AGREEMENT BETWEEN SURVEY AND INTERVIEW MEASURES OF WEIGHT CONTROL PRACTICES IN ADOLESCENTS

Objective: The present study examined agreement between survey and int erview measures of weight control practices in a nonclinical sample of adolescents. Method: Surveys were administered in three school health classes. Clinical interviews were conducted in a student subsample (N = 43). Results: Survey-based prevalences for eating behaviors in the past month were: trying to lose weight, 44%; binge eating, 41%; vomiti ng, 4.1%; laxative use, 0%; and fasting 14%. Interview-based prevalenc es were 30%, 11.6%, 0%, 0%, and 0%, respectively. Sensitivity was high for all behaviors assessed. However, positive predictive values were low. Discussion: Surveys may be useful as preliminary screening tools for prevention programs, but may yield inflated estimates of unhealthy weight control practices in nonclinical adolescent populations. Resea rch is needed to examine whether adolescents overreport weight control practices on surveys or whether they are less willing to disclose suc h practices in a private interview. (C) 1998 by John Wiley & Sons, Inc .