Assessing the spatial structure of population variables in England and Wales

Areas within England and Wales have population profiles that make them distinct from other locales; some areas have lower unemployment rates than others, while, in some places, there is a greater mix of ethnic groups than elsewhere. Thus, the degree of difference between areas differs geographically and between population sub‐groups. Being able to measure change in these differences is crucial in assessing whether the population has become more or less similar over time. The spatial distribution of the population by, for example, ethnicity or employment status can be characterised and the resulting measures show how the population is geographically organised, and how this changes through time. For example, spatial concentrations of the population by age may be less obvious locally (e.g. within a town or city) or regionally (e.g. the north west of England) than by housing tenure. This paper makes two key contributions: (i) it introduces methods for the analysis of spatial distributions of population sub‐groups and (ii) enhances our understanding of the characteristics of population sub‐groups in England and Wales and how they have changed over time. Based on Census data for Output Areas, the analysis uses the index of dissimilarity, Moran's I autocorrelation coefficient and the variogram to measure (spatial) variation in variables representing population sub‐groups by age, ethnic group, housing tenure, car or van ownership, qualifications, employment, limiting long term illness and National Statistics Socio‐economic Classification. The analysis shows that, between 2001 and 2011, unevenness in most population sub‐groups reduced and the populations in individual Census zones across England and Wales became more similar. Neighbouring Census zones also became more similar (more ‘clustered’). The findings suggest that there were decreased differences both within and between regions for many population variables between 2001 and 2011