Measuring the Impact of Youth EFNEP: Questionnaire Development and Validation

Background: Given the problem of childhood obesity and food insecurity among low-income children, the Youth Expanded Food and Nutrition Education Program (EFNEP) needs an appropriate, valid and reliable evaluation tool to determine the program effectiveness. Purpose: To describe the development and preliminary validation of EFNEP Youth Quest questionnaire, an impact assessment tool designed for Youth EFNEP program. Use of theory: The Community Nutrition Education logic model adapted with constructs of Social Cognitive Theory and Theory of Planned Behavior. Targeted audience: Low income-ethnically diverse children, in third, fourth and fifth grade. Design: The development of the questionnaire included six phases: preliminary curricula content analysis, conceptualization, construction, expert reviews, cognitive interviews, and revisions. The selected measures were: outcome expectations, self-efficacy, intentions and knowledge related to nutrition, physical activity and food safety. Each measure was assessed with different topics that emerged from a content analysis of multiple Youth EFNEP curricula. Items were selected through the literature review and/or existing instruments; new items were created as necessary. Evaluation: Content validity and face validity were assessed through expert reviews (n=5) and cognitive interviews (n=14), respectively. Data from 452 children was collected for factor analysis, internal consistency and item difficulty analysis. Test-retest reliability was also assessed (n=75). Predictive validity of the nutrition and physical activity scales were assessed using direct measures of food intake (n=62) and physical activity (n=47). Results: Content analysis, expert reviews and cognitive interviews were used to develop the questionnaire and to confirm the content and age appropriateness of the questionnaire. Factor analysis revealed interpretable factors for each of the content domains and served as strategy for item reduction. Item difficulty for knowledge items ranged between 6-92%. Internal consistency for most of the final psychosocial scales was acceptable. Kappa statistics for test-retest reliability ranged between 0.06-0.70. For predictive validity, only 3 out of the 12 hypothesized correlations were significant. Conclusion and Implications: Although further work is needed, the preliminary results of this study suggest that EFNEP Youth Quest could be used for evaluating Youth EFNEP programs. This study could serve as framework for designing similar assessment tools for different age groups