The Impact of the Bystander’s Relationship With the Victim and the Perpetrator on Intent to Help in Situations Involving Sexual Violence

Abstract

A large body of research has explored the individual and situational factors that influence bystander intervention for sexual violence. However, little research has explored the how the bystander’s relationship to the victim and the perpetrator impacts helping. To explore this gap in the literature, the present study used vignettes to experimentally manipulate the bystander’s relationship to the victim, and the bystander’s relationship to the perpetrator to examine how these factors impact intent to help in low (i.e., unwanted contact) and high (i.e., situation at high risk of rape) severity situations of sexual violence. The gender of the bystander was also examined by recruiting a sample of women and men. Results suggested that bystanders were more likely to intervene when the situation was more severe and when the bystander was female. Results were mixed regarding intent to help when the bystander knew the victim or the perpetrator. Moreover, these factors interacted in complicated ways such that, for example, women are equally likely to intend to help a victim whether they know the perpetrator or not while men are more likely to help a victim if the perpetrator is someone they do not know. The results of this study suggest that bystander intervention for situations involving sexual violence are complex and future research should further tease out the moderating effects. Prevention programs using a bystander framework may need some segments of training that are more gender specific and that directly address the relationship between the bystander and the victim and perpetrator.

Keywords

sexual violence bystander intervention relationship to victim relationship to perpetrator

A growing body of research has explored the role of bystanders (i.e., individuals with the potential to help someone before, during, or after an event) in preventing and/or intervening in situations involving sexual violence (e.g., Banyard, Moynihan, & Crossman, 2009; Bennett, Banyard, & Garnhart, 2014; Brown, Banyard, & Moynihan, 2014; Coker et al., 2011). This research suggested that that there are number of factors that impact bystander intervention for sexual violence; these factors include the individual bystander (e.g., gender, knowledge about sexual violence), the situation (e.g., severity of the situation), and the relationship with the parties involved (i.e., relationship between the victim and the perpetrator; see Banyard, 2011, for a review). Although this research has been informative, it has largely neglected the importance of the bystander’s relationship with the victim and the perpetrator on helping in situations involving sexual violence. Moreover, the little research that has examined the impact of the bystander’s relationship with the victim and perpetrator in situations involving sexual violence (e.g., Bennett & Banyard, 2016¹; Katz, Pazienza, Olin, & Rich, 2015; Nicksa, 2014) and has also lacked an examination of how/whether the bystander’s gender can interact with both the bystander’s relationship to the victim and the bystander’s relationship to the perpetrator to impact helping. The purpose of the present study was to experimentally examine how the bystander’s gender and the bystander’s relationship with the victim and the perpetrator influenced several forms of intent to help during situations involving sexual violence.

Facilitators and Barriers to Bystander Intervention for Sexual Violence

Research examining facilitators and barriers to bystander intervention for sexual violence has identified a number of factors that influence a bystander’s likelihood to help. Some of the factors relate specifically to an individual bystander. For example, research has suggested that bystanders are more likely to help in situations involving sexual violence if they are female, are Black, have more knowledge about sexual assault, have greater prosocial tendencies, endorse few rape myths, have high self-efficacy, and report prior victimization (Banyard, 2008; Bennett, Banyard, & Garnhart, 2014; Brown et al., 2014; Burn, 2009; Frese, Moya, & Megias, 2004; MacNab & Worthley, 2008; McMahon, 2010). In addition, researchers have also identified situational factors that influence the likelihood of intervening in situations involving sexual violence; namely, perceiving the situation as unambiguously dangerous. Fischer and colleagues (2011) conducted a meta-analysis of studies on bystander intervention from 1960 to 2010 and found that bystanders were more likely to help when the situation was perceived as dangerous and when the victim was at risk of having physical consequences from the incident. Lastly, research has also suggested that the relationship between the victim and the perpetrator has an influence on bystander intervention related to sexual violence. More specifically, studies have suggested that bystanders are more likely to blame the victim when the victim and the perpetrator have a relationship (Bieneck & Krahé, 2011; Krahé, Temkin, & Bieneck, 2007). Similarly, Shotland and Straw (1976) found that bystanders were less likely to intervene when a man was attacking at woman if they believed the victim and the perpetrator were married than if they believed the victim and the perpetrator were strangers.

Relationship Between the Bystander and the Victim and the Perpetrator

Little research has explored how the bystander’s own relationship with the victim and the perpetrator influences helping. Research that does exist has focused on more general types of helping rather than the unique situation of sexual violence. Bystanders are more likely help when they have had any interaction with the victim (ranging from a conversation to a friendship; for example, Bell, Grekul, Lamba, Minas, & Harrell, 2001; Howard & Crano, 1974; Levine, Cassidy, Brazier, & Reicher, 2002). Bystanders may be less likely to intervene if they know the perpetrator. For instance, Curphy and colleagues (1998) found that bystanders were less likely to report someone for a code violation at a military academy if they were a friend than a stranger.

Only a handful of studies have explored how knowing the victim and the perpetrator influences intervention for situations involving sexual violence. In relation to knowing the victim, Bennett & Banyard (2016) found that bystanders reported more positive bystander perceptions (i.e., perceived the situation as more of a problem and safer to intervene) when they knew the victim than when they did not know the victim. Similarly, Katz and colleagues (2015) found that bystanders reported a greater intent to help friends than strangers in situations involving sexual violence. In relation to knowing the perpetrator, Bennett & Banyard (2016) found that bystanders reported the situation as less of a problem but safer to intervene if the perpetrator was friend than a stranger. In addition, Nicksa (2014) found that bystanders reported a lower intent to report a crime (i.e., physical assault, theft, and sexual assault) if the offender was known.

