Evaluations of bystander intervention education programs demonstrate that this approach results in students’ increased willingness to intervene in prosocial ways to prevent sexual violence (e.g., Moynihan, Banyard, Arnold, Eckstein, & Stapleton, 2010). These programs often focus on first-year college students, though theories and research on student development raise questions about when students are best trained as bystanders (Bennett, Banyard, & Garnhart, 2014). While some authors argue that bystander intervention is more effective for older secondary school students (Polanin, Espelage, & Pignott, 2012), other researchers highlight how traditional-aged students in their early years of college are focused on forming new relationships, a stage of growth that may make the bystander role of confronting or challenging peers difficult (e.g., Bowman, 2010).
Bystander intervention has generally been treated as a monolithic construct, where individuals are encouraged to step in and “do something” to prevent sexual violence. Viewing bystander intervention in this way is more applicable in the field of social psychology, where it has been studied in situations that are typically fairly straightforward, such as witnessing a crime or a health emergency. In the field of sexual violence, however, the bystander situation is more complicated. Sexual violence occurs on a continuum that includes different types of opportunities for individuals to take action, ranging from emergency situations to low-risk opportunities to address sexist behavior that supports sexual violence (Stout, 1991). In higher education, bystander programs have been critiqued for emphasizing how to intervene without first helping students recognize when they can intervene. McMahon and Banyard’s (2012) conceptual model includes a “continuum of opportunity” for preventing sexual violence on campus in both reactive and proactive ways. High-risk bystander opportunities occur directly before a sexual assault where there is a high level of harm facing the potential victim, such as witnessing a potential perpetrator giving alcohol to render someone incapacitated in order to have sex with that individual (McMahon & Banyard, 2012). In low-risk opportunities, derogative attitudes toward women may be expressed by the bystander’s peers (e.g., using sexist language, telling rape jokes) but do not present immediate risk to any potential victims. In opportunities to respond during or after an assault, bystanders may know or suspect that a sexual assault is occurring, such as directly observing the assault or hearing about it after it occurs. Proactive opportunities are “positive actions that students can take to demonstrate a commitment to addressing sexual violence” and can include taking a class to learn more about sexual violence or becoming a peer educator on campus (McMahon & Banyard, 2012, p. 10). It is important for the field to begin to distinguish among these various types of bystander opportunities because they may be perceived differently by students and therefore may require different types of skill building.
Designing appropriate bystander educational efforts for students as they enter college requires understanding the base of knowledge and behavior they already possess. To date, however, research into this question is sparse. In this article we examine the patterns of bystander behavior reported by incoming college students to provide a better view of the landscape into which bystander-focused prevention on college campuses is placed.
Study participants were recruited from students attending a new student orientation program at a large public university in the Northeast. The current data were collected as part of a larger longitudinal study, funded by the Centers for Disease Control and Prevention. Students were invited to complete a paper-and-pencil survey and were entered in a raffle to win TVs and iPads.
A total of 6,003 new students entered the university and 4,386 students completed the survey, for a 73% response rate. A total of 299 participants failed to correctly answer two reliability check questions, and an additional 417 participants had duplicate participant codes; all of these participants were removed from the data set prior to data analysis, for a final sample size of 3,670. This total included 46.9% males and 52.9% females (0.2% missing). Among the sample, 47.2% were White, 26.7% Asian or Asian American, 5.7% Black/African American, 7.5% Latino, 5.3% Multiethnic, 2.0% Middle Eastern, and 5.3% Other (0...