This study prospectively evaluated perpetrator risk factors for sexual assault perpetration, including peer influences, beliefs and attitudes about sexuality, alcohol use, and token resistance. Perpetration of sexual assault was evaluated at three time periods: pretest, 3-month follow-up, and 7-month follow-up. Retrospective and prospective analyses yielded differential predictors of sexually aggressive behavior. However, perpetration of sexual assault at any particular assessment period was a predictor of perpetration during the subsequent follow-up period. Furthermore, several variables that have previously been demonstrated in the literature to be related to the perpetration of sexual assault were not significant in regression analyses, indicating that these variables may be rendered insignificant when accounting for past perpetration in prospective analyses. These findings may have significant potential impact on development of sexual assault prevention programming with men.