in families in the general population. In all, 76 per cent of adults in families in England have at least GCSE qualifications or equivalent, compared to only 45 per cent of adult respondents in families accepted as homeless. The main difference centred on the proportion of adult respondents who were graduates, which was much lower than the levels found among adults in families with children in England.370 Hoxhallari, L; Conolly, A. and Lyon, N. (2007) Families with Children in Britain: Findings from the 2005 Families and Children Study (FACS), Department for Work and Pensions research report 424. This is not a direct comparison because the FACS data covers the respondent adult and their partner (where present) whereas the Survey 1 data on homeless families covers only the adult respondent.Chapter 10 Employment, income and expenditure 225Vocational qualifications were also much less common among adult 10.29 respondents in families accepted as homeless than was the case for adults in families across England. More than one half of families contained an adult with one or more NVQ level qualifications in England during 2005 (54 per cent) 371, compared to just 24 per cent of adult respondents to Survey 1.Overall, just 10 per cent of families in England contained one or more adults 10.30 with no vocational or academic qualifications during 2005372. By contrast, over one third (36 per cent) of the adult respondents in families accepted as homeless had no such qualifications. Adult respondents were generally 10.31 unlikely to report that they had any difficulty in reading English in their daily life (90 per cent reported that their reading skills were ‘very good’ or ‘fairly good’). Only 5 per cent described their reading skills as ‘below average’, and 3 per cent described them as ‘poor’. Two per cent of respondents described themselves as ‘unable to read English’.Income, debt, and major expenditure This section reviews the income levels reported amongst families accepted 10.32 as homeless, and any changes in this since they left their last settled accommodation. It also considers the state benefits and lump sum loans and grants received by these families. It then moves on to analyse the debts faced by families accepted as homeless, and in particular any debts which are overdue. It finishes by examining any major expenditure incurred by these families associated with having to leave their last settled accommodation.IncomeCurrent household income among families accepted as homelessTable 10.4 shows the self-reported average and median weekly income 10.33 (exclusive of Housing Benefit)373 of families accepted as homeless at point of survey. As can be seen, couples with children and families in settled housing tended to receive larger average and median incomes.371 FACS (2005) CHP analysis.372 FACS (2005) CHP analysis.373 Housing Benefit was excluded from all income analysis because data on this benefit tends to be highly unreliable as it is often paid direct to landlords rather than to recipients.226 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 10.4: Total family income per week (excludes Housing Benefit)Household type Average Median BaseWoman lone parentIn temporary accommodation In settled housingAll£149£177£164£126£147£137744659 1,403CoupleIn temporary accommodation In settled housing All£219£249£233£199£212£201386264650AllIn temporary accommodation In settled housingAll£173£196£186£142£160£1501,130 9232,053Source: Survey 1 Data refer to net household income. Regression analysis showed that, other things being equal10.34 374, the following factors had an independent effect on the likelihood of a family receiving more than the median weekly income (of £150 per week, see Table 10.4):• working status of household: approximately three-quarters (78 per cent) of families containing someone in employment had a weekly income above the median, compared to just 32 per cent of workless families.• household size: approximately three-quarters of larger families (72 per cent), with four or more members, had incomes above the weekly median, compared to 39 per cent of other families (a finding that almost certainly reflected their eligibility for larger benefit payments).• age of adult respondent: families in which the adult respondent was aged under 25 were less likely to have incomes above the weekly median than other families (32 per cent, compared to 59 per cent).Household incomes for families accepted as homeless in comparison with the general population of families in England Table 10.5 compares average and median net family income per week for 10.35 families accepted as homeless with that of equivalent household types in the general population of England. As can be seen, incomes among families accepted as homeless were generally lower than for families in the general population. The differences between couple households accepted as homeless and couples in the general population were particularly 374 Factors controlled for in this regression analysis included: demographic characteristics; geographical variables; current accommodation type; temporary accommodation experiences; whether household was workless; and child and adult vulnerability clusters.Chapter 10 Employment, income and expenditure 227pronounced. This is consistent with the far greater rates of worklessness amongst couple households accepted as homeless than their equivalents in the general population, as noted earlier (see para 10.13).Table 10.5: Total family income per week compared with families in general population (excludes Housing Benefit)Household type Average weekly incomeMedian weekly incomeBaseSurvey 1 women lone parents £164 £137 1,403Women lone parents in England £264 £233 5,079Difference -£100 -£96 –Survey 1 couples with children £233 £201 650Couples with children in England £635 £546 5,079Difference -£402 -£345 –All Survey 1 families £186 £150 2,053All families with children in England £537 £461 6,990Difference -£351 -£311 –Source: Survey 1 and (FACS) (2005) (England only) (FACS data include all sources of weekly income with the exception of Housing Benefit which has been excluded) (CHP analysis.) Data refer to net household income.375FACS (2005) data indicate that the bottom 20 per cent of families in England 10.36 had a net weekly income, excluding Housing Benefit, of £231 or under during 2005376. Overall, 74 per cent of families accepted as homeless had the same weekly net income of £231 or under a week. Changes in income since last settled accommodation One third of adult respondents (32 per cent) reported that their family’s 10.37 weekly income was higher at point of survey than it had been in their last settled accommodation, 27 per cent reported that it was now lower, and 41 per cent reported that it had remained the same. There was no association between increases or decreases in family income 10.38 and whether a family was in temporary or settled accommodation at point of interview. Likewise, there was no evidence that particular experiences of temporary accommodation were associated with increases or decreases in family income. 375 In previous waves of FACS, data for families where mothers or partners were self-employed was excluded from the income tables. As from FACS 2005, families with a self-employed parent have been included.376 CHP analysis. Figures exclude Housing Benefit.228 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsRegression analysis suggested that, other things being equal10.39 377, the following factors had an independent effect on the likelihood of increases in a families’ income since last settled accommodation378: • age of adult respondent: adult respondents aged under 25 were more likely to report an increase in household income since their last settled home (42 per cent compared to 26 per cent of other respondents). This may relate to increased benefit entitlement as these respondents get older and/or have children. • cause of homelessness: those for whom relationship breakdown was a cause of homeless were less likely to report an increase in household income (20 per cent compared to 35 per cent of other adult respondents). • deteriorations in general health of adult respondent: adult respondents whose general health had deteriorated since their last settled accommodation were also less likely to report improvements in household income (20 per cent compared to 37 per cent of other adult respondents).Reductions in family income were found by regression analysis to be 10.40 independently associated with the following factors, other things being equal379:• becoming a workless household: most families that had moved from working to workless status since their last settled accommodation reported a deterioration in household income (65 per cent, compared to 34 per cent of other families). This was the strongest independent influence on reductions in family income.• departure of an adult from household: most adult respondents who reported that an adult (usually their partner) had left their household since their last settled accommodation also reported a deterioration in their income (63 per cent, compared to 34 per cent of other adult respondents).• deteriorations in the general health of adult respondent: where adult respondents’ health was self-reported to have worsened since last settled accommodation, decreases in family income were also more likely (58 per cent compared to 38 per cent of other adult respondents). 377 Factors controlled for in this regression analysis included: changes in families’ working status; whether family currently workless; changes in adult respondents’ health; changes in household composition (gained or lost an adult); demographic characteristics; geographical variables; where applied from as homeless; current accommodation type; temporary accommodation experiences; causes of homelessness.378 There also appeared to be an independent (positive) association with households moving from a workless to a working status, but the numbers to whom this was relevant (6 per cent of all families accepted as homeless) were too small to make this finding statistically robust.379 Factors controlled for in this regression analysis included: changes in families’ working status; whether family currently workless; changes in adult respondents’ health; changes in household composition (gained or lost an adult); demographic characteristics; geographical variables; where applied from as homeless; current accommodation type; temporary accommodation experiences; causes of homelessness.Chapter 10 Employment, income and expenditure 229Benefits and tax credits receivedTable 10.6 provides an overview of the main forms of state benefits and 10.41 tax credits received by families accepted as homeless at point of survey, and compares this to receipt of these payments amongst families with children in the general population in 2005380. The two main payments linked to the presence of children in a household, Child 10.42 Benefit (a universal benefit for households with dependent children) and Child Tax Credit (a means-tested tax credit, but with a relatively high upper income limit), were claimed by a substantial proportion of both families accepted as homeless and families in the general population. Child Tax Credit was claimed at a somewhat higher rate by families accepted as homeless; a finding that is consistent with their generally low incomes as noted above (see para 10.36). Table 10.6: Benefits received by families accepted as homeless and families in general populationFamilies accepted as homeless (Survey 1)All families with children in England DifferenceChild Benefit 94% 91% +3%Child Tax Credit 69% 59% +10%Income Support 53%* 13% +40%Working Tax Credit 25% 19% +6%Jobseeker’s Allowance 7% 1% +6%Disability Living/Attendance Allowance6% 6% 0%Incapacity Benefit 5% 4% +1%Base 2,039 6,500 – Sources: Survey 1 and (FACS) (2005) (England only). CHP analysis *7 per cent of families accepted as homeless in receipt of Income Support received Disability Premium. The most notable difference between families accepted as homeless and 10.43 those in the general population was the much higher rate at which the former were in receipt of Income Support (53 per cent as compared with 13 per cent). However, this must be seen in the context of 42 per cent of all lone parent households in England, and 57 per cent of those with a child under 5, being in receipt of Income Support in 2005381. 380 Again, Housing Benefit is excluded from this analysis because data on this derived from recipients tends to be very unreliable (see footnote 373 above).381 FACS (2005) (England only) CHP analysis.230 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsWhile the differences were much narrower, families accepted as homeless 10.44 were also more likely to be in receipt of Working Tax Credit and Jobseeker’s Allowance than were families in the general population. Again, all of these findings are consistent with the relatively low incomes and low rates of paid work among families accepted as homeless reported above (see para 10.36).Patterns of benefit receipt did not vary according to whether a family was in 10.45 settled housing or in temporary accommodation at point of survey. Table 10.7 shows the proportion of families accepted as homeless who 10.46 had received benefits in the form of grants and loans since they applied as homeless. As can be seen, families who were in settled housing at point of survey were more likely to have received a Community Care Grant; this grant was in all likelihood provided to them in order to buy furniture, white goods and other household items for their new settled home. The take-up of Social Fund Crisis Loans and Sure Start Maternity Grants was similar amongst families in temporary accommodation and those in settled housing at point of survey, but the latter were slightly more likely to have received a Social Fund Budgeting Loan.Table 10.7: Grants and loans received by families accepted as homeless, by whether in temporary or settled housingTA Settled housingAll Community Care Grant 12% 29% 21%Sure Start Maternity Grant 20% 20% 20%Social Fund Budgeting Loan 15% 22% 19%Social Fund Crisis Loan 12% 13% 13%Social Fund Loan (maternity expenses) 2% 2% 2%Base 1,118 921 2,039Source: Survey 1 Families in the general population of England were much less likely than 10.47 families accepted as homeless to have received a Community Care Grant or a Social Fund loan. Only 2 per cent of families with children in England received a Community Care Grant during 2005, compared to 21 per cent of families accepted as homeless (since their application as homeless)382. Only 7 per cent of families with children in England received a Social Fund Crisis Loan or Budgeting Loan in 2005, compared to 27 per cent of families accepted as 382 FACS (2005) CHP analysis.Chapter 10 Employment, income and expenditure 231homeless (since their application as homeless)383. Again, as these grants and loans are subject to means-testing, these figures and those in Table 10.7 are indicative of the low incomes of many families accepted as homeless. One in ten families accepted as homeless (11 per cent) received regular child 10.48 maintenance payments from a previous partner, and 6 per cent of all families accepted as homeless were given regular financial help by friends or family. DebtTable 10.8 shows the outstanding loans held by families accepted as 10.49 homeless, and the proportion of families that had fallen behind with their repayment of these various forms of debt. As can be seen, the most commonly held debts in families accepted as 10.50 homeless were Social Fund loans (24 per cent) or, less frequently, bank overdrafts (15 per cent), bank or building society loan (12 per cent), or money borrowed on a credit/store card (11 per cent). Overall, 54 per cent of families accepted as homeless possessed at least one of these types of outstanding loan (Table 10.8). Table 10.8: Outstanding loans and repayment problems among families accepted as homelessRecipient familiesFallen behind with paymentsSocial Fund Loans (Crisis or Budgeting) 24% 2%Agreed overdraft from a bank 15% 4%Bank or building society loan (excludes mortgages)12% 5%Money borrowed on a credit/store card 11% 5%Money owed to mail order/catalogue company 8% 3%Loan collected from you in your home 7% 3%Hire-Purchase agreement 6% 2%Loan from family or friends 5% 2%Other 2% <1%Any of the above forms of debt 54% 17%Base 2,047 2,047Source: Survey 1 383 Figures exclude grants and loans for maternity. Source: FACS (2005) CHP analysis.232 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 10.8 demonstrates that the proportion of all families accepted as 10.51 homeless who had fallen behind with repayments on any particular type of loan was low. However, one third (32 per cent) of families who reported having one or more loans said that they were behind with the repayment of at least one of these loans; a group that represented 17 per cent of all families accepted as homeless. Table 10.9 shows the regular utility bills and housing-related costs which 10.52 families accepted as homeless reported being responsible for at point of survey. Responsibility for at least some regular bills was near-universal, with most families making regular payments for power, heating and water. However, large numbers of adult respondents did not report that their families were responsible for rent and council tax payments384. One third (35 per cent) of all families accepted as homeless had overdue 10.53 payments on one or more of their regular utility and/or housing-related bills. Water charges and rent payments were the most frequently cited by families as overdue (each were reported as overdue by 13 per cent of all families accepted as homeless). Table 10.9: Regular utility and housing-related bills paid by families accepted as homelessFamilies with relevant responsibility Fallen behind with paymentsElectricity 92% 7%Phone/mobile phone 89% 6%Gas 84% 8%Water 80% 13%Rent 53% 13%Council Tax 39% 9%Anything else requiring regular payments 42% 7%Any of the above payments 97% 35%Base 2,047 2,047Source: Survey 1384 This will be partly explained by direct payment of Housing Benefit to landlords meaning that some adult respondents will be unaware that their families are in fact paying rent. As noted above, this is why Housing Benefit was excluded from all income and benefits analysis. Likewise, the operation of Council Tax Benefit, as effectively a ‘rebate’ system, means that adult respondents may not always be aware that they are in fact paying Council Tax with the help of this benefit.Chapter 10 Employment, income and expenditure 233The number of regular bills a family was responsible for was higher if it was 10.54 in settled housing at point of survey than was the case in some forms of temporary accommodation. On average, adult respondents in both settled housing and in self-contained temporary accommodation reported being responsible for five regular bills; the average was three regular bills for those in hostels or B&B hotels and for those staying temporarily with friends or relatives. Families staying with friends or relatives were less likely to report overdue bills than families in other forms of temporary accommodation, or in settled housing, at point of survey. Overall, two in five (41 per cent) of all families accepted as homeless were 10.55 behind with at least one loan repayment and/or a regular bill. However, it should be noted that multiple types of overdue bills/debts were relatively uncommon. Thus, this 41 per cent of families accepted as homeless who had at least one such overdue payment was comprised of: 19 per cent of all families accepted as homeless who had one such overdue payment; 10 per cent who had two types of overdue payment; and 12 per cent who had three or more types of overdue payments. The presence or absence of overdue loans/bills was not associated with 10.56 whether a family was in settled housing or temporary accommodation at point of survey, nor were any particular temporary accommodation experiences associated with overdue debts/bills. Likewise, there was no association with household types, or with the working or workless status of the family. However, regression analysis indicated that, other things being equal10.57 385, the following factors had an independent effect on the likelihood of a family being behind with at least one loan repayment and/or a regular bill: • being accepted as homeless in the South: families accepted in this broad region were more likely to report one or more overdue bills or loan repayments (50 per cent) than families accepted elsewhere (37 per cent).• being a former asylum seeker: families in which the adult respondent was a former asylum seeker were less likely to report overdue debts or bills (29 per cent) than other families (43 per cent).Major expenditureHouseholds were asked if they had had to buy or pay for a range of items 10.58 because they had had to leave them behind or get rid of them when they 385 Factors controlled for in this regression included: demographic characteristics; geographical variables; where applied from as homeless; current accommodation type; causes of homelessness; temporary accommodation experiences; working or workless status of household; whether family has become ‘workless’ since last settled accommodation; changes in household composition (departure of an adult member of household since last settled accommodation).234 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsleft their last settled accommodation. Table 10.10 shows the percentages of households reporting such expenditure, by whether in temporary or settled housing at point of survey. As Table 10.10 indicates, around three in five of all families accepted as 10.59 homeless (60 per cent) had had to purchase white goods, and around half had incurred costs on curtains/blinds (53 per cent), carpets (50 per cent), furniture (57 per cent), and repairing/redecorating accommodation (45 per cent). Families had had to pay for the costs of moving their furniture and belongings from their last settled accommodation in 41 per cent of cases. The numbers reporting all of the other specified forms of expenditure were smaller. Table 10.10: Expenditure on specific items as a result of leaving last settled accommodationItem TA Settled housingAllWhite goods 46% 69% 60%Furniture 44% 66% 57%Curtains/blinds 33% 67% 53%Carpets/floor coverings 24% 68% 50%Repairing/ redecorating accommodation 24% 59% 45%Costs of moving furniture/ belongings 39% 42% 41%Replacing school uniforms 14% 19% 17%Replacing adults clothes/ shoes 13% 13% 13%Replacing children’s clothes/shoes 12% 12% 12%Replacing children’s toys 13% 11% 12%Connection charges for utilities 10% 10% 10%Other major items 5% 7% 6%Base 678 657 1,335Source: Survey 1 Base: all adult respondents with a ‘last settled accommodation’ that was valid for comparison purposes. As might be expected, given that they were in the process of setting up 10.60 a long-term home, families in settled housing were much more likely to have made many types of expenditure than those still in temporary accommodation. This was especially true with regards to white goods, Chapter 10 Employment, income and expenditure 235furniture, curtains/blinds, carpets/floor coverings, and repair/redecoration386. There was, on the other hand, little variation between those in temporary and settled accommodation with regards to spending money on replacing items such as children’s clothes, shoes, toys or school uniforms, and adults’ clothes and shoes.Overall, 29 per cent of families had spent at least £200 on the items listed in 10.61 Table 10.10. Regression analysis indicated that, other things being equal387, the only independent effect on the likelihood of incurring this level of expenditure was whether or not the family had moved into settled housing. One third (34 per cent) of families who had moved into settled housing reported spending at least £200 on the items listed in Table 10.10, as compared to one quarter (25 per cent) of those who were still in temporary accommodation at point of survey. Financial exclusionOverall, 15 per cent of adult respondents reported that they did not have 10.62 a current account with a bank, building society or other organisation, as compared with 7 per cent of parents with dependent children in the general population, who lacked a current or savings account in 2005388. Women lone parents were more likely to lack a current account than other 10.63 adult respondents (17 per cent compared to 10 per cent). A similar pattern was reported by FACS (2005) in relation to families with children in England (15 per cent of women lone parents as compared to 5 per cent of other families did not have a current or savings account)389. Overall financial circumstances This section reviews adult respondents’ overall self-assessment of how well 10.64 their family was coping financially, and any changes in this since they left their last settled accommodation.386 As noted above (see para 10.46), those families in settled housing were also more likely than those still in temporary accommodation to have received a Community Care Grant, most probably to help with these types of costs.387 Factors controlled for in this regression included: demographic characteristics; geographical variables; where applied from as homeless; current accommodation type; causes of homelessness; temporary accommodation experiences; working or workless status of household; whether family has become ‘workless’ since last settled accommodation; changes in household composition (loss of an adult member of household since last settled accommodation).388 These comparisons are approximate. Adult respondents to Survey 1 were asked ‘Do you have a current account with a bank, building society or other organisation?’ whereas respondents to the FACS (2005) survey were asked ‘Do [you/you and your partner] have any current accounts or savings accounts? This could be in your own name only, or held jointly with someone else.’389 FACS (2005.) CHP analysis236 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsCurrent financial circumstances Table 10.11 shows adult respondents’ overall financial self-assessment at 10.65 point of survey, alongside comparative national data.This reveals that adult respondents, on the whole, considered themselves to 10.66 be in greater financial difficulty than did parents with dependent children in the general population. Only 22 per cent of adult respondents considered their family to be managing ‘very well’ or ‘quite well’, compared with 58 per cent of parents in England. Correspondingly, the proportion of adult respondents considering themselves to ‘not manage very well’, ‘have some financial difficulties’ or to be ‘in deep financial trouble’ (at 34 per cent) was around three times greater than that of the wider population of parents (10 per cent). Table 10.11: How families accepted as homeless were managing financially compared with families in the general population Survey 1 Families in EnglandDifference Manage very well 5% 24% -19%Manage quite well 17% 34% -17%Get by all right 45% 31% +14%Don’t manage very well 14% 3% +11%Have some financial difficulties 15% 5% +10%In deep financial trouble 5% 2% +3%Total 100% 100% –Base 2,031 6,498 –Source: Survey 1 and (FACS) (2005) (England only) CHP analysis.As can be seen from Table 10.12 below, adult respondents in settled housing 10.67 were less likely to report current financial problems (27 per cent)390, than adult respondents in temporary accommodation (41 per cent).390 i.e. ranging from ‘don’t manage very well’ through to ‘deep financial trouble’.Chapter 10 Employment, income and expenditure 237Table 10.12: How families accepted a homeless were managing financially at point of survey, by whether in temporary or settled accommodationTA Settled housing All families Manage very well 4% 6% 5%Manage quite well 14% 21% 17%Get by all right 42% 47% 45%Don’t manage very well 16% 12% 14%Have some financial difficulties 19% 11% 15%In deep financial trouble 6% 4% 5%Total 100% 100% 100%Base 1,113 918 2,031Source: Survey 1 Regression analysis confirmed that, other things being equal10.68 391 current financial difficulties were more common among families in temporary than in settled accommodation at point of survey. In addition, the following factors were identified as having an independent influence on the likelihood of self-reporting current financial difficulties: • having overdue loans/bills: adult respondents who reported one or more overdue loans or bills were more than twice as likely as other respondents to say that they had current financial difficulties (48 per cent as compared to 23 per cent). This was the strongest independent effect. • being a former asylum seeker: adult respondents who had claimed asylum at some point were more likely to report current financial difficulties than other adult respondents (46 per cent compared to 31 per cent).• being accepted in London: families accepted in London were more likely than those accepted elsewhere to report current financial difficulties (44 per cent compared to 29 per cent).• current mental health problems: adult respondents with current mental health problems reported having current financial difficulties more often than other adult respondents (43 per cent compared to 29 per cent).• age: young adult respondents (under 25) were less likely to report current financial difficulties (28 per cent) than was the case for older adult respondents (36 per cent). 391 Factors controlled for in this regression included: demographic characteristics; geographical variables; where applied from as homeless; current accommodation type; causes of homelessness; temporary accommodation experiences; working or workless status of household; whether family has become ‘workless’ since last settled accommodation; changes in household composition (loss of an adult member of household since last settled accommodation); current mental health problems; ever having had a drug or alcohol problem; whether had any overdue loans/bills.238 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsChanges in overall financial circumstances since last settled accommodationFamilies were asked to report on their financial position in their last settled 10.69 accommodation as compared to their financial position at point of survey. Table 10.13 summarises the responses to this question. Table 10.13: How families accepted as homeless were managing financially in their last settled home compared to their accommodation at point of surveyFinancial situation in last settled accommodationCurrent accommodation at point of survey TA Settled housing All families at point of surveyManage very well 16% 3% 7% 5%Manage quite well 30% 14% 21% 18%Get by all right 33% 43% 47% 45%Don’t manage very well10% 14% 11% 12%Have some financial difficulties7% 19% 12% 15%In deep financial trouble5% 6% 3% 4%Total 100% 100% 100% 100%Base 1,334 679 655 1,334Source: Survey 1 Base: all adult respondents with a ‘last settled accommodation’ that was valid for comparison purposes.As can be seen, there was an overall deterioration in financial circumstances 10.70 compared to last settled accommodation for families in temporary accommodation at point of survey. In their last settled accommodation, 79 per cent of families were at least ‘getting by all right’, or were managing ‘quite well’ or ‘very well’. This was only true for 60 per cent of families in temporary accommodation at point of survey. For families in settled housing at point of survey, there was some slippage compared to the reported situation in last settled accommodation, in that they were half as likely to report doing ‘very well’, and less likely to report managing ‘quite well’, than was reported for families in their last settled accommodation (Table 10.13)392.392 Please note that there were no differences in the financial circumstances reported in last settled accommodation between those in temporary accommodation and those in settled housing at point of survey.Chapter 10 Employment, income and expenditure 239Overall, 47 per cent of families reported some deterioration in their financial 10.71 circumstances between their last settled accommodation and their current (temporary or settled) housing. Regression analysis indicated that, other things being equal393, the following factors had an independent influence on the likelihood of reporting a deterioration in financial circumstances: • becoming a workless household: two-thirds (67 per cent) of families that had had an adult in employment in their last settled accommodation but no longer did at point of survey reported a deterioration in their financial situation, compared to 42 per cent of other families. This was the strongest independent effect.• relationship breakdown as a cause of homelessness: adult respondents who reported relationship breakdown as a cause of their homelessness were more likely to report a deterioration in their family’s financial situation than other adult respondents (55 per cent compared to 42 per cent). • being accepted in the North and Midlands: families accepted in this broad region were less likely to report a deterioration in their financial situation than those accepted elsewhere (39 per cent compared to 54 per cent).This regression analysis indicated that, once other factors were taken into 10.72 account, there was no relationship between a household being in temporary or settled accommodation at point of survey and their likelihood of reporting a deterioration in their financial situation since their last settled home. Improvements in a families’ financial situation since their last settled 10.73 accommodation were, as can be seen from Table 10.13, more unusual than deteriorations. Overall, 18 per cent of adult respondents reported that their families’ financial situation had improved since their last settled accommodation. Regression analysis indicated that, when other factors were held constant394, families who had stayed at least six months in temporary accommodation were less likely to report an improvement in their financial circumstances. Those who had experienced a temporary accommodation stay of this length reported an improvement in 14 per cent of cases, compared to 21 per cent of other families. No other independent effects were found.393 Factors controlled for in this regression included: demographic characteristics; geographical variables; where applied from as homeless; current accommodation type; causes of homelessness; temporary accommodation experiences; working or workless status of household; whether family has become workless since last settled accommodation; changes in household composition (loss of an adult member of household since last settled accommodation); current mental health problems; ever having had a drug or alcohol problem.394 Factors controlled for in this regression included: demographic characteristics; geographical variables; where applied from as homeless; current accommodation type; causes of homelessness; temporary accommodation experiences; working or workless status of household; whether family has become workless since last settled accommodation; changes in household composition (loss of an adult member of household since last settled accommodation); current mental health problems; ever having had a drug or alcohol problem.240 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsGiven that, in overall terms, the incomes of families had not declined 10.74 since leaving their last settled accommodation (see para 10.37), it appears that changes in expenditure must explain these (net) increases in financial difficulties amongst families accepted as homeless. Families in temporary accommodation for more than one year This section reviews the evidence on employment, income and expenditure 10.75 amongst families who had stayed in temporary accommodation for over one year (Survey 4 families), and compares this to the findings on Survey 1 families who had generally spent a shorter period in temporary accommodation.Economic status, worklessness and barriers to work amongst Survey 4 families As can be seen in Table 10.14, adults in Survey 4 families had a similar profile 10.76 with regards to economic status as adults in Survey 1 families, though they were somewhat more likely to be in training or education and slightly less likely to be in employment. Table 10.14: Economic status of all adults in Survey 4 families compared with all adults in Survey 1 familiesStatus Adults in Survey 4 families Adults in Survey 1 families DifferenceIn paid work (including self-employed) 22% 29% -7%Looking after the home or family 37% 38% -1%In education or training 18% 8% +10%Unemployed and seeking work 10% 12% -2%Sick or disabled 6% 7% -1%Other (not in employment) 7% 6% +1%Total 100% 100% –Base 1,020 3,015 –Source: Survey 4 and Survey 1 Base: for Survey 4 is 1,020 adults across 571 households; for Survey 1 is 3,015 adults across 2,053 households.As with Survey 1, adults in Survey 4 families were much less likely to be in 10.77 paid work than adults in families in the general population of England (22 per cent of adults in Survey 4 families were in work compared to 70 per cent of adults in families in the general population, see Table 10.1). Chapter 10 Employment, income and expenditure 241One third of Survey 4 families contained one adult in work (32 per cent), with 10.78 another 4 per cent containing two or more adults in work. This meant that most (64 per cent) Survey 4 families were workless. These figures are almost identical to those found among Survey 1 families (see para 10.11 above). As was the case among Survey 1 families, when effects associated with the 10.79 household composition of Survey 4 families were taken into account, these families were still more likely to be workless than their equivalents in the general population395: • the great majority (86 per cent) of Survey 4 families headed by women lone parents with children aged under five were workless, compared to 78 per cent of similar households in Survey 1, and 55 per cent of this group in the general population.