Early identification of potentially harmful eating patterns is critical in the effective remediation of such behaviors. The purpose of this investigation was to examine the degree lo which various factors including gender, family history, and athletic status predict disordered eating behavior; social physique anxiety and percent body fat were added as potential predictor variables. The eating behaviors of student-athletes and nonathlete students were also compared. One hundred eighty undergraduate students (males = 49, females =131) provided demographic information and completed the Eating Attitudes Test (EAT) and the Social Physique Anxiety Scale (SPAS). Stepwise multiple-regression analysis indicated that social physique anxiety, gender, and body fat (%Fat) combined to predict 34% of disordered eating behaviors: EAT = 0.921 SPA - 1.05 %Fat + 10.95 Gender (1 = M. 2 = F) - 17.82 (R 2 = .34, SE = 4.68). A one-way ANOVA comparing ihe eating behaviors of athletes and nonathletes revealed no significant difference between these groups.