Changing Attitudes About Being a Bystander to Violence

Abstract

Bystander approaches to reducing sexual violence train community members in prosocial roles to interrupt situations with risk of sexual violence and be supportive community allies after an assault. This study employs a true experimental design to evaluate the effectiveness of Bringing in the Bystander™ through 1-year post-implementation with first-year students from two universities (one rural, primarily residential; one urban, heavily commuter). We found significant change in bystander attitudes for male and female student program participants compared with the control group on both campuses, although the pattern of change depended on the combination of gender and campus.

Keywords

bystander response prevention evaluation sexual violence prevention

Sexual assault is a pervasive and serious problem facing college campus communities, with estimates that one in four to five women experience an attempted or completed sexual assault in their college career (Fisher, Cullen, & Turner, 2000). Sexual assault survivors may experience a host of negative mental and physical health consequences (Gidycz, Orchowski, King, & Rich, 2008; Jordan, Campbell, & Follingstad, 2010). This underscores the need for colleges and universities to have prevention and intervention programs to reduce the incidence of sexual assault on campus and provide needed services and support for survivors. Sexual assault prevention programs, including those based on a bystander approach, have proliferated on college campuses. This study examines the long-term impact for men and women of the Bringing in the Bystander™ sexual violence prevention program on two college campuses.

A bystander approach to the prevention of sexual violence focuses on changing community attitudes and norms and providing all community members a specific role they can identify with and adopt in preventing sexual violence (e.g., Banyard, Moynihan, & Plante, 2007; Katz, Heisterkamp, & Fleming, 2011; Langhinrichsen-Rohling, Foubert, Brasfield, Hill, & Shelley-Tremblay, 2011; Potter, 2012). While addressing that the majority of perpetrators are men and that women are disproportionately victims of sexual violence on campus, a bystander prevention approach is designed to create a community in which all individuals are willing and able, each in their own way, to interrupt an assault or a situation that could lead to assault, to speak out against social norms that support sexual violence, and to be an effective and supportive ally to survivors of sexual violence. To date, the powerful role bystanders may play in preventing victimization and assisting survivors remains a largely untapped resource (e.g., West & Wandrei, 2002). For example, K. M. Anderson and Danis (2007) found sorority sisters unprepared to help friends who experienced sexual or relationship violence. However, friends are often the first and only people college student survivors tell about their experiences (Banyard, Moynihan, Walsh, Cohn, & Ward, 2010), underscoring the importance of engaging members of the college community in being effective bystanders.

With the growth of campus-based bystander approaches to sexual violence prevention, research has yielded promising findings for in-person education program effects on attitudes and behaviors, including perceptions of wrongfulness of coercive behaviors, reduced association with sexually aggressive peers, and greater bystander efficacy and intent to intervene as an active bystander (e.g., Gidycz, Orchowski, & Berkowitz, 2011; Katz et al., 2011; Langhinrichsen-Rohling et al., 2011; Moynihan, Banyard, Arnold, Eckstein, & Stapleton, 2011; O’Brien, 2001). As with most prevention efforts, this work has centered mainly on attitude change (Lonsway et al., 2011), although recent research has highlighted the importance of these attitudes as correlates of behavior (Banyard, 2008; Banyard & Moynihan, 2011). However, to date, research has focused on short-term change and has been limited to study of one campus at a time, typically the campus where the program was developed. A necessary next step is to translate existing programs to new campus environments and to address the following question: Will programs be effective on a new campus and in the same way? On one hand, programs grounded in empirical research and theory about what motivates positive bystander behavior should easily translate from one context to another. On the other hand, theories of community readiness for change in prevention suggest that communities may differ in their receptivity to and understanding of prevention messages and thus program tailoring will be needed (Edwards, Jumper-Thurman, Plested, Oetting, & Swanson, 2000). In addition, research on helping behavior indicates that extent of helping varies by gender, by culture, by rural or urban location, and even by gender composition of the group of bystanders present (see Banyard, 2011, for a review of ecological/contextual variables that impact active bystander intervention). For colleges and universities, location (rural vs. urban), size, racial and ethnic composition of the student body, and whether it is a largely residential or commuter campus may impact receptivity to prevention efforts. Campuses may also vary in how much they have engaged with the issue of sexual violence and its prevention (Karjane, Fisher, & Cullen, 2005).

This study investigates the question of translation and uses an experimental design to examine the effectiveness of the Bringing in the Bystander in-person sexual violence prevention program with first-year students on two campuses. We focused on first-year students for two reasons: First, because research has found that first-year female students are at particularly high risk for experiencing a sexual assault (Banyard, Cohn, Moynihan, Walsh, & Ward, 2007; Gross, Winslett, Roberts, & Gohm, 2006); and second, first-year students were targeted to capitalize on the start of the developmental transition to adulthood. The transition to college in late adolescence is a developmental moment when many young people begin to live independently. It is a time of continued identity exploration and a time of opportunity; a chance for prevention messages and new behavioral norms to be explored and adopted.

In this first examination of translating Bringing in the Bystander, we focused on assessing attitude change by measuring students at pretest, posttest, and 12-month follow-up, which is a longer follow-up period than past research. We focused on attitude change in this article because it is the first hurdle to overcome in translating a program to a new setting, and understanding the factors that influence these changes is a critical component of any evaluation. Although the campuses had many differences (described below), we hypothesized the program would work similarly on both campuses—that students who participated in the program would be more ready to act as an effective bystander to sexual violence compared with those in the control group (at first posttest) and that this change would remain for 12 months after the program. Specifically, we hypothesized that on both campuses, students who participated in the program would show decreases in rape myth acceptance and would show increases in knowledge about sexual violence, awareness of sexual violence as a problem (often referred to as precontemplation in the readiness to help model described below, feeling they share responsibility for solving the problem of sexual violence (often referred to as contemplation in the readiness to help model described below), confidence they can effectively intervene (i.e., bystander efficacy), and intent to intervene as a helpful bystander.

Method

Participants

We recruited three cohorts of first-year students into the study on Campus 1, and two cohorts of first-year students on Campus 2, for a total of 1,236 participants (711 from Campus 1 and 535 from Campus 2).¹ Although both were medium-sized (12,000-15,000 students) New England campuses, Campus 1 is rural and residential, whereas Campus 2 is urban, a mix of residential and commuter, and had greater racial and ethnic diversity. Student recruitment was via flyers posted on recruitment tables at dining halls and academic buildings, announcements in classes and on student online bulletin boards, weekly meetings with Resident Advisors, and e-mail announcements. When qualified participants (between ages 18 and 24, in their first semester ever in college, and not trained as an advocate for a rape crisis center or domestic violence agency) expressed interest in the study, they were randomly assigned to the control group (n = 482) or program group (n = 466; participants unable to attend the program were given the opportunity to join the control group).

Nine hundred forty-eight first-year students took the pretest (76.7% of the recruited sample). Five hundred fifty were from Campus 1, and 398 were from Campus 2. The sample was evenly distributed between men and women (51.5% male, 47.8% female, while 3 participants identified as transgender), 85.2% of the sample identified as White, 73.2% reported father’s education of at least some college (20% of the sample reported their fathers had graduate school or professional degrees), and most were living on campus (89.6%). Given the study qualifications for inclusion, the mean age of participants was 18.2 years (SD = .49).