• Survey 4 families headed by women lone parents whose children were all aged over five were less likely to be workless than those with younger children (66 per cent were workless), a similar figure to that found for this type of household in Survey 1 (61 per cent), but again much higher than for this group in the general population (32 per cent were workless in 2005). • couple-headed households in Survey 4 were less likely to be workless than lone women parents (44 per cent were workless), as was also in the case with Survey 1 couples (47 per cent were workless), though their rates of worklessness were very much higher than among couples with children in the general population (4 per cent were workless in 2005). Survey 4 adult respondents not in work tended to identify very similar barriers 10.80 to work as Survey 1 adult respondents who were not in work, though they were somewhat more likely to mention not being able to afford childcare and not being able to pay the rent (see Table 10.15). They were as unlikely as Survey 1 adult respondents to report that homelessness-specific issues, such as ‘living in temporary accommodation’ (1 per cent) or the ‘disruption caused by homelessness’ (1 per cent), formed a barrier to seeking work.395 The comparisons used here are drawn from FACS (2005) (England only) CHP analysis.242 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 10.15: Barriers to employment reported by adult respondents in Survey 4 families and Survey 1 families Reasons Survey 4 adult respondents Survey 1 adult respondents DifferenceNone, already looking 11% 11% 0%No child care available 14% 11% +3%Cannot afford child care 19% 13% +6%Don’t have the skills/ qualifications 10% 7% +3%Do not want to spend more time apart from children28% 25% +3%No work available 3% 2% +1%Would not be able to pay rent or mortgage10% 3% +7%Better off not working 9% 5% +4%Studying/ on a training course 9% 8% +1%Own illness/ disability 13% 13% 0%Transport problems 2% 1% +1%Child’s illness/ disability 4% 3% +1%Maternity leave/pregnancy 3% 6% -3%Other 2% 2% 0%Base 365 1,323 –Source: Survey 4 and Survey 1 (all adult respondents not in work) Multiple responses were possible.Qualifications amongst Survey 4 adult respondents Table 10.16 reports the highest academic qualification held by Survey 4 10.81 adult respondents as compared to the highest qualification held by adult respondents to Survey 1 and by adults in families across England. It shows that Survey 4 adult respondents tended, like Survey 1 adult respondents, to have an overall lower level of academic qualifications than adults in families in the general population, particularly with respect to degree level qualifications.Chapter 10 Employment, income and expenditure 243Table 10.16: Qualifications held by Survey 4 adults, in comparison with Survey 1 adult respondents and adults in families in the general populationHighest educational qualification held in householdSurvey 4 adult respondents Survey 1 adult respondents Adults in families in EnglandGCSE(s) grade A-C or equivalent 31% 35% 35%A levels or equivalent 4% 6% 14%Degree and/or higher degree 6% 4% 27%Base 571 2,053 6, 500Source: Survey 1 and (FACS) (2005) (England only).Survey 4 adult respondents also mirrored Survey 1 adult respondents in 10.82 possessing NVQ or equivalent vocational qualifications at a lower rate than that found in the general population (23 per cent held them, compared to 24 per cent among Survey 1 respondents, and 54 per cent of adults in families in England397). Overall, 41 per cent of Survey 4 adult respondents lacked any formal 10.83 qualifications, (this compared with 36 per cent of Survey 1 adults respondents, and 10 per cent of adults in families with children in England398). Income of Survey 4 families Table 10.17 shows the self-reported average and median weekly income 10.84 (exclusive of Housing Benefit) of families accepted as homeless. Both the average and median figures were slightly higher for Survey 4 than Survey 1 families (see Table 10.4). This difference is explicable through the typically larger household size (and thus higher benefit entitlements) found among Survey 4 families relative to Survey 1 families (see para 2.52). Table 10.17: Total family income per week for Survey 4 families (excludes Housing Benefit)Household type Average Median BaseWoman lone parent £183 £155 304Couple £266 £250 181All £213 £190 485Source: Survey 1 Data refer to net household income. 396 Hoxhallari, L; Conolly, A. and Lyon, N. (2007) Families with Children in Britain: Findings from the 2005 Families and Children Study (FACS) Department for Work and Pensions research report 424. This is not a direct comparison because the FACS data covers the respondent adult and their partner (where present) whereas the Survey 4 and Survey 1 data only covers the adult respondent.397 FACS (2005) CHP analysis.398 FACS (2005) CHP analysis.244 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsIncomes were markedly lower among Survey 4 couples with children than 10.85 was the case for couples in the general population (see Table 10.5). Weekly average net incomes of £266 (excluding Housing Benefit) were £369 less than the average for couples with children in the general population. As was the case for Survey 1 respondents, differences with women lone parents families in the general population were less stark, though Survey 4 women lone parents’ weekly average net income (excluding Housing Benefit) of £183 was £81 less than the average for women lone parent families in the general population (see Table 10.5). Benefits received by Survey 4 families Table 10.18 shows that Survey 4 families were claiming a very similar range 10.86 of benefits to Survey 1 families. Child Tax Credit was less commonly claimed, which seems a surprising finding given that Survey 4 families had more children than Survey 1 families overall (see para 2.55), but the survey did not collect data that would allow this to be explored further. Table 10.18: Benefits received by Survey 4 families, compared to Survey 1 familiesSurvey 4 familiesSurvey 1 familiesDifferenceChild Benefit 94% 94% 0%Child Tax Credit 52% 69% -17%Income Support 50% 53% -3%Working Tax Credit 23% 25% -2%Jobseeker’s Allowance 8% 7% +1%Disability Living/Attendance Allowance9% 6% +3%Incapacity Benefit 4% 5% -1%Base 571 2,039 – Sources: Survey 4 and Survey 1 As Table 10.19 indicates, Survey 4 families were less likely than Survey 1 10.87 families to have received a Community Care Grant (6 per cent compared to 21 per cent). This seems to have been the result of their not yet being in settled housing, as the bulk of Community Care Grants received by Survey 1 households had been for families in settled housing. Social Fund Budgeting Loans (14 per cent) and Crisis Loans (13 per cent) had been received by Survey 4 families at a similar rate to Survey 1 families (19 per cent and 13 per cent respectively) (Table 10.19).Chapter 10 Employment, income and expenditure 245Table 10.19: Grants and loans received by Survey 4 families, compared to Survey 1 familiesSurvey 4 Survey 1 DifferenceCommunity Care Grant 6% 21% -15%Sure Start Maternity Grant 20% 20% 0%Social Fund Budgeting Loan 14% 19% -5%Social Fund Crisis Loan 13% 13% 0%Social Fund Loan (maternity expenses) 2% 2%Base 571 2,039 –Sources: Survey 4 and Survey 1Overall financial circumstances of Survey 4 familiesSurvey 4 families had a very similar profile to Survey 1 families in terms of the 10.88 kinds of utility bills and loans that they had. Overall, 44 per cent of Survey 4 families were behind with at least one loan repayment and/or a regular bill, a very similar level to that found among Survey 1 families (41 per cent)399. At 15 per cent, the proportion of Survey 4 families lacking access to a bank or 10.89 building society current account was identical to that in Survey 1.Table 10.20 shows how Survey 4 families reported they were managing 10.90 financially at the point of survey compared to Survey 1 families. As can be seen, Survey 4 families were more likely than Survey 1 families to report that their financial situation was difficult (49 per cent reported they were ‘not managing very well’, ‘had some financial trouble’ or were ‘in deep financial trouble’, compared to 34 per cent of Survey 1 respondents).399 As with Survey 1, multiple types of overdue debts/bills were relatively uncommon among Survey 4 families. Thus, the 44 per cent of all families accepted as homeless who had at least one such overdue payment included 19 per cent of all families who had one overdue type of payment, 8 per cent who had two types of overdue payments, and 17 per cent with three or more types of overdue payments. These figures are all very close to those for Survey 1 (see para 10.55).246 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 10.20: How Survey 4 families were managing financially at point of survey, compared to Survey 1 familiesSurvey 4 families Survey 1 families Difference Manage very well 2% 5% -3%Manage quite well 11% 17% -6%Get by all right 39% 45% -6%Don’t manage very well 19% 14% +5%Have some financial difficulties 24% 15% +9%In deep financial trouble 6% 5% +1%Total 100% 100% –Base 571 2,031 –Source: Survey 4 and Survey 1 The pattern shown in Table 10.20 for Survey 4 families appeared to have 10.91 more in common with the patterns found among Survey 1 families still in temporary accommodation at point of survey, rather than those in settled housing (see Table 10.12). As Survey 4 families were all in self-contained temporary accommodation by point of survey, Survey 1 responses for families in self-contained temporary accommodation were examined to see if this interpretation bore scrutiny. Survey 1 families in self-contained temporary accommodation were more 10.92 likely to report difficulties in managing financially (42 per cent) than were Survey 1 families either in settled housing or in other forms of temporary accommodation (28 per cent). The similarities between Survey 4 families, 49 per cent of whom reported difficulty managing, and Survey 1 families in self-contained temporary accommodation, 42 per cent of whom reported difficulty managing, do seem to indicate that financial pressures were most common in self-contained temporary accommodation400. Further investigation revealed this to be partly but not fully explained by the concentration of self-contained temporary accommodation in London, where families were generally more likely to report that they were struggling financially (see para 10.68). 400 Young people accepted as homeless 16-17 year olds who were living in self-contained temporary accommodation were also more likely than other young people to report financial difficulties (see para 12.131).Chapter 10 Employment, income and expenditure 247Regression analysis on Survey 4 could not control for the effects of self-10.93 contained temporary accommodation on the financial situation of Survey 4 families because all of these families were in this form of accommodation. However, it did indicate that, other things being equal401, the following factors had an independent effect on the likelihood of self-reporting current financial difficulties: • having overdue loans/bills: Survey 4 adult respondents who reported one or more overdue loans or bills were more likely than other adult respondents to say that they had current financial difficulties (57 per cent as compared to 42 per cent). A similar relationship was found among Survey 1 adult respondents (see para 10.68). • being a former asylum seeker: Survey 4 adult respondents who had claimed asylum at some point were more likely to report current financial difficulties than other adult respondents (58 per cent compared to 44 per cent). Again, a similar relationship was found among Survey 1 adult respondents (see para 10.68). Conclusions This chapter has reviewed the employment and financial circumstances 10.94 of families accepted as homeless. It has confirmed that these families are living on very low incomes, with the average incomes for couple-headed households in particular lagging well behind that of their equivalents in the general population. Receipt of means-tested benefits (especially Income Support) and/or tax credits was high. The proportion of adults in paid work in families accepted as homeless was 10.95 (at 29 per cent) somewhat higher than might have been anticipated, but nonetheless levels of worklessness amongst families accepted as homeless far outstripped that of families with children in the general population. In total, 64 per cent of all families accepted as homeless were ‘workless’, as compared with only 14 per cent of families with children in the general population. This disparity was only partly accounted for by the high proportion of lone mothers with young children amongst families accepted as homeless. There was a net 15 percentage point increase in worklessness reported since families left their last settled accommodation.401 Factors controlled for in this regression analysis included: working or ‘workless’ status of household; whether any overdue loans/bills; household type; ethnic background of adult respondent; whether adult respondent had ever claimed asylum in the UK; and household size.248 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFamilies accepted as homeless had often seen their financial circumstances 10.96 deteriorate since they left their last settled accommodation. A deterioration in financial circumstances was particularly likely if a family had moved from a working to workless status. One third of families (35 per cent) were behind with their payments on at least one regular household bill. However, expenses directly associated with moves due to homelessness, and consumer debt, seemed to be relatively minor problems in the context of the overall weak economic position of these families. In almost all respects, Survey 4 families who had been in temporary 10.97 accommodation for over one year reported similar employment, educational and financial circumstances as those in Survey 1. However, they were more likely to report that, overall, they were struggling financially than were Survey 1 families. Closer inspection revealed that Survey 1 families in self-contained temporary accommodation (the form of accommodation in which almost all Survey 4 families were living) were more likely than those in settled housing or other forms of temporary accommodation to report financial problems. While some of this discrepancy is accounted for by the concentration of self-contained temporary accommodation in London (where families were generally more likely to be struggling financially), this form of temporary accommodation does seem associated with particular financial pressures for families.The next chapter moves on to consider the needs and experiences of children 10.98 in families accepted as homeless. Chapter 11 Children’s experiences 249Chapter 11:Children’s experiencesIntroduction The overall objective of this study was to understand the causes, experiences 11.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data was collected in five separate surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys).Concern about the effect of temporary accommodation on children’s well-being 11.2 and life chances has been a key driver of both policy and research in this area for many years. Negative impacts of temporary accommodation on children’s health (particularly their mental health), behaviour and development have been suggested by a range of research402. Previous research has also produced evidence of disruption to children’s education when they lost time at school or had to move schools as a result of homelessness, and has highlighted problems such as lack of space to study or to do homework and other conditions which may undermine concentration in temporary accommodation settings403. However, research on the effects of temporary accommodation has often 11.3 been small in scale, and has usually been narrowly focused with regards to the type of impacts investigated, and the forms of temporary accommodation studied. Thus, as noted in Chapter 1, much early research focused on the very poor living conditions in B&B hotels, especially in London, in the late 1980s and early 1990s (see para 1.10). Particular concerns about the detrimental effects of this form of temporary accommodation on children resulted in legislation so that privately owned B&B accommodation cannot be used to accommodate families with children accepted as homeless, except when no other accommodation is available, and then for no more than six weeks404. 402 Amery, J., Tomkins, A. and Victor, C. (1995) ‘The prevalence of behavioural problems amongst homeless primary school children in an outer London borough: a feasibility study’, Public Health, 109, (6) 421-424; Cummella, S., Grattan, E. and Vostanis, P. (1998) ‘The mental health of children in homeless families and their contact with health, education and social services’, Health and Social Care in the Community, 6, 331-342; Mustafa, Z. (2004) Listen Up: The voices of homeless children, London: Shelter; Vostanis, P. and Cumella, S. (1999) Homeless Children: Problems and needs, London: Jessica Kingsley; Vostanis, P., Gratten, E., Cumella, S. (1998) ‘Mental health problems of homeless children and families: longitudinal study’, British Medical Journal, 316, 899-902; Hall, S., Powney, J. and Davidson, P. (2000) The Impact of Homelessness on Families, Edinburgh: Scottish Council for Research in Education.403 HM Inspectorate of Schools (1990) A Survey of the Education of Children Living in Temporary Accommodation, London: Department of Education and Skills; Power, S., Whitty, G. and Youdell, D. (1995) No Place to Learn: Homelessness and education, London: Shelter; University of Edinburgh/Shelter (1998) Homelessness and Children’s Education. Scotland: Scottish Homes; Shelter (2000) Where’s Home? Children and homelessness in Bristol, London: Shelter.404 Homelessness (Suitability of Accommodation) (England) Order 2003 (SI 2003/3326).250 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThe main purpose of this chapter is to examine whether longstanding 11.4 concerns about the negative impacts of temporary accommodation on children are supported by the evidence from this nationally representative study of families accepted as homeless. The chapter draws upon data provided by children and their parents within families accepted as being owed the main homelessness duty between 1 January 2005 and 30 June 2005 (Surveys 1 and 2). Data from these surveys is used to examine children’s:• experience of leaving their last settled accommodation;• views on their current accommodation;• physical health, mental health and behaviour;• social support networks;• experience of school and their educational performance; and• service use405.The chapter then provides an overview of the experiences of children within 11.5 families accepted as owed the main homelessness duty who had lived in temporary accommodation for more than one year. Here, relevant data from Surveys 4 and 5 (data derived from adults and children in temporary accommodation for more than one year) is compared to the results from Surveys 1 and 2 in order to investigate any differences in findings for children who have spent longer and shorter periods in temporary accommodation. As elsewhere in this report, we present two types of statistical analysis in 11.6 this chapter: bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a range of other factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected by regression analysis406.This survey evidence suggests that children within these families were 11.7 generally happy at school and home. Some positive (net) changes were evident for children as compared with when they lived in their last settled accommodation (especially with regards to their school performance and relationships with parents), but some negative (net) changes were also apparent (with regards to loneliness and reduced participation in clubs/405 Children’s views regarding their overall quality of life are discussed in Chapter 13.406 See Chapter 1 and Appendix 2 for more detail.Chapter 11 Children’s experiences 251activities). Changing schools because of homelessness appeared to be a far more important influence on impacts on children (both positive and negative) than any particular aspect of their experience of temporary accommodation.Key points • Generally, children within families accepted as homeless were happy at school and home, and only a small minority seemed to have extremely difficult or fractured family relationships at the point of survey. • Some positive (net) changes were reported for children compared with when they lived in their last settled accommodation – especially with regards to improvements in their school performance and relationships with parents. • However, some negative (net) changes were also apparent – particularly with regards to loneliness and reduced participation in clubs/activities. • Changing schools as a result of homelessness (experienced by one third of school-age children in households accepted as homeless) had a powerful influence on both positive and negative impacts on children. • Parents accorded the initial disruption of leaving their last settled accommodation far more importance in any negative impacts on their children than specific physical attributes of families’ current (temporary or settled) accommodation. • Likewise, moving between temporary accommodation addresses, and extended stays in temporary accommodation (over one year), appeared to have little effect on the experiences of children. • However, children living in temporary accommodation were far more likely than those in settled housing to be unhappy with aspects of their accommodation and to want to move to somewhere else.BackgroundThis chapter draws upon data from four different sets of respondents: 11.8 • adult respondents in Survey 1; • child respondents in Survey 2; • adult respondents in Survey 4; and • child respondents in Survey 5. 252 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThis range of data was collected in order to provide the fullest possible picture 11.9 of the experiences of children in families accepted as homeless. Data was gathered from both children and adults because some questions (e.g. views of accommodation and neighbourhood) were most appropriately asked of children; some were more appropriately asked of adults (e.g. regarding children’s health status); and some could usefully be asked of both adults and children and their responses compared (e.g. social support networks and school performance)407. As noted in Chapter 1, Surveys 4 and 5 were required because the ‘time-window’ approach taken in Surveys 1 and 2 would, by definition, exclude those families in temporary accommodation for extended periods (see also Appendix 1). This section provides basic contextual information regarding each of these four datasets.Survey 1: data from adult respondentsThe total number of children (aged 0-17 years) reported on by adults in 11.10 families accepted as owed the main homelessness duty between 1st January 2005 and 30 June 2005 was 3,272408. More than 99 per cent of these were the respondents’ own children (natural or adopted), the remaining few (less than 1 per cent) were the children of the respondent’s partner. Survey 1 adult respondents are therefore referred to as ‘parents’ henceforth in this chapter.The age composition of all children reported upon by Survey 1 parents is 11.11 given in Table 11.1. Half (50 per cent) were of preschool age (under 5), including almost a third (30 per cent) aged under 2; a further 32 per cent were aged 5-11; 13 per cent aged 12-15; and only 5 per cent aged 16-17 years. There were approximately equal numbers of boys and girls. Table 11.1: Child age band, as reported by parentsAge band (years) Percentage0-1 30%2-4 20%5-7 15%8-11 17%12-15 13%16-17 5%Total 100%Base 3,272Source: Survey 1407 Clearly some caution has to be exercised in interpreting data derived from parents on matters such as their children’s ability to form peer relationships.408 See Appendix 1 for survey methodology.Chapter 11 Children’s experiences 253At the point of survey, 56 per cent of all these children were living in settled 11.12 accommodation and 44 per cent were living in temporary accommodation. Of the children living in temporary accommodation, 83 per cent were in self-contained temporary accommodation, 10 per cent were living with friends or relatives, and 7 per cent were in hostels or B&B hotels.Approximately two thirds (67 per cent) of all children in these families had 11.13 a ‘last settled accommodation’ to which their current circumstances could be compared409. All of the relevant comparisons below are therefore limited to this group, or as many of them as valid responses were available for on specific questions. Survey 2: data from child respondentsSurvey 2 involved interviewing one child aged 8-15 years (where such a 11.14 child was present) in each family accepted as owed the main homelessness duty between 1st January 2005 and 30 June 2005410. There were 450 Survey 2 child respondents in total411. Approximately half (53 per cent) were girls and 47 per cent boys. In terms of age composition, 51 per cent were aged 8-11 years, and 49 per cent aged 12-15 years. Some of the more ‘difficult’ questions (requiring more complex vocabulary or retrospective reflection) were asked of the 12-15 year olds only.At the point of survey, 57 per cent of Survey 2 child respondents were 11.15 living in settled accommodation and 43 per cent were in temporary accommodation. Of those still in temporary accommodation, the majority (88 per cent) were living in self-contained temporary accommodation, with only 7 per cent in hostels or B&B hotels, and 5 per cent staying with friends or relatives412. Two-thirds (67 per cent) of all of Survey 2 child respondents had a ‘last settled 11.16 accommodation’ before their family was accepted as homeless to which their current accommodation experiences could be compared. Again, therefore, all comparisons below to last settled accommodation are limited to this group, or as many of them as valid responses were available for on specific questions.409 See Chapter 1 and Appendix 1 for an explanation of ‘last settled accommodation’.410 Virtually all (99 per cent of) survey 2 child interviews took place inside the child’s home. There was usually someone else present during the interview – only 9 per cent of child respondents were interviewed alone. The child’s mother (or female partner of the father) was present in three quarters of the interviews, with younger or elder siblings present in 20 per cent and 15 per cent of interviews respectively. In the vast majority of these cases it was assessed by the interviewer that the adult(s) present had little or no influence over children’s responses other than to encourage them (94 per cent of 12-15 yr olds; 89 per cent of 8-11 yr olds). Whilst no parent dictated entirely what their child said, interviewers reported that 8 per cent ‘sometimes told child what to say’ (this was true for 11 per cent of 8-11 yr olds, and 6 per cent of 12-15 yr olds).411 See Appendix 1 for survey methodology.412 Given the small number of Survey 2 child respondents living in these latter two forms of temporary accommodation it was not possible to ascertain any influence that current temporary accommodation type may have on children’s experiences with the required degree of statistical robustness.254 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsSurvey 4: (data from adult respondents) and Survey 5 (data from child respondents)Similar data was gathered from Survey 4 adult respondents, and Survey 11.17 5 child respondents, to allow systematic comparisons of some of the experiences of children who had been in temporary accommodation for more than one year to those of children in similar forms of accommodation for shorter periods of time413. The comparisons made were restricted to children living in self-contained temporary accommodation in Surveys 1 and 2 as, at the point of survey, all children in Surveys 4 and 5 were living in self-contained temporary accommodation.Survey 4 provided adult-derived data regarding 1,066 such children aged 0-17 11.18 years, while Survey 5 provided data from 180 child respondents aged 8-15 years. Differences between these children and those described above (Surveys 1 and 2) are discussed at the end of this chapter.Children’s views on leaving their last settled accommodationMore than half (60 per cent) of Survey 2 child respondents said that they 11.19 missed someone or something from their last settled accommodation. Table 11.2 shows that friends were by far and away the people most commonly missed, reported by half (53 per cent) of child respondents. Extended family members were missed by approximately one in eight, as were significant places (e.g. gardens) and pets.413 We could not compare all aspects of children’s experiences as there was no data collected on ‘last settled accommodation’ for children in Surveys 4 and 5 (see Chapter 1 and Appendix 1).Chapter 11 Children’s experiences 255Table 11.2: People, places and things missed by children since moving from their last settled accommodation, as reported by children (8-15 years)Person/place/thing missed PercentageFriend(s) 53%Family who no longer live with respondent (excl. parents/carers) 12%Garden/ yard/ any other special places 11%Pet(s) 11%Own bedroom 6%Parent/ carer who no longer lives with respondent 4%Toys or possessions 3%Old school/teacher 2%Other person 2%Other thing or place 15%Base 300Source: Survey 2 Base: All children who reported missing someone or something. Multiple responses were possible. With regards to material possessions, 59 per cent of all Survey 2 child 11.20 respondents reported that they were able to take everything with them when they left; 37 per cent had had to leave behind at least some things; and 5 per cent had had to leave everything behind.Approximately two-thirds (65 per cent) of Survey 2 child respondents had a 11.21 pet when living in their last settled accommodation. Of these, 36 per cent (22 per cent of all Survey 2 children) were unable to take their pet with them when they left. The most common reason (affecting 23 per cent of children with pets in their last settled accommodation, and 8 per cent of all Survey 2 children) was that rules in their new accommodation prevented the keeping of pets.Children’s views on their current accommodationThis section summarises Survey 2 child respondents’ views on:11.22 • space, sharing, and places to play;• quality of accommodation;• perceived safety in the home and neighbourhood; and256 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds• an overall assessment of their current accommodation, including whether or not they wanted to move somewhere else414.Space, sharing and places to play While the great majority (86 per cent) of all Survey 2 child respondents 11.23 reported having somewhere inside where they could go when they wanted to be by themselves, this was true of only 77 per cent of those in temporary accommodation, compared to 93 per cent of those in settled accommodation.One third (34 per cent) of Survey 2 child respondents had nowhere 11.24 outside to play or relax. Again, a greater proportion of those in temporary accommodation suffered from outdoor space restrictions: 40 per cent of those in temporary accommodation, as compared with 29 per cent of those who were in settled accommodation, had no outdoor play/relaxation space.As Figure 11.1 demonstrates, gardens (49 per cent) and parks (41 per cent) 11.25 were the most common outdoor spaces available to Survey 2 children for play/relaxation. One quarter (23 per cent) of child respondents played in the street.Figure 11.1: Places outside available for play or relaxation, as reported by children (8-15 years)49%41%23%10% 9% 9% 8%3%1%6%0%10%20%30%40%50%60%OtherWoodsCarparkBackyardAtfriends'housePlayingfields/greenPlay-groundStreetParkGardenper cent of child respondentsSource: Survey 2 Base: 296 (children with access to outdoor space for play or relaxation). Multiple responses were possible.414 See Chapter 7 and 8 for adults’ views on conditions in temporary and settled accommodation respectively.Chapter 11 Children’s experiences 257About half (47 per cent) of all Survey 2 child respondents shared a bedroom 11.26 in their current accommodation (54 per cent of 8-11 year olds, and 38 per cent of 12-15 year olds). Children in temporary accommodation were more likely to share a bedroom than those living in settled accommodation: a total of 55 per cent as compared with 40 per cent respectively. Whilst two-thirds (65 per cent) of the Survey 2 child respondents sharing 11.27 a bedroom reported no problems with sharing, older children were less comfortable with this arrangement (44 per cent of the 12-15 year olds sharing were ‘not very happy’ or ‘not at all happy’ about this, as compared with 29 per cent of 8-11 year olds).The main problems cited by those Survey 2 children (aged 8-15) who disliked 11.28 having to share centred on a lack of privacy (21 per cent), space restrictions (20 per cent), conflict with the other person sharing (14 per cent), and difficulty sleeping (10 per cent). It should be noted that in the vast majority of cases, sharing will have been with members of their immediate family (see para 7.16).As one might expect, given the small number of Survey 2 child respondents 11.29 living in hostels and B&B hotels at the point of survey, very few (only 3 per cent of all Survey 2 children) shared either a kitchen or a bathroom with people from other families.Quality of accommodationWhen asked whether they agreed or disagreed with a range of statements 11.30 regarding the quality of their current accommodation, it was clear that the majority of Survey 2 child respondents considered their home to be both ‘clean’ (87 per cent) and ‘warm enough’ (88 per cent) (see Figure 11.2). A total of 88 per cent considered their home to be ‘comfortable’, but Survey 2 children in temporary accommodation were less likely to make such a claim (76 per cent as compared with 98 per cent). One quarter (24 per cent) of all Survey 2 child respondents in temporary accommodation considered their accommodation to be ‘scruffy’, and 12 per cent thought it ‘smelly’ (as compared with 15 per cent and 6 per cent respectively of children in settled housing).258 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 11.2: Perceptions of current accommodation, as reported by children (8-15 years)8984928398761524612Smelly ScruffyComfortable*per cent of child respondentsWarm enoughClean0%20%40%60%80%100%TASettled accommodationSource: Survey 2 Base: 223 children in settled accommodation, 224 children in temporary accommodation. *The ‘comfortable’ criterion question was asked of children aged 12-15 only (Base 223). All others were asked of 8-15 year olds (Base 447). Safety in the home and neighbourhoodThe majority (81 per cent) of all Survey 2 child respondents reported that 11.31 they ‘always felt safe’ when inside their current accommodation. However, 17 per cent ‘sometimes felt safe, sometimes not’, and 2 per cent reported never feeling safe’. Children’s perceptions of safety indoors accorded relatively closely with that of Survey 1 adult respondents, as 21 per cent of all Survey 1 parents reported that they (and/or other family members) did not feel safe inside their current accommodation (see para 7.62). Most of the Survey 2 child respondents who reported feeling unsafe in their 11.32 accommodation explained that this was because they were fearful of other people in ‘rough’ neighbourhoods.There was no difference in perceptions of safety inside accommodation 11.33 between Survey 2 child respondents living in temporary and settled accommodation, boys and girls, or children of different age cohorts. Almost two-thirds (62 per cent) of 8-11 year olds, and the great majority (91 11.34 per cent) of 12-15 year olds, reported that they were allowed to go outside on their own in their local neighbourhood. When asked about walking on their own in the daytime, 60 per cent of these children (or 45 per cent of all Survey 2 child respondents) reported that they ‘always felt safe’, 33 per cent (25 per cent) ‘sometimes felt safe, sometimes not’, and 5 per cent (4 per cent) ‘never felt safe’ (the rest reported that they ‘never walk on their own’) (Figure 11.3).Chapter 11 Children’s experiences 259While a lesser proportion of children reported feeling unsafe when walking 11.35 with friends in the daytime in their local neighbourhood, levels of fear were not trivial, with 12 per cent of all Survey 2 child respondents reporting that they ‘sometimes felt safe, sometimes not’ and 3 per cent that they ‘never felt safe’ (Figure 11.3).Figure 11.3: Perceptions of safety when walking on own or with friends in the local neighbourhood, as reported by children (8-15 years)9312762533600%20%40%60%80%100%Always feel safe Sometimes feel safe,sometimes notNever feel safe Don't walk on own/with friendsWalking with friendsWalking alone Source: Survey 2 Base: 334 (all children who are allowed to walk alone); 450 (all children allowed to walk with friends). There were no gender or age variations with regard to perceptions of safety 11.36 outdoors; nor were there differences between the perceptions of children in temporary or settled accommodation. For many of the Survey 2 child respondents who said that they felt unsafe at 11.37 least some of the time in their neighbourhood (when walking on their own or with friends), fear was based upon generalised concerns regarding ‘stranger danger’, that is, fear of people unknown to them415. For a number, however, fear was centred upon attributes more broadly associated with ‘unsafe’ areas – such as crime, gangs and the presence of drunk people. A small proportion felt unsafe because of the threat presented by other children in the area (e.g. bullies).415 See also Mayhew, E., Finch, N., Beresford, B., and Keung, A. (2005) ‘Children’s Time and Space’. In Bradshaw, J. and Mayhew, E. (eds) (2nd edn.) The Well-being of Children in the UK, London: Save the Children, pp161-181.260 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsOverall assessment of current accommodationSurvey 2 child respondents aged 12-15 were asked to rate their current 11.38 accommodation on a scale from 1 to 10 where 1 is ‘terrible’ and 10 ‘excellent’. The average overall rating given was 6.2, and the median was 7. At 7.0, the average rating given by children in settled accommodation was notably greater than that of those still in temporary accommodation (5.3). All Survey 2 child respondents (8-15 years) were asked whether they 11.39 preferred their current accommodation or their last settled accommodation. Children in settled accommodation were much more likely to prefer their current accommodation (56 per cent did so) than they were their last settled accommodation (true of only 25 per cent). Opinion was more evenly split amongst children still resident in temporary accommodation – with 41 per cent liking their current accommodation more and 36 per cent liking it less than their last settled accommodation.All Survey 2 child respondents were also asked whether they would like to 11.40 stay in their current accommodation or move somewhere else if they were given the choice. Opinion was relatively evenly divided overall, with 47 per cent saying they would like to stay and 45 per cent that they would like to move elsewhere (with the remaining 8 per cent being undecided). However, the proportion of those living in temporary accommodation wishing they could move somewhere else (74 per cent) was more than twice that of those living in settled accommodation (31 per cent).One quarter (28 per cent) of Survey 2 children who wanted to stay in their 11.41 current accommodation explained that this was because they had friends living nearby. Likewise, a desire to live nearer to (old) friends was one of the most common reasons given for wanting to move somewhere else (reported by 20 per cent of children wanting to move). However, a larger proportion of children wanting to move identified a desire for more space as their key motivation (36 per cent) – particularly more personal space (e.g. own bedroom) and space suitable for play. Nearly one fifth (18 per cent) of those wanting to move attributed their desire to live somewhere else to their dislike or fear of their current neighbourhood.Health This section reviews data from Survey 1 adult respondents regarding their 11.42 children’s general health at the point of survey, and changes in general physical and mental health, and behaviour, since leaving their last settled accommodation. Chapter 11 Children’s experiences 261General healthSurvey 1 parents considered the vast majority of their children to be in good 11.43 general health at the point of survey: a total of 88 per cent were deemed to be in ‘very good’ or ‘good’ health. The overall health profile of the survey children did in fact broadly reflect that of children aged 0-17 in the wider population416 (Figure 11.4).Figure 11.4: General physical health status of children (0-17 years), as reported byparents (with national comparison)56%Very badBadFairGoodVery good32%8%3% 1%58%36%5%1% 0%HSE 2003 ComparatorSource: Survey 1 and Health Survey for England 2003 (weighted data, children aged 0-17 years, CHP analysis) Base: 3,272 Survey 1 children, and 4,137 HSE (2003) children. There were no differences noted between the health of children in temporary 11.44 or settled accommodation, nor according to age, gender, or temporary accommodation experience.Changes in general physical healthIn the opinion of Survey 1 parents, the general physical health status of 11.45 almost three quarters (72 per cent) of their children did not alter between leaving their last settled accommodation and the point of survey; the health of one fifth (20 per cent) improved after leaving their last settled accommodation; and there was a deterioration in only 8 per cent of cases. Thus, amongst the minority of children for whom general physical health status had changed, it was more than twice as likely to improve as it was to deteriorate.416 The HSE (2003) children’s health data are not fully comparable to ours as the equivalent question was asked directly of children aged 13 and older in the HSE, but was asked of parents of all children (0-17 yrs) in our survey. The relevant HSE (2003) statistics do nevertheless provide a useful broad comparison with children’s overall health status nationally.262 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsImprovements in children’s general physical health were most commonly 11.46 attributed by parents to the fact that children felt settled and happy in their new accommodation, had left problems associated with their previous accommodation, or that their previous accommodation had been of poor quality.Among the minority of households where parents reported a deterioration 11.47 in their child’s general physical health, this was most commonly attributed to the stress associated with moving to a new home, a move into poor quality housing, and/or a lack of peer friendships in a new area.There was no substantial difference between the changes in general physical 11.48 health status experienced by children in temporary or settled accommodation, boys or girls, or by children in different age groups. However, children in families for whom violent relationship breakdown417 was a cause of homelessness were more likely than other children to have experienced an improvement in general health status since leaving their last settled accommodation (26 per cent as compared with 19 per cent). Changes in mental healthSurvey 1 parents reported that one half (50 per cent) of children aged 5 and 11.49 over never appeared anxious, stressed or depressed in either their last settled accommodation or in their current accommodation; a further 7 per cent had problems in both contexts. Since leaving their last settled accommodation 18 per cent of children had either: a) begun exhibiting symptoms of anxiety, stress or depression, or b) become more anxious, stressed or depressed than they had been before418. However, a higher proportion of children (25 per cent) were reported to have improved mental health status after leaving their last settled accommodation419. There was little gender variation in mental health status change, but children 11.50 aged 15 years and older had a greater tendency to experience change in mental health status than younger children (problems got worse for 24 per cent and better for 39 per cent). 