In a previously published study that used this same data set but examined different outcome variables (i.e., perceptions of the situation as a problem and safety; Bennett & Banyard, 2016), we found that the relationship with the victim and the perpetrator interact in complicated ways in relation to how bystanders appraised the situation. Bystanders reported perceiving the situation as a problem most when they knew the victim only, an effect which was diminished if they also knew the perpetrator or only knew the perpetrator. Conversely, knowing both the victim and the perpetrator was more beneficial in perceiving the situation as safe to intervene, an effect that was diminished if participants knew only the victim or only the perpetrator. Although these findings are informative, they address only one piece of the process of bystander action, how a bystander appraises the situation (i.e., perceptions of the situation as a problem, safety to intervene). Models of bystander intervention also highlight the important variables of intent to help as another key step in the process of bystander action (Banyard & Moynihan, 2011).

Bystander Gender and Bystander Relationship to the Victim and the Perpetrator

The impact of gender on bystander intervention has been studied mainly in relation to general helping of friends (e.g., Eagly & Crowley, 1986; George, Carroll, Kersnick, & Calderon, 1998). In general, these studies found that women provided more helping to friends. Eagly and Crowley (1986) also discussed how gendered scripts affect how men and women provide help (e.g., men were more likely to help in situations where there was a call to be assertive than women). Similarly, Levine and Crowther (2008) examined the impact of gender roles on bystander intervention by manipulating the gender of fellow bystanders in situations involving direct need (e.g., a physical assault). They found that when fellow bystanders were women, men increased their likelihood to help, an effect that may have occurred due to men assuming traditional gender roles around female bystanders. In addition, when the fellow bystanders were men, women decreased their likelihood of helping, also potentially due to women deferring the responsibility of intervening to the fellow male bystanders, who may have been seen as more capable or able of being heroic and assertive following traditional gender scripts. Gender may not only influence whether individuals help, but may also influence how individuals help. For instance, Banyard and Moynihan (2012) found gender differences in the type of helping that men and women reported doing in the past 2 months, with women more likely to engage in behaviors that revolved around safety (e.g., walking a drunk friend home) whereas men were more likely to engage in behaviors that involved direct intervention (e.g., telling friends concerns about their relationship).

Only two known studies have explored how the bystander’s gender and the bystander’s relationship to the victim or the perpetrator influences helping for sexual violence. Katz et al. (2015) used a vignette study to manipulate the bystander’s relationship to victim in a party rape situation. Contrary to expectations, they did not find gender differences in intent to help and did not find an interaction between the bystander’s gender and the bystander’s relationship to the victim. However, Nicksa (2014) found gender differences in reporting sexual assault, physical assault, and theft to police (women had a greater likelihood of reporting than men). In addition, women and men differed in their likelihood of reporting a sexual assault depending on their relationship with the perpetrator; such that men were more likely to report if they were friends with the perpetrator than if the perpetrator was a stranger. This effect was not present for women. To our knowledge, no study has explored whether the gender of the bystander interacts with the bystander’s relationship to both the victim and the perpetrator to influence helping.

The Current Study

The purpose of the present study was to examine whether the gender of the bystander and the bystander’s relationship with the victim and the perpetrator has an impact on their intent to help in a low and high severity situation involving sexual violence. Prior research suggested that intent to help is an important factor in predicting bystander behavior (Banyard & Moynihan, 2011). The current study was a 2 (severity: low and high) × 2 (gender of the bystander/participant: female or male) × 2 (the bystander/participant’s relationship to the victim: stranger or friend) × 2 (the bystander/participant’s relationship to the perpetrator: stranger or friend) factorial design. The between-subject variables were the bystander’s gender, the bystander’s relationship with the victim, and the bystander’s relationship with the perpetrator whereas the within-subject variable was severity.

Given the practical and ethical difficulties with conducting real-world experiments involving bystander intervention for sexual violence, the present study used vignettes to manipulate the variables of interest (with the exception of bystander gender which was not experimentally manipulated but rather examined by recruiting a sample of female and male participants). Prior research has also used vignettes to explore similar questions (e.g., Nicksa, 2014). The dependent variable was a bystander’s expressed intent to help in the situation. Consistent with recommendations by McMahon and Banyard (2011), several types of behavioral intent were measured—intent to take action to help the victim, intent to take action to confront the perpetrator, and intent to seek help via outside professional resources. Grounded in previous research (e.g., Authors, 2014; Banyard & Moynihan, 2011; Curphy et al., 1998; Fischer et al., 2011; Nicksa, 2014), we hypothesized as follows:

Hypothesis 1: Participants will report a greater intent to help in the high severity than the low severity condition.

Hypothesis 2: Women will report a greater intent to help than men.

Hypothesis 3: Participants with a close relationship to the victim will report a greater intent to help than participants who have a stranger relationship to the victim across types of actions.

Hypothesis 4: Participants with a close relationship with perpetrator will report a lower intent to help than participants who have a stranger relationship to the perpetrator.

Hypothesis 5: The relationship to the perpetrator will differentially affect men’s and women’s likelihood of helping. Men will report a greater intent to help if they know the perpetrator than if they do not know the perpetrator; an effect that will not be present for women. The extent to which gender will affect intent to help victims or to seek resources is exploratory.

Hypothesis 6: The impact of the bystander’s relationship to the victim and the perpetrator will have lesser impact in high severity condition as compared with the low severity condition.