At pretest, there were no significant differences between control and program group participants on gender, race, father’s education, whether the participant was living on campus, or on any of the attitude or behavior measures that were outcomes for this study. There were a number of differences between the campuses, which was by design when we selected a second campus for translation of the program. Campus 2 had a higher proportion of participants who were students of color (24.9% vs. 10.5% on Campus 1) and commuter participants (24.9% vs. 1.6% for Campus 1), and a lower proportion with fathers having a college degree (61% compared with 81.8% for Campus 1). The samples did not differ significantly by gender.

Of the 948 participants who filled out pretests, 607 took the posttest and had posttests that could be matched to pretest data (52 surveys could not be matched to pretest data), for a retention rate of 64%. At posttest, 47.4% of the sample was male, 88.3% of participants lived on campus, and 88.2% of the sample was White. A MANOVA analysis revealed those retained at posttest did not differ on pretest bystander behavior or attitudes when compared with those who dropped out of the study. There were no differences between participants who continued in the study and those who did not on race or commuter status. Women and those who had a father without a college degree were more likely to be retained to posttest. The only significant difference between the experimental groups was that the control group had a slightly higher percentage of commuters.

Three hundred forty-six participants took the 12-month follow-up survey and could be matched to pretest and posttest data (44 surveys were unmatched), for a retention rate of 36.5%. At 12-month follow-up, 53.5% of the sample was female, 72.8% had fathers with at least a college degree, 88.9% lived on campus, and 88.8% identified as White. Participants who completed the 12-month survey were more likely to be women. There were no significant differences between those retained and not on pretest demographics of race, father’s education, and commuter status at pretest. There were significant differences on pretest attitudes and behaviors, with 12-month follow-up participants higher on attraction to sexual violence and rape myth acceptance, lower on pretest bystander behaviors, and less likely to correctly identify the prevalence of sexual violence at pretest than participants who did not take the final survey.

The final sample is composed of participants who completed pretests and posttests that could be matched together. Because we examined gender as a covariate, the three transgender participants were removed from further analyses below due to insufficient sample size.²

Procedures

The Bringing in the Bystander in-person program is similar to and builds on other bystander-focused gender-based violence prevention programs, such as Mentors in Violence Prevention (Katz, 1994). The program teaches participants how to play important prevention roles as bystanders when they observe risky situations before and during acts of sexual violence (e.g., observing a very intoxicated person being led into a bedroom at a party by a group of people), and after if a friend discloses that she or he has been a victim. Like other programs (e.g., Gidycz et al., 2011; Katz et al., 2011), this involves addressing individual and community attitudes and norms surrounding sexual violence, such as dispelling myths about sexual violence, as well as attitudes about the costs and benefits of engaging in helpful bystander behavior. The program content covers the prevalence, causes, and consequences of sexual violence, and building empathy for victims. The full continuum of sexual violence is discussed (from inappropriate jokes and touching to legally defined rape), including the ways in which dating violence is often a context for sexual violence on campus and that most perpetrators are known to victims. Curriculum content addresses the gendered nature of sexual violence, with victims being disproportionately women victimized by men, and gender as a key facet of understanding this community problem. The program also includes discussion of men as victims of sexual violence in same-gender relationships. Active learning exercises (e.g., role plays) help participants build skills and think about how to intervene safely and be a supportive ally to survivors (e.g., Gidycz et al., 2011).

The Bringing in the Bystander program moves away from the limitations in previous work of approaching men as potential perpetrators and women as potential victims. Instead, it emphasizes the role everyone plays in preventing sexual violence as a member of the community, which is underscored by the training being administered by pairs of one male and one female co-facilitator. The program is administered to single-gender groups (e.g., Katz, 1994) in recognition that men and women experience sexual violence prevention messages differently (e.g., Heppner et al., 1995), so that engagement and discussion may be more fruitful and more likely to impact attitudes and behavior in a single-gender setting. (For more detail on the program, including more specifics on how it compares to other bystander-based sexual violence prevention programs, see Banyard, Eckstein, & Moynihan, 2010; Banyard, Moynihan, & Crossman, 2009; Banyard, Plante, & Moynihan, 2004.)

Bringing in the Bystander was originally designed for Campus 1. For translation to Campus 2, presentation materials were changed to fit a new campus in a different state, such as support resources (e.g., information for the local rape crisis center) and relevant sexual assault statistics, policies, and laws. During the co-facilitator training, a primary focus was on exploring what forms bystander attitudes and behavior may take in the context of their campus, and how that should shape their presentation. As part of the program, co-facilitators and participants share their experiences as bystanders, engage in discussion around bystanding and sexual violence, and brainstorm ways they could intervene in a situation on their campus. This provided opportunity for co-facilitators at Campus 2 to provide relevant examples and lead discussion around issues pertinent to Campus 2. For example, Campus 2 is in an urban community, closer than Campus 1 to areas with gang activity and high crime rates, so what it means to intervene safely as a bystander has to address the concerns participants have around those issues. Because co-facilitators were undergraduate and graduate students from Campus 2, they were able to understand and address those concerns in ways program participants saw as legitimate. Campus 2 has no campus-based sexual violence programming or support services, so staff from local rape crisis centers went through the training with the co-facilitators. This gave the co-facilitators on Campus 2 the opportunity to understand the larger community context for sexual violence in their area, which helped shape how they facilitated the program. For example, if one recommendation for how to help survivors is to accompany them to a service provider, that may be viewed differently by students if it is an on-campus versus off-campus resource.

The program duration was 4½ hours, split over two sessions. Participants completed surveys prior to implementation of the program (pretest), and three more times: approximately 2 weeks (posttest), 5 months, and 1 year after program implementation. (The 5-month follow-up focused on different issues, and findings from that data are presented elsewhere—Banyard, Potter, et al., under review). Following the first cohort on Campus 1 returning to take the surveys in person in a classroom, we determined students preferred accessing surveys online, and all other surveys were web administered. Participants received e-mails with a link to each survey. They were paid US$10 for each survey, and those who took all four surveys were entered in a raffle to win a US$150 Best Buy gift card. In addition, on Campus 1, cohort 2 and 3 participants were offered extra incentives to complete 5- and 12-month surveys. They were given the chance to win a lottery of several US$50 gift cards. On Campus 2, this additional incentive was offered to cohort 2 participants at the 5-month survey. Participants who completed the program were paid US$40 for their time in the program.

Four months after program implementation, both campuses were exposed to the Know Your Power™ Bystander Social Marketing Campaign (SMC) for 6 weeks (for more detail on the SMC, see Potter, 2012). This campaign is a series of images portraying “typical” college scenes explicitly modeling prosocial bystander behavior in the prevention of sexual and relationship violence. Images were used on various mediums, including bookmarks, table tents, and posters that were displayed in residence halls, dining halls, student centers, libraries, and academic buildings. On Campus 1, images were also used for bus wraps and on computer splash screens. Images were tailored to each campus to reflect the racial and ethnic composition of students and social situations typical of the campus (for the importance of images specific to the target audience, see Potter, Moynihan, & Stapleton, 2011; Potter & Stapleton, 2011). Images were tailored to Campus 2 through a multi-stage process, similar to the one used on Campus 1 to develop the original images (Potter & Stapleton, 2011). First, focus group participants provided feedback on images without the image dialogue. Images were then revised based on student feedback. Second, dialogue boxes for the revised images were edited by focus group participants to best reflect the language and terminology of students on that campus. As a part of this campaign, first-year students were provided with promotional items, such as water bottles, flash drives, bookmarks, and backsacks that included the campaign slogan and website. This means that by the 12-month follow-up, the program group was exposed to the program and SMC, and the control group was only exposed to the SMC. Past research on SMCs on this topic have found them to be effective in increasing college students’ awareness of sexual violence as a problem and one they share responsibility for solving (Potter, 2012; Potter et al., 2011).