417 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.418 Both of these experiences are henceforth referred to as symptoms of ‘deterioration of mental health status’.419 Our survey focussed on changes in children’s general psychological wellbeing perceived by parents since acceptance as homeless. It did not attempt to compare the prevalence of specific clinically diagnosed mental disorders to that of children in the general population, as existing national data is derived from intensive assessments of children’s mental health based upon detailed questionnaires such as the General Health Questionnaire and/or Strengths and Difficulties Questionnaire (see Green, H., McGinnity, A., Meltzer, H., Ford, T. and Goodman, R. (2005) Mental Health of Children and Young People in Great Britain, 2004, Basingstoke: Palgrave Macmillan; Quilgars, D., Searle, B. and Keung, A. (2005) ‘Mental Health and Well-being’, in Bradshaw, J. and Mayhew, E. (eds) (2nd edn). The Well-being of Children in the UK, London: Save the Children, pp 134-160.)Chapter 11 Children’s experiences 263Regression analysis indicated that, other things being equal11.51 420, the main factor which exerted an independent influence on deteriorations in a child’s mental health was living in an area perceived to be unsafe: 18 per cent of children living in a neighbourhood which their parent felt was unsafe, compared with 8 per cent of other children, were reported to have experienced a deterioration in mental health. Living in temporary accommodation at point of survey had no independent effect on the likelihood of a child experiencing a deterioration in mental health.Regression analysis on improvements in mental health found that, other 11.52 things being equal421, the only factor exerting an independent effect was violent relationship breakdown422 as a reason for applying as homeless: 28 per cent of children for whom this was a cause of homelessness, compared with 12 per cent of other children, were reported to have experienced an improvement in mental health status. Parents most commonly attributed any increases in their child(ren)’s levels of 11.53 anxiety, stress or depression to the stress of moving home, conflicts/issues within the family, problems settling in a new area, and/or having to leave old friends. Correspondingly, parents felt that improvements in children’s mental health 11.54 status were largely accounted for by children feeling settled in their new home or being away from (former) family problems.Changes in behaviour Survey 1 parents were asked about behavioural changes perceived in children 11.55 aged 2 and older since leaving their last settled accommodation. Overall, parental assessments indicated that there had been no observable change in the behaviour of more than half (59 per cent) of these children. There was no net change in the pattern of behaviour of the remaining children, as the behaviour of 20 per cent was said to have improved since leaving their last settled accommodation, but the behaviour of approximately the same proportion (21 per cent) had deteriorated. There were no differences in perceived behavioural change according to children’s age or gender.420 Factors controlled for in this regression analysis included: gender; age; whether accepted in a rural area; whether accepted in London, the South, or North and Midlands; moved school because of homelessness; current accommodation type; temporary accommodation experiences; causes of homelessness; accommodation conditions; parent has a current mental health problem; household is workless, and household has become workless since leaving last settled accommodation; family has current financial difficulties; and changes in ability to manage financially since leaving last settled accommodation.421 Factors controlled for in this regression analysis included: gender; age; whether accepted in a rural area; whether accepted in London, the South, or North and Midlands; moved school because of homelessness; current accommodation type; temporary accommodation experiences; causes of homelessness; accommodation conditions; parent has a current mental health problem; household is workless, and household has become workless since leaving last settled accommodation; family has current financial difficulties, and changes in ability to manage financially since leaving last settled accommodation.422 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.264 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsRegression analysis indicated that, other things being equal11.56 423, the factors which exerted an independent influence which made reported deteriorations in a child’s behaviour more likely were:• their parent having a current mental health problem: 22 per cent of children whose parent self-reported a current mental health problem, compared with 10 per cent of other children, were said to be exhibiting worse behaviour than in their last settled accommodation.• living in an area perceived to be unsafe: 22 per cent of those children living in a neighbourhood which their parent felt was unsafe, compared with 11 per cent of other children, were said to be exhibiting worse behaviour than in their last settled accommodation.Once these and other factors were taken into account, living in temporary 11.57 accommodation at point of survey had no independent effect on the likelihood that a child’s behaviour had deteriorated.Regression analysis indicated that, other things being equal11.58 424, the single factor which exerted an independent effect on improvements in a child’s behaviour was violent relationship breakdown as a reason for their family’s homelessness425 (24 per cent of these children, compared to 10 per cent of other children, were said to be behaving better than in their last settled accommodation).Parents most commonly attributed any improvement in behaviour to 11.59 the fact that children felt more settled and less stressed in their current accommodation, or that they had moved away from problems associated with previous accommodation. Less frequently identified as causes for behavioural improvement was the better quality, or quantity, of space available in their new accommodation. Worsening behaviour was normally blamed by parents on the stress caused 11.60 by moving home and children feeling unsettled in a new area, although problems within the family were also cited, as was having to change schools (see below for information on school changes). Interestingly, whilst a move 423 Factors controlled for in this regression analysis included: gender; age; whether accepted in a rural area; whether accepted in London, the South, or North and Midlands; moved school because of homelessness; current accommodation type; temporary accommodation experiences; causes of homelessness; accommodation conditions; parent has a current mental health problem; household is workless, and household has become workless since leaving last settled accommodation; family has current financial difficulties, and changes in ability to manage financially since leaving last settled accommodation.424 Factors controlled for in this regression analysis included: gender; age; whether accepted in a rural area; whether accepted in London, the South, or North and Midlands; moved school because of homelessness; current accommodation type; temporary accommodation experiences; causes of homelessness; accommodation conditions; parent has a current mental health problem; household is workless, and household has become workless since leaving last settled accommodation; family has current financial difficulties, and changes in ability to manage financially since leaving last settled accommodation.425 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.Chapter 11 Children’s experiences 265into better quality accommodation was sometimes cited as a reason for improvements in a child’s behaviour, the standard of the physical environment (including space to play) was seldom mentioned as an explanation for the deterioration in behaviour of children.Support networksThis section draws on data both from parents (Survey 1) and children (Survey 11.61 2) to assess children’s: general levels of support; relationship with parents and peers; contact with wider family; and participation in clubs and activities. General levels of supportMore than one quarter (29 per cent) of all Survey 2 child respondents 11.62 reported feeling lonely at least some of the time in their current accommodation. There was no difference in experiences of loneliness between Survey 2 children in temporary accommodation and those in settled accommodation, nor between age groups. The proportion of girls feeling lonely in their current accommodation was however greater than that of boys (31 per cent as compared with 20 per cent respectively).A greater proportion of Survey 2 children reported feeling lonely at least 11.63 some of the time at the point of survey than was the case in their last settled accommodation (29 per cent as compared with 17 per cent) – possibly a reflection of ruptured peer and/or family relationships following the move to new accommodation.Nevertheless, almost all (93 per cent of) Survey 2 child respondents reported 11.64 that they had someone to talk to when they felt upset or worried. Parents were the most common source of such support – cited by 81 per cent of child respondents. Friends and siblings were identified as important sources of support for 31 per cent and 23 per cent of children respectively. Very few (only 2 per cent) identified professional support workers as a source of this sort of support.The great majority of Survey 2 children (85 per cent) had someone to talk 11.65 to when they felt upset or worried in both their last settled accommodation and current accommodation; 2 per cent had no-one in either; 4 per cent had someone in their last settled accommodation but not in their current accommodation; and (more encouragingly) 8 per cent had no-one in their last settled accommodation but did at the point of survey. 266 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsRelationships with parents The findings reported here are based upon Survey 2 self-completion questions 11.66 (i.e. children were able to answer these without the risk of other family members overhearing their responses) asked of 12-15 year olds only. When asked how well they got along with the adult (Survey 1) respondent (typically their mother) at the point of survey, the great majority reported getting on well (78 per cent), or at least okay (15 per cent), with only 7 per cent reporting that they did not get on with their parent very well.Those 12-15 year olds who had moved from their last settled accommodation 11.67 were asked if the nature of their relationship with that parent had altered since doing so. Their responses indicated that their relationships with parents were four times as likely to have improved (this was the case for 41 per cent of these children) as they were to have deteriorated (true for 10 per cent); although nearly half (47 per cent) reported that there had been no change, and an additional 2 per cent of respondents reported being ‘not sure’. There was no variation in responses between those in temporary accommodation and those in settled accommodation. This data gives the impression that only a very small proportion of the children 11.68 in these families faced extremely difficult or fractured family relations at point of interview. This impression is reinforced by the fact that very few (only 3 per cent) of 12-15 year olds reported that they had run away from their last settled accommodation (and stayed away for at least one night), and only 6 per cent had done so subsequently.Contact with wider family Survey 1 parental assessments indicated that the (overall) amount of contact 11.69 children had with extended family members such as aunts, uncles or grandparents since leaving their last settled accommodation had increased for more than a third (36 per cent) of households, had not altered at all for 42 per cent, and had decreased for just over one fifth (22 per cent). Only 1 per cent were no longer able to see family that they had previously had contact with, whereas 5 per cent had initiated contact since leaving their last settled accommodation. The proportion who reported that their children had no contact with extended family in either their last settled accommodation or currently was also small (3 per cent).Survey 1 parents identifying violent relationship breakdown11.70 426 as a reason for applying as homeless were more likely to report that their child(ren) now had more contact with family (e.g. aunts, uncles or grandparents) than were families who did not (43 per cent as compared with 29 per cent).426 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.Chapter 11 Children’s experiences 267Peer relationshipsSurvey 1 parents were asked about any changes in the quality of relationships 11.71 between children (aged 5 and older) and their peers since they had moved from their last settled accommodation. They reported that there had been no observable change for three quarters (73 per cent) of these children. Amongst the minority for whom there had been a change, twice as many (18 per cent of all children) were said to have ‘got better’ at establishing and maintaining peer relationships as had ‘got worse’ at doing so since leaving their last settled accommodation (9 per cent of all children). There was no variation noted by children’s age or gender. Regression analysis indicated that, other things being equal11.72 427, the only independent factor associated with children reportedly having improved relations with their peers was moving schools as a result of homelessness: 22 per cent of those school-age children who had had to change school as a result of homelessness, as compared with 14 per cent of those who had not, were reported to have improved relations with their peers.Regression analysis indicated that, other things being equal11.73 428, the single independent factor associated with children reportedly having worse relations with their peers was living in an area perceived to be unsafe: 10 per cent of those children living in a neighbourhood which their parent felt was unsafe, compared with 2 per cent of other children, were reported to have worse relations with their peers. There was no independent effect of being in temporary accommodation on the likelihood of a child having worse relations with their peers once this and other factors were controlled for.The disruption and stress of moving were deemed by parents to have been 11.74 the key contributory factor for the minority of children whose peer relations were deemed to have deteriorated. Dislocation from existing friendship networks into a new environment where children had no friends, problems at (typically a new) school, together with the general stress of moving/change, were the main causes of difficulties identified by Survey 1 parents.Conversely, the creation of a more settled stress-free environment, and/or 11.75 move to a more spacious home enabling children to invite friends around, were the predominant explanations provided by Survey 1 parents for improvements in the peer relationships of their children.427 Factors controlled for in this regression analysis included: gender; age; whether accepted in a rural area; whether accepted in London, the South, or North and Midlands; moved school because of homelessness; current accommodation type; temporary accommodation experiences; causes of homelessness; accommodation conditions; parent has a current mental health problem; household is workless, and household has become workless since leaving last settled accommodation; family has current financial difficulties, and changes in ability to manage financially since leaving last settled accommodation.428 Factors controlled for in this regression analysis included: gender; age; whether accepted in a rural area; whether accepted in London, the South, or North and Midlands; moved school because of homelessness; current accommodation type; temporary accommodation experiences; causes of homelessness; accommodation conditions; parent has a current mental health problem; household is workless, and household has become workless since leaving last settled accommodation; family has current financial difficulties, and changes in ability to manage financially since leaving last settled accommodation.268 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsChild respondents were also asked about any changes to their friendship 11.76 networks experienced since leaving their last settled accommodation. One third (34 per cent) said that they no longer saw their old friends. The proportion of children in temporary accommodation no longer seeing their old friends was approximately the same as that of children who had moved into settled accommodation429. Of the children who did still see their old friends, slightly more than half (54 per cent, or 35 per cent of all Survey 2 children) saw them less often, 33 per cent (21 per cent of all Survey 2 children) about the same amount, and 14 per cent (10 per cent of all Survey 2 children) more often. Most of the Survey 2 child respondents who no longer saw their old friends 11.77 or saw them less frequently since moving into their new home reported that this was because they now lived too far away, although some noted that changing school limited their ability to spend time with the friends they had before leaving their last settled home.In total, three quarters (74 per cent) of all Survey 2 child respondents had 11.78 friends around to their current accommodation after school or at weekends. For those who did not, the predominant reason was that they did not have any friends living nearby; other reasons cited included being ashamed of their home, a lack of space, and rules disallowing visits from friends (which may have been imposed by the housing provider or parents/carers).A majority (70 per cent) of Survey 2 child respondents reported no change 11.79 in the frequency of having friends around after school or at weekends since leaving their last settled accommodation; and while 14 per cent stopped inviting friends around after moving, this was balanced by the 15 per cent who started doing so. Children in settled accommodation were more likely to have friends around after school or at weekends than those in temporary accommodation (83 per cent as compared with 64 per cent).Clubs and activitiesMore than half (55 per cent) of the 12-15 year old Survey 2 child respondents 11.80 had been involved in clubs and activities after school or during weekends when resident in their last settled accommodation. Worryingly, although 15 per cent of this group reported that they participated in such activities more often at the point of survey than they had in their last settled accommodation, nearly half (45 per cent) had decreased the frequency of their participation in such activities, and a further 25 per cent had discontinued involvement completely. Thus children were almost five times as 429 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1 para 1.27).Chapter 11 Children’s experiences 269likely to have decreased or ceased their participation in clubs and activities as they were to have increased it since leaving their last settled accommodation.There was no difference with regards to involvement in clubs or activities 11.81 between children who were in temporary and settled accommodation.School and educational performance This section reviews children’s school attendance, school performance and any 11.82 moves they have made between schools as a result of homelessness. School status and attendanceSurvey 1 parents reported that, of their 5-17 year old children (who had not 11.83 left school), 7 per cent were currently excluded temporarily (‘suspended’), fewer than 1 per cent had been excluded permanently (‘expelled’), and fewer than 1 per cent were awaiting a school place.Whilst Survey 1 parents reported that some of their children were unhappy 11.84 at school – with 12 per cent often bullied/unhappy at school and 7 per cent often refusing to go to school – only a small minority regularly missed school for other reasons (e.g. truancy, transport problems, staying at home to help the family etc.) (Table 11.3).Table 11.3: Child’s (5-17 years) school attendance and experience, as reported by parentsPercentageChild often bullied or unhappy at school 12%Child often refuses to go to school 7%Transport problems often prevent child from going to school 3%Child often stays at home because of physical or mental health problems2%Child often does not go to school because housing situation too disruptive2%Child often plays truant 2%Child often stays at home to help rest of family 1%Child often misses school for other reasons 4%Base 1,591Source: Survey 1 Base: All school-age children who had not left school, been excluded, or were awaiting a school place. Multiple responses were possible.270 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsWhen asked about their school attendance, one quarter (25 per cent) of 11.85 Survey 2 child respondents reported that they had missed some school because they moved to a new home after leaving their last settled accommodation. However, the experience of homelessness appeared to have little impact on other reasons for missing school, such as truancy and refusing to go to school, as both affected very small numbers of Survey 2 children. School moves due to homelessnessSurvey 1 parents reported that one third (33 per cent) of children aged 5 and 11.86 older had had to change school because their family had moved after being accepted as homeless430. Primary school aged children (5-11 years) were more likely to have changed school because of homelessness than were secondary-aged children (38 per cent as compared with 23 per cent). As noted above (and below) moving school as a result of homelessness appeared to have a powerful impact on children, both positive and negative. Regression analysis indicated that, other things being equal11.87 431, the following factors made homelessness-related school moves more likely for children:• having experienced hostels or B&B hotels: 50 per cent of those children who had stayed in these forms of temporary accommodation, as compared with 27 per cent of other children, had had to move school. • violent relationship breakdown as a cause of homelessness432: 43 per cent of children for whom this was a reason for applying as homeless, as compared to 30 per cent of other children, had had to move school. Once these and other factors were taken into account, neither the age of 11.88 the child nor whether they were living in temporary or settled housing had any independent effect on the likelihood of their having made homelessness-related school moves.School performance We investigated parents’ perceptions regarding changes in their children’s 11.89 performance at school. One third (34 per cent) of children aged 5-17 (who were at school both in their last settled accommodation and in their current accommodation) were, according to their parents, now performing better at school. Just over half (56 per cent) were said to be performing to approximately the same standard, and only 7 per cent were said to be performing less well than they had in their last settled accommodation. 430 Amongst those families still in temporary accommodation, further school moves as a result of homelessness may have occurred after point of survey, so the total percentage of children who will experience such school moves will be somewhat higher.431 Factors controlled for in this regression analysis included: gender; age; whether accepted in a rural area; whether accepted in London, the South, or North and Midlands; current accommodation type; temporary accommodation experiences; causes of homelessness; accommodation conditions.432 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.Chapter 11 Children’s experiences 271Perceived changes in school performance did not vary by gender, but varied 11.90 by age in that a greater proportion of older children experienced a decline in performance than did younger children (13 per cent of 15-17 year olds, as compared with 6 per cent of 5-9 year olds and 8 per cent of those aged 10-14 years). There was no relationship between parental assessments of changes in school performance and whether or not a child lived in settled or temporary accommodation. Regression analysis indicated that, other things being equal11.91 433, the only independent influence on improvements in the reported school performance of a child was whether they had had to move school as a result of homelessness (40 per cent of those children who had had to move school, as compared to 28 per cent of other children, were said to be performing better). Regression analysis also indicated that, other things being equal11.92 434, there was also a relationship between deteriorations in school performance and school moves due to homelessness (11 per cent of those children who had had to move school, as compared to 5 per cent of other children, were said to be performing less well). Improvements in school performance since leaving their last settled 11.93 accommodation were rarely attributed by Survey 1 parents to specific aspects of the new accommodation itself, but rather to either the fact that children felt more settled and/or less stressed in their new environment, and/or because they had moved to a better school. Conversely, for the far smaller number whose performance was said to have 11.94 deteriorated, the stress of moving and problems with a new school (e.g. bullying or a different teaching system) were most commonly identified as the main causes of change by Survey 1 parents. Parents also often blamed ‘family problems’, feelings of ‘unsettledness’, and dislocation from peer support networks for their child’s difficulties at school.Survey 2 child respondents’ self-assessments of changes in their school 11.95 performance followed a similar pattern to that of parents. Only 12-15 year olds were asked about changes in school performance since leaving their 433 Factors controlled for in this regression analysis included: gender; age; whether accepted in a rural area; whether accepted in London, the South, or North and Midlands; moved school because of homelessness; current accommodation type; temporary accommodation experiences; causes of homelessness; accommodation conditions; parent has a current mental health problem; household is workless, and household has become workless since leaving last settled accommodation; family has current financial difficulties, and changes in ability to manage financially since leaving last settled accommodation.434 Factors controlled for in this regression analysis included: gender; age; whether accepted in a rural area; whether accepted in London, the South, or North and Midlands; moved school because of homelessness; current accommodation type; temporary accommodation experiences; causes of homelessness; accommodation conditions; parent has a current mental health problem; household is workless, and household has become workless since leaving last settled accommodation; family has current financial difficulties, and changes in ability to manage financially since leaving last settled accommodation.272 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldslast settled accommodation. Of these, only 10 per cent thought that their performance had deteriorated since leaving their last settled accommodation, the remainder of responses were split relatively evenly between ‘about the same’ (47 per cent) and ‘better now’ (43 per cent). There was no variation according to children’s age, gender, or whether they were in settled or temporary accommodation.When asked how well they thought they were doing with their school work 11.96 at the point of survey, the overwhelming majority (94 per cent) of Survey 2 child respondents attending school thought they were doing ‘well’ or ‘okay’, with only 6 per cent considering themselves to be ‘having problems’. There was no variation in self-assessed school performance by age, gender, or whether children were in settled or temporary accommodation.Survey 2 child respondents were, on the whole, positive about school, with 11.97 82 per cent reporting that they liked it ‘a lot’ or at least ‘a little’. Only 8 per cent disliked the school they were currently attending (either ‘a lot’ or ‘a little’), with the remaining 10 per cent being ambivalent (reporting that they ‘neither like it nor dislike it’).It was notable in the Survey 1 parental responses above that it was not the 11.98 specific attributes of their accommodation (e.g. availability of space for homework) that were deemed to affect children’s educational performance, but rather the general disruption associated with moving home. However, this is not to say there are no problems with regard to physical accommodation. Of the 12-15 year old Survey 2 child respondents who had to do homework or coursework, approximately one quarter (24 per cent) reported having difficulty finding somewhere at home to concentrate on their work. This problem was more prevalent for those in temporary accommodation than those who had moved into settled accommodation (37 per cent as compared with 13 per cent respectively).Service use Considering education-related services to begin with, Survey 1 parents 11.99 reported that 6 per cent of their children aged 5 and older had been given a Special Educational Needs (SEN) statement435 at some point within the preceding year – a proportion approximately double that of all children in English schools with SEN statements in January 2005 (3 per cent)436.435 SEN statements are issued where a child finds learning significantly more difficult than the majority of children their age and thus requires special educational interventions. A wide range of conditions may lead to the identification of SEN including: visual, hearing or other physical impairments; cognitive disorders; learning difficulties; or social, emotional or behavioural difficulties. See Coles, B. and Richardson, D. (2005) ‘Education’, in Bradshaw, J. and Mayhew, E. (eds) (2nd edn.) The Well-being of Children in the UK, London: Save the Children, pp 262-288.436 Department for Education and Skills Special Educational Needs in England, January 2005, London: Department for Education and Skills.Chapter 11 Children’s experiences 273Survey 1 parents reported that only a very small minority of their children 11.100 aged 5 and older had attended either a special school (2 per cent) or Pupil Referral Unit (1 per cent) in the preceding year. These children may nevertheless be over-represented in these institutions, as only 1 per cent and less than one per cent (0.2 per cent) of England’s school population were registered in a special school or pupil referral unit respectively at January 2005437. Survey 1 parents reported that 3 per cent of their children had seen an Educational Welfare Officer in the preceding year. Table 11.4 outlines the proportion of children using a variety of other services, 11.101 as appropriate to their age group. Teenagers living in settled accommodation were nearly twice as likely to utilise Connexions services than were those in temporary accommodation (39 per cent of those in settled accommodation as compared to 22 per cent of those in temporary accommodation). There were no other differences in children’s utilisation of services according to whether they were in temporary or settled accommodation.Table 11.4: Services used by children (0-17 yrs), as reported by parentsPercentage BaseNursery place (0-4 yrs) 28% 1,681Children’s centre, Out of School Club or Kid’s Club (0-14 yrs)16% 3,020Connexions Scheme (13-17 yrs) 30% 474Youth worker, social worker or Youth Offending Team worker (13-17 yrs)17% 474Source: Survey 1 According to Survey 1 parents, only 0.2 per cent of their children had 11.102 ever lived in a children’s home, and only 1.9 per cent had lived with foster carers. Overall, 2.0 per cent of children had ever been looked after by a local authority. Whilst national statistics provide a ‘snapshot’ rather than ‘prevalence’ measure (thus limiting their comparability)438, they do serve to suggest that the proportion of children in families accepted as homeless with experience of care may be greater than in the population at large, as less than 0.6 per cent of 0-17 year olds were looked after by their local authority in England as at 31 March 2005439.437 Department for Education and Skills Schools and Pupils in England, January 2005 (Final), London: DfES.438 ‘Snapshot’ data reports the number of individuals at a given point in time (e.g. on a specified day), whereas ‘prevalence’ data reports the number of individuals during a defined time period (e.g. one year).439 Department for Education and Skills (2006a) Statistics of Education: Children Looked after by Local Authorities: Year Ending 31 March 2005 (Volume 1), London: DfES.274 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsA somewhat larger proportion (5.1 per cent) of children had been on the 11.103 Child Protection Register (CPR)440 at some point. Although the comparability of national statistics for this is also limited (for it too provides ‘snapshot’ rather than ‘prevalence’ data), it also serves to suggest that the proportion of children in families accepted as homeless who have been on the CPR is likely to be greater than that of children in England as a whole (0.2 per cent as at March 31 2005)441.Children within families in temporary accommodation for more than one yearAs noted above, parent-derived data from Surveys 1 and 4, and child-derived 11.104 data from Surveys 2 and 5, were used to compare the experiences of children, currently in self-contained temporary accommodation, who had spent longer (more than one year) and shorter periods in temporary accommodation. This analysis revealed that there were no differences between these two groups of children with regard to:• children’s views regarding the overall quality of their current accommodation, the adequacy of space inside or outside, and access to appropriate space for doing homework or coursework;• the children’s feelings of safety inside their accommodation and outside in the local neighbourhood;• the proportion of children reporting that they would choose to leave their current accommodation if they were given the choice;• parental assessments of the children’s current general health;• children’s school attendance;• the proportion of children having to change school because their family had moved after being accepted as homeless; and• the proportion of children in receipt of special education measures (e.g. having seen an Educational Welfare Officer, received a SEN statement, or attended a special school or Pupil Referral Unit in the preceding year).440 Children on the CPR are those who have been identified by professionals as being at continuing risk of significant harm and hence in need of a child protection plan. It includes some who are not abused but thought to be at risk. See Hooper, C. (2005) ‘Child maltreatment’. In Bradshaw, J. and Mayhew, E. (eds) (2nd edn.) The Well-being of Children in the UK, London: Save the Children.441 DfES (2006b) Statistics of Education: Referrals, Assessments and Children and Young People on Child Protection Registers: Year Ending 31 March 2005. London: DfES.Chapter 11 Children’s experiences 275• The proportion of children using specified services were also broadly similar between the two datasets. The one exception was that 0-4 year olds who had been in temporary accommodation for more than one year (as reported by Survey 4 parents) were more likely to have a nursery place than those in Survey 1 families (40 per cent as compared with 28 per cent).In addition, Survey 4 parents were less likely to report that their child 11.105 was unhappy or bullied at school than were Survey 1 parents (4 per cent of children in Survey 4 families as compared with 10 per cent in Survey 1 families)442. There were however no differences between the self-assessed performance at, and enjoyment of, school between Survey 5 child respondents and Survey 2 child respondents. ConclusionsThis study has painted a rather more positive picture of children in families 11.106 accepted as homeless than has previous research. Generally, children within these families were happy at school and home, and only a small minority seem to have extremely difficult or fractured family relationships. Also, some positive (net) changes were reported for children as compared with when they lived in their last settled accommodation – especially with regards to their school performance and relationships with parents. However, negative (net) changes were also apparent – particularly with regards to loneliness and reduced participation in clubs/activities. More than one third of children aged 5 years and over had had to change 11.107 school because of homelessness. Changing schools could have a powerful impact on children – both positive and negative.Children in temporary accommodation were more likely to be unhappy with 11.108 aspects of their accommodation and to want to move somewhere else than children in settled accommodation. However, parents felt that the initial disruption of leaving their last settled 11.109 accommodation was a far more important influence with regards to any negative experiences of children than were specific physical attributes of families’ current (temporary or settled) accommodation. The number of moves between different temporary accommodation addresses also appeared to have little effect on children’s experiences. Likewise, those in temporary accommodation for extended periods (over one year) demonstrated very few 442 Both this finding and that on nurseries may reflect the fact that Survey 4 families had often been resident at their current temporary accommodation address for a relatively long period (average of 2.5 years at point of survey, see Chapter 6, para 6.52), and as a consequence the children in these families may be more settled in the local area than children who had been in temporary accommodation for shorter periods.276 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsdifferences in experiences than was the case for other children currently in self-contained temporary accommodation. The next chapter will explore the characteristics, needs and experiences of the 11.110 young people accepted as 16-17 year olds owed the main homelessness duty.