Method

Participants

Participants were recruited through a psychology subject pool at a northeastern university. Although 641 surveys were completed, only 481 surveys were used due to some participants not meeting study requirements (i.e., below 18 years of age; n = 5), not completing more than one fourth of the survey (n = 14), providing duplicate surveys (n = 77), or not paying attention to the survey as demonstrated when not correctly answering a minimum of two of the three manipulation checks (n = 64). The 481 participants were slightly overrepresented in terms of women (55.5% women and 44.5% men), were mostly young (M = 19.06, SD = 1.88), in the first 2 years of college (53.6% freshman and 29.7% sophomore), and, consistent with the university population where the study was conducted, were mostly Caucasian (90.9%).

Procedures

Participants were randomly assigned by gender to read vignettes in which they had a relationship with the victim and the perpetrator (close to the victim only, close to the perpetrator only, close to both the victim and the perpetrator, close with neither the victim nor the perpetrator). The vignettes involved a low and high severity situation depicting sexual violence. Following each vignette, participants were asked to rate their intent to help in the situation.

In an effort to prevent an ordering effect, the order of the low and high severity vignettes was randomly assigned. In addition, the use of random assignment should have eliminated the need to control for extraneous variables. Nonetheless, data were also collected on variables that prior research has suggested influence bystander behavior in situations involving sexual violence to make sure participants in each condition were comparable on these variables (i.e., bystander efficacy, prior victimization, social desirability, and rape myth endorsement; Banyard, 2008; Koss et al., 2007; McMahon & Farmer, 2011; Stöber, 2001).² Analyses confirmed that there were no differences between participants in different relationship conditions or different severity order on these variables.

Measures

Vignettes

The vignettes were developed by the first two authors who are expert researchers in sexual violence bystander intervention in consultation with professionals in the field working on rape prevention and described situations involving sexual violence.³ Sample vignettes are provided in Appendix A. Participants were randomly assigned by gender to a relationship with the victim and the perpetrator (close to the victim only, close to the perpetrator only, close to the victim and the perpetrator, close to neither the victim nor the perpetrator). Before reading the vignettes, participants were asked to write the initials of a close female and a close male friend at the university. Participants received the initials of their friend(s) in the vignette if they were assigned a close relationship with the victim and/or perpetrator. In both vignettes, the victim was always female and the perpetrator was always male.

The two vignettes varied in severity, such that the low severity conditions involved unwanted contact (i.e., unwanted rubbing of the victim’s butt) whereas the high severity condition involved a situation of risk for sexual assault (i.e., the perpetrator taking an intoxicated victim back to his house). Both the low and high severity situation occurred at a party and involved the victim and the perpetrator consuming alcohol.

Manipulation check

A manipulation check was provided to examine whether participants could correctly identify their relationship with the victim and the perpetrator for each vignette (i.e., Based on the scenario described, what is your relationship with Jesse/John/friends initials?). In addition, a manipulation check was also provided to examine whether the high severity vignette was perceived as more severe than the low severity vignette (i.e., Which scenario carries more risk that the woman could get hurt?).

Bystander intent to help

Adapted from Banyard, Moynihan, Cares, and Warner (2014), a shortened set of bystander intent to help items was used to better access the likelihood of doing specific helping behaviors for each vignette. For each of the listed nine behaviors, participants responded on a 5-point Likert-type scale (1 = not at all likely, 2 = slightly likely, 3 = somewhat likely, 4 = very likely, 5 = extremely likely). Items were chosen to reflect several types of action—actions directed at victims, actions directed at perpetrators, and actions directed at seeking outside help. See Appendix B for the full measure.

Data Analysis Plan

First, a series of chi-square tests were used to determine whether the manipulation of closeness and severity was successful. Second, a repeated-measures MANOVA was used with severity as the within-subject condition, relationship to the victim, relationship to the perpetrator, gender as the between-subject condition, and intent to help as the dependent variable. Hypothesis 1 was tested by examining whether there was a main effect for severity; Hypothesis 2 was tested by examining whether there was a main effect for gender; Hypothesis 3 was tested by examining whether there was a main effect for the bystander’s relationship to the victim; and Hypothesis 4 was tested by examining whether there was a main effect for the bystander’s relationship to the perpetrator. In addition, a repeated-measures MANOVA was used to explore for a two-way interaction between the bystander’s relationship to the victim and the bystander’s relationship to the perpetrator (Hypotheses 3 and 4). Finally, a repeated-measures MANOVA also was used to explore for a three-way interaction between bystander gender, relationship to victim and perpetrator (Hypothesis 5), the bystander’s relationship to the victim, the bystander’s relationship to the perpetrator, and severity (Hypothesis 6).

Results

Manipulation Check

To determine whether the manipulation of closeness and severity were effective, two sets of analyses were conducted. Participants were asked what their relationship was with the victim and perpetrator in both the low severity and high severity conditions. They were also asked which scenario (i.e., low or high severity vignette) posed the greatest risk that the woman could be hurt. As reported in Bennett & Banyard (2016), these checks were successful and indicated that participants perceived the relationship variables and severity in the vignettes as described. Nonetheless, in an effort to control for participants that were not paying attention during the study, participants were only retained for analyses if they correctly answered a minimum of two of the three manipulation checks.