Measures

Social Desirability

This scale consisted of 16 of the original 17 items, like Stöber (2001). Participants scored “1” for true or “0” for false; a number of items were reverse scored. An example of one item in the scale is as follows: “In traffic I am always polite and considerate of others.” Cronbach’s alpha for the current sample was low (α = .63). Given that social desirability is only being used as a control variable, we kept this measure in the study. Scores were computed by taking the mean across items. Higher scores indicate greater socially desirable responding. The mean for the current sample at pretest was .53, SD = .19, range = 0 to 1.

Illinois Rape Myth Acceptance Scale—Short form

The original 20-item scale (Payne, Lonsway, & Fitzgerald, 1999) was developed to assess participants’ endorsement of common myths about sexual assault. Our surveys used 17 of the items (we excluded three filler items). However, item analysis indicated that all but three of the items had a floor effect, with most participants not endorsing the rape myth. For that reason, the items from the scale used in our analysis were Rape Myth 1: “If a woman is raped while she is drunk, she is at least somewhat responsible for letting things get out of control” (M = 2.13, SD = 1.13); Rape Myth 7: “Rape accusations are often used as a way of getting back at men” (M = 2.06, SD = 1.00); and Rape Myth 15: “Men don’t usually intend to force sex on a woman, but sometimes they get too sexually carried away” (M = 2.22, SD = 1.09). Participants indicated on a 5-point scale (range 1-5) the extent to which they agreed with each item.

Attraction to Sexual Violence Scale

This 6-item scale (Malamuth, 1989) has been used to measure willingness to engage in sexual aggression (O’Donohue, Yeater, & Fanetti, 2003). In this study, an item analysis indicated that only two items did not have a floor effect. For that reason, we only used those two items in our analysis: “What percent of men would find forcing someone to do something sexual arousing” (M = 3.65, SD = 1.94) and “What percent of women would find forcing someone to do something sexual arousing” (M = 2.52, SD = 1.75). Participants selected the percentage they chose for the answer (e.g., 1 = 0%, 2 = 10%, to 11 = 100%).

Readiness to Help Scale

This 33-item scale (Banyard, Moynihan, Cares, & Warner, 2014) is based on Prochaska and DiClemente’s Transtheoretical Model (Grimley, Prochaska, Velicer, Blais, & DiClemente, 1994). Previous work has highlighted this model as useful for conceptualizing bystanders and prevention (Banyard, Eckstein, & Moynihan, 2010). As reported in Banyard, Moynihan, et al. (2014), a principal components factor analysis with varimax rotation was computed for the original items to assess attitudes toward and actions of engaging in prevention of sexual abuse, intimate partner abuse, and stalking. This study uses the two factors that correspond to attitudes about readiness to help (referred to, in some previous research, as readiness for change): precontemplation and contemplation. Examples of items in these subscales include “I don’t think stalking is a problem on this campus” (precontemplation) and “Sometimes I think I should learn more about stalking” (contemplation). Participants responded on a 5-point scale (1 = not at all true and 5 = very much true) to indicate how much each of the statements was true of them. The precontemplation factor included 12 items (11.05% of variance) and the contemplation factor (17.02% of the variance) included 9 items. Cronbach’s alpha for the precontemplation scale was .87 and for contemplation was .91. The two subscale scores were created by taking the mean across items. Descriptive statistics are as follows: precontemplation (M = 2.55, SD = .69, range = 1-4.92) and contemplation (M = 2.78, SD = .83, range = 1-5).

Bystander Efficacy Scale

This scale (Banyard, 2008), used previously to evaluate prevention efficacy, includes 18 statements. A participant rates her or his confidence to perform various bystander behaviors on a scale from 0 (can’t do) to 100 (very certain that he or she can do). An example is, “Ask a friend if they need to be walked home from a party.” The mean across all 18 items becomes the total score used. Cronbach’s alpha for this sample was .93. Previous research with different samples of participants has established the validity of this measure (Banyard, 2008). For the current sample, M = 72.60, SD = 17.64, range = 0 to 100.

Intent to Help a Friend Scale

This scale (Banyard, Moynihan, et al., 2014) includes 38 items that assess participants’ self-reported likelihood to engage in certain helpful bystander behaviors with someone they know. Each participant rated his or her likelihood to perform the behaviors using a 5-point scale (1 = not at all likely to 5 = extremely likely). For example, “I talk to people I know to make sure we don’t leave an intoxicated friend behind at a party.” Higher scores indicate that the participant feels more likely to perform the behavior listed. The mean at pretest was 3.79, SD = .72, range = 1 to 5, α = .97.

Intent to Help a Stranger Scale

This scale (Banyard, Moynihan, et al., 2014) includes 41 items that assess participants’ self-reported likelihood to engage in certain helpful bystander behaviors with someone they do not know. Each participant rated his or her likelihood to perform the behaviors using a 5-point scale (1 = not at all likely to 5 = extremely likely). For example, “I express disagreement with someone I don’t know who says having sex with someone who is passed out or very intoxicated is okay.” Higher scores indicate a participant feels more likely to perform the behavior. At pretest, the mean for the current sample was 3.09, SD = .87, range = 1 to 5, α = 98.

Knowledge items (2nd edition)

From the original set of knowledge questions (Banyard, Moynihan, & Plante, 2007), we used the four core items that were general enough to apply to both campuses. The items were as follows: “What percentage of women experience sexual assault in their lifetime?” “What percentage of men experience sexual assault in their lifetime?” “What percentage of reported sexual assaults are false reports?” “The most commonly used drug in sexual assault is ________.” The first three of these items were asked using a multiple-choice format. The fourth included a blank space into which participants typed their answer. Responses were scored “0” for an incorrect response or “1” for a correct response. At pretest, small segments of the sample with valid pretest and posttest scores got each item correct (Item 1: 22.4%, Item 2: 5.6%, Item 3: 13.6%, and Item 4: 35.2%). Cronbach’s alpha for these items together was quite low, so each was considered as a separate outcome.

Data Analysis

Repeated measures MANOVA using Generalized Linear Model(GLM) in SPSS 19.0 was used to examine changes in bystander attitudes from pretest to posttest and to 12-month surveys, with campus and gender as covariates. Follow-up paired-sample t tests were used to examine specific changes by group for individual attitudes for outcomes identified as significant at the univariate level in the repeated measures analysis. The analyses were run on the sample composed of participants who completed pretests and posttests that could be matched together.

Results

Table 1 presents pretest scores for all outcome measures by gender and by campus. Independent-samples t tests show many outcomes on which women differed from men at pretest and several outcomes on which the campuses differed. At pretest, women were higher on bystander efficacy, contemplation, and intent to help friends and strangers. Women believed a higher percentage of men would find forcing someone to do something sexual arousing and were more knowledgeable about the percentage of women who experience sexual assault in a lifetime. Men had higher endorsement of Rape Myths 7 and 15, higher precontemplation scores, and were more knowledgeable that alcohol is the most common drug used in sexual assault. They believed a higher percentage of women would find forcing someone to do something sexual arousing. Campus 1 reported lower contemplation scores, higher intent to help a friend, and higher knowledge of the most common drug used in sexual assault and the percentage of sexual assault reports that are false. Campus 2 believed a higher percentage of women and of men would find forcing someone to do something sexual arousing.

