DUTCH EATING BEHAVIOR QUESTIONNAIRE: VALIDATION AND EXPLOITATION IN ADULTS WITH OBESITY

Aimed the validation of the Dutch Eating Behaviour Questionnaire (DEBQ) in a sample of adults with obesity, and analyse the differences between participants from São Miguel, Açores, and mainland Portugal, and between men and women regarding emotional and external eating, and restrained eating. A total of 187 adults with obesity (70 from São Miguel; 117 from mainland Portugal), with a mean age of 42 years old (SD = 14,36) and a mean BMI of 36 kg/m 2 (SD = 4,98) completed the DEBQ and the Binge Eating Scale (BES). The construct validity (factorial, convergent and discriminant), criterion, reliability and sensitivity of the DEBQ were Rua Jardim do Tabaco, n.34, 1491-041 Lisboae-mail: malbergaria@ispa.pt R. Albergaria , F. Pimenta, C. M. Galvão, A. B. Leal , I. T. Guilherme, J. Maroco, & I. Leal www.sp-ps.pt 145 evaluated. The results showed a good fit (X 2 /df = 1,83; CFI = 0,922; TLI = 0,916; RMSEA = 0,067 C.I. 90% ]0,060; 0,074[, p = 0,001), with appropriate values of factorial weights (except items 21 and 29), sensitivity and reliability (α ≥ 0,88). The average variance extracted is adequate (AVE ≥ 0,55), except for the subscale of restrained eating (0,42); the subscales of emotional and external eating have no discriminant validity. The correlation between BES and the subscales of DEBQ is positive and significant (r = ]0,53; 0,61[, p < 0,001), with an exception of restrained eating (r = 0,03; p < 0,668). The QHCA, an instrument widely used, presents adequate psychometric properties in this Portuguese sample.