Practice Profile: Bystander Education Programs for Sexual Assault Prevention on High School and College Campuses

Katz and Moore (2013) evaluated the effectiveness of bystander education programs on reducing sexual assault on college campuses. The meta-analysis was restricted to include bystander programs on college campuses that recruited individuals by approaching them as allies in the prevention of sexual assault. Moreover, at least some part of the program had to educate participants on how to help reduce others’ risk of sexual assault and/or train participants on how to support a victim of sexual assault. To identify studies, specific keywords related to sexual assault prevention were used to search bibliographic databases. Studies were also located through internet searches and contacting authors for copies of unpublished studies or studies that were not yet available online. Relevant journals in community psychology, prevention, and interpersonal violence were also searched, and bibliographies of studies were screened to locate other relevant studies.

To be eligible for inclusion in the meta-analysis, the study had to include North American college students who attended a bystander education program for sexual assault prevention. The study also had to provide quantitative data on the relevant outcomes, which included bystander efficacy, rape-supportive attitudes, intent to help, rape proclivity, bystander helping behaviors, and perpetration behaviors. Bystander efficacy was related to how competent individuals felt in responding to sexual assault risk. Rape-supportive attitudes focused on acceptance of rape myths (for example, agreement with statements like “if a woman doesn’t physically fight back, you can’t really say that it was rape”). Intent to help measured how likely an individual would be to help in a sexual assault situation. Bystander helping behavior measured actual instances of engaging in bystander helping behavior since the beginning of the program. Data on these outcome variables had to be gathered at pretest and posttest, and compared with a comparison or control group that did not receive the bystander education program.

The search yielded 41 potential studies. Of these 41 studies, 12 studies met the inclusion criteria and were included in the meta-analysis. The 12 studies included 10 that used random assignment and 2 that used a quasi-experimental design, and provided 32 effect sizes. The studies were conducted between 1997 and 2011 and had a total sample size of 2,926 participants. The total sample size comprised 1,474 treatment participants (i.e., those in bystander education programs) and 1,474 comparison/control individuals. The average age of participants was approximately 19, and about 66.7 percent of participants were men. All education programs were conducted in same-sex groups, and, on average, the programs lasted 140 minutes.

To analyze the impact of bystander education programs on sexual assault prevention, a random effects model was used. Each study outcome was weighted and then combined to create a summary effect size, also known as the standardized mean difference.

Kettrey, Marx, and Tanner-Smith (2019) conducted a meta-analysis to evaluate the effectiveness of bystander education programs in reducing sexual assault on college and high school campuses. The researchers identified studies through searches of electronic databases, relevant academic journals, and grey literature searches, which included contacting leading authors and experts on bystander programs to identify any current/ongoing research that may have been eligible for review, screening bibliographies of eligible students and relevant reviews to identify additional studies, and conducting forward-citation searches on Google Scholar.

The search produced 27 eligible studies from between 1997 and 2017. Twenty-one studies used randomized controlled trial designs, and six used quasi-experimental designs. Of these studies, 22 evaluations took place on a college or university campus, and five were delivered in secondary/high schools (grades 7–12). Twenty-two studies were published in peer-reviewed journals. Twenty-five studies were conducted in the United States, one was conducted in Canada, and one took place in India.

Most studies reported outcomes using continuous measures of treatment effects, which were converted using a standardized mean difference effect size metric with a small sample correction (i.e., Hedges’ g). Several studies reported binary outcomes, which were transformed from log odds ratio effect sizes into standardized mean difference effect sizes. The meta-analyses were conducted using random-effects inverse variance weights.