The Impact of Victim Intoxication and Perpetrator Status on Bystander Responses

As many as one in four women report some form of sexual victimization during their undergraduate years (Fisher, Cullen, & Turner, 2000). The high prevalence of sexual assault on college campuses has led many universities to implement bystander interventions. These interventions encourage students to intervene to prevent risky sexual situations from escalating to assault. Although data evaluating bystander interventions are promising (e.g., Banyard, Moynihan, & Plante, 2007; Coker et al., 2011), little is known about the situational factors that may interfere with bystanders taking action. The present study addressed this gap by examining the effects of victim alcohol intoxication and perpetrator social status (represented as a college athlete) as potential barriers to effective bystander intervention. Using an experimental design and vignette methodology, bystander responses were assessed at two time points: (a) in response to a situation involving imminent sexual risk (T1), and; (b) in response to a victim’s disclosure of sexual assault (T2). A moderated mediation model was tested to determine whether bystander responses to sexual risk at T1 would mediate the interactive effects of victim alcohol intoxication and perpetrator social status on bystander responses to disclosure at T2. A statistical interaction was expected such that participants observing a sexual risk scenario involving a highly intoxicated victim and an athlete as the perpetrator would be the least likely to intervene. Those who were less likely to intervene would, in turn, respond less positively and more negatively to disclosure. The hypotheses were partially supported. Victim alcohol intoxication—though not perpetrator status—predicted bystanders’ likelihood to intervene at T1 such that those in the high intoxication condition were more likely to intervene than were those in the low and no intoxication conditions. Further, bystanders’ likelihood to intervene in the moment predicted a number of responses to the victim’s disclosure at T2. Those who were more likely to intervene provided more supportive and less distracting responses, reported fewer rape supportive attributions, and attributed lesser blame to the victim (for the assault). The utility of these results, limitations of the present study, directions for future research, and implications for prevention are discussed