Factor Analysis of Bystander Intent to Help Scale

To assess the dimensionality of the nine different bystander intent items (as suggested by Banyard et al., 2014; McMahon & Banyard, 2011), two exploratory factor analyses were conducted (for the low and high severity conditions separately) with a promax rotation requested and the criterion that only factors with eigenvalues greater than 1 were retained. We did not expect the analyses to differ by severity but did two analyses to be sure. Three factors were retained from the analyses for both the low and high severity vignettes. Factor 1 had three items related to bystander behavior directed toward the victim (labeled “Intent to Help Victim”), Factor 2 had three items related to bystander behavior directed toward the perpetrator (labeled “Intent to Confront Perpetrator”), and Factor 3 had three items related to bystander behavior directed toward accessing outside resources (labeled “Intent to Contact Outside Resources”). Thus, the bystander intent to help scale was organized into three subscales (Intent to Help Victim, Intent to Confront Perpetrator, and Intent to Contact Outside Resources), which were used for all future analyses. These findings are presented in Table 1.

Table 1.

Rotated Factor Loadings for Intent to Help Items.

	Nine Bystander Intent to Help Items
	Factor 1		Factor 2		Factor 3		Communality
	Low	High	Low	High	Low	High	Low	High
Start conversation with victim	.72	.74	.11	.26	.07	.10	.54	.57
Get people together to take victim home	.87	.90	.37	.35	.33	.34	.77	.82
Find victim’s friends to help	.87	.87	.26	.27	.28	.32	.76	.77
Tell perpetrator that is unwanted sexual contact	.27	.35	.93	.96	.42	.41	.86	.92
Tell perpetrator he could get in serious trouble	.29	.35	.92	.91	.36	.38	.86	.83
Tell perpetrator this could be rape	.25	.29	.90	.91	.47	.45	.82	.82
Call formal support system on campus (e.g., resident assitant, coach )	.25	.30	.39	.35	.92	.87	.84	.77
Call crisis center	.23	.24	.45	.42	.91	.89	.83	.80
Call campus police	.26	.23	.40	.37	.90	.84	.81	.70
Sum of squared loadings	1.38	1.33	4.09	4.09	1.63	1.57
% explained variance	15.30	14.80	45.42	45.48	18.07	17.48

Repeated-Measures MANOVA

A repeated-measures MANOVA was used to assess whether the mean scores on the three intent to help measures differed between participants by relationship condition with the victim, relationship condition with the perpetrator, and gender within the low and high severity conditions.

As presented in Table 2, the main effect was significant for severity with a large effect size, gender with a large effect size, relationship with the victim with a large effect size, and relationship with the perpetrator with a very large effect size (Warner, 2013). Thus, the responses on the intent to help items differed by the severity of the situation, gender, and the relationship the participant had with the victim and the perpetrator. Participants had greater intent to take all three types of action in the high severity vignette, supporting Hypothesis 1 (see Table 3 for univariate effects). In partial support of Hypothesis 2, women were significantly more likely than men to intend to help victims, but in contrast they were also significantly less likely than men to confront perpetrators (though there was no difference between men and women in intent to contact resources). As presented in Table 3, participants who had a close relationship with the victim reported higher intent to help the victim and contact outside resources than participants who did not know the victim. Hypothesis 3 was partially supported, in that having a relationship with the victim was related to a greater intent to help in some areas (i.e., intent to help the victim, intent to contact outside resources), but had no impact on intent to help in other areas (i.e., intent to confront the perpetrator). Participants who had a close relationship with the perpetrator reported a lower intent to help the victim, higher intent to confront the perpetrator, and a lower intent to contact outside resources than participants who did not know the perpetrator. Thus, Hypothesis 4 was partially supported, in that knowing the perpetrator was related to a lower likelihood of helping in some ways (i.e., intent to help the victim, intent to contact outside resources) but was related to a greater likelihood of confronting the perpetrator (i.e, intent to confront the perpetrator).

Table 2.

Significant Effects Independent Variables on Dependent Variables.

Independent Variables	df	df Error	F	Partial η²
Severity	3	471	14.42***	.10
Gender	3	471	34.46***	.18
Victim Closeness	3	471	17.95***	.10
Perpetrator Closeness	3	471	42.08***	.21
Gender × Perpetrator Closeness	3	471	3.88**	.02
Victim Closeness × Perpetrator Closeness	3	471	7.39***	.05

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

Table 3.

Significant Univariate Effects of Independent Variables on Dependent Variables.

Measure	df	df Error	F	Partial η²	Condition	M
Severity
Intent to help victim	1	473	39.67***	.08	Low	3.50
Intent to help victim	1	473	39.67***	.08	High	3.76
Intent to confront perpetrator	1	473	11.62**	.02	Low	2.60
Intent to confront perpetrator	1	473	11.62**	.02	High	2.76
Intent to contact outside resources	1	473	7.67**	.02	Low	1.69
Intent to contact outside resources	1	473	7.67**	.02	High	1.79
Gender
Intent to help victim	1	473	47.55***	.09	Women	3.88
Intent to help victim	1	473	47.55***	.09	Men	3.38
Intent to confront perpetrator	1	473	11.66**	.02	Women	2.52
Intent to confront perpetrator	1	473	11.66**	.02	Men	2.84
Intent to contact outside resources	1	473	0.44	ns	Women	1.76
Intent to contact outside resources	1	473	0.44	ns	Men	1.71
Victim Closeness
Intent to help victim	1	473	48.15***	.09	Stranger	3.38
Intent to help victim	1	473	48.15***	.09	Friend	3.88
Intent to confront perpetrator	1	473	1.28	ns	Stranger	2.63
Intent to confront perpetrator	1	473	1.28	ns	Friend	2.73
Intent to contact outside resources	1	473	4.09*	.01	Stranger	1.66
Intent to contact outside resources	1	473	4.09*	.01	Friend	1.81
Perpetrator Closeness
Intent to help victim	1	473	19.31***	.04	Stranger	3.79
Intent to help victim	1	473	19.31***	.04	Friend	3.47
Intent to confront perpetrator	1	473	28.84***	.06	Stranger	2.43
Intent to confront perpetrator	1	473	28.84***	.06	Friend	2.93
Intent to contact outside resources	1	473	9.52**	.02	Stranger	1.85
Intent to contact outside resources	1	473	9.52**	.02	Friend	1.62