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 277Chapter 12:The experience of young people accepted as homeless 16-17 year olds IntroductionThe overall objective of this study was to understand the causes, experiences 12.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data was collected in five separate surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys).This chapter examines the characteristics and experiences of young people 12.2 accepted as 16-17 year olds owed the main homelessness duty between 1st January 2005 and 30 June 2005 (‘Survey 3’). Sixteen and 17 year olds were included in this study because they are considered children for many legal and policy purposes until they attain the age of 18, and therefore it did not seem appropriate to exclude them from a study of statutory homelessness amongst families with children simply because they were not living with their parents or other carers. Furthermore, 16 and 17 year old homeless applicants – apart from those 12.3 who fall into the exception categories noted below – now have a ‘priority need’ for accommodation443. The Homelessness Code of Guidance444 urges local authorities to consider the possibilities for reconciling homeless 16-17 year olds with their families, so that they may return home to live unless it is unsafe or undesirable for them to do so. Where 16-17 year olds are provided with accommodation, local authorities should ensure that their care and support needs are addressed. As part of a package of measures delivered through a National Youth Homelessness Scheme, announced in November 2006, the Government is now committed to ending the use of B&B hotels for 16 and 17 year olds, except in an emergency, by 2010. In practice, one would expect that a high proportion of 16-17 year olds accepted as homeless will be placed in supported accommodation for a period of time rather than placed directly into their own tenancy. The benefits of a period in supported accommodation for these young people – most of whom will not previously 443 The only exceptions are 16-17 year olds who are either a ‘relevant child’ or a ‘child in need’ in terms of the Children Act 1989, where responsibility for arranging suitable accommodation rests with the children’s services authority.444 DCLG, Department of Health (DoH), Department for Education and Skills (DfES) (2006) Homelessness Code of Guidance for Local Authorities. London: DCLG.278 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldshave had a tenancy of their own – means that ‘temporary accommodation’ may have a different meaning and significance for this group than for families accepted as homeless. In 2006/7, 5,650 households were accepted as being owed the main homelessness duty primarily because the applicant was in priority need through being aged 16 or 17445; this represents a substantial reduction in absolute numbers from 10,060 in 2003/4, but this group has consistently accounted for around 8 per cent of total acceptances since the legal change to their priority need status in 2002.There is a significant body of research on homeless young people in Britain12.4 446, much of it prompted by the dramatic and very visible escalation in youth homelessness in the late 1980s and early 1990s, which many commentators attributed to high levels of youth unemployment and reductions in young people’s social security entitlements (particularly for 16-17 year olds) at that time 447. The youth homelessness literature has consistently highlighted the vulnerability of homeless teenagers, including 16-17 year olds, indicating that many have experienced family disruption; parental neglect or abuse; local authority care; poverty; poor educational experiences; and other forms of childhood trauma and disadvantage448. It also highlights the risks they face, both before and after leaving home or care449, of drug or solvent abuse450; sexual abuse (particularly young women)451; mental health problems (particularly young women)452; involvement in crime (particularly young men)453; and rough sleeping (particularly young men)454. However, most of the existing research on homeless young people has been qualitative in nature, at least within the UK, and there has been little attention paid to their experience of temporary accommodation (other than specialist provision such as foyers455). 445 This is limited to those 16-17 year olds given priority need status primarily on the grounds of age. There will be other 16-17 year olds accepted as owed the main homelessness duty but whose priority need is based primarily on other grounds (e.g. having a child in their household).446 Fitzpatrick, S., Kemp, P. and Klinker, S. (2000) Single Homelessness: An overview of research in Britain, Bristol: Policy Press.447 Fitzpatrick, S. (2000) Young Homeless People, London: Macmillan.448 Hutson, S. and Liddiard, M. (1994) Youth Homelessness: The construction of a social issue, London: MacMillan; Smith, J., Gilford, S. and O’Sullivan, A. (1998) The Family Background of Homeless Young People, York: Joseph Rowntree Foundation; Bruegel, I. and Smith, J. (1999) Taking Risks: An analysis of the risks of homelessness for young people in London, London: Safe in the City; Randall, G., and Brown, S. (2002) Trouble at Home: Family conflict, young people and homelessness, London: Crisis.449 Stein, M. (2004) What Works For Young People Leaving Care? London: Barnados; Mendes, P. and Moslehuddin, B. (2006) ‘From dependence to interdependence: Towards better outcomes for young people leaving State care’, Child Abuse Review, 15 (2): 110.450 Wincup, E., Buckland, G., and Bayliss, R. (2003) Youth Homelessness and Substance Use: Report to the Drugs and Alcohol Research Unit, London: Home Office; Mallett, S., Rosenthal, D., and Keys, D. (2005) ‘Young people, drug use and family conflict: Pathways into homelessness’, Journal of Adolescence, 28 (2): 185-199.451 Hendessi, M. (1993) ‘Supporting homeless young women survivors’, Childright, Nov, 101, 9 -10.452 Jones, A. (1999) Out of Sight, out of Mind: The experiences of homeless women, London: Crisis; Stephens, J. (2002) The Mental Health Needs of Homeless Young People, Bright Futures: Working with vulnerable young people, London: Mental Health Foundation; Vasiliou, C. (2006) Making The Link Between Mental Health And Youth Homelessness: A pan-London study, London: Mental Health Foundation.453 Carlen, P (1996) Jigsaw: A political criminology of youth homelessness, Buckingham: Open University Press; Wardhaugh, J. (2000) Sub City: Young people, homelessness and crime, Aldershot: Ashgate.454 Fitzpatrick, S. (2000) Young Homeless People, London: Macmillan.455 Anderson I. and Quilgars, D. (1995) Foyers for Young People: Evaluation of a pilot project, York: Centre for Housing Policy; Allen, C. (2001) ‘On the social consequences (and social conscience) of “the Foyer industry”: A critical ethnography’, Journal of Youth Studies, 4, (4) 471-494.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 279This chapter therefore seeks to examine whether concerns regarding the 12.5 vulnerability and needs of homeless 16-17 year olds are supported by this nationally representative survey of those accepted as owed the main homelessness duty by English local authorities, as well as to investigate their experience of accommodation provided under the statutory homelessness system. The chapter discusses: • the demographic characteristics and personal histories of young people accepted as homeless 16-17 year olds;• their reasons for seeking assistance from the local authority;• their accommodation experiences since being accepted as homeless;• their health;• their access to social and professional support; and • their economic status and financial circumstances.The data presented in this chapter comprises the most statistically robust 12.6 information available to date on young people accepted as homeless in England. It must be borne in mind, however, that it is limited to those accepted by local authorities as homeless 16-17 year olds and thus is not representative of other groups of homeless young people. Moreover, some technical limitations of the data, and contextual factors affecting its interpretation, should be noted.First, the sample size of young people accepted as homeless 16-17 year olds 12.7 is smaller than that for the families accepted as homeless: this is appropriate because the overall population of 16-17 year olds accepted as homeless is much smaller than that of families (and, as noted above, actually fell during the research period) (see para 12.3). However, the relatively smaller sample size does limit the degree of detailed statistical analysis that can be undertaken456. Second, the response rate was lower amongst young people accepted as 12.8 homeless 16-17 year olds than for families accepted as homeless; this was due mainly to ‘non-contacts’ rather than refusals to participate in the survey (see Appendix 1). There is therefore a greater risk of ‘non-response’ bias than is present in the family surveys, and it is not possible, for example, to be certain whether there is an over- or under-representation of those young people who are most vulnerable in the achieved sample (see Appendix 1). 456 It also means that, more so than in other chapters, we are unable to provide the percentage estimates to illustrate some of the associations identified because the margins of error on these percentages exceed +/-10 per cent on account of the relatively small sample sizes. However, all of the relationships noted are statistically significant at the 95 per cent confidence level. See Chapter 1, para 1.27 for a full explanation.280 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAs elsewhere in this report, we present two types of statistical analysis in 12.9 this chapter: bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a range of other factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected by regression analysis457.This survey evidence indicates that young people accepted as homeless 12.10 16-17 year olds are a far more disadvantaged and socially excluded group than families accepted as homeless. These young people’s childhoods were often marred by extremely difficult family relations, and many had also had a severely disrupted education. Large proportions had experienced mental health and/or substance misuse problems, and they were far likelier than their peers in the general population to not be in education, employment or training (NEET). Many had become NEET since leaving their last settled accommodation, and young people often reported that their financial circumstances had deteriorated. However, young people accepted as homeless 16-17 year olds made greater use of professional support services, including supported accommodation, than families accepted as homeless, and this support may have mitigated potentially worse outcomes for many of them. Key points• The majority of young people accepted as 16-17 year olds owed the main homelessness duty had experienced some form of trauma during childhood. In particular, experiences of family disruption and violence were widespread, as were disruptions to schooling.• Half (52 per cent) of young people reported that they had experienced anxiety, depression or other mental health problems. One third (33 per cent) had current mental health problems – a rate approximately three times that of young people the same age in the general population.• More than one third (37 per cent) of young people reported that they had experienced substance misuse; 16 per cent reported a current substance misuse problem.457 See Chapter 1 and Appendix 2 for more detail.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 281• Relationship breakdown between young people and their parent(s)/step-parent(s) was the dominant reason for their homelessness applications. Violence featured in around two-fifths (41 per cent) of these relationship breakdowns with parents/step-parent(s).• A large majority (85 per cent) of young people had tried to do something to address their housing problem before approaching the council. Two thirds (64 per cent) had been concerned about making a homelessness application; the main concern identified was the possibility of having to live in a ‘rough’ area.• Almost all of the young people had spent some time in temporary accommodation, and 47 per cent had experienced some form of supported accommodation. Many young people seemed to appreciate the company of other young people and the help from staff in such accommodation.• Young people’s access to emotional and instrumental support had improved overall since they left their last settled accommodation (primarily due to increased access to professional sources of support), but was still poorer than that of people the same age in the general population. A much greater proportion of young people were in receipt of practical support services than were Survey 1 adult respondents. • Over half (57 per cent) of young people were not in education, employment or training – a rate around five times that of young people the same age in the national population.• Approximately one in three young people (34 per cent) had discontinued participation in education, employment or training since leaving their last settled accommodation.• Young people had very low incomes. They were typically finding it much more difficult to cope financially in their current accommodation than they had in their last settled accommodation. As with families accepted as homeless, financial difficulties were particularly associated with living in self-contained temporary accommodation.Demographic profileAge, gender, and household typeAlmost two thirds (62 per cent) of young people accepted as homeless 16-17 12.11 year olds were female, and just over one third (38 per cent) were male458. 458 This finding is consistent with youth homelessness research which suggests that, while the young homeless population is probably roughly even in gender terms, young women are less likely to sleep rough than young men and readier to approach local authorities when they find themselves homeless. Over the age of 18, the single young homeless population becomes increasingly male-dominated, because many young homeless women have children in their late teens and early twenties and are thereafter enumerated as homeless families. See: Smith, J., Gilford, S., Kirby, P. O’Reilly, A. and Ing, P. (1996) Bright Lights and Homelessness: Family and single homelessness among young people in our cities, London: YMCA; Fitzpatrick, S. (2000) Young Homeless People, London: Macmillan.282 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsOver half (53 per cent) had turned 18 by the time they were interviewed, 42 per cent were 17 years old, and only 5 per cent were aged 16 years at the point of survey.As Figure 12.1 reveals, the majority of Survey 3 households comprised either a 12.12 single female (49 per cent) or a single male (37 per cent) with no children. Six per cent of Survey 3 young people were single women living with a child459. A further 6 per cent lived with a partner (but no children), and only 2 per cent with both their partner and child(ren). All of the children living within these households were (female) respondents’ own children.Figure 12.1: Household type of Survey 3 young people49%37%6% 6%2%0%10%20%30%40%50%Partners +child(ren)Single female +child(ren)Partners,no childrenSingle male,no childrenSingle female,no childrenSource: Survey 3 Base: 350 It should also be noted that 10 per cent of young female respondents were 12.13 pregnant at the time of survey.Ethnicity As Figure 12.2 indicates, 82 per cent of Survey 3 young people were White. 12.14 Black or Black British respondents were the next largest group (10 per cent)460. Ethnic minorities comprised a smaller proportion of young respondents than they did adult respondents to Survey 1 (18 per cent as compared with 24 per cent)461.459 Survey 3 was restricted to those 16-17 year olds who were accepted as having primary need (and therefore owed the main homeless duty) primarily because of their age, and so those young people with a child at the time of acceptance would not be in Survey 3. However, Survey 3 young people may have had a child subsequently, and as in all surveys in this study, eligibility for inclusion in the sample was defined at the point of acceptance as owed the main homelessness duty, and so their having a child by point of interview would not affect their eligibility for inclusion in Survey 3.460 Within this group, the proportion reporting Black African and Black Caribbean ethnic origin was roughly equal (at 6 per cent, and 4 per cent, of all Survey 3 young people respectively).461 By way of a general comparator, Census data indicates that 13 per cent of all 16-17 year olds in England and Wales are from an ethnic minority group (2001 Census, Office for National Statistics, Dataset C4902, Table S101, CHP analysis).Chapter 12 The experience of young people accepted as homeless 16-17 year olds 283Figure 12.2: Ethnic origin of Survey 3 young people4%82%2%10%2%Chinese or other ethnic groupBlack or Black British – Caribbean,African, any otherBlack backgroundAsian or Asian British – Indian,Pakistani, Bangladeshi, any otherAsian backgroundMixed – White and BlackCaribbean, White and Black African,White and Asian or any otherMixed backgroundWhite – British, Irish, or any otherWhite backgroundSource: Survey 3 Base: 350 Around one in twenty (6 per cent) of Survey 3 young people were former 12.15 asylum seekers (compared to 11 per cent of Survey 1 adult respondents)462. Most of these former asylum seekers were from ethnic minority groups.Where young people were accepted as homelessOne in five (20 per cent) of young respondents were accepted as homeless in 12.16 London, with the remainder being relatively evenly split between the North and Midlands (42 per cent) and the South (38 per cent) broad regions.Only 9 per cent of young people were accepted as homeless in ‘rural’ areas 12.17 (as compared with 23 per cent of families with children who were accepted in rural areas, see para 2.45)463. Given the small numbers of young people living in London, and in rural areas, 12.18 geographical analysis of Survey 3 data was not possible.462 Current asylum seekers are ineligible for assistance under the homelessness legislation.463 See Appendix 2 for an explanation of how rural and urban were defined for the purposes of this research.284 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsPersonal historyFigure 12.3 portrays the personal experiences reported by these young people12.19 464. As in Survey 1, self-completion questions were used when dealing with the most sensitive issues, including those relating to sexual matters, experiences of violence, and involvement in crime or anti-social behaviour (ASB).Clearly, a prominent issue affecting Survey 3 young people was family 12.20 disruption and restructuring during their childhoods. The parents of 67 per cent of young respondents had separated or divorced, and a step-parent had moved into the home of 51 per cent. Half (52 per cent) had been brought up by a lone parent for most of their childhood, and one or both parents of 12 per cent had died.Six in ten (61 per cent) young people stated that they ‘did not get on’ with 12.21 their parent(s) when they were growing up, and nearly half (47 per cent) had run away (and stayed away for more than one night) during their childhood. A large number (44 per cent) reported that at least one of their parents had problems with depression, anxiety or other mental health problems. Many young people had either witnessed or experienced violence within their 12.22 families when they were growing up. Two fifths (40 per cent) reported that their parents had been violent towards one another, and a similar proportion (39 per cent) had themselves been victims of violence in their childhood home. Seventeen per cent had been sexually abused as a child. One in eight (12 per cent) reported that they had been on the CPR465, and a quarter (28 per cent) reported that they had had their own social worker when growing up. A relatively large number (14 per cent) had lived in foster care and 5 per cent had lived in a children’s home. In total, 18 per cent had been ‘looked after’ by a local authority as a child466.At the same time, one quarter (25 per cent) of young people admitted that 12.23 they had been violent toward a parent and a similar proportion (28 per cent) had stolen from a parent. Over one third (39 per cent) of young people reported that they had been involved in criminal behaviour or ASB, and 9 per cent had served sentences in a prison or Young Offenders Institution.464 As with adult respondents, these experiences were selected on the basis that all have been discussed as possible factors which may heighten an individual’s risk of homelessness.465 Children on the CPR are those who have been identified by professionals as being at continuing risk of significant harm and hence in need of a child protection plan. It includes some who are not abused but thought to be at risk.466 There is no comparable data on children being ’looked after’ in the general population, as published statistics are based on ‘snapshots’ at particular points in time rather than lifetime prevalence. See Chapter 11 (para 11.102–11.103) for discussion of these national statistics. It should be borne in mind that these figures do not include any young people currently ‘looked after’ as they are the responsibility of the local children’s services rather than the housing authority.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 285Approximately one third (35 per cent) stated that their family had ‘moved 12.24 around a lot’ during their childhood, and 15 per cent said that their family had spent some time homeless when they were children. Almost one in ten (9 per cent) had themselves applied as homeless before their current homelessness application. The educational experiences of young people often seemed to have been as 12.25 disrupted as their home life. More than half (58 per cent) reported that they had ‘missed a lot of school’, and a similar proportion (54 per cent) had been suspended or excluded from school on at least one occasion.The violence experienced by many young people in childhood appeared to 12.26 afflict a notable proportion in young adulthood too. One quarter (28 per cent) had been the victim in a violent relationship with a boyfriend/girlfriend or partner. In addition, 12 per cent had been sexually assaulted since turning 16 years of age.Approximately half (52 per cent) self-reported experience of anxiety, 12.27 depression or other mental health problems, and more than one third (37 per cent) reported experiencing problems with drugs, solvents or alcohol467.Only two per cent of young people had not had any of the experiences listed 12.28 in Figure 12.3.467 Detailed discussion of self-reported current mental health problems amongst young respondents, as well as problems associated with substance misuse, is provided later in this chapter.286 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 12.3: Personal history of Survey 3 young people0% 10% 20% 30%Percent of respondents40% 50% 60% 70%Has had sex in exchange formoney, food, drugs, shelterHas lived as a 'traveller'Has lived in a children's homeHas been in prison or YOIHas been sexually assaulted as an adultOne or both parents diedHas been on the Child Protection RegisterHas lived with foster carersFamily spent some time homelessduring childhoodWas sexually abused as childHas been been violent to a parentHas been in a violent relationship (as abused)Had own social worker when growing upHas stolen from a parentFamily moved house a lot during childhoodExperience of drug, solvent oralcohol problemsHas been involved in crime/ASBExperienced domestic violence as a childParents were violent towards eachother during childhoodParents had mental health problemsRan away from home and stayed away formore than one night during childhoodA step-parent moved into the family homeduring childhoodWas brought up by a lone parent formost of childhoodHad ever experienced anxiety, depressionor other mental health problemsWas suspended or excluded fromschool at least onceMissed a lot of schoolDidn't get on with parents during childhoodParents separated or divorcedduring childhood67%61%58%54%52%52%51%47%44%40%39%39%37%35%28%28%28%25%17%15%5%3%3%14%12%12%12%9%Source: Survey 3 Base: 350 Multiple responses were possible. Chapter 12 The experience of young people accepted as homeless 16-17 year olds 287Several of the experiences described above were strongly associated with 12.29 gender. Notably, young men were more likely to have been suspended or excluded from school (68 per cent of young male respondents as compared with 46 per cent of young female respondents), involved in crime or ASB (59 per cent compared to 27 per cent of young women), or incarcerated in prison or a Young Offenders Institution (17 per cent as opposed to 4 per cent of young women). In contrast, young women were more likely to report that they did not get on with their parents when growing up (70 per cent of females compared to 47 per cent of males), to have been victims of a violent relationship with a partner (35 per cent as opposed to 16 per cent of young men), or to have been sexually assaulted as an adult (17 per cent compared to 2 per cent of young men).A number of these and other experiences were associated with ethnicity12.30 468. White young people were more likely than those from ethnic minority backgrounds to have run away from home, to have been suspended or excluded from school, to have missed a lot of school, to have had their own social worker, to have been violent toward a parent, or to have stolen from a parent. The family backgrounds of White young people and those from ethnic minority groups also tended to be very different. White young people more commonly reported that their parents had divorced or separated, that a step-parent had moved into the family, that a parent suffered from mental health problems, or that their parents had been violent towards one another469. The inter-relationships between many of the personal experiences discussed 12.31 above and other aspects of vulnerability reported by young people, were assessed using K-means cluster analysis (see Appendix 2). Four groups of respondents were identified, henceforth referred to as ‘vulnerability clusters’. These included:• Cluster One – ‘multiple problems’. Almost all of the young people in this group had suffered from mental health problems at some point in their lives; they were more than twice as likely to have had problems with drugs, solvents or alcohol as other young respondents; nearly three quarters had been involved in crime or ASB; and the great majority had witnessed or experienced violence at home when growing up. One quarter (26 per cent) of young people were classified in this vulnerability cluster.• Cluster Two – ‘mental health and other problems’. Young people in this group were also likely to have suffered mental health problems, but were much less 468 Some of these ethnicity-related findings may be affected by the relatively high proportion of ethnic minority Survey 3 respondents who were former asylum seekers (around one third). However, the sub-sample numbers were too small for detailed analysis to test this.469 The margins of error are too wide with respect to ethnic minority young people for the actual percentages to be given in this paragraph, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).288 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldslikely to have been involved in crime or ASB, or to have had problems related to substance abuse than respondents in Cluster One. Approximately one fifth (21 per cent) of young people fell into this group.• Cluster Three – ‘offending and other problems’. This group is differentiated from the others by a high level of self-reported involvement in crime and/or antisocial behaviour, but low incidence of mental health problems. One fifth (20 per cent) of young people were classified in this group.• Cluster Four – ‘fewest problems’. Respondents in this group were the least likely to report the following: having been involved in crime or ASB, substance abuse or mental health problems, spending time in care, being homeless as a child, running away from home, witnessing or experiencing violence in the home, being on the CPR, having had their own social worker, or having had their education disrupted when a child. This was the largest group, comprising one third (33 per cent) of all young people.These vulnerability clusters were strongly associated with gender. Young 12.32 women were disproportionately represented in the ‘mental health and other problems’ vulnerability cluster (29 per cent of females as compared with 10 per cent of males). In contrast, the proportion of young men in the ‘offending and other problems’ vulnerability cluster was almost twice that of women (28 per cent as compared with 15 per cent).These four vulnerability clusters were used throughout the analysis of Survey 12.33 3 data and are therefore referred to periodically in the discussion below.Seeking help from a local authorityAlthough 72 per cent of Survey 3 young people identified their parental 12.34 home as their ‘last settled accommodation’470, only one quarter (27 per cent) were living with their parents when they sought help from a local authority. A further 10 per cent were living in their partner’s house or flat. Almost one third (30 per cent) were living with friends or relatives; a total of 23 per cent were living in a hostel or other form of supported accommodation, a B&B hotel, or supported lodgings; and a further 5 per cent were sleeping rough.470 See footnote 486 for further details on young people’s last settled accommodation.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 289Table 12.1: Accommodation Survey 3 young people were living in when sought help from the local authorityType of accommodation PercentageFriends/ relatives house or flat 30%Parent/s house or flat 27%Hostel or other supported accommodation 15%Partner’s house or flat 10%B&B hotel 6%Sleeping rough 5%Council/ housing association rented accommodation 4%Supported lodgings 2%NASS accommodation <1%Squatting <1%Private rented accommodation <1%Other <1%Total 100%Base 350Source: Survey 3 Around two thirds (69 per cent) of young people knew they were going to 12.35 apply as homeless when they approached a local authority. These young people found out how to apply as homeless from two main 12.36 sources: family and/or friends (43 per cent) and professional workers (39 per cent)471. Very few (less than 5 per cent) of those who knew they would apply as homeless found out about doing so from any other source (e.g. leaflets/booklets). There was no association between the sources of information about applying as homeless and young peoples’ demographic characteristics or vulnerability clusters.Approximately one third (31 per cent) of young people, therefore, did not 12.37 know they were going to apply as homeless before going to the council. Of these, most (82 per cent) said they went to the local authority as they ‘needed help with their housing situation but did not know what to do’. Far less commonly, they approached the council to get on the housing waiting list/register (16 per cent). 471 The proportions for both categories are very similar to that of adult respondents to Survey 1 (see Figure 4.4).290 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAwareness of the statutory homelessness arrangements was not associated 12.38 with demographic characteristics or vulnerability clusters.As one might expect, given their young age, only a minority (13 per cent) of 12.39 Survey 3 respondents reported that they were already on the housing register or a waiting list before applying as homeless.Young people were also asked whether they had done anything to address 12.40 their housing situation before they sought assistance from the council. Like Survey 1 adult respondents, the vast majority (85 per cent) had tried at least one of a number of strategies, as listed in Figure 12.4. Seeking informal help was by far the most common action taken, with about 12.41 half of young people having asked friends to let them stay and also about half having asked family to let them stay. A smaller proportion had sought professional help by seeking advice or assistance from a Connexions personal advisor (17 per cent), their own social worker (13 per cent), a youth worker (12 per cent), housing advice centre (10 per cent), social services (8 per cent), young people’s advice centre (5 per cent), or a teacher or other professional at school (4 per cent). Around one in five young people (18 per cent) had tried to rent a flat/house from a private landlord472.There was no gender variation in the proportion of young people employing 12.42 at least one of the strategies listed above before seeking assistance from their local authority, but young women were more likely than young men to have approached social services for help with their housing situation (14 per cent compared to 4 per cent).472 Survey 3 young people were therefore notably less likely to try to secure housing in the private rented sector before seeking help from the local authority than were Survey 1 adult respondents (18 per cent had done so, as compared with 33 per cent of Survey 1 adult respondents).Chapter 12 The experience of young people accepted as homeless 16-17 year olds 291Figure 12.4: The actions undertaken by Survey 3 young people before approaching a local authority for help57%56%18%17%13%12%10%8%5%4%2%2%1%1%1%0% 10% 20% 30%% of young people40% 50% 60%Looked on the internet for adviceSpoke to someone at young people'sday-centre/club/groupSpoke to a family mediation serviceSpoke to a telephone service for young peopleTried to get help from a rent deposit schemeSpoke to a teacher or other professional at schoolSpoke to a young people's advice centreabout housing problemsAsked social servicesSpoke to a housing advice centreSpoke to youth worker about housing problemsSpoke to own social workerSpoke to a Connexions personal adviserTried to get a flat/house to rentfrom a private landlordTried to get (other) friends to let them stayTried to get (other) family to let them staySource: Survey 3 Base: 350 Multiple responses were possible.Young people in the ‘multiple problems’ vulnerability cluster were most likely 12.43 to have adopted one or more of the strategies described above, with those in the ‘fewest problems’ vulnerability cluster being the least likely to have done so. This might be explained by those in the ‘multiple problems’ vulnerability cluster being more aware of the different services available to those who need support because of previous contact with social services and/or the criminal justice system, etc.All young people were asked about any concerns they may have had about 12.44 making a homelessness application. Whilst just over one third (36 per cent) stated that they were not concerned about applying as homeless, the rest (64 per cent) identified at least one concern about making a homeless application. Figure 12.5 shows that the most common concern reported by young people was that they might have to live in a rough area (reported by 32 per cent). In addition, 23 per cent were concerned about the quality 292 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsof housing that they were likely to be given. Young women were more concerned than young men about both of these factors (39 per cent of young women compared to 20 per cent of young men for living in a rough area, and 28 per cent compared to 16 per cent for quality of housing).The possibility of having to live far from family or friends was cited as a 12.45 concern about applying as homeless by 21 per cent of young people. They also had some concerns with respect to having to accept the first house/flat given (17 per cent); that they would not be accepted as homeless (15 per cent); and/or that they would be ‘labelled’ as homeless (14 per cent) (Figure 12.5).473 Figure 12.5: Survey 3 young peoples’ concerns about making a homelessness application32%23%21%17%15%14%0% 10% 20%% of young people30% 40%Being labelled 'homeless'Would not be acceptedas homelessWould have to accept firstplace that was offeredWould have to live farfrom friends/familyWould be given a poorquality flat/houseWould have to livein a 'rough' areaSource: Survey 3 Base: 350 Multiple responses were possible. More than half (57 per cent) of young people felt that they were very or 12.46 fairly well informed whilst the council was making a decision about their application. However, 23 per cent felt ‘not very well informed’ and 19 per cent felt ‘not informed at all’. These findings are very similar to those for adult respondents in Survey 1 (see para 4.33).473 See Figure 4.5 for concerns about making a homelessness application expressed by adult respondents in Survey 1.