College Students’ Perceptions of Barriers to Bystander Intervention

Abstract

Sexual violence is a major problem on college campuses and is associated with a range of negative health consequences for victims. Teaching students to intervene as prosocial bystanders has become a common element of sexual assault prevention efforts; although these programs have demonstrated positive effects on participants’ beliefs and knowledge, their impact on actual behavior is weaker. Understanding the factors that inhibit intervening in risky situations may enhance the effectiveness of bystander programs by identifying material that addresses these barriers. A sample of 281 first-year college students indicated whether they had encountered 10 situations that may present elevated risk of sexual or physical assault since arriving on campus, and if so, whether they had done something to intervene. If they had not intervened, they were asked to identify the barriers that had inhibited them. Participants also completed measures of two factors proposed to predict bystander behavior, self-efficacy and emotion regulation. A majority of participants intervened in most of the situations, but only 27% of participants intervened in every situation they encountered. Men and women differed in the barriers they identified most frequently across situations, with men endorsing Perceived Responsibility more often than women, and women reporting Skill Deficits more often than men. Neither men nor women perceived Audience Inhibition to be a significant barrier; it was salient in only one of the 10 situations. Students higher in global bystander self-efficacy were more likely to intervene and less likely to report barriers related to skill deficits and perceived responsibility. These results suggest that existing bystander intervention programs efforts can be improved by fostering a greater sense of collective responsibility in students and teaching specific intervention behaviors.

Keywords

sexual assault intervention prevention situational factors

Sexual violence, including sexual coercion and sexual assault, is a major problem on college campuses nationwide. Approximately 20% of women and 6% of men report experiencing sexual assault while in college (Krebs, Lindquist, Warner, Fisher, & Martin, 2007), and victims report a range of mental health problems, including depression, posttraumatic stress disorder (PTSD) symptoms, and suicidal ideation (Gilboa-Schechtman & Foa, 2001; Najdowski & Ullman, 2009; Ulman & Brecklin, 2002; Ullman & Filipas, 2001). For many years, the dominant approaches to reducing sexual assault on college campuses were to target knowledge of and beliefs about sexual consent and coercion and to teach self-protection strategies (e.g., Borges, Banyard, & Moynihan, 2008; Hollander, 2014; Orchowski, Gidycz, & Raffle, 2008). However, these methods have not proven to be effective in reducing sexual violence (DeGue et al., 2014; Lonsway et al., 2009). Recent efforts to prevent sexual violence on college campuses have shifted from focusing on potential perpetrators and victims to focusing on individuals who witness violent or potentially violent interactions (aka “bystanders”). Bystander intervention programs teach members of the college community to stop situations they observe from escalating into unwanted sexual or violent encounters and to help victims after such incidents (Banyard, 2008; Banyard, Moynihan, & Crossman, 2009; Banyard, Moynihan, & Plante, 2007; Coker et al., 2011). Evaluations of bystander intervention programs provide promising support for their capacity to change beliefs and knowledge about sexual consent and coercion (Banyard et al., 2009; Banyard, Plante, & Moynihan, 2005; Moynihan & Banyard, 2008), but relatively little evidence that they lead to more responsive behavior. The goal of the present study was to better understand factors that inhibit bystander behavior by examining college students’ perceptions of barriers to intervening in situations that may present risk of physical or sexual violence to another person.

Bystander Intervention

Bystander intervention programs are designed to help participants learn to recognize potentially risky situations, instill a sense of responsibility for helping others in their community, and teach effective intervention behaviors. This approach is based on Latane and Darley’s (1970) theory of the determinants of responsive bystander behavior. Their model outlines a series of steps required for intervention to occur: bystanders need to notice the event, identify it as dangerous, take responsibility for intervening, know how to intervene, and then take action. Empirical evaluations of campus bystander programs indicate that they have moderate-sized effects on participants’ knowledge about sexual assault, self-efficacy for intervening, and intent to intervene, but only small effects on actual behavior (Banyard et al., 2009; Banyard et al., 2007; Coker et al., 2011; Foubert, Langhinrichsen-Rohling, Brasfield, & Hill, 2010; Katz & Moore, 2013; Kleinsasser, Jouriles, McDonald, & Rosenfield, 2015; Miller et al., 2013; Moynihan & Banyard, 2008). Several factors appear to increase the probability of engaging in responsive bystander behavior; some are consistent with Latane and Darley’s (1970) model (feeling responsible for acting, greater self-efficacy), whereas others reflect personal or social factors. For example, individuals are more likely to intervene in a risky situation if they have a relationship with the victim or perpetrator, are younger, female, have a history of victimization, and report peer norms that are unsupportive of sexual coercion (Banyard & Moynihan, 2011; Bennett, Banyard, & Garnhart, 2014; Brown, Banyard, & Moynihan, 2014; Katz, Pazienza, Olin, & Rich, 2015; Palmer, 2016).

However, there also are barriers that can inhibit observers from intervening. Getting involved in a potentially dangerous interaction between other people presents some risk for the bystander. Their involvement may not be welcomed by one or more of the participants or by others who are also present, and could be met by adverse social or physical consequences. Furthermore, many potentially risky situations are ambiguous; it may not be clear if there is a threat, and if so, what the appropriate course of action would be. Most college students, especially those in their first year, are seeking acceptance and friendships in their new context and may view the costs of intervening as outweighing its potential benefits.