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

There was also a significant two-way interaction of victim closeness and perpetrator closeness with a small effect size and a two-way interaction of gender and perpetrator closeness with a small effect size. All other interactions (i.e., two-way interactions: severity and gender, severity and victim closeness, severity and perpetrator closeness, gender and victim closeness; three-way interactions: victim closeness, perpetrator closeness, and severity; gender, victim closeness, and perpetrator closeness; severity, victim closeness, and gender; severity, perpetrator closeness, and gender; four-way interaction: severity, victim closeness, perpetrator closeness, and gender) were not significant. Thus, given that there was not a three-way interaction between victim closeness, perpetrator closeness, and severity, Hypothesis 6 was not supported.

Two-way interaction between gender and perpetrator closeness

The two-way interaction between gender and perpetrator closeness was significant for intent to help the victim, F(1, 473) = 12.68, p < .01; the effect size was medium with a partial η² of .02 and intent to confront the perpetrator, F(1, 473) = 4.91, p < .05; and the effect size was small with a partial η² of .01 (Warner, 2013). To further explore the significant interaction, two MANOVAs were used to assess whether the mean scores on intent to help the victim and intent to confront the perpetrator differed between participants who were randomly assigned to relationship with the perpetrator (friend or stranger) separately for women and men. The main effect of perpetrator closeness was significant both for women, F(2, 264) = 21.49, p < .001, and men, F(2, 221) = 22.40, p < .001. As demonstrated in Table 4, both women and men were more likely to confront the perpetrator if they were close to the perpetrator. However, men were more likely to help the victim if they were strangers with the perpetrator, an effect that was not present for women. Thus, Hypothesis 5 was partially supported. Although men did report a greater intent to confront the perpetrator if they knew the perpetrator than if they did not know the perpetrator, they also reported a lesser intent to help the victim if they knew the perpetrator than if they did not know the perpetrator. Moreover, women also reported a greater intent to confront the perpetrator if they were close to the perpetrator, but knowing the perpetrator did not affect women’s intent to help the victim.

Table 4.

Significant Interaction Between Gender and Perpetrator Closeness.

Measure	df	df Error	F	Partial η²	Perpetrator Closeness	M
Women
Intent to help victim	1	265	0.60	ns	Stranger	3.92
Intent to help victim	1	265	0.60	ns	Friend	3.84
Intent to confront perpetrator	1	265	29.17***	.10	Stranger	2.16
Intent to confront perpetrator	1	265	29.17***	.10	Friend	2.87
Men
Intent to help victim	1	212	24.95***	.11	Stranger	3.68
Intent to help victim	1	212	24.95***	.11	Friend	3.10
Intent to confront perpetrator	1	212	4.14*	.02	Stranger	2.71
Intent to confront perpetrator	1	212	4.14*	.02	Friend	2.99

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

Two-way interaction between victim closeness and perpetrator closeness

The two-way interaction between victim closeness and perpetrator closeness was significant for intent to help the victim, F(1, 473) = 5.67, p < .05; the effect size was small with a partial η² of .01 and intent to confront the perpetrator, F(1, 473) = 19.53, p < .001; and the effect size was medium with a partial η² of .04 (Warner, 2013). To further explore the significant interaction, two MANOVAs were used to assess whether the mean scores on intent to help the victim and intent to confront the perpetrator differed between participants who were randomly assigned to relationship with the perpetrator (friend or stranger) separately for those close to the victim and those not close to the victim. The main effect of perpetrator closeness was significant both for those that were not close to the victim, F(2, 233) = 38.44, p < .001, and those that were close to the victim, F(2, 242) = 11.96, p < .001. As demonstrated in Table 5, those not close to the victim reported a greater intent to confront the perpetrator if they were friends with the perpetrator. In contract, those close to the victim reported a greater intent to help the victim if they were strangers to the perpetrator.

Table 5.

Significant Interaction Between Victim Closeness and Perpetrator Closeness.

Measure	df	df Error	F	Partial η²	Perpetrator Closeness	M
Stranger to victim
Intent to help victim	1	234	0.93	ns	Stranger	3.46
Intent to help victim	1	234	0.93	ns	Friend	3.35
Intent to confront perpetrator	1	234	51.11***	.18	Stranger	2.13
Intent to confront perpetrator	1	234	51.11***	.18	Friend	3.09
Close to victim
Intent to help victim	1	243	20.82***	.08	Stranger	4.14
Intent to help victim	1	243	20.82***	.08	Friend	3.69
Intent to confront perpetrator	1	243	0.37	ns	Stranger	2.68
Intent to confront perpetrator	1	243	0.37	ns	Friend	2.76

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

Discussion

A number of interesting findings emerged from this study, which was the first to systematically explore the role of the relationship between the bystander and the victim and the perpetrator and the bystander’s gender in relation to intent to take action to stop a sexual assault. Consistent with McMahon and Banyard (2011) and others (e.g., Banyard et al., 2014), bystander action in relation to sexual violence seems to be a multidimensional construct. The factor analyses of the brief measure of bystander intent used here produced a consistent three factor solution representing action to support the victim, to confront the perpetrator, or to call in outside resources. Moreover, the current study captured these dimensions in relatively few items; suggesting that the bystander intent to help measure used in the present study may be helpful to field based program evaluations that often have little time or resources to give participants long surveys. The finding that different aspects of intent to take action were differentially associated with the bystander’s relationship with the parties involved and gender suggests that future research is needed to better understand how other factors (e.g., race, relationship between the victim and the perpetrator) influence different forms of bystander action. Rather than the general Latané and Darley (1970) situational model where a core set of variables predicts overall bystander action, the present study suggests that the relationship between the bystander and the victim and the perpetrator can influence the type of action utilized to address sexual violence (i.e., helping the victim, confronting the perpetrator, and contacting outside resources).