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 293Reasons for applying as homelessTable 12.2 lists all of the reasons reported by young people for applying to 12.47 the council as homeless474. The overwhelming reason – reported by 70 per cent – was relationship breakdown. One fifth (19 per cent) reported that they had overstayed their welcome or could no longer be accommodated, while overcrowding was a reason given by 13 per cent of young people. A range of other reasons were identified but never by more than five per cent of young people. Notably, only five per cent stated that one of the reasons for applying as homeless was because this was the only way to get rehoused, and only 3 per cent felt that this was the quickest way to get rehoused.Table 12.2: Reasons for applying as homeless given by Survey 3 young peoplePercentageRelationship breakdown 70%Overstayed welcome/ could no longer be accommodated 19%Housing was overcrowded 13%Applying as homeless was the only way to get rehoused 5%Applying as homeless was the quickest way to get rehoused 3%Eviction or threatened with eviction 3%Harassment, anti-social behaviour or crime 2%Had to leave NASS accommodation 2%Mental health problems or physical health problems 2%Drug or alcohol problems 2%Social worker thought they needed to move somewhere else 2%Pregnancy – had to move out of previous home 2%Tenancy came to an end 1%Housing was in poor condition 1%Problems with paying the mortgage or rent <1%Other 4%Base 350Source: Survey 3 Multiple responses were possible.When asked to identify a single ‘main reason’ for applying as homeless, the 12.48 responses given by young people closely resembled that for ‘all reasons’ shown in Table 12.2 – with, for example, 65 per cent identifying relationship 474 Like Survey 1 adult respondents, Survey 3 young people were asked to identify all the reasons that they felt had contributed to their application as homeless from a pre-defined list of prompts.294 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsbreakdown, 10 per cent reporting that they had outstayed their welcome or could no longer be accommodated, and 10 per cent that their housing was overcrowded.The reasons for applying as homeless reported by young people did not vary 12.49 by demographic characteristics or vulnerability cluster.A closer look at relationship breakdownYoung people identifying relationship breakdown as a cause of their 12.50 homelessness were asked who their relationship had broken down with. As Table 12.3 shows, over three quarters (78 per cent) of these individuals reported that the relationship breakdown had occurred between themselves and their (own) parent or parents. A further 15 per cent stated that the relationship breakdown had been with a step-parent, and 13 per cent reported that the relationship breakdown involved other members of the family.Table 12.3: Who Survey 3 young peoples’ relationships had broken down withSource of relationship breakdown PercentageParent(s) 78%Step-parent(s) 15%Other member(s) of the family 13%Boyfriend/girlfriend/partner 5%Partner’s parent(s) 2%Friend(s) 1%Foster carer(s) -%Other 3%Base 238Source: Survey 3 Multiple responses were possible Where the relationship breakdown was with a parent or step-parent, young 12.51 people were asked why the relationship broke down. The most common response, given by 77 per cent, was that they ‘just could not get along with each other’ (Table 12.4). A range of other more specific reasons were given by far smaller proportions of young people, including: their parent/step-parent did not like their choice of partner (12 per cent); the young person was involved in anti-social behaviour or crime (10 per cent); their parent/step-parent(s)’ drug or alcohol problems were hard to live with (8 per cent); and they did not like the young person being involved with drugs or alcohol (9 per cent).Chapter 12 The experience of young people accepted as homeless 16-17 year olds 295Table 12.4: Reasons for relationship breakdown of Survey 3 young people with parent(s) or step-parent(s)Reason PercentageJust could not get along with each other 77%They did not like my choice of partner/girlfriend/boyfriend 12%I was involved in anti-social behaviour or crime 10%They did not like me being involved with drugs or alcohol 9%I was told it was time I got my own place 9%Parent’s/step-parent’s drug/alcohol problems 8%I was not paying my own way 8%They did not like me being unemployed and in the house all day 6%They did not like the company I kept 6%I/my partner became pregnant 3%They did not like me being gay/lesbian/bi-sexual 2%Other reason 7%Base 191Source: Survey 3 Multiple responses were possible Nearly half (45 per cent) of all the young people affected by relationship 12.52 breakdown reported that violence had been involved. Notably, 41 per cent of those identifying relationship breakdown with parents/step-parents reported that violence had been involved. Young men and women were equally likely to report experiencing violence within relationship breakdowns. There was no association between demographic characteristics or vulnerability clusters and the likelihood of violent relationship breakdown. Accommodation experiences since being accepted as homelessHousing situation at point of surveyAt the point of survey, 40 per cent of Survey 3 young people had moved into 12.53 settled accommodation, while the remaining 60 per cent were still living in temporary accommodation.As Table 12.5 shows, almost all (95 per cent) of those who had moved 12.54 into settled accommodation reported that they were living in social rented housing, with only 3 per cent in the private rented sector.296 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsOf those still in temporary accommodation, a total of 38 per cent were 12.55 living in ‘self-contained’ temporary accommodation – including 32 per cent in council/housing association accommodation, 4 per cent in their partner’s flat/house, and 2 per cent in private rented housing. The remainder (62 per cent) were living in ‘shared’ temporary accommodation – including 30 per cent in a hostel or other supported/managed accommodation, 10 per cent in supported lodgings, 11 per cent with their parents, 8 per cent in a B&B hotel, and 4 per cent with friends or relatives (see Table 12.5). This broad differentiation between ‘self-contained’ temporary accommodation and ‘shared’ temporary accommodation is utilised in the analysis throughout the remainder of this chapter475.Table 12.5: Current accommodation type of Survey 3 young people, by whether in settled or temporary accommodationAccommodation type Settled accommodationTA AllOwn/ jointly own (inc. mortgage) flat/house2% – 1%Rent from council/ housing association95% 32% 57%Rent from private landlord 3% 2% 2%Partner’s flat/house – 4% 2%Parent’s flat/house – 11% 7%Friend(s)’/ relative(s)’ flat/house – 4% 2%B&B hotel – 8% 5%Hostel or other managed/supported accommodation– 30% 18%Supported lodgings – 10% 6%Total 100% 100% 100%Base 129 219 348Source: Survey 3. There was no association between any demographic characteristics or 12.56 vulnerability cluster, and whether or not young people had moved into settled accommodation476.475 As with Survey 1 families accepted as homeless (footnote 205), the figures below on the split between social and private rented temporary accommodation may be inaccurate as respondents were generally unaware of the tenure of self-contained temporary accommodation.476 As discussed in Chapter 6, geographical variables were the key influence on whether or not families with children had been provided with settled housing by point of interview. However, as noted above (see para 12.18), it was not possible to conduct geographical analysis on the Survey 3 dataset.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 297Time spent in temporary accommodationThe vast majority of Survey 3 young people (94 per cent) reported that they 12.57 had spent some time in temporary accommodation, with only 6 per cent moving directly into settled accommodation from the place from which they were accepted as homeless.Table 12.6 shows that similar proportions of young people had spent less 12.58 than 6 months in temporary accommodation (41 per cent) as had spent 7-12 months in temporary accommodation (49 per cent). Only 3 per cent of all young people had spent over a year in temporary accommodation. The average total period of time spent in temporary accommodation by point of survey was 6.9 months (median was 7 months)477.Table 12.6: Time spent in temporary accommodation at point of survey by Survey 3 young people, by whether in settled or temporary accommodationTime spent in TA Settled accommodationStill in TA All0-6 months 59% 30% 41%7-12 months 29% 63% 49%More than 1 year – 5% 3%Moved directly into settled accommodation12% – 6%Total 100% 100% 100%Base 129 219 348Source: Survey 3. A higher proportion of those in settled accommodation had experienced 12.59 shorter stays in temporary accommodation (59 per cent had spent less than 6 months) compared to those still in temporary accommodation (30 per cent had spent less than 6 months in temporary accommodation) (Table 12.6). The average period of time spent in temporary accommodation by young people who had moved into settled accommodation was 5.2 months (median 5 months), whilst it was 7.9 months (median 8 months) for those still in temporary accommodation at the point of survey478.Type of temporary accommodation experiencedAs Figure 12.6 illustrates, one third (34 per cent) of all young respondents 12.60 (with experience of temporary accommodation) had stayed in self-contained 477 As with Survey 1, these figures are based on survey responses which are subject to recall problems and therefore should not be interpreted as exact periods of stay.478 It is not possible to predict the total length of time that this latter group will spend in temporary accommodation.298 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldstemporary accommodation. A slightly greater proportion (41 per cent) had lived in a hostel or other managed/supported accommodation, whilst a total of 38 per cent had stayed with friends and/or relatives, and 30 per cent had stayed with their parents. Around one quarter (28 per cent) had lived in a B&B hotel, and 11 per cent had lived in supported lodgings at some point.Figure 12.6: Type of temporary accommodation experienced by Survey 3 young people41%38%34%30%28%11%% of respondents0% 5% 10% 15% 20% 25% 30% 35% 40% 45%Supported lodgingsB&B hotelParentsSelf-contained TAFriends or relativesHostel or other managed/supported accommodationSource: Survey 3 Base: 329 (respondents with experience of TA) Multiple responses were possible. In total, 47 per cent of young people had spent a period in some form 12.61 of ‘supported accommodation’ (i.e. hostel, other managed/supported accommodation or supported lodgings). Young women were more likely to report having stayed in self-contained 12.62 temporary accommodation than were young men (38 per cent as compared to 25 per cent). There were no other associations between experience of different types of temporary accommodation and demographic characteristics, vulnerability cluster, or whether or not young people reported current mental health or substance misuse problems.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 299Moves between temporary accommodation addressesOf the young people who had stayed in temporary accommodation, nearly 12.63 half (46 per cent) had not moved between temporary accommodation addresses, 36 per cent had moved once, and 18 per cent moved two or more times479. As with Survey 1 families, those living in settled accommodation at the point of the survey were less likely to have moved between temporary accommodation addresses (43 per cent) than those still in temporary accommodation (61 per cent). There was no difference between the number of moves made by young people with different demographic characteristics, in different vulnerability clusters, or by those who did or did not report current mental health or substance misuse problems.Current accommodation conditions and overall satisfactionYoung people in Survey 3 were asked a series of questions regarding their 12.64 experience of, and satisfaction with, their current accommodation. This section compares the experience of those young people living in temporary accommodation with those who had moved into settled accommodation.A mixed picture emerged in terms of the relative advantages offered by 12.65 temporary and settled accommodation. As Figure 12.7 illustrates, a greater proportion of young people in settled accommodation reported that there was enough living space (92 per cent compared to 70 per cent in temporary accommodation), and that the rules and regulations were generally fair (97 per cent as compared to 89 per cent). Those living in settled accommodation were also less likely to report problems with infestations (5 per cent compared to 14 per cent). However, young people in temporary accommodation were more likely to report that their accommodation was well decorated when they arrived (67 per cent, compared to 34 per cent of those in settled accommodation), and less likely to report that it was damp (21 per cent, compared to 35 per cent).479 This means that young people were more likely to have moved between temporary accommodation addresses than families accepted as homeless (see Chapter 6), but multiple moves were still relatively rare.300 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 12.7: Current accommodation conditions experienced by Survey 3 young people, by whether in temporary or settled accommodation89%97%62%92%77%34%39%35%5%73%70%67%67%31%21%14%InfestationDamp in placesDirty when arrivedFairly well decoratedwhen arrivedSufficient controlover heatingEnough living spaceReasonable state ofrepair when arrivedRules and regulationsgenerally fair0% 20% 40% 60%% of respondents80% 100%Source: Survey 3 Base: 129 respondents in settled accommodation, 219 in temporary accommodationAs would be expected, experiences of sharing were almost totally confined 12.66 to those still in temporary accommodation, with 50 per cent of these young respondents sharing at least one room. About a third of those in temporary accommodation shared a lounge or living area (30 per cent) or kitchen (34 per cent), with a higher proportion sharing a bath or shower room (44 per cent). None of the Survey 3 young people were required to share a bedroom.Nearly half (49 per cent) of young people sharing at least one room identified 12.67 this as being a problem. The main problems identified were lack of privacy (33 per cent of all those sharing); other residents not being clean/hygienic (28 per cent); the need to wait/queue or availability at inconvenient times (18 per cent); too many people/not enough space for people to share (11 per cent); noise from other people (10 per cent); and theft of or damage to respondents’ property (9 per cent). Whilst sharing clearly presented a problem for some, seven in ten (71 per cent 12.68 of) young people living in a hostel, other supported accommodation, or B&B hotel, felt that other people living there were good company. In addition, 77 per cent of young people in such temporary accommodation settings reported that the staff were helpful, and only 17 per cent thought that the staff interfered too much.In total, one quarter (25 per cent) of young people reported that they 12.69 did not feel safe inside their current accommodation, and 29 per cent Chapter 12 The experience of young people accepted as homeless 16-17 year olds 301reported feeling unsafe in the area. There were no overall differences in perceptions of safety (indoors or in the neighbourhood) between those in settled and temporary accommodation, but young people in self-contained temporary accommodation were more likely to report feeling unsafe in their accommodation and/or in the area than those in shared temporary accommodation480. Young women were more likely to report feeling unsafe in the area than were young men (36 per cent compared to 17 per cent), as were young people with current mental health problems compared to other young people481.When asked to rank their accommodation on a scale of 1 to 10 (where 1 is 12.70 ‘terrible’ and 10 is ‘excellent’), marginally higher levels of overall satisfaction were reported by young people in settled accommodation (average of 6.4) compared to those in temporary accommodation (average of 5.9)482. Of those in temporary accommodation, individuals living in shared temporary accommodation were slightly more satisfied (average score of 6.1) than those living in self-contained temporary accommodation (average score of 5.6). Health General health status and longstanding illness or disabilityFigure 12.8 compares young respondent’s self-assessments of their current 12.71 general health status to that of young people in the equivalent age cohort from the HSE (2003)483. This indicates a clear discrepancy between the self-assessed general health of young respondents and that of their peers in the general population, as 32 per cent of young respondents felt that their health was ‘bad’, ‘very bad’ or simply ‘fair’, in comparison with only 9 per cent of 16-18 year olds nationally. Hence, whilst 91 per cent of 16-18 year olds nationally assessed their general health as ‘good’ or ‘very good’, only 68 per cent of young respondents were so positive about their health.480 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).481 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).482 These figures are very similar to the satisfaction ratings given by Survey 1 adult respondents, although Survey 3 young people in settled accommodation gave their housing a slightly lower average score (6.4) than did Survey 1 adult respondents in settled accommodation (7.0).483 National Centre for Social Research, Health Survey for England, 2003, SN: 5098, (16-18 year olds, weighted data, CHP analysis).302 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 12.8: Current general health status of Survey 3 young people, with national comparison (16-18 year olds)29%Very badBadFairGoodVery good39%22%9%1%48%43%8%1% 0%HSE 2003 ComparatorSource: Survey 3 and HSE (2003) (16-18 years) Base: Survey 3 350, HSE (2003) 803. CHP analysisThere was no variation in current general health status by demographic 12.72 characteristics, current accommodation type, or vulnerability cluster.At 25 per cent, the proportion of Survey 3 young people reporting a long-12.73 standing illness was not substantially different to that of 16-18 year olds nationally (21 per cent)484, but the percentage of young people reporting a limiting long-standing illness was greater than that of young people the same age in the population as a whole (17 per cent as compared with 7 per cent respectively)485.A mixed picture emerged with regards to changes in general health, as the 12.74 percentage of young people for whom this had improved since leaving their last settled accommodation was approximately equal to the proportion for whom it had deteriorated (18 as compared with 17 per cent respectively)486. There was no variation in changes in general health status experienced between individuals in settled or temporary accommodation, or those living in different types of temporary accommodation. 484 National Centre for Social Research, Health Survey for England, 2003, SN: 5098. (16-18 year olds, weighted data, CHP analysis.)485 Defined as any long-term illness or disability that ‘limits your activities in any way’.486 Two-thirds (66 per cent) of Survey 3 young people had a ‘last settled accommodation’ which could be used for comparative purposes (see Chapter 1 and Appendix 1). For almost three-quarters of these young people (72 per cent), their last settled accommodation had been their parents’ homes. For the others, it was the home of friends or relatives (14 per cent); a social rented flat (7 per cent); or their partner’s home (3 per cent). Four per cent claimed that it was a property that they owned or jointly owned themselves, but this seems unlikely to be accurate.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 303Around one in eight (13 per cent) of Survey 3 young people self-reported 12.75 having dyslexia, and 10 per cent said that they had other learning difficulties. Dyslexia and other learning difficulties were therefore much more prevalent amongst young people than amongst Survey 1 adult respondents (at 5 per cent and 2 per cent respectively, see para 9.21). As noted in Chapter 9, estimates of dyslexia in the general population vary between 2 and 15 per cent, depending on whether mild or severe dyslexia is being recorded487, but learning difficulties of other kinds affect a very low proportion of the general population (around 2 per cent)488.Changes in dietAt 36 per cent, the proportion of Survey 1 young people reporting a 12.76 deterioration in quality of diet since leaving their last settled accommodation was greater than the proportion for whom it had improved (29 per cent). There were no differences between those in settled and temporary accommodation, or between young people in different types of temporary accommodation, with respect to changes in diet.Mental health problemsHalf (52 per cent) of young people reported that they had 12.77 ever experienced depression, anxiety or other mental health problems, with one third (33 per cent) reporting that they had current mental health problems489. Amongst those with current mental health problems, these were likelier to have got worse rather than better since leaving their last settled accommodation490. Broadly indicative comparative data from the HSE (2003) suggests that the 12.78 prevalence of current mental health problems amongst young people could be nearly three times that of their peers nationally491. Only 12 per cent of the 16-18 year olds responding to the HSE (2003) self-reported that they were anxious or depressed492.487 Source: Parliamentary Office of Science and Technology (2004) Postnote: Dyslexia and Dyscalculia July 2004 Number 226 available at: http://www.parliament.uk/documents/upload/POSTpn226.pdf488 Source: British Institute of Learning Difficulties (undated) Factsheet – learning disabilities http://www.bild.org.uk/docs/05faqs/Factsheet%20Learning%20Disabilities.pdf489 The proportion of Survey 3 young people who had ever experienced mental health problems was almost exactly the same as for adult respondents in Survey 1 (53 per cent). However, the proportion of young people with a current mental health problem was marginally higher (33 per cent as compared with 27 per cent amongst Survey 1 adult respondents), and substantially higher if only women are considered (40 per cent of Survey 3 young women as compared with 28 per cent of Survey 1 female adult respondents reported current mental health problems).490 However, we don’t know if any of those without current mental health problems had such problems in the their last settled accommodation (see also para 9.30 for a discussion of this point with regards to adult respondents in Survey 1).491 Slight differences in the question about mental health in Survey 3 and the HSE 2003 dictate that any comparisons made should be interpreted as indicative only. The HSE asked respondents whether they were currently ‘not anxious or depressed’, ‘moderately anxious or depressed’ or ‘extremely anxious or depressed’. Affirmative responses to either of the latter two options are here combined to create the total 12 per cent cited above.492 National Centre for Social Research, Health Survey for England 2003, 16-18 year olds, weighted data, CHP analysis.304 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsRegression analysis indicated that, other things being equal12.79 493, the factors that were independently associated with Survey 3 young people having a current mental health problem were:• experiencing violence as a child: 45 per cent of young people who said that they had experienced violence as a child reported current mental health problems, as compared with 20 per cent of other young people.• being a young woman: 40 per cent of young women reported a current mental health problem, as compared to 24 per cent of young men. • having a current substance misuse problem: young people who reported current substance misuse were also more likely to report a current mental health problem494. • living in shared temporary accommodation: young people living in shared temporary accommodation were less likely to report a current mental health problem than other young people495. This possibly reflects benefits from the company and professional support provided in (some) such accommodation. Few young people identified mental health problems as their main reason for 12.80 applying to the council as homeless (see Table 12.2), but the prevalence of such problems – and their influence in defining the four vulnerability clusters (see para 12.31) – indicates that mental health is a crucially important issue for some members of this population.Drug and alcohol problemsMore than one third (37 per cent) of young people reported that they had 12.81 ever had problems due to drug and/or alcohol and/or solvent use, and 16 per cent still did.493 Factors controlled for in this regression analysis included: demographic characteristics; sexually abused as a child; violence within home as a child; been involved in crime or ASB; current substance misuse problems; spent time in care as a child; current accommodation type; temporary accommodation experiences; accommodation conditions; current financial difficulties, and ability to manage financially has deteriorated; whether NEET; social networks.494 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).495 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).Chapter 12 The experience of young people accepted as homeless 16-17 year olds 305Considering drug use specifically, over one quarter (28 per cent) of young 12.82 people reported that they had ever had problems due to drugs; 11 per cent still did. No national comparative data is available496, but this figure is far greater than the percentage of Survey 1 adult respondents self-reporting drug problems (see Table 9.3).Almost one fifth (19 per cent) of young people had 12.83 ever experienced problems due to alcohol use; 7 per cent still did. Whilst 11 per cent had ever had problems due to solvent use, only 1 per cent reported that they still did.Table 12.7 presents the prevalence of current poly-substance misuse. This 12.84 reveals that 8 per cent of young people reported having a current drug problem only, 3 per cent reported a current alcohol problem only, and 1 per cent a current solvent problem only. A further 4 per cent self-reported a current drug and alcohol problem.Table 12.7: Prevalence of current poly-substance problems amongst Survey 3 young peoplePercentageDrug problem only 8%Solvent problem only 1%Alcohol problem only 3%Drug and alcohol problem 4%None of the above 84%Total 100%Base 350Source: Survey 3. Regression analysis indicated that, other things being equal12.85 497, three factors exerted an independent influence which made it more likely that a young person would have a current substance misuse problem:496 Other surveys of self-reported drug-use typically report the proportion of young people who have ever used illicit drugs rather than measure the prevalence of problematic drug use. See for example Beckett, H., Heap, J., McArdle, P., Gilvarry, E., Christian, J., Bloor, R., Crome, I and Frischer, M (2004) Understanding Problem Drug Use Among Young People Accessing Drug Services: A multivariate approach using statistical modelling techniques, Home Office Online Report 15/04, London: Home Office; Neale, J. (2005) ‘Children, Crime and Illegal Drug Use’, in Bradshaw, J. and Mayhew, E. (eds) The Well-being of Children in the UK, London: Save the Children, pp 239-261.497 Factors controlled for in this regression analysis included: demographic characteristics; sexually abused as a child; violence within home as a child; been involved in crime or ASB; current mental health problem; spent time in care as a child; current accommodation type; temporary accommodation experiences; accommodation conditions; current financial difficulties, and ability to manage financially has deteriorated; whether NEET; social networks.306 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds• having a current mental health problem (this was the strongest effect)498; • not being in employment, training or education (i.e. being NEET)499; • having been involved in crime or ASB500; • gender: 22 per cent of young men, and only 12 per cent of young women, reported a current substance misuse problem. Social and professional support This section looks at the social and professional support available to young 12.86 people. It examines the extent to which their practical and personal support needs have been met by professional services, before exploring their access to emotional and instrumental support and changes to their social networks since leaving their last settled accommodation.Practical and personal support Table 12.8 summarises the proportion of young people in receipt of particular 12.87 services, together with those (self)-identifying unmet support needs.498 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).499 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).500 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27)Chapter 12 The experience of young people accepted as homeless 16-17 year olds 307Table 12.8: Survey 3 young people receiving and needing support servicesReceiving helpUnmet need for helpDon’t need helpTotal BasePractical support needsSomeone to help fill in official forms or apply for benefits43% 10% 48% 100% 350Practical or financial help getting furniture or other household equipment34% 12% 53% 100% 350Advice or help finding a job 33% 8% 59% 100% 350Advice or help with repairs to accommodation32% 7% 61% 100% 350Someone to speak to official people like social services, the council, or landlord30% 10% 59% 100% 350Advice or help getting into education or training30% 6% 65% 100% 350Advice or help with managing money, budgeting, or dealing with debts26% 14% 60% 100% 350Advice or help getting to see a doctor or accessing other health services16% 4% 80% 100% 350Advice or help with cooking, cleaning or doing the laundry9% 3% 87% 100% 350Personal support needsAdvice or help dealing with mental health problems, including depression and anxiety17% 5% 79% 100% 350Advice or help dealing with drug problems7% 2% 91% 100% 350Advice or help dealing with alcohol problems0% 0% 100% 100% 350Source: Survey 3 308 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsReceipt of practical support was overall greater amongst young people than 12.88 it was Survey 1 adult respondents – most notably the provision of assistance with official forms and benefit applications: 43 per cent of young people reported receiving this help as compared with 21 per cent of Survey 1 adult respondents (Table 9.4).Levels of unmet need reported by young people were relatively low (as 12.89 was also true for Survey 1 adult respondents, see Table 9.4). Assistance with financial management was the largest self-reported unmet support need, with 14 per cent of young people wanting help (which they were not currently receiving) to manage their money, budget and/or deal with debt. Approximately one third (34 per cent) of young people had received 12.90 assistance setting up home in their new accommodation (e.g. acquiring furniture and/or with repairs), but 12 per cent felt that help acquiring furniture or other household goods remained, for them, an unmet need.Young people were approximately three times more likely to be in receipt of 12.91 services to facilitate participation in employment (33 per cent) or education and training (30 per cent) than were Survey 1 adult respondents (11 per cent and 10 per cent respectively) (see Table 9.4). Yet, 8 per cent of young people wanted advice or help to find a job but were not getting it, and 6 per cent felt that assistance to get into education and training was an unmet need for them.Only a very small minority (3 per cent) of young people felt that they were not 12.92 receiving the help and advice they needed with regard to general ‘life skills’ such as cooking, cleaning or doing the laundry.Whilst one third (33 per cent) of young people reported that they currently 12.93 suffered from depression, anxiety or other mental problems, only 17 per cent were in receipt of support services for these problems, though only 5 per cent felt that they needed help or advice but were not receiving it.Use of the NHS and care and support servicesTable 12.9 shows the professional and other support services ever seen by 12.94 the young people, together with the proportion using such services in their current (temporary or settled) accommodation.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 309Table 12.9: Use of formal support services by Survey 3 young peopleEver seenSeen in current accommodationTA Settled accommodationAllKey worker or housing support worker78% 63% 38% 53%GP 78% 68% 64% 66%Social worker 40% 19% 15% 17%Health visitor 18% 8% 12% 10%Psychologist or psychiatrist 16% 7% 4% 6%Community Psychiatric Nurse/Counsellor13% 2% 6% 4%Drug or alcohol worker 12% 7% 0% 4%Family mediation service 7% 3% 2% 2%Women’s Aid or refuge worker 3% 1% 2% 1%Base 350 219 129 348Source: Survey 3. The majority (78 per cent) of young people had seen a key worker or housing 12.95 support worker at some point, but those in temporary accommodation were much more likely than those in settled accommodation to have done so in their current accommodation (63 per cent as compared with 38 per cent). Young people were approximately twice as likely to be in receipt of key worker/housing support services in their current accommodation as Survey 1 families (see Table 9.6).Use of GP services was widespread (66 per cent of all young people had seen 12.96 a GP while living in their current accommodation), but less than that reported by Survey 1 families (79 per cent), as might be expected given that few young people were in households containing small children.Two in five (40 per cent of) Survey 3 young people had had contact with 12.97 a social worker at some point (much higher than for adult respondents in families accepted as homeless 22 per cent, see Table 9.6).310 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsWhilst 16 per cent of young people reported ever having seen a psychologist 12.98 or psychiatrist, and 13 per cent a community psychiatric nurse or counsellor, very few had accessed such services since living in their current accommodation (6 per cent and 4 per cent respectively501). Approximately one in eight (12 per cent of) young people had ever seen a 12.99 drug or alcohol worker (as compared with 3 per cent of adult respondents in Survey 1), but only 4 per cent had done so in their current accommodation.A minority (7 per cent) of young people had ever used a family mediation 12.100 service, with only 2 per cent doing so since moving into their current accommodation.Emotional and instrumental supportFigure 12.9 compares the proportion of young respondents reporting that 12.101 they had someone who they could count on to listen if they “needed to talk” (i.e. emotional support) and/or to help out in a crisis (i.e. instrumental support) to that of the equivalent age cohort participating in the BHPS502. The proportion of young respondents reporting that they currently had someone to count on to listen when they need to talk (85 per cent), and to help out in a crisis (80 per cent), is notably less than that of people the same age nationally (97 per cent for each)503. However, access to both forms of support had improved for many young people since they had left their last settled accommodation, and the proportion with access to emotional support in their current accommodation had increased by 12 percentage points, and instrumental support by 6 percentage points (Figure 12.9). 501 8 per cent of Survey 1 adult respondents had ever seen either community psychiatric nurses/counsellors or a psychiatrist/psychologist (but see Table 9.6).502 The questions used in our survey to assess emotional and instrumental support were not harmonised completely with those used in the BHPS (2004) (as ours combined the ‘yes, one person’ and ‘yes, more than one person’ response options into a single affirmative response).503 University of Essex. Institute for Social and Economic Research, British Household Panel Survey 2004. Wave 13, England only, 16-18 yrs, weighted data, CHP analysis.