Validity and Reliability of the Child Feeding Questionnaire in School-Aged Children

Objective: The aim of this study was to test the reliability and validity of the Turkish version of the Child Feeding Questionnaire (CFQ), which was revised by Birch et al. in 2001. Methods: The study sample included parents of 495 students who were studying in the third, fourth, and fifth grades of three public elementary schools located in Fatih district of Istanbul. Data were collected using CFQ, which was developed for measuring families' beliefs, approaches, and practice regarding child nutrition. The questionnaire includes seven sub-dimensions and 31 items. For reliability analysis, the internal consistency coefficients of the CFQ sub-dimensions, test-retest coefficient, and the item-total correlation coefficient were calculated. Content and construct validity were also tested. Results: The internal consistency of sub-dimensions was between 0.59 and 0.81, while the test-retest reliability was between 0.41 and 0.64 and item-total correlation coefficients were between 0.39 and 0.70. It was determined that expert opinions regarding the items were consistent. Construct validity of the 30-item CFQ, which included seven sub-dimensions, was tested using confirmatory factor analysis (CFA). Fit indices obtained from the initial CFA were not satisfactory. After modifying the error variance between items 8-9, 21-22, and 23-24, CFA was repeated, and satisfactory fit indices were found. Factor loadings of all items were between 0.31 and 0.84. Conclusion: The internal consistency, test-retest reliability, item analysis, content validity, and DFA results indicated that CFQ is valid and reliable and can be used for measuring parents' perceptions and worries regarding their children's obesity predisposition and nutrition practices