Table 1.

Pretest Scores and Independent-Samples t Tests for Group Differences by Gender and by Campus.

	Gender		Campus
	Female (n = 430-453)	Male (n = 467-485)	Campus 1 (n = 536-546)	Campus 2 (n = 362-392)
Variable name	M (SD)	M (SD)	M (SD)	M (SD)
Rape Myth 1	2.06 (1.07)	2.19 (1.18)	2.17 (1.12)	2.08 (1.14)
Rape Myth 7	1.88 (0.93)	2.22 (1.03)***	2.09 (1.00)	2.01 (1.00)
Rape Myth 15	2.06 (1.04)	2.35 (1.11)***	2.26 (1.08)	2.14 (1.09)
% men sexually aroused	3.78 (1.94)	3.52 (1.92)*	3.53 (1.85)	3.80 (2.04)*
% women sexually aroused	2.30 (1.61)	2.71 (1.84)***	2.34 (1.52)	2.75 (2.00)**
Precontemplation	2.42 (0.65)	2.67 (0.71)***	2.52 (0.67)	2.59 (0.73)
Contemplation	2.97 (0.80)	2.60 (0.83)***	2.71 (0.81)	2.87 (0.86)**
Efficacy	75.38 (15.89)	69.86 (18.78)***	73.34 (17.25)	71.37 (18.16)
Intent to help friend	3.99 (0.64)	3.60 (0.73)***	3.82 (0.71)	3.73 (0.73)*
Intent to help stranger	3.15 (0.87)	3.03 (0.86)*	3.05 (0.87)	3.14 (0.87)
Lifetime sexual assault for women	0.26 (0.44)	0.19 (0.40)*	0.21 (0.41)	0.24 (0.43)
Lifetime sexual assault for men	0.05 (0.20)	0.06 (0.24)	0.06 (0.24)	0.05 (0.22)
% false sexual assault reports	0.12 (0.32)	0.15 (0.36)	0.17 (0.38)	0.09 (0.28)***
Date rape drug	0.25 (0.43)	0.44 (0.50)***	0.41 (0.49)	0.27 (0.44)***

p < .05. **p < .01. ***p < .001.

A repeated measures MANOVA was computed for 10 attitude outcomes (3 rape myth items, 2 attraction to sexual violence items, bystander efficacy, precontemplation, contemplation, intent to help friends, and intent to help strangers) for three time points (pretest, posttest, and 12-month). The main contrast was group (program vs. control group) and time. Gender and campus were covariates with social desirability as a control variable. Between subjects, there was a main effect of gender F(10, 254) = 11.10, p < .001, Wilks’s Lambda = .70, $η_{p}^{2}$ = .30; campus F(10, 254) = 4.09, p < .001, Wilks’s Lambda = .86, $η_{p}^{2}$ = .14; group F(10, 254) = 3.37, p < .001, Wilks’s Lambda = .88, $η_{p}^{2}$ = .12; and social desirability F(10, 254) = 3.05, p < .01, Wilks’s Lambda = .89, $η_{p}^{2}$ = .11. There were no significant between-subjects interaction effects (group × campus, group × gender, campus × gender, group × campus × gender). Within subjects, there was a main effect of time F(20, 244) = 2.13, p < .01, Wilks’s Lambda = .85, $η_{p}^{2}$ = .15. There was no time by gender, nor time by social desirability interactions. There was a significant time by group interaction F(20, 244) = 4.16, p < .001, Wilks’s Lambda = .75, $η_{p}^{2}$ = .25, a large effect. There was a significant time by campus interaction F(20, 244) = 1.79, p < .05, Wilks’s Lambda = .87, $η_{p}^{2}$ = .13, a large effect. There was a significant time by campus by gender interaction F(20, 244) = 1.64, p < .05, Wilks’s Lambda = .88, $η_{p}^{2}$ = .12, a medium to large effect. There were no significant three-way interactions between time by group by gender or time by group by campus. However, there was a significant time by group by campus by gender four-way interaction F(20, 244) = 1.98, p < .01, Wilks’s Lambda = .86, $η_{p}^{2}$ = .14, a medium to large effect. In sum, there were differences in attitudes between the treatment and the control group, between the campuses, and between men and women. Furthermore, there were differences over time in attitudes that depended on which group, gender, and campus a participant was in.

Univariate analyses using Greenhouse–Geisser as an indicator revealed significant time by group interactions for six outcomes: Rape Myth 1, precontemplation, contemplation, efficacy, intent to help friends, and intent to help strangers. Table 2 presents follow-up paired-sample t tests examining differences between pretest and posttest scores (a comparison between participants who participated in the in-person program and participants in the control group) and between pretest and 12-month scores (which compared participants who by 12 months had only seen the SMC and those who had seen the SMC and participated in the program). These t tests show attitude change for the program group compared with the control group on most outcomes, and persistence in this attitude change to 12 months.

Table 2.

Paired-Sample t Tests of Bystander Attitudes by Prevention Exposure Group.

	Pre to Posttest				Pre to 12 month
	Control (n = 294-301)		Program (n = 289-293)		SMC-only (n = 180-188)		Program and SMC (n = 154-158)
	Pre	Post	Pre	Post	Pre	12 Month	Pre	12 Month
Variable name	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)
Rape Myth 1	2.15 (1.13)	1.95 (1.06)**	2.07 (1.08)	1.68 (1.00)***	2.34 (1.16)	1.74 (0.95)***	2.15 (1.12)	1.67 (1.02)***
Efficacy	71.94 (18.49)	69.57 (20.99)*	72.54 (16.62)	77.45 (18.62)***	71.96 (16.94)	70.59 (21.56)	74.02 (16.15)	79.05 (18.53)**
Precontemplation	2.57 (0.69)	2.49 (0.74)	2.53 (0.66)	2.17 (0.70)***	2.60 (0.66)	2.51 (0.77)	2.59 (0.66)	2.40 (0.78)**
Contemplation	2.83 (0.84)	2.78 (0.83)	2.77 (0.82)	3.01 (0.87)***	2.80 (0.81)	2.72 (0.85)	2.79 (0.82)	2.93 (0.92)*
Intent to help friend	3.77 (0.73)	3.60 (0.92)***	3.82 (0.70)	3.95 (0.86)**	3.80 (0.63)	3.52 (0.97)***	3.92 (0.67)	3.84 (0.89)
Intent to help stranger	3.05 (0.87)	2.99 (0.89)	3.10 (0.88)	3.38 (0.92)***	3.05 (0.85)	2.97 (0.94)	3.13 (0.91)	3.23 (0.93)

Note. SMC = Social Marketing Campaign.

p < .05. **p < .01. ***p < .001.

Univarate analyses using Greenhouse–Geisser as an indicator revealed five outcomes for which the four-way interaction between time, group, campus, and gender was significant (using p < .10 as a significance level, given the sample size and nature of the analyses). These outcomes were scores on Rape Myth 15, precontemplation, efficacy, intent to help friends, and intent to help strangers. Tables 3 and 4 present the results of paired-sample t tests for these outcomes separately by campus, gender, and group. (Table 3 examines differences from pre to posttest and Table 4 from pretest to 12-month follow-up.) The results show not only some testing effects (rape myths decline over time for most groups) but also intervention group effects (attitudes change over time for program group participants but not for the control/SMC-only group). They also show that while men and women on Campus 1 showed some persistent attitude change for 12 months, only women at Campus 2 showed such positive effects as a result of participating in the program. Overall, both groups, those who were in the control group at posttest and who by 12 months had only seen the SMC, and those who participated in the program showed reduction in rape myths that persisted to 1 year. Participants who attended the program showed increases in bystander efficacy for 12 months following the pretest, lower scores on precontemplation, and higher scores on contemplation and intent to help (suggesting progression to higher stages of readiness to help in the group that also received the prevention program), although these effects were not seen for men on Campus 2.