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 311Figure 12.9: Proportion of Survey 3 young people with access to emotional and instrumental support, with national comparison (16-18 years)Last settled accommodationCurrent accommodation85%73%97%80%74%97%0%20%40%60%80%100%Have someone to count tohelp out in a crisisHave someone to count on tolisten when need to talkPer cent of respondentsBHPS 2004 EnglandComparatorSources: Survey 3 and BHPS (2004) (wave 13, England only, 16-18 years, CHP analysis) Base: Survey 3 – 350 in current accommodation, 223 in last settled accommodation; BHPS (2004) – 337. Young people from ethnic minority groups were more likely to report that 12.102 they did not have access to emotional or instrumental support than were White young people504; as were those reporting current mental health problems in comparison to those without current mental health problems505. Current availability, and changes in the availability, of emotional and instrumental support were not related to any other demographic variable or vulnerability cluster, nor to current accommodation type or temporary accommodation experiences. Figure 12.10 compares the various sources of emotional and instrumental 12.103 support available to young people in their current accommodation and last settled accommodation. As might be expected, the proportion identifying parents as a source of both emotional and instrumental support reduced considerably after they were accepted as homeless – a gap apparently filled by a combination of friends, partners and social workers or other professional support workers. An increase in use of and/or reliance upon professional sources of support was particularly marked. Only 8 per cent of respondents identified such individuals as a key source of emotional support, and 4 per cent a source of instrumental support in their last settled accommodation, 504 This may, again, be related to the relatively high proportion of former asylum seekers amongst the ethnic minority respondents, but the sub-sample numbers were too small for detailed analysis to test this.505 These are similar to the findings for Survey 1 adult respondents (see para 9.58). The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).312 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsbut these figures rose to 20 and 17 per cent respectively in their current accommodation.Figure 12.10: Sources of emotional and instrumental support for Survey 3 young people0% 10% 20% 30%per cent of respondents40% 50% 60%Instrumental support: Last settled accommodationInstrumental support: Current accommodationEmotional support: Last settled accommodationEmotional support: Current accommodationOtherCollege staff/teacherPeople from religious orlocal community groupsSocial worker or otherprofessional support workerPartnerOther familyParent(s)Friends50453635294242522425273225181711208174232401003232Source: Survey 3 Base: 350 in current accommodation, 223 in last settled accommodation. Multiple responses were possible. Chapter 12 The experience of young people accepted as homeless 16-17 year olds 313Changes in social networks Most young people had experienced changes in their social networks: at 12.104 point of survey only one quarter (28 per cent) had the same level of contact with their family, and 38 per cent the same level of contact with friends, as they had had in their last settled accommodation. The balance of change was skewed in a negative direction. Thus 46 per cent of young people had less contact with family and 41 per cent less contact with friends, as compared with 25 per cent who had increased contact with family and 19 per cent who had increased contact with friends.There was a strong association between gender and changes in contact 12.105 with family (but not friends), in that male respondents were more likely to see family members less (or not at all) after leaving their last settled accommodation than were female respondents (66 per cent as compared with 36 per cent). Changes in contact with family and friends were not related to any other demographic characteristics, vulnerability cluster, or current accommodation type.Economic status and financial circumstancesEconomic activityFigure 12.11 illustrates the main current economic activity of 17 and 18 year 12.106 old young people, together with national comparative data (for England and Wales) from the Youth Cohort Study (YCS)506. Sixteen year old young people have been excluded from the analysis as there were too few to establish patterns or enable comparisons with their peers in the general population (see para 12.11)507.506 It should be borne in mind that the YCS is a self-completion survey whereas Survey 3 was a face-to-face administered survey. This may impact on the pattern of responses to some questions.507 Of those few 16 year olds that did participate in this survey, almost half were in full-time education, with the majority of the others being unemployed or ‘taking a break from work or study’.314 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 12.11: Current economic activity of Survey 3 young people, with national comparators0%20%40%60%80%100%Full-time education at school or college Modern apprenticeship/other govt. supported trainingFull-time job (over 30 hours a week) Part-time job (if main activity)Looking for work/unemployed Something else/not stated18 year old YCScomparator18 year old surveyrespondents17 year old YCScomparator17 year old surveyrespondents718313646634664830101178110643186633Sources: (1) Survey 3; (2) Department for Education and Skills (2005) Youth Cohort Study: The Activities and Experiences of 17 Year Olds: England and Wales 2005. Statistical First Release SFR 48/2005 (revised), Department for Education and Skills, London; (3) Department for Education and Skills (2004) Youth Cohort Study: The Activities and Experiences of 18 Year Olds: England and Wales 2004. Statistical First Release SFR 43/2004 (revised), Department for Education and Skills, London. Base: 154 17 year old young people, 182 18 year old young people, 9,820 17 year old YCS respondents, 7,777 18 year old YCS respondents.Considering the current activity of 17 year olds first, it is apparent that the 12.107 proportion of young people in full-time education was half that of 17 year olds nationally (33 as compared with 66 per cent), and the proportion of those unemployed and looking for work was six times greater than that of people the same age in the wider population (34 per cent as compared with 6 per cent) (Figure 12.11). With regard to 18 year old young people, the proportion in full-time 12.108 education or on government supported training schemes was less than half that of 18 year olds nationally (19 per cent as compared with 51 per cent) (Figure 12.11). Only 16 per cent of 18 year old young people were in full- or part-time employment (as compared with 36 per cent in England and Wales), and the proportion of young people who were unemployed and looking for work (46 per cent) was approximately eight times greater than that of 18 year olds in the general population (6 per cent).Of the young people still in full-time education or government supported 12.109 training schemes, over two thirds (68 per cent) were training for vocational or professional qualifications, 13 per cent were doing basic/key skills courses, 13 per cent were studying toward A/AS levels, and 8 per cent for GCSEs.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 315NEET (not in education, employment or training)The most striking finding with respect to the economic status of young 12.110 people (aged 16-18) was that 57 per cent were not in education, employment or training (NEET) at point of survey508 – a figure more than five times that of the national estimate for 16-18 year olds in England in 2005 (11 per cent)509.Regression analysis, which controlled for a wide range of variables12.111 510, detected no independent effects on the propensity of young people to be NEET at point of survey.Figure 12.12 compares the current main economic activity of young people 12.112 to their economic activity in their last settled accommodation. It reveals that a total of 34 per cent had discontinued their participation in education, employment or training – and only 4 per cent had entered it – since leaving their last settled accommodation. This means that there was a substantial net increase in NEET status amongst young people amounting to 30 percentage points511.508 Young people are considered to be NEET if they are unemployed, looking after a family (including informal care of adults as well as children), disabled, in part-time education (but not part-time work), or otherwise not active in the labour market. See Godfrey, C., Hutton, S., Bradshaw, J., Coles, B., Craig, G. and Johnson, J. (2002) Estimating the Cost of Being “Not in Education, Employment or Training” at Age 16-18, Research Report RR346, London: DfES.509 Department for Education and Skills (2006) Participation in Education, Training and Employment by 16-18 Year Olds in England: 2004 and 2005. Statistical First Release SFR 21/2006, London: DfES.510 Factors controlled for in this regression analysis included: demographic characteristics; sexually abused as a child; violence within home as a child; been involved in crime or ASB; current mental health problem; current substance misuse; spent time in care as a child; current accommodation type; temporary accommodation experiences; accommodation conditions; current financial difficulties, and ability to manage financially has deteriorated; having a disrupted education; social networks.511 This may reflect in part some young people leaving school after leaving their last settled accommodation and failing to move on to other education, training or employment.316 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsFigure 12.12: Changes in economic status of Survey 3 young people since leaving last settled accommodation34%28%34%4%Entered employment, educationor training after leaving lastsettled accommodationDiscontinued education,employment or training afterleaving last settledaccommodationNEET in both settingsIn education, employmentor training in both settingsSource: Survey 3 Base: 223Regression analysis indicated that, other things being equal12.113 512, the only independent factor associated with becoming NEET was duration of stay in temporary accommodation. Young people who had been in temporary accommodation for more than six months were more likely to report having become NEET since leaving their last settled home (46 per cent) than those who had not been in temporary accommodation for as long as six months (26 per cent).Young people (aged 16-18) who were NEET were asked about a number of 12.114 potential barriers to education, employment and training. Their responses are shown in Table 12.10.512 Factors controlled for in this regression analysis included: demographic characteristics; sexually abused as a child; violence within home as a child; been involved in crime or ASB; current mental health problem; current substance misuse; spent time in care as a child; current accommodation type; temporary accommodation experiences; accommodation conditions; current financial difficulties, and ability to manage financially has deteriorated; having a disrupted education; social networks.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 317Table 12.10: Barriers to education, employment and training identified by NEET Survey 3 young peopleReason PercentageNeed more qualifications and skills to get a job or education or training place59%Have not found a suitable job or course 58%Have family problems 37%Too much disruption due to homelessness 37%Have housing problems 35%Not yet decided what sort of job or course I want to do 34%Would be worse off financially in work or on a course 30%No decent jobs or courses available where I live 30%Have poor health or a disability 21%Currently looking after the home or children 18%Having a break from work or study 13%Would find it difficult to travel to work or college because of poor transport13%Looking after other family members 4%Other reasons 25%Base 205Source: Survey 3 Multiple responses were possible. Their responses were compared to those of NEET YCS participants in the same 12.115 age range513. Notably, a far greater proportion of young people identified multiple reasons for their non-participation in education, employment or training. As with 16-18 year olds in the general population, some of the most common barriers identified by young people included the inability to find a suitable job or course (58 per cent), the local unavailability of jobs/courses (30 per cent), or their own indecision regarding what kind of job or course they wanted to do (34 per cent).However, the proportion of young people identifying personal problems as 12.116 barriers – including housing problems (35 per cent), family problems (37 per cent), and poor health or disability (21 per cent) – was far greater than that of their peers nationally. Moreover, a greater proportion of young people felt 513 Comparative data was drawn from sweeps one (16 year olds), two (17 year olds) and three (18 year olds) of YCS 11, provided by the DfES, March 2006. Comparative percentages are not given for YCS respondents, as the unpublished YCS data was broken down by age in such a way that it was not directly comparable to our dataset. Even so, the differences between Survey 3 young people and YCS respondents were so marked that we can be confident in highlighting those identified above.318 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsthat they needed additional qualifications and skills before entering work or education (59 per cent), or that transport problems were a barrier (13 per cent). Importantly, a higher percentage of young people felt that they would be worse off financially if they were in work or on a course (30 per cent)514.Finally, although comparative data was not available on this particular issue12.117 515, more than one third (37 per cent) of NEET young people cited the disruption caused by homelessness as a barrier to their participation in education, employment and training.QualificationsFigure 12.13 portrays the qualifications held by young people (aged 17-18) at 12.118 the point of survey. The proportion of 17 and 18 year old young people with five or more GCSEs graded A-C was approximately one third of that of people the same age in England and Wales generally (at 18 per cent and 16 per cent, as compared to 54 per cent and 51 per cent respectively)516. Figure 12.13: Qualifications of Survey 3 young people0 5 10 15 20 2518 year olds17 year oldsA level(s), AS level(s) orScottish Higher(s)NVQ, City And Guilds or anyother practical 'vocational'qualifications5 or more GCSEs orO levels graded A*-C(or Scottish equivalents)1-4 GCSEs graded A*-C/O levels (or Scottish equivalents)24%18%16%16%17%4%1%20%Source: Survey 3 Base: 154 17 year olds, 182 18 year olds. Multiple responses were possible. 514 Although it is important to remember that the majority of YCS respondents will have been living at home with their parents, so costs such as rent and service charges will not in all likelihood affect them.515 The question about disruption due to homelessness was included in Survey 3 but is not used by the YCS.516 Comparative statistics calculated from tables published in: (1) DfES (2005) Youth Cohort Study: The Activities and Experiences of 17 Year Olds: England and Wales 2005. Statistical First Release SFR 48/2005 (revised), London: DfES; and (2) Department for Education and Skills (2004) Youth Cohort Study: The Activities and Experiences of 18 Year Olds: England and Wales 2004. Statistical First Release SFR 43/2004 (revised), London: DfES.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 319When asked how good they were at reading English when needed in daily 12.119 life, 59 per cent of young people answered ‘very good’, 28 per cent ‘fairly good’, 7 per cent ‘below average’ and 5 per cent ‘poor’. Less than 1 per cent could not read English at all. These figures were very similar to those of Survey 1 adult respondents (see para 10.31).IncomeTable 12.11 shows that young people received a median income of £45.00 12.120 per week (average £74.68) (exclusive of Housing Benefit). Young people still living in temporary accommodation had higher median weekly incomes than those who had secured settled housing (£53.20 compared with £45.00). Table 12.11: Income per week (excluding Housing Benefit) of Survey 3 young people, by whether in temporary or settled accommodationAccommodation situation Average Median BaseAll young people £74.68 £45.00 350Young people still in TA £82.65 £45.00 219Young people in settled housing £62.61 £53.20 129Source: Survey 3Table 12.12 shows the proportions of young people with a last settled 12.121 accommodation experiencing (self-assessed) increases and reductions in income in different housing situations. Almost two-thirds of young people were receiving higher incomes currently than they had in their last settled accommodation (60 per cent for young people living in settled accommodation at point of survey and 63 per cent for those still in temporary accommodation)517. However, between about a fifth and a quarter of young people had lower incomes than in their last settled accommodation (18 per cent still in temporary accommodation and 27 per cent in settled accommodation).517 Bear in mind that for most Survey 3 young people their last settled accommodation was the parental home, and therefore some of this increase may be attributable to increased benefit entitlements as a consequence of estrangement from their parents. It may also reflect young people’s increased benefit entitlement on turning 18, and the fact that some had had children by point of survey.320 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 12.12: Comparisons of current income and income in last settled accommodation received by Survey 3 young people, by whether in temporary or settled accommodationTA Settled accommodationCurrent income higher 63% 60%Current income lower 18% 27%Current income the same 18% 13%Base 126 85Source: Survey 3. Table 12.13 shows the main forms of benefit received by young people12.122 518, and demonstrates that the receipt of benefits was in most respects very similar between young people in settled and temporary accommodation. Thus about two-fifths (39 per cent) received Income Support; just over a quarter (28 per cent) received Job Seeker’s Allowance; and 8 per cent received an Education Maintenance Allowance. Approximately one in ten (9 per cent) received Child Benefit, with 6 per cent receiving Child Tax Credit. Only 2 to 3 per cent received Disability Living or Attendance Allowance and Incapacity Benefit.Table 12.13: Benefits currently received by Survey 3 young people, by whether in temporary or settled accommodationTA Settled accommodationAllIncome Support 38% 40% 39%Jobseeker’s Allowance 27% 31% 28%Child Benefit 9% 9% 9%Child Tax Credit 7% 5% 6%Education Maintenance Allowance 6% 10% 8%Disability Living/Attendance Allowance3% 2% 2%Incapacity Benefit 1% 7% 3%Base 219 129 348Source: Survey 3 Multiple responses were possible 518 This also includes benefits received by anyone else in their family group. However, as is indicated in Figure 12.1, only 8 per cent of all of these young people were living with a partner who might also have been claiming benefits.Chapter 12 The experience of young people accepted as homeless 16-17 year olds 321Table 12.14 illustrates the proportions of young people receiving a variety of 12.123 grants and loans since applying as homeless. This demonstrates that around a quarter had received a Community Care Grant and/or a Social Fund Crisis Loan (23 per cent for both). About one in ten (11 per cent) had received a Social Fund Budgeting Loan and 6 per cent a Sure Start Maternity Grant. None had received a Social Fund loan to cover maternity expenses. Table 12.14: Grants and loans received by Survey 3 young people, by whether in temporary or settled accommodationTA Settled accommodationTotalCommunity Care Grant 13% 38% 23%Sure Start Maternity Grant 5% 7% 6%Social Fund Budgeting Loan 10% 13% 11%Social Fund Crisis Loan 23% 24% 23%Social Fund Loan to cover maternity expenses0% 0% 0%Base 219 129 348Source: Survey 3 Multiple responses were possible. As might be expected, young people in settled housing were much more 12.124 likely than those still living in temporary accommodation to have received a Community Care Grant (38 per cent compared with 13 per cent)519, but the other funds were received by similar proportions of young people in settled and temporary accommodation.Fourteen per cent of young people (both those in settled and in temporary 12.125 accommodation) received regular financial help from friends or family. Financial exclusionThe survey found that 14 per cent of young people did not have a current 12.126 account with a bank, building society or other organisation (as was the case for 15 per cent of Survey 1 adult respondents). The proportion holding such accounts did not vary by any demographic characteristics, vulnerability cluster, or current accommodation type.Overall financial situationYoung people’s self-assessments regarding how well they were coping 12.127 financially are portrayed in Figure 12.14. This reveals that a total of 35 per cent felt that they had current financial difficulties (i.e. were not managing 519 At 23 per cent, the percentage of all Survey 3 young people in receipt of a Community Care Grant was slightly higher than that for Survey 1 families, 21 per cent (see para 10.46).322 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsvery well financially, were having some financial difficulties, or were ‘in deep financial trouble’). These results are very similar to those for Survey 1 adult respondents (see para 10.66)520.Figure 12.14: Survey 3 young peoples’ assessment of their current financial situation17%40%18%4%In deep financial troubleHave some financial difficultiesDon't manage very wellGet by all rightManage quite wellManage very well8%13%Source: Survey 3 Base: 350 Regression analysis indicated that, other things being equal12.128 521, the following factors had independent effects which made it more likely that a young person would report current financial difficulties522: • becoming ‘NEET’523; and • living in self-contained temporary accommodation524 (this was also associated with financial difficulties amongst families accepted as homeless – see para 10.92).520 National comparisons are not possible because too few people in the relevant age group participated in the FACS 2005 survey (from which this question was derived).521 Factors controlled for in this regression analysis included: demographic characteristics; sexually abused as a child; violence within home as a child; been involved in crime or ASB; current mental health problem; current substance misuse; spent time in care as a child; current accommodation type; temporary accommodation experiences; accommodation conditions; whether NEET, and became NEET since last settled accommodation; having a disrupted education; social networks.522 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).523 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).524 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).Chapter 12 The experience of young people accepted as homeless 16-17 year olds 323Table 12.15 compares young people’s ability to manage financially at the 12.129 point of survey to when they were in their last settled accommodation. This reveals that, as with Survey 1 adult respondents (see para 10.71), many had experienced a decline in their ability to cope financially after being accepted as homeless. The percentage stating that they managed ‘very well’ or ‘quite well’ dropped from 54 per cent of individuals in their last settled accommodation to just 25 per cent in their current accommodation. Similarly, the proportion reporting that they were currently experiencing some financial difficulties or were ‘in deep financial trouble’ increased from 2 per cent to 22 per cent.Table 12.15: Survey 3 young peoples’ assessment of their financial management in current accommodation and last settled accommodationCurrent accommodationLast settled accommodationManage very well 8% 24%Manage quite well 17% 30%Get by all right 40% 36%Don’t manage very well 13% 7%Have some financial difficulties 18% 2%In deep financial trouble 4% 0%Total 100% 100%Base 350 223Source: Survey 3 In fact, young people were more than four times as likely to report that 12.130 their ability to manage financially had declined since leaving their last settled accommodation (true for 56 per cent) than they were to report that it had improved (true for 12 per cent). Regression analysis indicated that, other things being equal12.131 525, the following factors had independent effects which made it more likely that a young person would report a deterioration in their financial circumstances since their last settled accommodation: 525 Factors controlled for in this regression analysis included: demographic characteristics; sexually abused as a child; violence within home as a child; been involved in crime or ASB; current mental health problem; current substance misuse; spent time in care as a child; current accommodation type; temporary accommodation experiences; accommodation conditions; whether NEET, and became NEET since last settled accommodation; having a disrupted education; social networks.324 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds• living in self-contained temporary accommodation (which was also associated with financial difficulties amongst families accepted as homeless, see para 10.92)526;• becoming ‘NEET’527; and • having experienced a disrupted education: 60 per cent of young people who reported that their education had been disrupted by missing a lot of school, suspensions or exclusions, compared with 47 per cent of other young people, reported worsening current financial circumstances.As with the adult respondents to Survey 1, young peoples’ ability to manage 12.132 financially was a key factor affecting their perceptions of their overall quality of life (see Chapter 13). Conclusions This chapter has explored the characteristics and experiences of young 12.133 people accepted as 16-17 year olds owed the main homelessness duty (over half of whom were 18 by the point of survey). It covered a wide range of dimensions of these young people’s lives – including their demographic characteristics, personal histories, experiences of temporary accommodation, levels of satisfaction with current accommodation, health, access to social and professional support, qualifications, economic status and financial circumstances.The overriding finding from this chapter is that young people accepted as 12.134 homeless 16-17 year olds are, on the whole, a far more disadvantaged and socially excluded group than families accepted as homeless. It confirms the previous (mainly qualitative) research evidence on their vulnerability. These young people’s childhoods were very often marred by extremely difficult and fractured family relations – with family restructuring, violence, parents with mental health problems, and frequent moves commonly experienced – and many had also had a very disrupted education. Large proportions had experienced mental health and/or substance misuse problems, and they were far likelier than their peers in the general population not to be in education, employment or training. They also often faced a number of negative experiences since leaving 12.135 their last settled accommodation. In particular, one third had discontinued participation in education, employment and training since leaving their last 526 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).527 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).Chapter 13 Overall quality of life 325settled accommodation, and many reported increased difficulties in managing financially.On the other hand, young people made considerably greater use of 12.136 professional support services than families accepted as homeless, and their access to emotional and instrumental support had improved overall since leaving their last settled accommodation. The utilisation of professional support, including supported accommodation, may well have mitigated potentially worse outcomes for many young people.The next chapter will provide an overview of the general quality of life of 12.137 all groups participating in this study, including: adults (Surveys 1 and 4) and children (Surveys 2 and 5) in families accepted as homeless, and young people accepted as homeless 16-17 year olds (Survey 3).326 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsChapter 13: Overall quality of life IntroductionThe overall objective of this study was to understand the causes, experiences 13.1 and impacts of statutory homelessness amongst families and 16-17 year olds in England. Data was collected in five separate surveys covering parents, children and young people assisted under the homelessness legislation (see Chapter 1 and Appendix 1 for details of all five surveys).This chapter considers the overall quality of life of adults and children in 13.2 families accepted as homeless, and the quality of life of young people accepted as homeless 16-17 year olds. It assesses quality of life in a number of ways, including:• respondents’ views on whether their life was better or worse as compared with when they were living in their last settled accommodation528; • whether respondents’ felt that their life was ‘on hold’;• whether respondents’ were worried about the future; and• respondents’ general levels of happiness.The chapter draws on data from all five surveys: Survey 1 and Survey 2 13.3 (adults and children respectively in families accepted as being owed the main homeless duty between 1st January 2005 and 30 June 2005); Survey 3 (young people accepted over the same time period as 16-17 year olds owed the main homelessness duty); and Survey 4 and Survey 5 (adults and children respectively in families accepted as owed the main homelessness duty and in temporary accommodation for more than a year)529. As in other chapters, we present two types of statistical analysis here: 13.4 bivariate analysis, which indicates whether there is a statistically significant association between two variables, when their relationship is considered in isolation; and regression analysis, which explores which variables have an independent effect in determining the likelihood of a given finding, when a 528 See Chapter 1 and Appendix 1 for definition and explanation of ‘last settled accommodation’ as used in this research.529 Surveys 4 and 5 were required because the ‘time-window’ design for Surveys 1,2 and 3, while delivering a representative sample of those accepted as homeless over a six month period, by definition excluded those with prolonged stays in temporary accommodation. See Appendix 1 for a full explanation.Chapter 13 Overall quality of life 327range of other factors are held constant. However, it should be noted that bivariate statistics are often used to illustrate relationships between variables where an independent effect has been detected530. This survey evidence indicates that both families with children and young 13.5 people were far likelier to report an improvement than a deterioration in their quality of life as compared with when they were living in their last settled accommodation. For the minority of families and young people for whom quality of life was said to have declined, key factors included living in accommodation or a neighbourhood perceived to be unsafe, and deteriorating financial circumstances. Quality of life at point of survey was consistently reported to be better amongst families who had been provided with settled housing than for those still living in temporary accommodation. For young people, living in settled or temporary accommodation seemed less critical to their quality of life. Quality of life findings for Survey 4 families (in temporary accommodation for over one year) were very similar to those for Survey 1 families still living in self-contained temporary accommodation at time of survey531. Key pointsAdults• Encouragingly, those adult respondents who reported that life had got better since leaving their last settled accommodation heavily outnumbered those for whom it had got worse (57 per cent as compared with 19 per cent). • Across every measure, adult respondents who had been provided with settled accommodation reported having a better quality of life than those still living in temporary accommodation. • Most notably, while the majority (64 per cent) of adult respondents living in temporary accommodation considered their lives to be ‘on hold’, this was true of only 18 per cent of those who had been provided with settled housing.• Adult respondents staying in hostels or B&B hotels, or in temporary arrangements with friends and relatives, consistently reported a poorer quality of life than those living in self-contained temporary accommodation532.532530 See Chapter 1 and Appendix 2 for more detail.531 The comparison here is to Survey 1 families in self-contained temporary accommodation as all Survey 4 families were in this form of accommodation by point of survey.532 While the small number of responses from those living in the first two of these forms of temporary accommodation compromises their statistical reliability, this pattern of results was so uniform across every measure that it seems reasonable to conclude that it is a fair representation of reality.328 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds• Poor quality of life was also often related to mental health problems, financial difficulties, feeling unsafe in accommodation or neighbourhood, and having insufficient living space. • Survey 4 adult respondents in temporary accommodation for over one year, all of whom were in self-contained temporary accommodation, reported a similar quality of life to other adult respondents in self-contained temporary accommodation. Children• Parents were far likelier to report an improvement (57 per cent) than a decline (12 per cent) in their child(ren)’s overall quality of life as compared with when they were living in their last settled accommodation.• This was especially true for children living in settled housing; those for whom violent relationship breakdown533 was a cause of homelessness; those whose family’s financial situation had improved; and those whose parents’ felt that they had enough living space. • Child respondents were generally happier about life overall than were adults respondents. However, children still in temporary accommodation, and older children (aged 12-15), were less positive than other children, with the latter group frequently reporting boredom, anger and/or worrying about their parents. • The reported quality of life for Survey 5 children in temporary accommodation for over one year was very similar to that of other children in self-contained temporary accommodation at time of survey. Young people• Young people accepted as homeless 16-17 year olds were also much more likely to report that life was better (52 per cent), rather than worse (25 per cent), than it had been in their last settled accommodation. • For the minority of young people who perceived their quality of life to have worsened since leaving their last settled accommodation, this was associated with feeling unsafe in their current neighbourhood, and also with deteriorations in their ability to cope financially.• Young people still living in temporary accommodation, and those who felt that they had insufficient living space, were much more likely than other young people to report that they felt that their life was ‘on hold’.• Worrying about the future and levels of general (un)happiness were influenced, respectively, by living in accommodation or an area perceived to be unsafe.533533 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.Chapter 13 Overall quality of life 329AdultsChanges in quality of life We asked adult respondents about changes in their overall quality of life 13.6 since they had left their last settled accommodation534. Encouragingly, those who felt that life had got better (57 per cent) heavily outnumbered those for whom life had got worse (19 per cent), with only 24 per cent feeling that life was about the same. However, perceptions of changes to quality of life were strongly associated 13.7 with accommodation circumstances, with adult respondents in settled housing most likely to note a positive change (68 per cent), and those still in temporary accommodation far less likely to do so (42 per cent) (though there was still net positive change for those in temporary accommodation). Regression analysis indicated that, other things being equal13.8 535, the following factors were independently associated with feeling that life had improved:• being in settled rather than temporary accommodation (see para 13.7 above);• an improvement in financial circumstances: 81 per cent of those who reported that their financial position had improved, as compared with 52 per cent of other adult respondents, reported that their life had got better overall.• violent relationship breakdown as cause of homelessness: 72 per cent of those who reported violent relationship breakdown as a reason for applying as homeless, as compared with only 54 per cent of other adult respondents, said that life had got better536. • having enough living space: 62 per cent of those who had enough living space, as compared with 40 per cent of those who did not, said that life had got better. Regression analysis also indicated that, other things being equal13.9 537, the factors that were independently associated with a feeling that life had got worse were:534 See Chapter 1 and Appendix 1 for an explanation of ‘last settled accommodation’ and when this was considered valid for comparison purposes.535 Factors controlled for in this regression analysis included: demographic characteristics; geographical variables; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether household was workless, and whether household had become ‘workless’ since leaving last settled accommodation; how managing financially, and managing better or worse financially since leaving last settled accommodation; current mental health problems; and whether had ever had a problem with drugs/alcohol.536 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.537 Factors controlled for in this regression analysis included: demographic characteristics; geographical variables; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether household was workless, and whether household had become ‘workless’ since leaving last settled accommodation; how managing financially, and managing better or worse financially since leaving last settled accommodation; current mental health problems; and whether had ever had a problem with drugs/alcohol.330 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds• being in temporary accommodation at time of survey (see para 13.7 above).• a deterioration in financial circumstances: 40 per cent of those who reported a worsening financial position, as compared with 12 per cent of other adult respondents, felt that life had deteriorated for them.• feeling unsafe inside current accommodation: 41 per cent of those who felt unsafe in their accommodation also felt that life had got worse, as compared with 15 per cent of other adult respondents.