Consistent with several hypotheses derived from previous work, bystanders expressed greater intent to help when the situation was more severe. Latané and Darley’s (1970) original work showed that awareness that there is a problem is a key step in helping behavior. It is likely that participants more clearly saw the severe situation as an instance of sexual violence and a situation where the victim was being harmed. Thus, they expressed greater intent to do something across types of actions they could take. These findings remind us that in prevention work, a key piece of bystander mobilization is helping bystanders learn and perhaps expand their definitions of what sexual assault is and its damaging effects so that they will see a broader range of situations as serious.

Not surprisingly, individuals were more likely to say they would intervene to help a victim and contact outside resources if they knew the victim. Revealing the complexity of different types of bystander action knowing the victim did not have an impact on intent to confront the perpetrator. Similar complexity was obtained with regard to knowing the perpetrator. Participants indicated greater intent to help a victim if the perpetrator was a stranger to them, more likely to take action by confronting the perpetrator if that perpetrator was a friend, and more likely to seek outside resources if the perpetrator was a stranger. These findings combined with analysis of the two-way interaction for two of the three outcomes (i.e., intent to help the victim and intent to confront the perpetrator) remind us that bystander actions are diverse and are influenced by many variables. Additional research is needed about what variables best leverage different types of actions. A more fine-grained analysis such as this could then be used to better tailor prevention activities.

The findings about gender were also consistent and discrepant with previous studies (Banyard & Moynihan, 2012; Levine & Crowther, 2008). Overall, women reported greater intent to help victims whereas men reported greater intent to confront perpetrators. This fits within frameworks of helping that posit that gender scripts and stereotypes influence the ways in which men and women choose to help (Eagly & Crowley, 1986); women may be more likely to be active bystanders with victims whereas men focus on interrupting perpetrator behavior. Overall, inconsistent with prior research, however, there were many patterns of gender similarity (e.g., men and women reported an equal intent to contact resources), and only the two-way interaction between gender and perpetrator closeness was significant. Within this analysis, men and women continued to look similar in that they were both more likely to confront perpetrators whom they knew. However, gender did moderate the relationship between perpetrator closeness and intent to help a victim. Women had similar levels of intent to help victims whether they knew the perpetrator or not. For men, however, there was a higher intent to help the victim if the perpetrator was unknown to them. This finding may be due to the expectation that men follow the “bro code,” such that men are expected not to tell on or intervene in situations when other male friends are involved.

Despite the knowledge gained from this study, there are several limitations that should be noted. There was limited ethnic/racial and age diversity in the sample. Consistent with the population in the university where the sample was collected, most of the participants (roughly 90%) were White. Prior research has been mixed about the impact of race on bystander intervention, with one study finding that Black students reported more bystander behaviors than Caucasian students (Brown et al., 2014) whereas other studies have found no relationship between race of bystander and bystander intervention (e.g., Christy & Voigt, 1994; Frye, 2007). Thus, future research should continue to explore how racial/ethnic identity may influence bystander behavior. The sample was also mainly drawn from first- and second-year students on one college campus. Given what we have learned about how community context may influence bystander action and how bystander action may change over 4 years in college (Banyard, 2011), more research is needed to replicate and expand these findings in different settings. Furthermore, the vignettes used described the victim as female and the perpetrator as male, suggesting that additional research is needed to further explore how gender role expectations for the victim and the perpetrator interact with gender role expectations of bystanders. Finally, the present study is limited in that we assessed bystander intentions rather than actual bystander behavior. Indeed, prior research has indicated that the correlates of intent to help can vary from the correlates of actual helping behavior (Banyard & Moynihan, 2011). Thus, additional research is needed to further understand how the bystander’s relationship with the victim and the perpetrator affects actual bystander behavior.

Despite the limitations, the present study adds to the existing literature by suggesting that the bystander’s relationship to the victim and the perpetrator affects helping and by suggesting that future research should continue examining the predictors of different types of helping. Moreover, findings from the current study underscore the importance of programming efforts focusing on addressing barriers to intervening. For example, prevention work should continue to help individuals be more aware of the continuum of sexual violence, helping them attend to lower severity situations where risk can potentially escalate to find low cost ways to check in and diffuse the situation. The current study highlights how difficult it is to step in to assist a victim or seek outside resources when you know the perpetrator. This may reflect limitations in individual’s bystander toolkits. Prevention programs should help participants brainstorm distracting or distancing strategies for removing a potential victim from a situation in ways that may feel less threatening to a friendship with a perpetrator, such as enlisting other friends and activating peer support or by encouraging conversations among friends about these issues and about preplanning before going to social events. While the patterns of gender similarity suggest that overall bystander action curricula can be used equally effectively with men and women, the results of the current study also suggest that programming with men needs to remind them to think about action beyond perpetrator confrontation, to make it more acceptable or develop helping scripts that include approaching victims or reaching out to others for help. Women may need more direct discussion of safe ways to approach potential perpetrators. An increased understanding of the complexity of bystander intervention and factors that help explain this complexity (e.g., gender, closeness to victim and perpetrator) can continue to inform bystander prevention initiatives and untimely contribute to the reductions of sexual violence in our society.