Table 3.

Paired-Sample t Tests for Group Differences in Attitudes Separately by Campus and Gender From Pretest to Posttest.

	Campus 1
	Men				Women
	Control (n = 78-80)		Program (n = 87-88)		Control (n = 74-76)		Program (n = 81-82)
	Pre	Post	Pre	Post	Pre	Post	Pre	Post
Variable name	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)
Rape Myth 1	2.10 (1.12)	2.18 (1.12)	2.27 (1.15)	1.81*** (1.10)	2.12 (1.12)	1.71** (0.91)	2.02 (0.98)	1.59** (0.93)
Precontemplation	2.62 (0.62)	2.49 (0.76)	2.74 (0.65)	2.15*** (0.65)	2.41 (0.59)	2.29 (0.64)	2.36 (0.61)	2.02*** (0.67)
Efficacy	70.57 (18.00)	66.17* (24.29)	69.83 (17.63)	74.49* (20.31)	75.22 (17.49)	75.25 (17.11)	75.26 (14.15)	81.86*** (17.53)
Intent to help friends	3.59 (0.70)	3.44 (0.94)	3.65 (0.69)	3.82* (0.87)	4.05 (0.59)	3.92 (0.83)	4.04 (0.66)	4.31*** (0.67)
Intent to help strangers	3.03 (0.88)	3.00 (0.90)	2.87 (0.84)	3.21** (0.86)	2.99 (0.85)	2.99 (0.91)	3.12 (0.86)	3.57*** (0.87)
	Campus 2
	Men				Women
	Control (n = 61-63)		Program (n = 52)		Control (n = 81-82)		Program (n = 69-72)
	Pre	Post	Pre	Post	Pre	Post	Pre	Post
Variable name	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)
Rape Myth 1	2.19 (1.18)	2.13 (1.16)	2.12 (1.26)	2.00 (1.09)	2.18 (1.11)	1.80** (1.01)	1.85 (0.93)	1.40** (0.80)
Precontemplation	2.79 (0.82)	2.59 (0.77)	2.72 (0.69)	2.55 (0.78)	2.50 (0.68)	2.58 (0.77)	2.34 (0.61)	2.07** (0.65)
Efficacy	65.71 (21.09)	63.47 (21.59)	71.10 (18.40)	71.97 (20.66)	74.94 (16.68)	72.28 (18.69)	73.80 (16.31)	80.05** (14.28)
Intent to help friends	3.44 (0.85)	3.23* (0.91)	3.52 (0.72)	3.55 (0.89)	3.95 (0.65)	3.71** (0.86)	4.00 (0.62)	4.01 (0.86)
Intent to help strangers	3.06 (0.84)	2.87* (0.86)	3.13 (0.91)	3.15 (0.98)	3.13 (0.89)	3.07 (0.87)	3.35 (0.86)	3.54* (0.93)

p < .05. **p < .01. ***p < .001.

Table 4.

Paired-Sample t Tests for Group Differences in Attitudes Separately by Campus and Gender From Pretest to 12 Month.

	Campus 1
	Men				Women
	SMC-only (n = 55-57)		Program + SMC (n = 54-56)		SMC-only (n = 55-57)		Program + SMC (n = 48)
	Pre	12 Month	Pre	12 Month	Pre	12 Month	Pre	12 Month
Variable name	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)
Rape Myth 1	2.33 (1.20)	1.79** (0.96)	2.34 (1.23)	1.95* (1.15)	2.21 (1.05)	1.68** (1.00)	2.02 (0.98)	1.54** (0.92)
Precontemplation	2.76 (0.61)	2.76 (0.75)	2.73 (0.64)	2.42** (0.72)	2.40 (0.63)	2.19 (0.73)	2.46 (0.65)	2.30 (0.68)
Efficacy	69.99 (15.61)	65.24 (20.98)	68.31 (17.97)	75.60** (18.14)	74.73 (16.71)	77.19 (18.21)	77.29 (13.13)	82.60* (14.02)
Intent to help friends	3.61 (0.59)	3.29** (0.92)	3.59 (0.73)	3.67 (0.76)	4.04 (0.55)	3.97 (0.85)	4.21 (0.45)	4.19 (0.70)
Intent to help strangers	3.01 (0.78)	2.80 (0.91)	2.84 (0.88)	3.02 (0.72)	2.96 (0.88)	3.17 (0.99)	3.22 (0.83)	3.45* (0.94)
	Campus 2
	Men				Women
	SMC-only (n = 27-30)		Program + SMC (n = 15-17)		SMC-only (n = 43-44)		Program + SMC (n = 36-37)
	Pre	12 Month	Pre	12 Month	Pre	12 Month	Pre	12 Month
Variable name	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)
Rape Myth 1	2.73 (1.26)	2.17* (1.02)	2.29 (1.21)	2.12 (1.22)	2.23 (1.14)	1.45*** (0.73)	1.97 (1.09)	1.22*** (0.53)
Precontemplation	2.90 (0.70)	2.77 (0.78)	2.99 (0.59)	3.13 (1.03)	2.46 (0.64)	2.42 (0.67)	2.40 (0.62)	2.20 (0.72)
Efficacy	65.11 (20.55)	65.13 (23.70)	79.00 (17.28)	65.21* (29.39)	75.17 (15.28)	72.27 (22.95)	76.09 (14.62)	85.07*** (15.21)
Intent to help friends	3.46 (0.66)	3.06* (1.07)	3.65 (0.73)	3.10 (1.03)	3.93 (0.59)	3.51** (0.88)	4.13 (0.56)	3.96 (1.01)
Intent to help strangers	3.25 (0.81)	3.05 (0.90)	3.21 (1.12)	2.61 (0.92)	3.10 (0.92)	2.88 (0.90)	3.43 (0.86)	3.56 (0.99)

Note. SMC = Social Marketing Campaign.

p < .05. **p < .01. ***p < .001.

A second repeated measures MANOVA was computed across three time points (pretest, posttest, and 12-month follow-up) for the four knowledge items. Between subjects, there was a significant effect of campus F(4, 142) = 4.09, p < .01, Wilks’s Lambda = .90, $η_{p}^{2}$ = .10, confirming (above and in Table 1) that there were some general differences by campus. There was also a significant effect of group F(4, 142) = 2.46, p < .05, Wilks’s Lambda = .94, $η_{p}^{2}$ = .07. There were no other significant between-subjects effects. Within subjects, the only significant effect was a significant time by group by campus interaction F(8, 138) = 2.09, p < .05, Wilks’s Lambda = .89, $η_{p}^{2}$ = .11. This suggests that the effect of the program over time differed by campus. There was no significant four-way time by group by campus by gender interaction. Univariate analyses using Greenhouse–Geisser as an indicator revealed that the three-way interaction between time, group, and campus was significant for the two knowledge questions related to the percentage of women and of men that experience sexual assault in a lifetime (using p < .10 as a significance level, given the sample size and nature of the analyses). Table 5 presents the results of paired-sample t tests to investigate these effects for the knowledge questions. It suggests that those in the program at Campus 1 increased their knowledge about lifetime sexual assault and that lasted until at least the 12-month follow-up.