• having insufficient living space: 35 per cent of those felt that they had insufficient living space, as compared with 14 per cent of other adult respondents, said that life had got worse.Whether life was ‘on hold’Adult respondents were asked whether they felt they could ‘get on with life’ 13.10 while they were living in their current accommodation, or whether their life was ‘on hold’538. Two-fifths (39 per cent) overall felt that their life was ‘on hold’. However, there were very sharp distinctions according to current 13.11 accommodation status: 64 per cent of those still living in temporary accommodation felt that life was on hold, compared with only 18 per cent of those in settled housing. The feeling that life was on hold was commonest amongst those staying temporarily with friends and relatives, or in hostels and B&B hotels539. Regression analysis indicated that, other things being equal13.12 540, the following factors were independently associated with feeling that life was on hold:• being in temporary accommodation (see para 13.11 above). • current financial difficulties: 58 per cent with financial difficulties, compared with 29 per cent without such difficulties, felt that life was on hold.• feeling that neighbourhood is unsafe: 61 per cent of those who felt that their neighbourhood was unsafe, but only 33 per cent of other adult respondents, felt that their life was on hold. 538 We asked this question because previous (qualitative) research had indicated that a sense that life was ‘on hold’ was a key factor which undermined the quality of life of families in temporary accommodation. See: Holder, T., Curteis, S., Griffiths, S., Hunter, G. and James, K. (2002) Life on Hold: “I can’t even think about tomorrow”: The housing and support needs of families in temporary accommodation in Leeds, Leeds City Council (unpublished report); Sawtell, M. (2002) Lives on Hold: Homeless families in temporary accommodation, London: The Maternity Alliance.539 This finding does not quite reach statistical significance at the 95 per cent confidence level because of the small number of families living in these two forms of temporary accommodation at point of survey. However, such a similar pattern of responses was detected across all aspects of quality of life that it seems reasonable to conclude that it is a fair representation of reality.540 Factors controlled for in this regression analysis included: demographic characteristics; geographical variables; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether household was workless; how managing financially; current mental health problems; and whether had ever had a problem with drugs/alcohol.Chapter 13 Overall quality of life 331• insufficient living space: 66 per cent of those who felt that they did not have enough living space, compared to 30 per cent of other adult respondents, felt that life was on hold.• current mental health problems: 55 per cent of those with current mental health problems, compared with 33 per cent of other adult respondents, felt that life was on hold.• being accepted in the North and Midlands: those accepted in this broad regions were less likely to feel that life was on hold (32 per cent), than those accepted elsewhere (45 per cent).Worrying about the futureWe also asked about the extent to which adult respondents worried about 13.13 the future. Overall, 45 per cent said that they worried about the future either often or all the time. Again responses here demonstrated a very strong relationship with 13.14 accommodation type. Thus, while only 36 per cent of those in settled housing said that they worried about the future either often or all the time, this rose to 55 per cent of those in temporary accommodation. As with ‘life on hold’, worrying about the future was commonest amongst those staying with friends or relatives, or in hostels and B&B hotels541.Regression analysis indicated that, other things being equal13.15 542, the following factors were independently associated with worrying about the future:• current mental health problems: 69 per cent of those with mental health problems, as compared with 36 per cent of other adult respondents, worried about the future. This was the strongest independent effect.• being in temporary accommodation (see para 13.14 above). • current financial problems: 61 per cent of those with financial problems worried about the future, as compared with only 36 per cent of other adult respondents.• feeling that neighbourhood is unsafe: 57 per cent of those who felt that their neighbourhood was unsafe, compared to 41 per cent of other adult respondents, worried about the future.541 While the small number of responses from those living in the first two of these forms of temporary accommodation compromises their statistical reliability, this pattern of results was so uniform across every measure that it seems reasonable to conclude that it is a fair representation of reality.542 Factors controlled for in this regression analysis included: demographic characteristics; geographical variables; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether household was workless; how managing financially; current mental health problems; and whether ever had a problem with drugs/alcohol.332 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsOverall levels of happinessAll adult respondents were asked to categorise themselves according to how 13.16 they felt about life overall at the moment (see Table 13.1). Once more, there was a clear pattern according to accommodation type. Thus while 57 per cent of adult respondents considered themselves to be very or fairly happy, including 68 per cent of adult respondents in settled housing, this was true of only 44 per cent in temporary accommodation. Yet again, those staying with friends and relatives, or in hostels and B&B hotels, reported the poorest quality of life, with far smaller proportions in these groups reporting being very or fairly happy543. Table 13.1: How adult respondents felt about life overallSettled accommodationTA AllVery happy 26% 11% 19%Fairly happy 42% 33% 38%Mixed feelings 24% 36% 30%Not very happy 6% 15% 10%Not at all happy 2% 6% 4%Total 100% 100% 100%Base 923 1118 2041Source: Survey 1 Regression analysis indicated that, other things being equal13.17 544, the following factors were independently associated with being very or fairly happy:• being in settled housing (see para 13.16 above).• having enough living space: 63 per cent who felt that their living space was sufficient were very or fairly happy, as compared to 37 per cent who felt that it was insufficient.• age of adult respondent: 63 per cent of adult respondents aged under 25, compared to 53 per cent of those over this age, were very or fairly happy. This regression analysis also indicated that the following factors had 13.18 independent effects which made feeling very or fairly happy less likely:543 While the small number of responses from those living in the first two of these forms of temporary accommodation compromises their statistical reliability, this pattern of results was so uniform across every measure that it seems reasonable to conclude that it is a fair representation of reality.544 Factors controlled for in this regression analysis included: demographic characteristics; geographical variables; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether household was workless; how managing financially; current mental health problems; and whether ever had a problem with drugs/alcohol.Chapter 13 Overall quality of life 333• current mental health problems: only 32 per cent of those with current mental health problems, as compared to 66 per cent of other adult respondents, were very or fairly happy.• current financial difficulties: only 36 per cent of adult respondents with financial difficulties, as compared with 67 per cent without such difficulties, felt very or fairly happy. • living in a workless household: only 53 per cent of those living in a workless household, as compared with 63 per cent of those in a household where there was at least one adult in work, felt very or fairly happy. • violent relationship breakdown as a cause of homelessness: only 50 per cent of those who reported violent relationship breakdown was a reason for applying as homeless, compared with 58 per cent of other adult respondents, felt very or fairly happy545. • feeling unsafe inside accommodation: only 35 per cent of those who felt unsafe inside their accommodation, compared with 61 per cent of other adult respondents, felt very or fairly happy.• feeling that neighbourhood is unsafe: only 38 per cent of those who felt that their neighbourhood was unsafe, as compared with 62 per cent of other adult respondents, felt very or fairly happy.Adult respondents in temporary accommodation for more than one yearAdult respondents in temporary accommodation for more than a year (Survey 13.19 4 adult respondents) reported having a very similar quality of life to Survey 1 adult respondents in self-contained temporary accommodation546. However, they were marginally less likely to report that they considered their life to be ‘on hold’: 52 per cent did so, as compared with 60 per cent of adult respondents in self-contained temporary accommodation in Survey 1. Children Parents’ views on changes in their children’s quality of life We asked Survey 1 adults about perceived changes in their child(ren)’s overall 13.20 quality of life547 since leaving their last settled accommodation548. Again, reports of improvements (57 per cent) far outnumbered reports of decline 545 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.546 Comparisons here are confined to those in self-contained temporary accommodation in Survey 1 because all Survey 4 families were living in this form of accommodation.547 As always with parental assessments, some caution must be exercised in interpreting this material.548 See Chapter 1 and Appendix 1 for an explanation of ‘last settled accommodation’ and when this was considered valid for comparison purposes.334 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds(12 per cent), and those provided with settled housing were far likelier than those still in temporary accommodation to cite positive changes (69 per cent as compared with 44 per cent) (see Figure 13.1). Those in hostels and B&B hotels, or staying with friends or relatives, were most likely to report that their child(ren)’s lives had got worse549. Regression analysis indicated that, other things being equal13.21 550, the following factors were associated with parents reporting positive change in their children’s overall quality of life:• being in settled rather than temporary accommodation (see para13.20 above);• an improvement in financial circumstances: 81 per cent of those who reported that their financial position had improved, as compared with 55 per cent of other adult respondents, reported that their children’s lives had got better. • violence as a cause of homelessness: 77 per cent of those who reported violent relationship breakdown as a reason why they had applied as homeless551, as compared with 56 per cent of other adult respondents, said that their children’s lives had got better.• having enough living space: 65 per cent of those who had enough living space, as compared with 41 per cent of those who did not, said that their children’s lives had got better. 549 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).550 Factors controlled for in this regression analysis included: demographic characteristics; geographical variables; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether household was workless, and whether household had become ‘workless’ since leaving last settled accommodation; how managing financially, and managing better or worse financially since leaving last settled accommodation; current mental health problems; and whether had ever had a problem with drugs/alcohol.551 In total, 13 per cent of all adult respondents reported violent relationship breakdown with a partner as reason for applying as homeless, and a further 3 per cent reported some other form of violent relationship breakdown.Chapter 13 Overall quality of life 335Figure 13.1: Parental assessments of changes in the quality of life of their childrenIn settled housingIn temporary accommodation0%20%40%60%80%100%About the sameWorseBetter44%69%24%5%32%25%Source: Survey 1 Base: 1,180 adult respondents with a last settled accommodation that was valid for comparison purposes and who had children resident with them in their last settled accommodation. Regression analysis indicated that, other things being equal13.22 552, the following factors were associated with parents reporting negative change in their children’s overall quality of life:• financial situation had deteriorated: 24 per cent of those who reported that their financial situation had worsened felt that their children’s lives had got worse, as compared with 9 per cent of other adult respondents.• feeling neighbourhood is unsafe: 26 per cent of those who felt the neighbourhood was unsafe, as compared with 8 per cent of other adult respondents, felt that their children’s lives had got worse.Children’s views on their quality of life We also asked all Survey 2 child respondents how they felt about life overall 13.23 at the point of survey. The majority (77 per cent) reported feeling either ‘very happy’ or ‘fairly happy’ (as compared with 57 per cent of adult respondents, see para 13.16 above). A further 16 per cent had mixed feelings, whilst a total of 6 per cent said that they were ‘not very happy’ or ‘not at all happy’ at the point of survey. Children in temporary accommodation were generally less happy than those in settled housing (only 34 per cent of the former, as compared with 52 per cent of the latter, reported being very happy) (see Table 13.2).552 Factors controlled for in this regression analysis included: demographic characteristics; geographical variables; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether household was workless, and whether household had become ‘workless’ since leaving last settled accommodation; how managing financially, and managing better or worse financially since leaving last settled accommodation; current mental health problems; and whether had ever had a problem with drugs/alcohol.336 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 13.2: How children feel about life overall, as reported by children (8-15 yrs) Settled accommodation TA All Very happy 52% 34% 44%Fairly happy 30% 38% 33%Mixed feelings 14% 19% 16%Not very happy 3% 7% 4%Not at all happy 2% 3% 2%Total 100% 100% 100%Base 223 224 447Source: Survey 2. There were no gender differences in general feelings about life, but the 13.24 younger Survey 2 children tended to be happiest overall, with 83 per cent of 8-11 year olds reporting that they were either ‘very happy’ or ‘fairly happy’, as compared with 71 per cent of children aged 12-15. All Survey 2 12-15 year old child respondents were asked whether or not they 13.25 agreed with a range of statements about their lives. Their responses, depicted in Figure 13.2, reveal that whilst approximately three quarters said that they enjoyed school (78 per cent) and living in their current accommodation (75 per cent), 59 per cent were often bored, and 54 per cent admitted to getting upset or angry often. Half (50 per cent) worried about their parent(s) ‘a lot’, and over one third (38 per cent) worried about the future.Chapter 13 Overall quality of life 337Figure 13.2: Children’s feelings about aspects of life, as reported by children (12-15 yrs)78%75%59%54%50%38%0% 20% 40% 60% 80%Worried aboutthe futureWorry aboutparents a lotGet upset orangry oftenOften boredEnjoy living hereEnjoy schoolSource: Survey 2 Base: 221 There were no differences in responses to the statements above (Figure 13.2) 13.26 between boys and girls, nor between children in settled housing and those still in temporary accommodation, with the one exception that children in settled housing were more likely to report that they enjoyed living in their current accommodation553. Children in temporary accommodation for more than one yearBoth parents and children in temporary accommodation for more than one 13.27 year (Surveys 4 and 5) reported a very similar quality of life for children as was reported for children living in self-contained temporary accommodation in Surveys 1 and 2554.553 The margins of error are too wide here the actual percentages to be given but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).554 Comparisons here were confined to those in self-contained temporary accommodation in Surveys 1 and 2 because all families in temporary accommodation for a year or longer (Surveys 4 and 5) were living in this form of accommodation.338 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsYoung peopleChanges in quality of life Around half (52 per cent) of young people reported that life had got better 13.28 since leaving their last settled accommodation, as compared with only 25 per cent who said that it had got worse. Thus there was a similarly positive net change as there was with families (see paras 13.6 and 13.20).Regression analysis, controlling for a wide range of variables13.29 555, detected no independent effects on the likelihood of young people feeling that life had got better since leaving their last settled accommodation. However, other things being equal556, the following independent factors were associated with their feeling that life had got worse:• a deterioration in financial circumstances; and• feeling that their neighbourhood was unsafe557.Unlike with families accepted as homeless, perceived changes in quality of 13.30 life since last settled accommodation were not associated with whether young people were living in settled or temporary accommodation at point of survey. Nor were changes in economic status (e.g. whether they had entered or discontinued education, employment or training) associated with changes in perceived quality of life (although changes in financial circumstances were clearly an influence, see above). Whether life was ‘on hold’In total, 39 per cent of young people felt that their life was ‘on hold’ (this was 13.31 identical to the figure for adult respondents, see para 13.10). As with adult respondents, responses to this question were very closely linked to current housing situation. Thus only 18 per cent of young people in settled housing felt that their life was ‘on hold’, as compared to 57 per cent of those still in temporary accommodation. 555 Factors controlled for in this regression analysis included: demographic characteristics; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether NEET, and whether have become NEET since leaving last settled accommodation; how managing financially, and changes in how managing financially since leaving last settled accommodation; social networks; vulnerability clusters.556 Factors controlled for in this regression analysis included: demographic characteristics; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether NEET (not in employment, education or training), and whether have become NEET since leaving last settled accommodation; how managing financially, and changes in how managing financially since leaving last settled accommodation; social networks; vulnerability clusters.557 Please note that we are unable to provide the percentage estimates to illustrate many of the associations identified with regards to the quality of life of young people because the margins of error on these percentages exceed +/-10 per cent on account of the relatively small sample sizes. However, all of the relationships noted are statistically significant at the 95 per cent confidence level. See Chapter 1, para 1.27 for a full explanation.Chapter 13 Overall quality of life 339Regression analysis indicated that, other things being equal13.32 558, the independent factors associated with feeling that life was on hold were:• being in temporary accommodation (see para 13.31 above).• insufficient living space559.Worrying about the futureOverall, 45 per cent of young people reported that they worried about 13.33 the future either ‘often’ or ‘all of the time’; a further 25 per cent worried ‘sometimes’; and 11 per cent worried ‘only now and again’. Regression analysis indicated that, other things being equal13.34 560, the main independent factor associated with worrying about the future was feeling unsafe in their current accommodation561.Notably, and unlike adult respondents, the extent of worry about the future 13.35 was not associated with whether a young person was in temporary or settled accommodation. Overall levels of happinessOverall, 47 per cent of young people reported feeling very or fairly happy 13.36 (compared to 57 per cent of Survey 1 adult respondents, and 77 per cent of Survey 2 child respondents, see paras 13.16 and 13.23). Young peoples’ general feelings about life appeared to be associated with 13.37 their current accommodation status, as illustrated in Table 13.3. Those still in temporary accommodation were less likely to report that they were either ‘very’ or ‘fairly’ happy than were young people in settled housing (40 per cent as compared with 58 per cent respectively).558 Factors controlled for in this regression analysis included: demographic characteristics; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether NEET, and whether have become NEET since leaving last settled accommodation; how managing financially, and changes in how managing financially since leaving last settled accommodation; social networks; vulnerability clusters.559 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).560 Factors controlled for in this regression analysis included: demographic characteristics; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether NEET, and whether have become NEET since leaving last settled accommodation; how managing financially, and changes in how managing financially since leaving last settled accommodation; social networks; vulnerability clusters.561 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1, para 1.27).340 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable 13.3: How young people feel about life overallSettled accommodationTA AllVery happy 18% 12% 14%Fairly happy 40% 28% 33%Mixed feelings 27% 44% 37%Not very happy 15% 10% 12%Not at all happy 1% 7% 4%Total 100% 100% 100%Base 129 219 350Source: Survey 3.However, regression analysis indicated that, other things being equal13.38 562, whether a young person was in temporary or settled housing did not in itself have an independent effect on their likelihood of reporting being very or fairly happy. Instead, the only independent factor detected, which made it less likely that they would report being very or fairly happy, was feeling that their neighbourhood was unsafe563. ConclusionsThis chapter has reviewed the overall quality of life of adults and children in 13.39 families accepted as homeless and of young people accepted as homeless 16-17 year olds. Encouragingly, amongst both families and young people, those who reported 13.40 that their quality of life was now better than it had been in their last settled accommodation heavily outnumbered those who reported that it was worse. Thus, it appears safe to conclude that the help provided through the statutory homelessness system has played an important role in assisting these families and young people to move on to a more satisfactory living environment than was the case before they were accepted as homeless. For the minority of families and young people for whom quality of life was said to have deteriorated, key factors were living in accommodation or a neighbourhood perceived to be unsafe, and worsening financial circumstances. 562 Factors controlled for in this regression analysis included: demographic characteristics; current accommodation type; accommodation conditions; temporary accommodation experiences; causes of homelessness; whether NEET, and whether have become NEET since leaving last settled accommodation; how managing financially, and changes in how managing financially since leaving last settled accommodation; social networks; vulnerability clusters.563 The margins of error are too wide here for the actual percentages to be given, but the results reported are statistically significant at the 95 per cent confidence level (see Chapter 1 para 1.27).Appendix 1: Technical report summary 341Families with children in settled housing consistently reported a far 13.41 better quality of life than those still living in temporary accommodation (though there was still net positive change for those living in temporary accommodation at point of survey). There was a particularly strong association between living in temporary accommodation and feeling that life was ‘on hold’. Staying with friends or relatives, or in hostels and B&B hotels, was associated with a particularly poor quality of life. However, those families experiencing extended stays in temporary accommodation (over one year), all of whom were in self-contained temporary accommodation, did not report a poorer quality of life than other families in self-contained temporary accommodation. For young people, there were more mixed findings on accommodation 13.42 status: living in temporary accommodation was associated with feeling that life was ‘on hold’, but not with worrying about the future or with general levels of (un)happiness. These findings on quality of life, together with the data presented in Chapter 12 with respect to young people’s accommodation experiences and levels of satisfaction, suggest that the meaning and significance of temporary accommodation may well be very different for young people than for families accepted as homeless. For young people, it is perhaps more accurate and helpful to view such accommodation as ‘transitional’ rather than simply as ‘temporary’. 342 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsAppendix 1:Technical report summaryPrepared by:Joel Williams and Angela Charlton at BMRB Social ResearchPart of BMRB Limited (British Market Research Bureau)Sample design Surveys 1, 2 and 3The ‘time window’ approachBecause the study was concerned with A1.1 all families and 16-17 year olds accepted as homeless, BMRB adopted a ‘time window’ approach to sampling rather than a ‘stock’ approach. The objective was to draw a random sample of families and 16-17 year olds A1.2 accepted as homeless over a set period (or ‘time window’). Some may still have been living in temporary accommodation at the time of interview but others will have been in settled housing. The alternative ‘stock’ approach would require a random sample of families/16-17 year olds who were in temporary accommodation at one particular point in time. Had BMRB adopted the ‘stock’ approach, the sample would have over-A1.3 represented those households that had spent a long time in temporary accommodation at the expense of those who had spent a short time in temporary accommodation.Source dataOnly local authorities hold complete information about the households they A1.4 accept as homeless. Therefore, BMRB employed a two-stage sample design, drawing a systematic sample of local authorities and then drawing samples of households within each co-operating local authority564.Given the general correlation between the number of families accepted as A1.5 homeless and the number of 16-17 year olds accepted as homeless, BMRB decided to draw a single sample of local authorities rather than one for each survey. Because the families survey was the larger of the two, BMRB elected to base its sampling work on data about family homelessness rather than young person homelessness.564 All local authorities that had accepted <60 families as homeless in 2004 were excluded from the sampling process for reasons of fieldwork efficiency. In total, 58 local authorities were excluded, reducing coverage to 97.1% of families accepted as homeless. This was felt to be a reasonable compromise between survey efficiency and survey quality.Appendix 1: Technical report summary 343Stratification of local authoritiesSampling frames are usually A1.6 stratified before the sample is drawn. Stratification essentially means that the units on the frame – in this case local authorities – are sorted into groups which share one or more characteristics in common. Stratification usually minimises sampling variance so long as the combination A1.7 of variables that is used is correlated with key survey variables or appropriate proxies. BMRB stratified a list of the eligible local authorities in England into four A1.8 groups on the basis of supplied P1E data from the twelve months 1st January 2004 to 31st December 2004 (the latest available at that time). The first level of stratification was by population density with those above the A1.9 average in one stratum and those below the average in another. The second level of stratification was on the basis of temporary A1.10 accommodation use. ‘Heavy’ use of temporary accommodation was indicated by a high ratio of households in temporary accommodation to new acceptances of families owed the main homelessness duty. Average ratios were calculated for each population density stratum, allowing BMRB to split them into two and create a total of four strata.The third stage of stratification was by region. BMRB also added a stage of A1.11 ‘implicit’ stratification by sorting the local authorities in each stratum by index of multiple deprivation.Selection of local authoritiesBMRB employed a different sampling fraction in each stratum to reflect A1.12 particular policy interest in: (a) experience of all types of temporary accommodation, including the less commonly used types; and (b) the experiences of ethnic minority populations. Both ethnic minority groups and unusual forms of temporary accommodation A1.13 tend to be concentrated in urban areas (particularly in London) and local authorities in these areas were therefore sampled at a disproportionately high rate. However, the need for robust data from less urbanised locations, and for a substantial overall effective sample size, led BMRB to recommend a fairly minor degree of disproportionate sampling. 344 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsBMRB suggested that around 80-100 local authorities should be involved in A1.14 the survey so that the impact of cluster effects could be kept to a minimum. BMRB felt this was the maximum number its staff could handle since both the feasibility and pilot studies demonstrated that the process of gathering raw data on statutory homeless households from local authorities was labour intensive.The involvement of 80-100 local authorities would mean average local A1.15 authority/cluster sizes of 20-25 interviews in the families survey565 and 5-7 interviews566 in the survey of 16-17 year olds. For sampling purposes, it was assumed that 20 family interviews would be carried out per local authority in strata 2-4 and 35 family interviews per local authority in stratum 1 (see below). BMRB assumed that ~60 per cent of sampled local authorities would co-A1.16 operate and that, on average, interviews would be achieved from 40-50 per cent of sampled households. A high non-contact rate was expected, particularly among 16-17 year olds. The refusal rate was expected to be fairly low, given there was a £10 cash incentive.Table A1.1 shows how many local authorities were sampled and how many A1.17 families were sampled within each. It also shows the target number of successes in each case.565 Total interview target = 2000.566 Total interview target = 500 interviews.Appendix 1: Technical report summary 345Table A1.1: Number of local authorities / families sampled by BMRB (across sampling strata)Stratum Target number of co-operating LasSampled number of LAsTarget no. of interviews per LASampled families per LA[1] Above average population density/above average ratio TA families/new acceptances25 38 [from 38]35 75+ or all if <75[2] Above average population density/ below average ratio TA families/new acceptances21 35 [from 58]20 50 or all if <50[3] Below average population density/ above average ratio TA families/new acceptances20 33 [from 70]20 50 or all if <50[4] Below average population density/ below average ratio TA families/new acceptances15 25 [from 130]20 50 or all if <50Total 81 131 [from 296]Within each stratum, BMRB sampled local authorities with a probability A1.18 proportionate to the average number of families accepted as homeless per quarter (using 2004 P1E data). A large number of local authorities – 55 – were selected with certainty under BMRB’s design. These local authorities were set aside before sampling from among the remaining local authorities in each stratum. If cluster sizes are kept equal in all sampled local authorities within the A1.19 stratum, this ‘PPS’ design will result in a sample in which all families have an equal probability of selection (an epsem sample). 346 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsSelection of families and 16-17 year oldsLocal authorities tended to find it very difficult to attach classificatory data to A1.20 their lists of households accepted as homeless. Where possible, BMRB sorted the lists of households accepted as homeless by current housing status (re-housed or still in TA) and by date of acceptance within these two groups. Occasionally other data were available such as ethnic group, household type and age. These were used to sort the list before sampling when available. Some families/16-17 year olds had to be excluded before sampling because A1.21 contact details were missing or insufficient and local authorities could provide no additional data. 4.0% of families and 10.8% of 16-17 year olds could not be sampled for these reasons.Selection of individuals to interviewIn most cases the interviewer conducted the interview with the person named A1.22 on the homelessness application. If the local authority provided two names, the interview was conducted with the person who was responsible for dealing with housing issues most often, and who could most easily comment on the position of the family as a whole. Alternatively if only one name was provided by the local authority and this person was difficult to contact or if they refused to take part in the survey, the interviewer had the option of interviewing an unnamed partner if he/she was part of the household at the time of the homelessness application. In short, a ‘purposive’ within-household sample design was employed.If a household sampled for Survey 1 contained one or more children aged A1.23 8-15, an interview was sought with one randomly selected child. Survey 2 comprises these interviews.For Survey 3, only the named individual could be interviewed.A1.24 Surveys 4 and 5 – families in temporary accommodation for more than one yearOverall designSurveys 4 and 5 were conceived as supplements to the main work of Surveys A1.25 1 and 2. Because the earliest date for acceptance as homeless was 1st January 2005, the design for Surveys 1 and 2 explicitly excluded all families in temporary accommodation for extended periods (a group of particular policy interest). Surveys 4 and 5 were designed to cover this population567.Sampling was carried out in early 2006 among households accepted as A1.26 homeless before 1st January 2005 that (a) contained child(ren) or a pregnant 567 An attempt was also made to survey young people accepted as homeless 16-17 year olds who had stayed in temporary accommodation for over one year. However, the achieved sample size was too small for robust statistical analysis.Appendix 1: Technical report summary 347woman at the time of acceptance, and (b) were still living in temporary accommodation at the time of sampling (in practice, all had been in temporary accommodation for over one year at the point of interview).The general sampling process was the same as that employed for Surveys A1.27 1 and 2: a sample of local authorities followed by a sample of eligible households within each selected local authority568.BMRB opted to set a general maximum A1.28 issued sample size per local authority of 50 families. This meant a minimum of 20 local authorities participating and BMRB suggested sampling 35 local authorities in order to guarantee this minimum.Sampling local authoritiesSection E7 of the P1E dataset shows the numbers leaving temporary A1.29 accommodation (“Households leaving accommodation secured under S.193 during the quarter”), excluding homeless-at-home.The households leaving temporary accommodation in a given quarter (or A1.30 longer period) should be a roughly representative sample of all those in temporary accommodation. E7 provides a rough breakdown in terms of the total length of time in temporary accommodation and is, therefore, a reasonable guide to the length of time households in each local authority generally spend in temporary accommodation. There was no information about how many of those in temporary A1.31 accommodation at the time of sampling had spent at least 6 months there. Nor was there any specific information about families or young people.Nevertheless, BMRB decided to use the E7 data to produce a number for A1.32 use as a ‘size measure’ for selecting local authorities for Survey 4 with a probability proportionate to size. The assumption was that this size measure would provide a good estimate of the relative position between local authorities.The formula was:A1.33 Number of ‘family’ households currently in temporary accommodation * pro-portion of households leaving temporary accommodation in last quarter that spent at least 12 months in temporary accommodation (e610d) * ((e78c+e78d+e78e)/e78f)568 Total target interview for Survey 4 = 500.348 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsBecause this size measure did not represent the actual number of survey-A1.34 eligible households in each local authority, BMRB needed to make an assumption about the number that would prove to be eligible and exclude local authorities that were unlikely to generate a minimum number of interviews. BMRB set a minimum size measure of 100 on the conservative assumption A1.35 that the size measure would be an overestimate of the number of survey-eligible households. Only 55 of the 354 local authorities met this criterion. However, this group A1.36 covered 91 per cent of the total sum of size measures so the exclusions are unlikely to introduce much bias. These 55 local authorities were listed in the stratification order compiled for Surveys 1 and 2.A sample of 35 local authorities was drawn from among the 55 with a A1.37 selection probability proportionate to the size measure.After the survey, it was possible to check the correlation between the size A1.38 measure and the actual number of survey-eligible families. On average, the size measure overestimated the number of survey-eligible families by a magnitude of 1.08. The correlation between the number of survey-eligible families in local authorities and the relevant size measures was .77. This suggests that the size measure was a very good proxy to use.Before selection, families were stratified much as they were for Surveys 1 A1.39 and 2. Where the total available sample in a local authority was 50 or less (families) the full database was issued to the interviewers569.Respondents were selected for interview within sampled households in the A1.40 same way as they were for Surveys 1 and 2. Questionnaire development and fieldwork measures (all surveys)Questionnaire developmentThe questionnaires for surveys 1, 2 and 3 were developed over the course of A1.41 six months in 2005. CHP developed a list of research questions in consultation with ODPM (now Communities and Local Government) and BMRB scripted drafts accordingly. 