Development and initial validation of an underreporting of Disordered Eating Behaviors Scale.

The purpose of this study was to develop the Underreporting of Disordered Eating Behaviors Scale (UDEBS) to detect attempts by respondents to underreport eating patterns that reflect symptoms of an eating disorder. In "Study 1", the scale and validity measures were administered via an online survey to 692 undergraduates. In "Study 2", these measures were administered to 810 undergraduates using a simulation design where participants were randomized to four conditions: (1) answering as truthfully as possible, or as if they are trying to hide (2) Binge Eating Disorder (BED), (3) Anorexia Nervosa (AN), or (4) Bulimia Nervosa (BN). "Study 1" results suggested that men and women have divergent UDEBS distributions and item endorsement rates; therefore, separate analyses were performed by gender. Individuals who scored above one standard deviation (SD) on the UDEBS reported less body dissatisfaction and disordered eating (ps < 0.05) than those who scored one SD below. In "Study 2" women who were instructed to hide an eating disorder reported significantly higher UDEBS scores than women instructed to answer truthfully. There were no significant differences between conditions for men. The final 15-item scale indicated that women who complied with instructions to underreport produced scores 1.45 (BN) to 1.72 (AN) times greater than the truthful responding condition. This novel scale can potentially assist in detecting women who underreport in relation to disordered eating, which may be useful in research and clinical contexts where underreporting is a concern. Future research is needed with men and clinical samples.Level I Evidence obtained from an experimental study.