Dispelling a Myth: Reevaluating the Predictive Validity of Rape Myth Acceptance for Likelihood of Engaging in Sexual Violence

Sexual violence prevention on college campuses has received significant recent attention. A prevalent intervention paradigm has centered around re-educating young people around consent and reduce endorsement of "rape myths," based on the correlation between rape myths and sexual violence incidents. Yet many of these programs have not measurably reduced sexual assaults. We evaluated the predictive value of a rape myth measure, as compared with other predictors (criminal history, childhood victimization, aggressive tendencies, substance use, and empathy), in predicting self-reported acts of forcible and incapacitated sexual assault in college-age men (N = 304) from 45 U.S. states. Across three logistic regression model pairs, rape myths were weakly associated with violence when considered as sole predictors. However, this predictive power dissipated when other predictors were included. Comprehensive models accounted significantly better for variability in outcomes; with criminal history demonstrating consistent predictive effects. Based on these findings, we recommend further research into prevention programming based on other predictors of violence.