In two multipart studies, we tested the effectiveness of an experiential learning-based intervention (Workshop Activity for Gender Equity Simulation–Academic [WAGES-Academic]) to reduce sexism endorsement. We randomly assigned undergraduates to either WAGES (n = 144) or one of two control conditions (n = 268): one where participants received the same information as WAGES but without experiential learning or another that included an experiential group activity but no gender equity information. WAGES participants (vs. both controls) reported less endorsement of sexist beliefs after completing the activity and/or at a follow-up 7–11 days later as measured by the Modern Sexism (Study 1), Neo-sexism (Study 2), Hostile Sexism (Study 2), and Gender-Specific System Justification (Studies 1 and 2) scales. Both studies demonstrated that these effects were attributable to WAGES providing more information, evoking less reactance, eliciting more empathy, and instilling more self-efficacy compared to the other conditions. Results suggest that programs to reduce sexist beliefs will be successful only insofar as they invite access to discussion in such a way that does not elicit defensive denial of the problem, create a context in which participants are readily able to empathize with other, and instill feelings of self-efficacy that one can address the problem.