Psychometric Testing of the Dutch Eating Behavior Questionnaire for Children.

Abstract

Background and Purpose: This study aimed to assess the validity and reliability of the Dutch Eating Behavior Questionnaire for Children (DEBQ-C) among 6-8-year-old Black girls. Methods: Investigators of this cross-sectional online survey of 112 Black mother-daughter dyads used principal component analysis (PCA) to assess evidence of the DEBQ-C's construct validity. Pearson's product-moment correlations were also used to compare the DEBQ-C's subscale scores to an independent measure of eating behaviors. A comparison of the subscale scores between participants with favorable and unfavorable eating behaviors was used to evaluate discriminant validity. Cronbach's alpha coefficient examined the DEBQ-C's internal consistency reliability. Results: PCA established evidence of validity with three subscales. Significant correlations were found between the independent measure of eating behavior and the emotional (r = -.25, p = .01) and external subscales (r = -.31, p = .001). Participants increased emotional (t = 2.53; df = 76; p = .007) and external (t = 3.98; df = 99; p < .001) eating scores had unfavorable eating behavior scores. Conclusions: Psychometric results demonstrate questionable support for the construct validity of two of the DEBQ-C subscales (emotional and restrained eating) and the reliability of the DEBQ-C questionnaire among 6-8-year-old Black girls. The questionnaire's three subscales, emotional, external, and restrained eating, can be used to examine eating behaviors.