Table 5.

Group by Campus Differences in Knowledge Using Paired-Sample t Tests.

	Campus 1				Campus 2
	Control (n = 156)		Program (n = 169-171)		Control (n = 141)		Program (n = 122)
Pre to Posttest	Pre	Post	Pre	Post	Pre	Post	Pre	Post
Variable name	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)
Lifetime sexual assault for women	.26 (.44)	.24 (.43)	.20 (.40)	.55*** (.50)	.21 (.41)	.31** (.46)	.25 (.43)	.35* (.48)
Lifetime assault for men	.06 (.23)	.06 (.23)	.05 (.21)	.15*** (.36)	.03 (.17)	.06 (.25)	.07 (.25)	.10 (.30)
	Campus 1				Campus 2
	SMC-only (n = 111-113)		Program + SMC (n = 103)		SMC-only (n = 72-73)		Program + SMC (n = 52-54)
Pre to 12 month	Pre	12 Month	Pre	12 Month	Pre	12 Month	Pre	12 Month
Variable name	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)
Lifetime sexual assault for women	.27 (.44)	.29 (.46)	.20 (.40)	.47*** (.50)	.30 (.46)	.32 (.47)	.27 (.45)	.29 (.46)
Lifetime sexual assault for men	.05 (.21)	.08 (.27)	.04 (.19)	.19*** (.40)	.00 (.00)	.15*** (.36)	.06 (.23)	.15* (.36)

Note. SMC = Social Marketing Campaign.

p < .05. **p < .01. ***p < .001.

Discussion

This study utilized an experimental design to evaluate the effectiveness of translating the Bringing in the Bystander in-person sexual violence prevention program to a new campus community. Overall, it appears the program, with minimal modifications, worked on a new campus, and many of the initial changes in attitude lasted at least as long as 12 months post program. This evaluation on two different campuses highlights that there are important distinctions to be made in terms of who it works for and how—in this case, how it works differently for women and men, how it works differently between campuses, and how it works differently by gender between campuses. This is an important lesson for the prevention field, which often wants one-time, low-resource, prevention tools that can easily apply to many communities. Social change efforts for problems as complex as sexual and relationship violence will require more. Whether seeking to enhance efforts on one campus, or translate efforts to another community, future goals for prevention should be for messages that are presented in an ongoing way, educational efforts that recur throughout a student’s time on campus, and are both institutionalized to promote longevity and flexible enough to adapt to changing needs of any given community. For example, prevention efforts are likely to be more effective if going through the program as a first-year student is the first step, with additional programming being provided each year as students continue their college careers.

As past research and many prevention programs acknowledge, women’s and men’s experiences with and attitudes toward sexual violence are very different, and this may affect how sexual violence prevention efforts work for them (L. A. Anderson & Whiston, 2005; Burn, 2009). This was underscored in this study by the differences in attitudes by gender at the outset, many of which persisted throughout the study. This may be a reflection of different gender realities. From a young age, women are socialized to think about the risk of sexual assault, so by the start of their first year of college (when our study participants took the pretest), they see sexual violence as a problem (hence lower precontemplation scores than men) and have already thought about what to do about it (reflected in higher contemplation scores than men; Burn, 2009). So, if the program engenders the same amount of change for men and women, in the end that translates to a persistence of gender differences, which is what we found on both campuses. Although we saw desired change in both groups and it persisted over time, at 12-month follow-up, men scored worse than women on many outcomes.

On Campus 1, both genders experienced significant attitude change associated with participation in the prevention program. However, on Campus 2, the program appears to have been more effective for women than men. While for women on Campus 2, the patterns of attitude change were similar to those on Campus 1, for men on Campus 2, the program resulted in limited attitude change initially (from pretest to posttest) and very little sustained attitude change (from pretest to 12-month follow-up). This raises the following questions: What was different for men at Campus 2? Was there something different about the program applicability to men on Campus 2, or about the climate on Campus 2, that impacted men and women differently? The answer may lie in the campus gender ratio. Unlike most colleges in the United States, Campus 2 is a heavily male campus. At the time of the study, the student body was almost 60% male, with the first-year class being over 60% male. This impacts the context for men and women in classrooms, residence halls, parties, and other spaces. In fact, during their training, the Campus 2 co-facilitators discussed this and how it differs from other campuses, and this issue was raised by participants during the program. While some male students commented that this imbalanced ratio gave women power in dating and social relations, it also may create a campus social context that is heavily determined by male peer norms (Schwartz & DeKeseredy, 1997), where attitudes and beliefs about women and about sexual assault may be more stereotyped and run counter to what is espoused by the Bringing in the Bystander program. If so, that may explain the lack of persistence of attitude changes because the messages of the program to males may have been largely overshadowed by the larger campus culture.

There are a number of other differences between campuses that may undermine the effectiveness of the program for men at Campus 2. While women may be “primed” to positively receive bystander programming on sexual assault, men may not be (as evidenced by the lower initial scores at pretest). In such cases, men may be more influenced by the broader community climate. Campus 1 has a long tradition of programming on and attention to issues of violence against women, including a campus-based center offering intervention and prevention services. Campus 2 has no such tradition or services. Thus, the program on Campus 2 was introduced in a community context that had less of a foundation on which to begin attitude change.

Campus 1 and Campus 2 also differ in their locations. Campus 2 is spread across an urban area, rather than being a geographically contiguous campus. The city has a high crime rate and some gang activity, of which students are aware. When men engage as active bystanders, they are much more likely than women to physically intervene (Burn, 2009). However, given how permeable the campus is to non-members and the potential perceived by students for those non-members to be dangerous, this may explain the lack of intent to help strangers for males (and females) on Campus 2. The options men on that campus may readily consider for intervening as a bystander are also ones that they perceive as more dangerous to their personal safety.

The differences between Campus 1 and Campus 2 were a key component in testing how Bringing in the Bystander would translate to a new campus. This was an initial attempt at programming on Campus 2, and it may be that more adaptation of the program to the realities of Campus 2 was needed, with particular attention to how that might differ by gender. The translation process for this study was purposefully circumscribed and focused mainly on changing content related to state laws and campus policies and having co-facilitators provide personal examples that were reflective of their campus. Future research is needed to more systematically unpack different components of the educational program to examine, for example, which components of the program resonated on each campus and which need a more rigorous process of modification. Such analysis was beyond the aims or the scope of what could be answered with the data collected for this study but highlights a key direction for future work. In relation to the current findings, it is also possible that the co-facilitators on Campus 2, which was a group of only seven, somehow did not connect as effectively with the male participants as they did with the females. These are issues to be explored in follow-up interviews with program participants. Also, follow-up interviews on both campuses may uncover areas for future program development and improvement that could increase program effects on attitude change on Campus 1 and Campus 2. What parts of the program resonated on both campuses, and with which groups? When there were differences in program effectiveness, was it because of the program design itself or an issue with implementation? This may help inform future programming efforts on both Campus 1 and Campus 2, as well as on other campuses where Bringing in the Bystander is being adapted or is serving as the foundation for more recently developed programs, including at Boston College, Clark University, Dartmouth College, and the University of Windsor in Ontario. Once there are answers to these questions, are there ways to find the likely answer to these questions on new campuses before implementing the program? As campuses already implementing the program change over time, how can they build a feedback mechanism for revisions into their process? This illustrates the powerful potential for prevention of working simultaneously between different communities at one time to create feedback loops that can strengthen all efforts to end sexual and relationship violence on campuses.

