Reported versus actual snacking behaviours: a note on methodology

Abstract

Self-reporting surveys have been used extensively in research to obtain information on dental health practices, snacking behaviours, motivation and lifestyles. However, how reliable and valid are questionnaire assessments of children's snacking behaviours? The purpose of this study is to do with methodology—to compare the use of questionnaires and activity sheets with the collection of ‘rubbish’ (an indicator of the actual snacks consumed) as a means of assessing the snacking behaviours in children in primary education. Four hundred and eighty-two children aged 5 and 11 years of age took part in the study. They completed a questionnaire assessment of the foods and drinks they consumed during their morning break. In addition they were required to collect the wrappers, cans or provide evidence of their actual snacks taken on each occasion in their named and dated ‘rubbish bag’. On visual inspection the results showed that there were discrepancies between the questionnaire responses and the rubbish bag collection. When the data were subjected to kappa analysis there were in general poor levels of correspondence between actual and reported snacking behaviours. The findings from this study suggest that the collection in ‘rubbish bags’ of the physical evidence of the snacks children devour may be a more appropriate indicator of the dietary habits of children in primary education.