Cross-Cultural Factorial Validation of the Children's Eating Behavior Questionnaire (CEBQ) in Samoan Children: A Methodological Note

To evaluate the factorial validity of the 35-item Children's Eating Behavior Questionnaire (CEBQ) among Samoan children and assess whether empirically derived subscales were associated with weight status.

Caregivers of n = 400 children aged 5-11 years in the ongoing Ola Tuputupua'e cohort study completed the CEBQ. We tested the original eight-factor structure using confirmatory factor analysis. Because model fit was poor, we conducted exploratory factor analysis and used multinomial logistic regression to examine associations between derived subscales and child weight status.

The original CEBQ structure fit poorly (chi-square/df = 3.53, CFI = 0.654, TLI = 0.613, RMSEA = 0.080). EFA identified a 14-item, three-factor solution: Enjoyment of Food, Emotional Undereating, and Emotional Overeating. Compared with children in the lowest Enjoyment of Food tertile, those in the second and third tertiles had higher adjusted odds of obesity (aOR = 4.59, 95% CI: 2.56-8.23; aOR = 10.56, 95% CI: 4.79-23.26, respectively). Emotional eating factors were not significantly associated with weight status.

The original CEBQ factor structure was not replicated among Samoan children. The alternative three-factor structure requires confirmation, but the findings underscore the need to establish measurement validity before applying eating behavior instruments across cultural settings.