569 Except cases where contact details were insufficient.Appendix 1: Technical report summary 349BMRB used its qualitative research team to carry out one-to-one interviews A1.42 testing cognitive understanding of key sections. They worked with known service providers to assist recruitment for these interviews. BMRB interviewed adults, children (as young as 8) and 16-17 year olds in the process. BMRB carried out cognitive testing in two distinct stages, allowing for A1.43 revisions between the two stages.A separate dress rehearsal, testing (1) sampling and fieldwork procedures and A1.44 (2) interview length, was carried out in Norwich and Peterborough (neither area was selected for the main sample).Only slight adjustments were made to the questionnaires for Surveys 4 & 5, A1.45 and none of these adjustments were subject to testing.Last settled accommodation One of the most important elements of questionnaire design was the A1.46 development of a definition of a ‘last settled accommodation’ for families (or, more precisely, adult respondents within these families) and 16-17 year olds, primarily to be used as a device to allow comparisons of these households’ circumstances and well-being at point of survey to those prior to their experience of homelessness. In the absence of any possibility of carrying out a longitudinal Randomised Controlled Trial, comparing households who had and had not experienced the statutory homelessness system, this was the best mechanism available to us to identify any possible impacts of homelessness and temporary accommodation on families and young people. In order to be a useful comparison point, it was important that families/A1.47 young people had lived long enough in this last settled accommodation for it to have constituted a stable base for them, and also that they had occupied it recently enough to avoid serious recall problems. The detailed criteria that were decided upon for previous accommodation to be deemed to constitute a valid comparison point were as follows:• it had to be ‘ordinary’ housing rather than managed accommodation (such as a hostel, supported accommodation, or a B&B hotel); • the family/young person had to have lived there within the last 2 years;• the family/young person had to have lived there for at least 6 months; • it had to be situated within the UK; and• it could not be the family/young person’s current accommodation.350 Statutory Homelessness in England: The Experience of Families and 16-17 Year Olds The process whereby last settled accommodation was identified for each respondent was as follows. Respondents were first asked to self-identify their last settled accommodation (and were prompted to consider whether this was the place from which they had applied as homeless or an earlier place). This ‘respondent-defined’ last settled accommodation was then interrogated to see if it met the objective criteria outlined above. If so, this was deemed to be their (respondent-defined) last settled accommodation for comparison purposes. If this respondent-defined last settled accommodation did not meet the above criteria, or if no last settled accommodation was self-identified by the respondent, then there was an interrogation of the respondents’ recent housing history to establish whether any (other) accommodation in which they had stayed met the above criteria (even though it was not defined by the respondent as their last settled accommodation). If such accommodation could be identified, then this was deemed to be their (questionnaire-defined) last settled accommodation for comparison purposes.Most (71 per cent) families in Survey 1 had a last settled accommodation A1.48 (either respondent-defined or questionnaire-defined) that fulfilled these objective criteria to be deemed a valid comparison point for the purposes of this research, and 29 per cent did not. Likewise, two thirds (66 per cent) of young people accepted as homeless 16-17 year olds had a last settled accommodation that was valid for comparison purposes. All analysis of ‘changes since last settled accommodation’ provided in the report is restricted to the families/young people who reported a last settled accommodation that provided such a valid comparison point. For almost all adult respondents in families accepted as homeless (99 per A1.49 cent), either a respondent-defined or a questionnaire-defined last settled accommodation could be identified, even though in a proportion of (respondent-defined) cases this was not valid as an (objective) comparison point. This broader definition of last settled accommodation (i.e. not restricted to those cases where it provided a valid comparison point) was used to investigate the ‘origins’ of family homelessness (see Chapter 4). Fieldwork measuresAdvance lettersAll selected families and young people were sent a letter addressed from A1.50 ODPM (now Communities and Local Government) in advance of the interviewer calling at the address. This letter gave the family/young person information about the survey and an opportunity to opt out. If the family or young person was living in accommodation with a gatekeeper (e.g. B&B hotel, a hostel and so on), a letter introducing the survey was sent addressed to the gatekeeper (usually the manager). This letter did not identify the family or individual BMRB were trying to contact.Appendix 1: Technical report summary 351An ‘opt in’ approach was taken with those living in a refuge. In these cases A1.51 the local authority routinely only provided a PO BOX address and therefore it was not possible to send an interviewer to the address. BMRB sent the manager of the refuge a letter introducing the survey and the advance letter addressed to the family or young person. If the family or young person was willing to take part in the survey, they were asked to contact BMRB with their contact details to pass onto the interviewer. Unfortunately, very few refuges co-operated with the survey request. A few refuges did respond to the letter but only called to say that the person had moved on and that they would forward the letter. According to P1E data from the first six months of 2005 only 1 per cent of families accepted as homeless living in temporary accommodation are in refuges. Therefore the lack of interviews with those living in refuges should not bias the results of the survey.Introducing the surveyInterviewers were provided with a doorstep introduction in the contact sheet: A1.52 “Good afternoon/evening. My name is …….from The Operations Centre calling on behalf of BMRB Social Research. We are carrying out a survey about your experiences of being homeless.”The subject of homelessness is a sensitive issue and if the named contact A1.53 was not at home and someone else at the address enquired about why the interviewer was calling, interviewers were briefed to say that the survey was about ‘housing issues’. It is standard practice on BMRB Social surveys to leave a ‘calling card’ if the interviewer is unable to contact the named person after 2-3 calls at an address. Calling cards are used to remind the household about the survey and provide interviewer contact details so member(s) of the household can contact the interviewer to arrange an appointment. Again as homelessness is a sensitive issue, rather than just leaving the card itself, interviewers were briefed to put the calling card into an envelope and address it to the named contact. Conducting the interviewThe majority of the interview was conducted using CAPI technology although A1.54 all surveys included a self-completion module in which the laptop was turned away from the interviewer so the respondent could enter answers in private. This self-completion section covered sensitive issues such as drug and alcohol use, mental health problems, experience of domestic and sexual abuse, etc. Respondents were given headphones so they could hear the questions and response options as well as read them. Respondents with limited EnglishTo help introduce the survey amongst those respondents with limited English, A1.55. 352 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsall interviewers were provided with a ‘Doorstep Language Card’. The card introduced the survey in 12 different languages. The card was used on the doorstep when the interviewer identified that the person was not able to understand the introduction. If the language was not listed and the interviewer was not able to A1.56 communicate at all with the household, a final outcome of ‘Inadequate English’ was noted for that respondent.There were two methods used for interviewing respondents with limited A1.57 English, (1) using a family member or friend aged 12 plus and (2) using a telephone interpreter. Both of these were explained on the ‘Doorstep Language Card’. The telephone interpreter was only used when no appropriate household interpreter was available. A total of 161 interviews were conducted using an interpreter, 14 by a telephone interpreter. The telephone interpreting service was provided by the National Interpreting A1.58 Service (NIS). A much shorter version of the interview was completed if a telephone A1.59 interpreter was used. The interview was restricted to collecting details of the respondent’s housing history. All questions about support needs, the impacts on children, and the self-completion module were excluded because interpreter-interviews lasted twice as long as unmediated interviews. The interview was also shorter if a household interpreter (family or friend) A1.60 was used. The self-completion section of the interview was removed from all interviews completed using a household interpreter, and if the interpreter was aged 12-17, the section about the impacts of homelessness on children was also removed. Surveys 2 and 5 – child interviews If a household sampled for Survey 1 or 4 contained one or more 8-15 year A1.61 olds, an interview was sought with one (randomly selected) child. Before approaching the child, the interviewer had to get informed consent from the parent/guardian. The parent/guardian was asked to read the parental permission card and if he/she agreed that the interviewer could approach the child, they were asked to sign the consent form in the child contact sheet to verify permission. The child was not asked to sign a consent form to state that they were A1.62 willing to take part in the interview. This was tested at the pilot stage but interviewers reported that many of the children were unhappy about providing written permission and it caused stress for the child. Before Appendix 1: Technical report summary 353conducting the interview with the child, the interviewer had to ensure the child understood the interviewing process and that they were also given the opportunity to ask any questions before starting the interview. This was scripted into the questionnaire to ensure all interviewers read out the explanation. It was also made clear to the child that they could refuse to answer any question.In 84 per cent of cases, at least one parent was in the room when the child A1.63 was interviewed. In 7 in 10 cases where a parent was present, he/she did not say anything to influence the child. Interviewers reported occasional interference in 1 in 10 cases. The remaining 2 in 10 occasionally encouraged their child but did not explicitly influence their answers. IncentivesAt the end of the interview all families and young people were given £10 cash A1.64 as a token of appreciation for taking part in the interview. Child respondents were not given an incentive.Helpline cardsInterviewers were provided with helpline cards containing contact details A1.65 for Shelter and Childline. If they asked the interviewer for advice or support, interviewers were specifically briefed not to provide any advice to respondents but to hand out a helpline card instead. Interviewers were asked to use their own judgement before giving out the cards. Field outcomes Surveys 1, 2 and 3In total, 72 out of 131 sampled local authorities agreed to take part in the A1.66 survey. Table A1.2. shows the response rate within each stratum. This varied from 48% in stratum 3 (16 out of 33) to 64% in stratum 4 (16 out of 25). Table A1.2: Field outcomes (LA)Stratum Sampled Co-operated Response rate1 38 20 53%2 35 20 57%3 33 16 48%4 25 16 64%All 131 72 55%354 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThe response rate among families was A1.67 58.8% and among 16-17 year olds was 33.9%. The vast majority of non-interview outcomes were non-contacts, particularly cases where the family/young person was known to have moved from their last known address but neither the current occupants, neighbours nor the local authority were able to provide a new address. This was a particular problem with the survey of 16-17 year olds: 46% of final outcomes were classified in this category (compared to 23% of final outcomes in the families survey).The cumulative (unweighted) response rate for the families survey was A1.68 32.3% (55%*59%) and for the 16-17s survey it was 18.6% (55%*34%). The cumulative response rate in each stratum was similar, despite differential response rates at each stage of fieldwork. The cumulative response rates (both surveys) in stratum 1 were lowest while those in stratum 4 were highest. Table A1.3 shows this data. Table A1.3: Cumulative response rate (unweighted)Stratum LA response rateFamilies response rateCumulative rr (families)16-17s response rateCumulative rr (16-17s)1 53% 56% 29% 29% 15%2 57% 58% 33% 31% 18%3 48% 68% 33% 36% 17%4 64% 57% 36% 39% 25%All 55% 59% 32% 33% 18%Where households in the families survey contained one or more 8-15 year A1.69 olds, an interview was sought with one (randomly selected) child, so long as an adult had already co-operated with the survey. The (unweighted) response rate was 66%. The figure of 66% does not factor in the A1.70 adult response rate. There were 1,476 cases where the adult final outcome was non-contact, refusal or other unsuccessful outcome. A proportion of these would have had child(ren) eligible for the survey. There are 2,105 cases where eligibility is known and 32.2% meet the criteria (677/2,105). If we assume that 32.2% of the cases with unknown eligibility would have been eligible, the overall (unweighted) response rate was 39.1%. If the local authority response rate of 55% is also factored in, the final cumulative response rate was 21% (55%*39%).Appendix 1: Technical report summary 355Weighted response ratesWeighted figures take into account the differential selection probabilities A1.71 among issued cases. This is a better guide to sample quality. Under this measure the local authority weighted response rate was 58.1%. The weighted families response rate was 54.4% and the weighted 16-17s response rate was 34.6%. This gives cumulative weighted response rates of 31.6% for the families survey and 20.1% for the 16-17s survey. These figures are only very slightly different from the unweighted figures.CHP removed 56 families interviews and 10 16-17s interviews that were felt, A1.72 on inspection, to be ineligible for the survey. The total used for analysis was therefore 2,053 adult respondents, 450 child respondents in families, and 350 16-17 year olds.Surveys 4 and 5In total, 22 out of 35 sampled local authorities agreed to take part in Surveys A1.73 4 and 5. This is a response rate of 63%. The (unweighted) response rate among families was 58%. As with Survey 1, A1.74 most of the non-interview outcomes were non-contacts, particularly cases where the family was known to have moved from their last known address but neither the current occupants, neighbours nor the local authority were able to provide a new address. The cumulative (unweighted) response rate for the families survey was 36.6% A1.75 (63%*58%). Weighted figures are often a better guide to sample quality. Under this A1.76 measure the local authority weighted response rate was 71.0% (unweighted = 62.9%). The weighted families response rate was 55.2% (unweighted = 58.2%). This gives a cumulative weighted response rate of 39.2% for the families survey (Survey 4). CHP/BMRB removed 10 probably ineligible cases from Survey 4 before A1.77 analysis, so data are based on 571 adult respondents and 180 child respondents.356 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsWeighting and design factors Surveys 1, 2 and 3WeightingBMRB applied design weights to the data but no additional non-response A1.78 weights. An analysis of the achieved sample against P1E distributions showed it to be broadly in line with expectations. Table A1.4 shows this analysis for a number of different characteristics of family homelessness570.However, it should not be inferred from this that there is no non-response A1.79 bias, merely that it is not detectable. 571Table A1.4: P1E data distributions and (design-weighted) family survey data distributionsCharacteristic P1E data Survey data Survey – P1ENumber of children– 1 child 53% 55% +2%– 2 children 28% 28% –– 3+ children 19% 17% -2%– 1 child (London only) 55% 56% +1%– 2 children (London only) 26% 28% +2%– 3+ children (London only) 19% 16% -3%(Broad) ethnic group of applicant where known (5% unknown in P1E)– White 80% 76% -4%– Black 11% 12% +1%– Asian 6% 7% +1%– Other 3% 5% +2%– White (London only573) 43% 44% +1%– Black (London only) 32% 30% -2%– Asian (London only) 12% 15% +3%– Other (London only) 14% 10% -4%570 BMRB found the same close connection between the P1E profile and the survey profile with the 16-17 year olds.571 5% unknown in London.Appendix 1: Technical report summary 357Table A1.4: P1E data distributions and (design-weighted) family survey data distributions (continued)LA type– District Council 37% 31% -6%– Inner London borough 12% 14% +2%– Outer London borough 11% 12% +1%– Metropolitan area 23% 23% –– Unitary Authority 17% 21% +4%Region– North East England 6% 11% +5%– Yorkshire/Humber 9% 7% -2%– East Midlands 8% 6% -2%– East of England 9% 8% -1%– London 23% 25% -2%– South East England 10% 11% +1%– South West England 8% 11% +3%– West Midlands 13% 6% -7%– North West England 14% 16% +2%Stratum– Stratum 1 20% 21% +1%– Stratum 2 31% 29% -2%– Stratum 3 13% 13% -– Stratum 4 33% 37% +4%– Excluded LAs 3% - -3%BMRB capped some of the largest design weights for Surveys 2 and 3 in an A1.80 effort to improve precision. No capping was required with the Survey 1 family homelessness adult data. Design factors BMRB used STATA to calculate design factors for a series of key variables both A1.81 for the full Survey 1 family homelessness sample and for sub-groups within the sample. Confidence intervals from a simple random sample of the same size should be A1.82 multiplied by the design factor to give confidence intervals for key variables. 358 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsBecause CHP used SPSS to analyse the data, BMRB devised a simple scaled A1.83 weight so that significance tests would assume a routine design factor of 1.41. This was considered a ‘safe’ weight, and a conservative approach which would enhance the statistical reliability of the findings in the study (see Appendix 2 for more detail). Clustering was by far the biggest contributor to these design factors.A1.84 Surveys 4 and 5As with Surveys 1-3, the only weights applied to the data in surveys 4 and 5 A1.85 were design weights. The absence of good population data made it difficult to apply non-response weighting with any confidence and BMRB eventually decided against this. 94.5% of the provided sample was from stratum 1 local authorities. The A1.86 design-weighted final sample has 86.1% in stratum 1 local authorities. This suggests that the response rate in non-stratum 1 local authorities was higher than in stratum 1 local authorities. The response rate was also slightly lower in London as Table A1.5 shows. Table A1.5: Sampling frame distribution against design-weighted distribution (Survey 3)LA type Sampling frame population (22 co-operating LAs = estimated 44% of survey-eligible population)Design-weighted dataDifference– Unitary authority 4.2% 6.5% +2.3%– Outer London 77.4% 69.3% -8.1%– Metropolitan authority0.0% 0.0% –– Inner London 12.4% 11.2% -1.2%– District council 6.0% 13.1% +7.1%– Stratum 1 94.5% 86.1% -8.4%– Strata 2-4 5.5% 13.9% +8.4%Nevertheless, these distribution differences do not indicate any major bias A1.87 in the design-weighted data even if the (very limited) real data was perfectly representative of the full population. Given the limited nature of this information, BMRB decided not to apply any non-response weights.Appendix 2: The analysis process 359Appendix 2: The analysis processMuch of the analysis presented in this report has been undertaken using A2.1 descriptive statistics, mainly in the form of frequencies (simple counts), bivariate analysis (crosstabulations), and various measures of dispersion (the average; the median, which is the middle value when all the values in a distribution are ranked; and the standard deviation, which represents the average distance of a set of values from the average). Bivariate analysis was conducted using crosstabulation, employing a range of A2.2 chi-square tests which include Pearson Chi Square, Continuity Correction, the Likelihood ratio, Fisher’s Exact Test and Linear-by-linear association. Results with a Chi-Square result beneath a 95 per cent confidence level were not reported as showing a statistical association, as is the standard convention within social science research. Where appropriate, binary logistic regression using the Forward Wald method A2.3 was also employed to explore which variables had an independent effect in determining the likelihood of a given finding, when a range of other factors were held constant. Binary logistic regression is one of a family of regression tests that allow for exploration of associations between dependent (outcome) variables and independent (explanatory) variables, while controlling for other potentially ‘confounding’ variables. While it is theoretically constructed for binary explanatory variables (i.e. variables coded as 0 or 1), nominal variables with non-ordered categories (for example different ethnic groups) can be included by converting them to one or more binary variables (for example White (0) and non-White (1)). Broadly speaking, binary logistic regression measures whether explanatory variables are associated with a given outcome occurring, controlling for the effects of all of the other variables entered in to the model. Its key output is described as the ‘adjusted odds ratio’. Possible relationships were therefore explored by examining the adjusted odds ratio (also known as the Exp(B) statistic); the research team also calculated the 95 per cent confidence interval (lower and upper) for the adjusted odds ratio, as well as examining the Wald statistic572. The findings of the regression analysis are reported very straightforwardly A2.4 in this report, by simply stating whether or not a relationship was identified 572 Pampel, F.C (2000) Logistic Regression: A primer (Quantitative Applications in the Social Sciences), London: Sage. 360 Statutory Homelessness in England: The Experience of Families and 16-17 Year Oldsbetween an outcome variable and an explanatory variable, when a range of other factors (identified in the relevant footnotes) were held constant. In order to maximise the accessibility of these results, where such an independent relationship was shown by regression analysis, this has been illustrated by showing contrasting crosstabulation percentages (bivariate statistics), rather than citing the Exp(B) and other regression statistics. For example, it was found that, when a range of geographical and demographic variables were held constant, the strongest independent effect on the likelihood of a family having moved between temporary accommodation addresses was length of time in temporary accommodation. This finding was illustrated by reporting that 58 per cent of families who had stayed in temporary accommodation for at least six months had made one or more moves between temporary accommodation addresses, as compared with only 30 per cent of those who had stayed in temporary accommodation for a shorter period. All bivariate and regression analysis was carried out using SPSS versions 14 A2.5 and 15. These versions of SPSS have limitations when analysing complex survey data because standard errors are computed assuming a simple random sample of the population, thus they do not take account the impact of disproportionate sampling probabilities, sample stratification and clustering. For the most part, this means that estimated standard errors are smaller than they should be. Other analysis programs such as STATA compute more accurate standard error estimates than SPSS. BMRB therefore used STATA to compute a range of standard errors and compared them with the SPSS estimates. On average, the STATA estimates were significantly larger so BMRB recommended that the research team inflate the SPSS standard errors by a standard margin of 1.4., which they did. This fits with a general design effect of 2.0 although, in reality, this varies from statistic to statistic and from sub-group to sub-group. So all differences and associations reported are statistically significant even when a very conservative effective sample size (half of the actual sample size) is assumed (see also Appendix 1). Cluster analysis was also undertaken by BMRB. This term refers to a family of A2.6 multivariate statistical techniques that aim to produce groupings (‘clusters’), in such a way that the members of each cluster are as alike as possible, but at the same time, as different as possible from members of other clusters. This method is used to segment a sample using behavioural or attitudinal data rather than the usual demographic descriptive data. It often proves informative about the natural groupings within a larger population. The specific cluster analysis method used in this study was K-means cluster A2.7 analysis. This method starts with a specified number of clusters – say five – and allocates every respondent randomly to one of five clusters. Using the Appendix 2: The analysis process 361responses from a battery of questions, it then makes changes iteratively to maximise the difference between the clusters and the homogeneity (sameness) of the members within a cluster.As reported in Chapters 3 and 12 respectively, this K-means cluster analysis A2.8 undertaken by BMRB resulted in two ‘adulthood’ and two ‘childhood’ ‘vulnerability clusters’ amongst the adult respondents in families accepted as homeless, and four ‘vulnerability clusters’ amongst young people accepted as homeless 16-17 year olds. These clusters were then used by CHP as key variables in interrogating the data on a range of outcomes for both families and young people. Commonly used variables in bivariate and regression analysis In both the bivariate and regression analysis employed in this report certain A2.9 groups of variables were routinely employed, and these groups of variables identified by ‘shorthand’ labels (e.g. ‘demographic characteristics’), described in the glossary of terms. The use of the relevant shorthand description to characterise the bivariate or regression multivariate analysis undertaken indicates that all of the relevant variables were employed. Where it was only some of them, the individual variables are specified and the shorthands are not used. Table A2.1: Groups of binary variables commonly employed for regression and bivariate analysis by the shorthand terms used to describe them (families accepted as homeless)Shorthand terms Variables included Accommodation conditions (settled and temporary accommodation)Family had to share bathroom and/or kitchen and/or living room with other households (only temporary accommodation)Problems reported due to sharing with other households (only temporary accommodation)Accommodation reported to have three or more of the following problems: damp, infestation, unsafe conditions which pose a risk to for children’s safety, was in poor repair when first arrived, dirty when first arrived, insufficient control over heating, difficult to enter with pram or buggy, not well decorated when first arrived Felt unsafe within accommodation Felt unsafe in area where accommodation locatedInsufficient living space reported362 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsTable A2.1: Groups of binary variables commonly employed for regression and bivariate analysis by the shorthand terms used to describe them (families accepted as homeless) (continued)Causes of homelessnessRelationship breakdown was a reason for applying as cause of homelessnessViolent relationship breakdown was a reason for applying as homeless cause of homelessnessOvercrowding was a reason for applying as homeless cause of homelessnessEviction/threatened with eviction was a cause of homelessness (including end of a fixed-term tenancy ending) was a reason for applying as homelessOverstayed welcome/could not longer be accommodated was a reason for applying as homeless cause of homelessnessCurrent accommodation typeFamily in settled housing Family in self- contained temporary accommodation Family in B&B hotel or hostel Family staying temporarily with friends or relatives familyDemographic characteristics Adult respondent had ethnic minority backgroundAdult respondent had ever sought asylum in UK Household size exceeded three membersHousehold size exceeded four members Adult respondent was aged under 25Two parent household Lone woman parent householdGeographical variablesFamily accepted in LondonFamily accepted in South East, South West or East of England (‘the South’)Family accepted in the East Midlands, West Midlands, Yorkshire and the Humber, the North East, or North West (‘the North and Midlands’)Family accepted in an area of ‘higher housing stress’ *Family accepted in an area of ‘relative deprivation’ *Family accepted in a ‘rural’ area* Appendix 2: The analysis process 363Table A2.1: Groups of binary variables commonly employed for regression and bivariate analysis by the shorthand terms used to describe them (families accepted as homeless) (continued)Housing history variablesAdult respondent had never had a settled home as an adult (also used under personal history) Adult respondent had never had an independent tenancy or been an owner occupierAdult respondent had made a previous homelessness application Temporary accommodation experience (since acceptance as homeless)Family had not stayed in temporary accommodation Family had stayed in temporary accommodation for more than 6 months Family had moved between temporary accommodation addresses Family had stayed in B&B hotel and/or hostelFamily had stayed with friends and/or familyFamily had stayed in self- contained temporary accommodation * see text below for more detail on these variablesHousing stress within areas was determined with reference to Wilcox’s A2.10 measure of housing market affordability573. The local authorities in which a respondent lived were assigned a ranking based on the affordability ratio of owner occupation for people aged 20-39 in that area (gross average house price in relation to gross average household income). Authorities were divided into quartiles: ‘least expensive’ (a ratio of less than 3.7574), the next least expensive (3.7 – 4.1), more expensive (4.2 – 4.8), and the ‘most expensive’ (a ratio of more than 4.8)575. In the analysis conducted for the present research, families or young people were deemed to be living in an area of ‘higher housing stress’ if that area was within the upper two quartiles. 573 Wilcox, S (2005) Affordability and the intermediate housing market: Local measures for all local authority areas in Great Britain, York: Joseph Rowntree Foundation. 574 i.e. average house prices were less than 3.7 times the average gross household income.575 These quartiles were devised on the basis of the 72 local authorities in which fieldwork took place, not on the basis of all local authorities in England.364 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsThe local authorities where families applied as homeless were also categorised A2.11 according to their rank in the 2004 Indices of Deprivation for England576. Four quartiles were created. One quartile was of ‘very deprived’ local authorities (those within the 29 most deprived authorities in England), ‘deprived’ (ranked between 30 and 79), ‘affluent’ (80 to 177), and ‘very affluent’ (ranked 178 or lower)577. For the purposes of analysis, families were defined as living in areas of ‘more deprivation’ if they were within a local authority area in the ‘very deprived’ or ‘deprived’ quartiles. A family was described as living in a ‘rural’ area if the local authority where A2.12 they lived was within the ‘Significant Rural’, ‘Rural-50’ or ‘Rural-80’ categories of the DEFRA Rural Definition and Local Authority Classification. Families living within local authorities that were classified as ‘urban’ were living in ‘Major Urban’, ‘Large Urban’ or ‘Other Urban’ within the DEFRA categories. These categories are defined as follows578: • Major Urban: districts with either 100,000 people or 50 percent of their population in urban areas with a population of more than 750,000. • Large Urban: districts with either 50,000 people or 50 percent of their population in one of 17 urban areas with a population between 250,000 and 750,000. • Other Urban: districts with fewer than 37,000 people or less than 26 per cent of their population in rural settlements and larger market towns. • Significant Rural: districts with more than 37,000 people or more than 26 per cent of their population in rural settlements and larger market towns. • Rural-50: districts with at least 50 per cent but less than 80 per cent of their population in rural settlements and larger market towns. • Rural-80: districts with at least 80 per cent of their population in rural settlements and larger market towns. Table A2.2 shows the variables that were commonly employed by the A2.13 shorthand terms used to refer to them for young people accepted as homeless 16-17 year-olds accepted as homeless. 576 ODPM (2005) The English Indices of Deprivation 2004 (Revised) London: ODPM available at: http://www.communities.gov.uk/pub/446/Indicesofdeprivation2004revisedPDF2198Kb_id1128446.pdf 577 These quartiles were devised on the basis of the 72 local authorities in which fieldwork took place, not on the basis of all local authorities in England. 578 See http://www.defra.gov.uk/rural/ruralstats/rural-definition.htmAppendix 2: The analysis process 365Table A2.2: Groups of binary variables commonly employed for regression and bivariate analysis by the shorthand terms used to describe them (16-17 year-olds accepted as homeless)Shorthand termsVariables included Accommodation conditions (settled and temporary)Accommodation was reported as dampInsufficient control over the heating reportedFelt unsafe inside accommodationFelt unsafe in area where accommodation locatedInsufficient living space reportedCurrent accommodation typeYoung person in settled housing Young person in self-contained temporary accommodation (council/housing association, private rented sector, or partner’s flat/house)Young person in shared temporary accommodation (B&B hotel, supported lodgings, staying with friends/relatives, or in hostel/foyer/refuge or other supported accommodation)Demographic characteristics Young person was aged 18 or over Young person was femaleYoung person had ethnic minority backgroundSocial networks Saw family less/more since leaving last settled accommodationSaw friends less/more since leaving last settled accommodationTemporary accommodation experiences (since acceptance as homeless)Had stayed in temporary accommodation Had stayed in temporary accommodation for more than 6 months Had moved between temporary accommodation addresses Had experience of shared temporary accommodation Had experience of self-contained temporary accommodation 366 Statutory Homelessness in England: The Experience of Families and 16-17 Year OldsRegression analysis on the merged Survey 1 and Survey 4 dataset As described in Chapter 1 and in Appendix 1, Survey 4 was conceived of as A2.14 supplementary to Survey 1 because the latter, by definition, excluded those families who had spent an extended period in temporary accommodation. This group were of particular policy interest because there were concerns that staying in temporary accommodation for long periods may have particularly negative effects. As reported in Chapter 2, once fieldwork was completed, it was found A2.15 that Survey 4 families had a very different geographical and demographic profile to those in Survey 1. For example, they were far more likely to have been accepted as homeless in London, and a higher proportion of the adult respondents in Survey 4 than in Survey 1 had an ethnic minority background and/or had sought asylum in the UK at some point. At the same time, analysis of Survey 1 data established that these sorts of geographical and demographic factors were very often associated with key outcomes investigated in the study. For example, adult respondents from ethnic minority backgrounds were less likely than other adult respondents to self-report mental health problems, and adult respondents in London were less likely to seek a private sector tenancy before approaching a local authority for help than those accepted elsewhere. This meant that we had to establish whether any statistical differences identified between Survey 1 and Survey 4 families could be attributed simply to the observed demographic and other geographical differences between them, rather than to the (longer) length of time spent in temporary accommodation by Survey 4 families. In order to investigate the existence of any such A2.16 independent effects of staying in temporary accommodation for over one year, we merged the Survey 1 and Survey 4 datasets and conducted regression analysis on this combined dataset. The factors controlled for in this series of regression analyses included: demographic characteristics; geographical variables; and whether a Survey 1 or Survey 4 adult respondent. If there were any differences unaccounted for by the distinct geographical and demographic profiles of the two groups, the variable ‘Survey 1 or Survey 4’ would appear as an independent explanatory variable in the regression results. If it did not (as was usually the case) this meant that the distinction between the Survey 1 and Survey 4 findings were fully accounted for by the other variables in the model (i.e. the geographical and demographic differences between the two groups). ISBN: 9781851128938Price: £45This report presents the findings of a major survey of homeless families and 16-17 year olds. The survey investigated the characteristics of homeless households and the causes of their homelessness. It also investigated their experiences of living in temporary accommodation, and the impacts of statutory homelessness and living in temporary accommodation on their health, welfare and well-being.ISBN 978-18511289389 781851 128938

Family Homelessness in Europe

Abstract

Extracted data

Family Homelessness in Europe

Abstract

Extracted data

Related items

Related items