Limitations

This study provides promising evidence that translation of the Bringing in the Bystander in-person sexual violence prevention program is possible. However, the study represents the evaluation of an initial effort at translation to one new campus with first-year students. Although there are good reasons to target first-year students (see above), it is important to realize where they are developmentally. As they are just launching into their college careers, a time when they are concerned with making friends and building community, one must ask how much of prevention messages that ask them to step up and stand out might they really be prepared to do? Furthermore, is it reasonable to expect to alter their attitudes long-term with several hours of programming when they have not yet formed their social groups and fully encountered the college social scene? In other words, we may be providing them with examples about a context they have not yet experienced and to which they cannot yet apply the examples discussed during the program. In the language of Prochaska and DiClemente’s Transtheoretical Model (Grimley et al., 1994), they may still be in the precontemplation stage of readiness to help, and all it is realistic to expect is to move them further along in that stage, so they start seeing sexual violence as a problem and a problem on their campus. Once they embrace the notion that it is a local problem, they then become more motivated to build new skills and practice them, suggesting the need for ongoing programming.

Focusing on a first-year sample may also have exacerbated the level of attrition. Students were recruited and participated in the program before they settled into their college routine, so they committed to at least a yearlong process while having little idea how that would fit into their lives. Study retention rates may also have been impacted by first-year student university retention rates and students who did not return for a second or third semester. Longitudinal research is typically plagued with difficulty in locating participants over time and keeping them in the study. By the end of the study (approximately 13 months after the pretest), the retention rate was 36.5%, although there were few differences in pretest demographics or bystander attitudes between those who dropped out and those who were retained. The retention rate means the size of some groups by the end of the study limited statistical power. In addition to the issue of working with first-year students, a number of other factors may have contributed to the retention rate. One was the move from in-person surveys to online surveys. While this theoretically may increase participation rates because it is easy, e-mail is also easy to ignore. This may be an important issue as due to the ease of fielding web-based surveys, students on many campuses are being inundated with them. Online survey fatigue was a factor at Campus 2, where students complained about the number of surveys they got in their e-mail inbox and trying to tell which survey went with this project and which did not (sometimes thinking they had filled out a survey for the project when they had not).

Directions for Future Research

This study shows promising results from the evaluation of one translation of a bystander sexual violence prevention program to a new campus. Relying on an experimental method, we found that the program appeared to work on both campuses in changing attitudes about sexual violence and the role each person can play in ending it. This reflects flexibility built into the Bringing in the Bystander program to allow the program materials to be customized where needed and to give co-facilitators the ability to tailor the presentation as appropriate. What this study illustrates as well is that to answer how such a program does and does not work with new populations, it is critical to examine the effectiveness in subgroups. If, for example, we had not looked at results by gender within each campus, it would have appeared instead that the program worked on Campus 1 and did not work very well on Campus 2. Instead, analyzing the data based on gender allowed us to see that men on Campus 2 had their own distinct pattern. Future work should continue this approach by generating large enough sample sizes to examine gender and campus in combination with additional distinguishing characteristics of students, such as resident versus commuter student, class standing (first year, sophomore, etc.), and race and ethnicity.

Further replications of programs, both Bringing in the Bystander and others, on additional, diverse campuses would broaden our understanding of how a bystander’s approach to sexual violence can be successful on a broad range of college campuses and can help create a community of prosocial bystanders willing and able to act to reduce sexual violence and support survivors of sexual violence.

Footnotes

Acknowledgements

Acknowledgments go to Mary Frederick, Robert Eckstein, and Angela Borges for their contributions to this project and to Rebecca Warner for her work as a statistical consultant, as well as the co-facilitators of the program on both campuses for their many hours of hard work.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Preparation of this manuscript was supported by the Centers for Disease Control and Prevention by Grant 5 R01 CE001388-02 (PI: Banyard). The content is solely the responsibility of the authors and does not necessarily represent the official views of the Centers for Disease Control and Prevention.

Notes

Author Biographies

Alison C. Cares, PhD, is an associate professor in the Department of Sociology, Criminology, & Anthropology at Assumption College in Worcester, Massachusetts. Her research focuses on consequences of sexual assault and intimate partner violence victimization, as well as evaluation of prevention efforts focused on violence against women and of general criminal justice policies and programs. She was recently co-principal investigator on a project to integrate crime victims’ issues into college curricula. She serves on the editorial boards of Victims and Offenders and Journal of Family Violence.

Victoria L. Banyard, PhD, is a professor in the psychology department at the University of New Hampshire. Her program of research includes the study of the long-term consequences of interpersonal violence, including the resilience of survivors and program evaluation of prevention efforts to address violence against women. She has been the principal investigator on several grants to evaluate prevention programs. She is on the editorial board of Psychology of Violence and Sexual Assault Report.

Mary M. Moynihan, PhD, is a research associate in the Women’s Studies Program and the prevention and evaluation coordinator of Prevention Innovations at the University of New Hampshire. Her research focuses on the attitudes and behaviors regarding prevention of and intervention in situations of sexual and relationship violence and stalking. She collaborates with other researchers to evaluate bystander-focused programs. She is also part of a research team that conducts research and publishes reports and articles on unwanted sexual, relationship, and stalking experiences on campus.

Linda M. Williams, PhD, is professor in the School of Criminology and Justice Studies at the University of Massachusetts, Lowell. For 40 years, she has directed longitudinal research on violence against women and children. Author of several books and dozens of scholarly publications, she has lectured in the United States and internationally on many topics, including child sexual abuse, trauma and memory, human trafficking, and researcher–practitioner collaborations. She has been principal investigator on 16 U.S. federally funded research projects and was recently funded to conduct a study of sexual violence case attrition in the United States. She was co-investigator on the CDC-funded research “Enhancing Bystander Efficacy to Prevent Sexual Violence: Primary Prevention on Two College Campuses.”

Sharyn J. Potter, PhD, MPH, is an associate professor in the Department of Sociology and co-director of Prevention Innovations, Research and Practices for Ending Violence Against Women on Campus, at the University of New Hampshire. She has a PhD in medical sociology and a MPH from Emory University. She co-leads the development, dissemination, and evaluation of the Know Your Power® Bystander Social Marketing Campaign that raises awareness regarding the problem of sexual and relationship violence and models prosocial behaviors that target audience members can use in situations where sexual and relationship violence and stalking are occurring, have occurred, or have the potential to occur.

Jane G. Stapleton, MA, is a research instructor in the Women’s Studies Program and the Department of Family Studies, and the Justice Studies Program at the University of New Hampshire. She is co-director of Prevention Innovations, Research and Practices for Ending Violence Against Women on Campus, a research and development unit at the University of New Hampshire. Her research focuses on developing and evaluating the Know Your Power® Bystander Social Marketing Campaign. She works on evaluation projects with practitioners from local, state, and national organizations.

References

Anderson

K. M.

Danis

F. S.

(2007). Collegiate sororities and dating violence: An exploratory study of informal and formal helping strategies. Violence Against Women, 13, 87-100.

Anderson

L. A.

Whiston

S. C.

(2005). Sexual assault education programs: A meta-analytic examination of their effectiveness. Psychology of Women Quarterly, 29, 374-388.

Banyard

V. L.

(2008). Measurement and correlates of prosocial bystander behavior: The case of interpersonal violence. Violence and Victims, 23, 85-99.

Banyard

V. L.

(2011). Who will help prevent sexual violence: Creating an ecological model of bystander intervention. Psychology of Violence, 1, 216-229.

