Measuring food literacy in 9 and 10 year old New Zealand children: questionnaire development, validity and reliability

by O'Sullivan, Teresa

Institution: University of Otago
Year: 0
Keywords: Food literacy; Children; Validity; Reliability; Questionnaire
Record ID: 1314511
Full text PDF: http://hdl.handle.net/10523/5522


Background: Food literacy is an emerging term used to describe the knowledge, skills and behaviours required to achieve a healthy and environmentally friendly diet. The term is increasingly used in policy and research as an approach to address complex problems such as obesity. Improving children’s food literacy in particular, has been the target of intervention studies and contemporary nutrition plans and policies. Yet the use of food literacy has been occurring without evidence for an association between food literacy and diet or other outcomes of interest. Though research in this area is growing, progression is limited by the lack of an accepted method to measure food literacy. Aim: To develop a questionnaire to assess food literacy in nine and ten year old New Zealand children and subsequently evaluate the questionnaire for validity and reliability. Methods: An online questionnaire was developed based upon the components of food literacy identified in the literature. The questionnaire consisted of 65 items in three sections: food origins, nutrition knowledge and food knowledge and skills, as well as a demographics section. The questionnaire underwent three phases of testing with revisions made after each phase. In Phase 1 the questionnaire was reviewed by an expert panel (n=11) to assess content validity. In Phase 2 the questionnaire was pretested with 9 and 10 year old children using cognitive interviews (n=4) and a focus group (n=4) to assess face validity. In Phase 3 the questionnaire was pilot tested in nine and ten year old children (n=85) from two New Zealand primary schools, to assess reliability and conduct item analysis. Internal reliability was assessed using Cronbach’s alpha and test-retest reliability using Kappa coefficients (for individualitems) and Intraclass Correlation Coefficients (for each section and the overall questionnaire). Item analysis involved determination of item difficulty and item discrimination for each item. Results: In Phase 1, the overall questionnaire demonstrated a Content Validity Index of 0.83, indicating acceptable content validity. In Phase 2, the questionnaire was revised to ensure items were clear and correctly interpreted by children to achieve face validity. In Phase 3, items that failed to meet item difficulty and discrimination criteria were removed from the questionnaire, unless essential for content validity. The internal reliability of each questionnaire section was acceptable to good (Cronbach’s α 0.69-0.79). Intraclass Correlation Coefficients showed acceptable to good test-retest reliability for each questionnaire section (0.69-0.83) and the overall questionnaire (0.89). Each phase of development progressively improved the questionnaire, resulting in a final 42-item food literacy questionnaire. Conclusions: The final questionnaire is a valid and reliable measure of food literacy in 9 and 10 year old New Zealand children. The questionnaire is a promising tool for describing children’s food literacy at a single time point, investigating any association…