Psychometric properties of the children's eating attitudes test

As interest in eating disorders has grown, there has been an increased need for psychometrically sound measures of young children's eating attitudes and behaviors. The present report examines one promising measure, the children's version of the Eating Attitudes Test (ChEAT; (Maloney, McGuire, Maloney, McGuire, Daniels & Specker. 1989. Pediatrics, 84, 482-489). Like the EAT-26, (Garfinkel & Garner. 1982. Anorexia nervosa: A multidimensional perspective. New York: Brunner/Mazel), this is a 26-item self-report questionnaire. As part of a larger study, the ChEAT was administered to 308 middle school girls. Analyses indicated that the 26-item version had adequate internal reliability, both for the entire sample and for each of the three grades. However, a 23-item version had even better internal reliability. Concurrent validity was demonstrated for the 26-item version in that correlations between the ChEAT and independent measures of weight management and body dissatisfaction were significant. Again, the concurrent validity was slightly better for the 23-item version. Factor analysis of the ChEAT-26 yielded factors quite similar to those of the EAT-26. However, an additional ChEAT factor emerged that tapped restricting and purging behaviors. On the whole, the ChEAT emerged as a promising instrument for measuring disturbed eating attitudes and behaviors in middle school girls.