Footnotes

Appendix A

Sample Vignettes

Low Severity	High Severity
You are at a party and you notice (victim’s name) across the room. Next to (victim’s name) is (perpetrator’s name). From what you can see, there is a lot of alcohol at the party, and (perpetrator’s name) and (victim’s name) appear to be drinking. (Perpetrator’s name) keeps grabbing (victim’s name)’s butt and rubbing up against her. (Victim’s name) is laughing but you can also tell she is trying to pull away from (perpetrator’s name). (Victim’s name) keeps removing his hands from her body and politely telling him to “cut it out.” Yet, (perpetrator’s name) continues to make advances.	You are at a party and you notice (victim’s name) laughing and having a great time. While at the party, you also notice (victim’s name) and (perpetrator’s name). Every time you have seen (victim’s name) and (perpetrator’s name), they have an alcoholic drink in their hands. At one point, you encounter them in the hallway and they are slurring their speech and declaring that they are wasted. (Victim’s name) and (perpetrator’s name) are kissing and you over hear (perpetrator’s name) tell (victim’s name) he is taking her back to his place. (Victim’s name) can barely walk on her own, and she seems reluctant to be kissing (perpetrator’s name). (Perpetrator’s name) begins to lead (victim’s name) away from the party.

Low Severity

High Severity

You are at a party and you notice (victim’s name) across the room. Next to (victim’s name) is (perpetrator’s name). From what you can see, there is a lot of alcohol at the party, and (perpetrator’s name) and (victim’s name) appear to be drinking. (Perpetrator’s name) keeps grabbing (victim’s name)’s butt and rubbing up against her. (Victim’s name) is laughing but you can also tell she is trying to pull away from (perpetrator’s name). (Victim’s name) keeps removing his hands from her body and politely telling him to “cut it out.” Yet, (perpetrator’s name) continues to make advances.

You are at a party and you notice (victim’s name) laughing and having a great time. While at the party, you also notice (victim’s name) and (perpetrator’s name). Every time you have seen (victim’s name) and (perpetrator’s name), they have an alcoholic drink in their hands. At one point, you encounter them in the hallway and they are slurring their speech and declaring that they are wasted. (Victim’s name) and (perpetrator’s name) are kissing and you over hear (perpetrator’s name) tell (victim’s name) he is taking her back to his place. (Victim’s name) can barely walk on her own, and she seems reluctant to be kissing (perpetrator’s name). (Perpetrator’s name) begins to lead (victim’s name) away from the party.

Appendix B

Acknowledgements

The authors thank Edward Lemay, PhD who provided feedback on the methodology used in the present study.

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) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Sidney Bennett is a doctoral candidate in the Psychology Department at the University of New Hampshire. She has a MA in Psychology and a MA in Justice Studies from the University of New Hampshire. Her research focuses on a variety of topics related to preventing and intervening in situations involving interpersonal violence, including: bystander intervention, case attrition, and victim motives for revealing and concealing interpersonal violence.

Victoria L. Banyard, PhD, is a Professor in the Department of Psychology at the University of New Hampshire and part of Prevention Innovations, a research center at the University of New Hampshire. Her program of research focuses on resilience among survivors of interpersonal violence and prevention of sexual assault and relationship abuse using a bystander framework.

Katie M. Edwards is an Assistant Professor of Psychology and Women’s Studies at the University of New Hampshire. Specific research interests include predictors and correlates of intimate partner violence (IPV) victimization, perpetration, and bystander intervention; leaving, disclosure, and recovery processes associated with IPV victimization; and ethics of IPV research. Professor Edwards is committed to using research to inform practice and policy.

References

Banyard

V. L.

(2008). Measurement and correlates of pro-social bystander behavior: The case of interpersonal violence. Violence and Victims, 23, 85-99. doi:10.1891/08866708.23.1.83

Banyard

V. L.

(2011). Who will help prevent sexual violence: Creating an ecological model of bystander intervention. Psychology of Violence, 1, 216-229. doi:10.1037/a0023739

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. doi:10.1037/a0023544

Banyard

V. L.

Moynihan

M. M.

(2012). Gender and bystander behavior to prevent sexual and relationship violence. Unpublished manuscript, Department of Psychology, University of New Hampshire, Durham, NH.

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, 101-115. doi:10.1037/a0033470

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. doi:10.1353/csd.0.0083

Bell

Grekul

Lamba

Minas

Harrell

W. A.

(2001). The impact of cost on student helping behavior. The Journal of Social Psychology, 135, 49-56. doi:10.1080/00224545.1995.9711401

Bennett

Banyard

V. L.

(2016). Do friends really help friends? The effect of relational factors and perceived severity on bystander perception of sexual violence. Psychology of Violence, 6, 64-72. doi:10.1037/a0037708

Bennett

Banyard

V. L.

Garnhart

(2014). To act or not to act, that is the question? Barriers and facilitators of bystander intervention. Journal of Interpersonal Violence, 29, 476-496. doi:10.1177/0886260513505210

10.

Bieneck

Krahé

(2011). Blaming the victim and exonerating the perpetrator in cases of rape and robbery: Is there a double standard? Journal of Interpersonal Violence, 26, 1785-1797. doi:10.1177/0886260510372945

11.

