A study of contextual factors influencing the perception of vegetables among primary school children

Low vegetable consumption among children, particularly in Scotland, is a cause for concern. Many factors may influence children’s vegetable consumption, including contextual factors such as socio-economic status and city of residence. It is also likely that associated sensory properties influence children’s perceptions of vegetables. The research presented in this thesis aimed to examine the influence of city of residence and socio-economic status on the vegetable perceptions and preferences ofprimary school children. The impact of the children’s frequency of vegetable consumption on their perceptions and preferences was also investigated. In total, 23 children aged between 8- 11-years-oldparticipated in the pilot study, and 94 children aged between 8-10-years-old in the main study. The samples used in the main study (baked beans, tomatoes, carrots, cabbage, cauliflower, turnip, sweetcom and peas) were all commonly consumed in Scotland and familiar to the subjects. Repertory grid method was used to obtain perceptual information of a free-choice nature relating to the vegetables, and additional hedonic data (without sample tasting) and contextual information was collected from each child. Data were principally analysed using generalised Procrustes analysis and internal preference mapping. Non-linear principal components analysis (PRINCALS) and homogeneity analysis (HOMALS) were employed to explore associations between children’s perceptions of the vegetables and demographic variables. Finally, consensus data were obtained from the free-choice data and an extended internal preference map was constructed to examine correlations between perceptions and preferences for the vegetables. In conclusion, the children’s preferences for and perceptions of the vegetables were influenced by socio-economic status, probably related to their experiences with/exposure to vegetables due to differences in the availability of different varieties between the socially disparate areas. Neither age, city of residence, gender, nor typical weekly consumption of vegetables were found to significantly influence the children’s preferences or perceptions of the sample vegetables. The texture of the vegetables strongly influenced the children’s perceptions and preferences of the sample vegetables. Despite the apparent influence of the samples ’ textural attributes, the flavour of the vegetables (i.e. sweet versus bitter) was additionally responsible for sample likes/dislikes. Repertory grid was found to be a suitable method for children aged 8- years-old and above. It was also determined during the course of this thesis that children aged 8-years-old and older could successfully understand the concept of linear bipolar scales and could rank samples accordingly, both for their liking of the samples andfor the intensity of particular descriptive attributes using such scales