RESUMO
OBJECTIVE: To develop and validate a questionnaire to assess diabetes knowledge, attitudes, beliefs, and intent to change among middle school students. DESIGN: A cross-sectional validation analysis. SETTING: Students in 3 urban middle schools in Utah participated in this study. PARTICIPANTS: The 25-item questionnaire was validated in independent samples of 277 and 304 students in the seventh and eighth grades. VARIABLES MEASURED: The questionnaire includes the following variables: diabetes knowledge, attitudes, beliefs, and intent to change dietary and physical activity behaviors. ANALYSIS: Rigorous statistical approaches, including Cronbach's α (goodness of fit) calculation, test-retest reliability, and exploratory and confirmatory factor analyses, were employed to examine the reliability and construct validity. RESULTS: The Cronbach's α coefficients were 0.7 for both subscales demonstrating acceptable internal consistency. All factor loadings were > 0.4, revealing close relations between factors and items. The confirmatory factor analysis model fit was 0.9 for the comparative fit index and Tucker-Lewis Index, indicating a reasonable model-data fit. CONCLUSIONS AND IMPLICATIONS: This study demonstrated the validity of a diabetes questionnaire for middle school students. Future work is needed to validate its use in a diabetes prevention program, given the growing need for diabetes education among young adolescents.