Brown

A. L.

Banyard

V. L.

Moynihan

M. M.

(2014). College students as helpful bystanders against sexual violence: Gender race, and year in college moderate the impact of perceived peer norms. Psychology of Women Quarterly, 38, 350-362. doi:10.1177/0361684314526855

12.

Burn

(2009). A situational model of sexual assault prevention through bystander intervention. Sex Roles, 60, 779-792. doi:10.1007/s11199-008-9581-5

13.

Christy

C. A.

Voigt

(1994). Bystander responses to public episodes of child abuse. Journal of Applied Social Psychology, 24, 824-847. doi:10.1111/j.1559-1816.1994.tb00614.x

14.

Coker

A. L.

Cook-Craig

P. G.

Williams

C. M.

Fisher

B. S.

Clear

E. R.

Garcia

L. S.

Hegge

L. M.

(2011). Evaluation of Green Dot: An active bystander intervention to reduce sexual violence on college campuses. Violence Against Women, 17, 777-796. doi:10.1177/1077801211410264

15.

Curphy

G. J.

Gibson

F. W.

Macomber

Calhoun

C. J.

Wilbanks

L. A.

Burger

M. J.

(1998). Situational factors affecting peer reporting intentions at the U.S. Air Force Academy: A scenario-based investigation. Military Psychology, 10, 27-43. doi:10.1207/s15327876mp1001_3

16.

Eagly

A. H.

Crowley

(1986). Gender and helping behavior: A meta-analytic review of the social psychological literature. Psychological Bulletin, 100, 283. doi:10.1037/0033-2909.100.3.283

17.

Fischer

Krueger

J. I.

Greitemeyer

Vogrincic

Kastenmüller

Frey

. . .Kainbacher

(2011). The bystander-effect: A meta-analytic review on bystander intervention in dangerous and non-dangerous emergencies. Psychological Bulletin, 137, 517-537. doi:10.1037/a0023304

18.

Frese

Moya

Megias

J. L.

(2004). Social perception of rape: How rape myth acceptance modulates the influence of situational factors. Journal of Interpersonal Violence, 19, 143-161. doi:10.1177/0886260503260245

19.

Frye

(2007). The informal social control of intimate partner violence against women: Exploring personal attitudes and perceived neighborhood social cohesion. Journal of Community Psychology, 35, 1001-1018. doi:10.1002/jcop.20209

20.

George

Carroll

Kersnick

Calderon

(1998). Gender-related patterns of helping among friends. Psychology of Women Quarterly, 22, 685-704. doi:10.1111/j.1471-6402.1998.tb00185.x

21.

Howard

Crano

W. D.

(1974). Effects of sex, conversation, location, and size of observer group on bystander intervention in a high risk situation. Sociometry, 37, 491-507. doi:10.2307/2786423

22.

Katz

Pazienza

Olin

Rich

(2015). That’s what friends are for: Bystander responses to friends or strangers at risk for party rape victimization. Journal of Interpersonal Violence, 30, 2775-2792. doi:10.1177/0886260514554290

23.

Koss

M. P.

Abbey

Campbell

Cook

Norris

Testa

. . .White

(2007). Revising the SES: A collaborative process to improve assessment of sexual aggression and victimization. Psychology of Women Quarterly, 31, 357-370. doi:10.1111/j.1471-6402.2007.00385.x

24.

Krahé

Temkin

Bieneck

(2007). Schema-driven information processing in judgements about rape. Applied Cognitive Psychology, 21, 601-619. doi:10.1002/acp.1297

25.

Latané

Darley

J. M.

(1970). The unresponsive bystander: Why doesn’t he help? New York, NY: Appleton-Century-Crofts.

26.

Levine

Cassidy

Brazier

Reicher

(2002). Self-categorization and bystander non-intervention: Two experimental studies. Journal of Applied Social Psychology, 32, 1452-1463. doi:10.1111/j.1559-1816.2002.tb01446.x

27.

Levine

Crowther

(2008). The responsive bystander: How social group membership and group size can encourage as well as inhibit bystander intervention. Journal of Personality and Social Psychology, 95, 1429-1439. doi:10.1037/a0012634

28.

MacNab

B. R.

Worthley

(2008). Self-efficacy as an intrapersonal predictor for internal whistleblowing: A U.S. and Canada examination. Journal of Business Ethics, 79, 407-421. doi:10.1007/s10551-007-9407-3

29.

McMahon

(2010). Rape myth beliefs and bystander attitudes among incoming college students. Journal of American College Health, 59, 3-11. doi:10.1080/07448481.2010.483715

30.

McMahon

Banyard

V. L.

(2011). When can I help? A conceptual framework for the prevention of sexual violence through bystander intervention. Trauma, Violence, & Abuse, 13, 3-14. doi:10.1177/1524838011426015

31.

McMahon

Farmer

G. L.

(2011). An updated measure for assessing subtle rape myths. Social Work Research, 35, 71-81. doi:10.1093/swr/35.2.71

32.

Nicksa

(2014). Bystander’s willingness to report theft, physical assault, and sexual assault: The impact of gender, anonymity, and relationship to the offender. Journal of Interpersonal Violence, 29, 217-236. doi:10.1177/0886260513505146

33.

Shotland

R. L.

Straw

M. K.

(1976). Bystander response to an assault: When a man attacks a woman. Journal of Personality and Social Psychology, 34, 990-999. doi:10.1037/0022-3514.34.5.990

34.

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. doi:10.1027//1015-5759.17.3.222

35.

Warner

R. M.

(2013). Applied statistics: From bivariate through multivariate techniques. Thousand Oaks, CA: Sage Publications Inc.