The aim of this study was to replicate and extend results of a previous trial that investigated the effectiveness of 2 peer-led eating disorders prevention interventions in reducing eating disorder risk factors in undergraduate women (C. B. Becker, L. M. Smith, & A. C. Ciao, 2006). To extend findings from the previous study by allowing for investigation of differential response, the authors randomly assigned a larger sample of both higher and lower risk sorority members (N = 188; age M = 18.64 years, range = 18-21; 20% minority) to either a cognitive dissonance (CD) or a media advocacy (MA) intervention under naturalistic conditions. Interventions were delivered by trained sorority peer leaders and consisted of two 2-hr group sessions. Participants completed questionnaires that assessed eating disorder risk factors at pretreatment, posttreatment, 7-week follow-up, and 8-month follow-up. Results indicate that both interventions reduced thin-ideal internalization, body dissatisfaction, dietary restraint, and bulimic pathology at 8 months, although higher and lower risk participants responded somewhat differently. Both CD and MA generally appeared effective for higher risk participants; only CD, however, appeared to benefit lower risk participants. Results further support the viability of using peer leaders in dissonance-based prevention. (Contains 4 tables, 1 figure, and 2 footnotes.)