Banyard

V. L.

Cohn

Moynihan

M. M.

Walsh

Ward

(2007). Unwanted sexual experiences at UNH: Incidence and prevalence, 2006. Durham: University of New Hampshire.

Banyard

V. L.

Eckstein

Moynihan

M. M.

(2010). Sexual violence prevention: The role of stages of change. Journal of Interpersonal Violence, 25, 111-135.

Banyard

V. L.

Moynihan

M. M.

(2011). Variation in bystander behavior related to sexual and intimate partner violence prevention: Correlates in a sample of college students. Psychology of Violence, 1, 287-301.

Banyard

V. L.

Moynihan

M. M.

Cares

A. C.

Warner

(2014). How do we know if it works?: Measuring outcomes in bystander-focused abuse prevention on campuses. Psychology of Violence, 4(1), 101-115.

Banyard

V. L.

Moynihan

M. M.

Crossman

M. T.

(2009). Reducing sexual violence on campus: The role of student leaders as empowered bystanders. Journal of College Student Development, 50, 446-457.

10.

Banyard

V. L.

Moynihan

M. M.

Plante

E. G.

(2007). Sexual violence prevention through bystander education: An experimental evaluation. Journal of Community Psychology, 35, 463-481.

11.

Banyard

V. L.

Moynihan

M. M.

Walsh

W. A.

Cohn

E. S.

Ward

(2010). Friends of survivors: The community impact of unwanted sexual experiences. Journal of Interpersonal Violence, 25, 242-256.

12.

Banyard

V. L.

Plante

E. G.

Moynihan

M. M.

(2004). Bystander education: Bringing a broader community perspective to sexual violence prevention. Journal of Community Psychology, 32, 61-79.

13.

Banyard

V. L.

Potter

S. J.

Cares

A. C.

Williams

L. M.

Moynihan

M. M.

Stapleton

J. G.

(under review). Ending sexual violence on campuses by improving bystander intervention: Multiple prevention tools working together. Manuscript submitted for publication.

14.

Burn

S. M.

(2009). A situational model of sexual assault prevention through bystander intervention. Sex Roles, 60, 779-792.

15.

Edwards

R. W.

Jumper-Thurman

Plested

B. A.

Oetting

E. R.

Swanson

(2000). Community readiness: Research to practice. Journal of Community Psychology, 28, 291-307.

16.

Fisher

B. S.

Cullen

F. T.

Turner

M. G.

(2000). The sexual victimization of college women. Washington, DC: Office of Justice Programs, National Institute of Justice and Bureau of Justice Statistics, U.S. Department of Justice.

17.

Gidycz

C. A.

Orchowski

L. A.

Berkowitz

A. D.

(2011). Preventing sexual aggression among college men: An evaluation of a social norms and bystander intervention program. Violence Against Women, 17, 720-742.

18.

Gidycz

C. A.

Orchowski

L. A.

King

C. R.

Rich

C. L.

(2008). Sexual victimization and health-risk behaviors: A prospective analysis of college women. Journal of Interpersonal Violence, 23, 744-763.

19.

Grimley

Prochaska

J. O.

Velicer

W. F.

Blais

L. M.

DiClemente

C. C.

(1994). The transtheoretical model of change. In Brinthaupt

T. M.

Lipka

R. P.

(Eds.), Changing the self: Philosophies, techniques, and experiences (pp. 201-227). Albany: State University of New York Press.

20.

Gross

A. M.

Winslett

Roberts

Gohm

C. L.

(2006). An examination of sexual violence against college women. Violence Against Women, 12, 288-300.

21.

Heppner

M. J.

Good

G. E.

Hillenbrand-Gunn

T. L.

Hawkins

A. K.

Hacquard

L. L.

Nichols

R. K.

. (1995). Examining sex differences in altering attitudes about rape: A test of the elaboration likelihood model. Journal of Counseling & Development, 73, 640-647.

22.

Jordan

C. E.

Campbell

Follingstad

(2010). Violence and women’s mental health: The impact of physical, sexual, and psychological aggression. Annual Review of Clinical Psychology, 6, 607-628.

23.

Karjane

H. M.

Fisher

B. S.

Cullen

F. T.

(2005). Sexual assault on campus: What colleges and universities are doing about it. Washington, DC: Office of Justice Programs, U.S. Department of Justice.

24.

Katz

(1994). Mentors in Violence Prevention (MVP) trainer’s guide. Boston, MA: Northeastern University’s Center for the Study of Sport in Society.

25.

Katz

Heisterkamp

Fleming

W. M.

(2011). The social justice roots of the Mentors in Violence Prevention model and its application in a high school setting. Violence Against Women, 17, 684-702.

26.

Langhinrichsen-Rohling

Foubert

J. D.

Brasfield

H. M.

Hill

Shelley-Tremblay

(2011). The men’s program: Does it impact college men’s self-reported bystander efficacy and willingness to intervene? Violence Against Women, 17, 743-759.

27.

Lonsway

K. A.

Banyard

V. L.

Berkowitz

A. D.

Gidycz

C. A.

Katz

J. T.

Koss

M. P.

. (2011). Rape prevention and risk reduction: Review of the research literature for practitioners. Harrisburg, PA: National Resource Center on Domestic Violence.

28.

Malamuth

N. M.

(1989). The Attraction to Sexual Aggression Scale: Part one. Journal of Sex Research, 26, 26-49.

29.

Moynihan

M. M.

Banyard

V. L.

Arnold

J. S.

Eckstein

R. P.

Stapleton

J. G.

(2011). Sisterhood may be powerful for reducing sexual and intimate partner violence: An evaluation of the Bringing in the Bystander in-person program with sorority members. Violence Against Women, 17, 703-719.

30.

O’Brien

(2001). The MVP program: Focus on student-athletes. In Ottens

A. J.

Hotelling

(Eds.), Sexual violence on campus: Policies, programs, and perspectives (pp. 98-199). New York: Springer.

31.

O’Donohue

Yeater

E. A.

Fanetti

(2003). Rape prevention with college males: The roles of rape myth acceptance, victim empathy, and outcome expectancies. Journal of Interpersonal Violence, 18, 513-531.

32.

Payne

D. L.

Lonsway

K. A.

Fitzgerald

L. F.

(1999). Rape myth acceptance: Exploration of its structure and its measurement using the Illinois Rape Myth Acceptance Scale. Journal of Research in Personality, 33, 27-68.

33.

Potter

S. J.

(2012). Using a multimedia social marketing campaign to increase active bystanders on the college campus. Journal of American College Health, 60, 282-295.

34.

Potter

S. J.

Moynihan

M. M.

Stapleton

J. G.

(2011). Using social self-identification in social marketing materials aimed at reducing violence against women on campus. Journal of Interpersonal Violence, 26, 1-20.

35.

Potter

S. J.

Stapleton

(2011). Bringing in the target audience in bystander social marketing materials for communities: Suggestions for practitioners. Violence Against Women, 17, 797-812.

36.

Schwartz

M. D.

DeKeseredy

W. S.

(1997). Sexual assault on the college campus: The role of male peer support. Thousand Oaks, CA: Sage.

37.

Stöber

(2001). The Social Desirability Scale-17 (SDS-17): Convergent validity, discriminant validity, and relationship with age. European Journal of Psychological Assessment, 17, 222-232.

38.

West

Wandrei

M. L.

(2002). Intimate partner violence: A model for predicting interventions by informal helpers. Journal of Interpersonal Violence, 17, 972-986.