Although many studies acknowledge that there are factors that can interfere with students intervening in high-risk situations, only a handful have empirically examined the links between potential barriers and bystander behavior (Bennett et al., 2014; Burn, 2009; Katz, Colbert, & Colangelo, 2015; Katz, Pazienza, et al., 2015b). Burn (2009) developed a questionnaire to assess obstacles to intervention that reflect each stage in Latane and Darley’s (1970) model: failing to notice the event, failing to recognize it as high risk, not feeling responsible for intervening, lacking the skills to intervene, and fearing negative evaluation from peers. She found that all five potential barriers were negatively correlated with college students’ reports of the likelihood that they would intervene in risky situations. Two were unique predictors when all five barriers were analyzed simultaneously: failure to take responsibility, and for women, perceived lack of skills (Burn, 2009). Bennett et al. (2014) used Burns’ measure to examine links between potential barriers and college students’ reports of bystander behaviors that they had engaged in over the prior 2 months. They found that two—failure to take responsibility and skills deficits—correlated with lower levels of intervention. Katz and her colleagues (Katz, Colbert, et al., 2015a; Katz, Pazienza, et al., 2015b) took a different approach to assessing potential barriers by creating hypothetical vignettes that described a situation in which a student was at risk of being sexually assaulted. They found that participants were less likely to indicate that they would help the potential victim when they did not believe it was their responsibility to intervene or reported concerns about whether others would support them if they did.

These studies provide initial evidence of the barriers that are most likely to impede even well-intentioned, well-informed students from engaging in responsive bystander behavior, and suggest that perceived responsibility, self-efficacy, and concerns about peers’ responses may be particularly salient for college students. However, the conclusions that can be drawn from these findings are limited because they assess barriers in relation to general behavioral intentions or beliefs about how participants would act in hypothetical situations rather than obstacles that inhibited participants from intervening in situations that they actually had experienced. Many of the questions on Burn’s (2009) measure are phrased hypothetically; although participants may have encountered some of the situations, they do not need to have personal experience in the situations to answer the questions. For example, one of the responsibility items is “If I saw someone I didn’t know who was at risk for being sexually assaulted, I would leave it up to his/her friends.” Research in social psychology has established that behavioral intentions and beliefs are not strong predictors of actual behavior (Bentler & Speckart, 1979; Kelley & Mirer, 1974; Webb & Sheeran, 2006) and the kinds of situations in which bystander intervention may be helpful often are stressful and anxiety provoking for bystanders and are more complex than can be captured in an item on a questionnaire. Consequently, it is not clear whether the results of these studies accurately represent how participants perceive and respond to potentially dangerous situations that they actually encounter.

Hoxmeier, Flay, and Acock (2018) addressed this issue by assessing whether college students had the opportunity to engage in 12 bystander behaviors that could occur prior to, during, or after a sexual assault and if so, whether they engaged in each behavior. Bystander actions included, “Help your friend who is passed out and being approached or touched by a guy or group of guys” and “Interrupt the situation when you walk in on your friend who is having sex with an intoxicated girl.” Participants also rated how easy it would be to perform each behavior, how approving their friends would be if they did, and how helpful the behavior would be. Compared with those who reported engaging in a given bystander behavior, those who had the opportunity but failed to enact the behavior reported lower efficacy to intervene for eight of the 12 items and lower perceived approval from friends if they engaged in the behavior for seven of the 12. The groups did not differ on how helpful 11 of the 12 behaviors were perceived to be. This study provides insight into factors that inhibited intervention in students who had experienced a particular situation, but assessed only a few potential barriers and only in the context of a sexual assault.

The Current Study

The goal of the present investigation was to increase understanding of barriers that prevent responsive bystander behavior in college students. Knowing why students who had the opportunity to intervene in potentially dangerous situations but did not do so can improve bystander training programs by identifying the factors mostly likely to inhibit intervention. We focused on students in their first semester on campus because they are at heightened risk for assault at this time and often receive training in sexual assault prevention (including bystander intervention) prior to or early in the first semester (Krebs, Lindquist, Warner, Fisher, & Martin, 2009). We built on prior research by examining a wider range of situations that could call for responsive bystander behavior, including both physical and sexual assaults, and by assessing barriers that occur at all of the steps described in Latane and Darley’s (1970) model. In contrast to most studies of bystander behavior on college campuses, we examined participants’ behavior only in situations that they had actually experienced.

We also assessed three factors that might predict whether or not students intervene. First, we examined whether there were gender differences in intervention or barriers to intervening. Prior research has shown that female college students are more likely to engage in bystander behavior than are males, but it is not known whether they perceive barriers to intervening differently. Given that women are more likely to be victims of sexual assault, we anticipated that they would be less likely to report not feeling responsible for intervening when they witnessed other women at risk. We did not make predictions for the other barriers. The second, self-efficacy, has been shown to predict greater willingness to engage in bystander behavior in college students (Banyard, 2008; Banyard & Moynihan, 2011; Banyard et al., 2007; Palmer, 2016), but its relation to behavior in situations that students had actually encountered is less clear. We anticipated that participants with greater self-efficacy for intervening would be less likely to report not knowing what to do as a barrier. Finally, given the potential for risky situations to evoke distress in bystanders, the capacity to effectively regulate emotion may be important for engaging in responsive intervention behavior. More specifically, feeling high levels of anxiety about how to help and/or the potential adverse consequences of intervening may inhibit bystander behavior. Emotion regulation has not been examined as a predictor of bystander behavior previously, and we predicted that students reporting more problems managing emotional arousal would be less likely to intervene and more likely to report lack of efficacy and fear of negative peer evaluations as barriers to intervention.

The present study aimed to answer the following research questions.

Research Question 1: How often do first-year college students intervene in situations that present a risk of sexual and physical violence?

Research Question 2: For students who did not take action in these situations, what do they report were the primary barriers for intervening?

Research Question 3: Do the perceived barriers to intervention relate to students’ (a) gender; (b) self-efficacy to intervene and (c) emotion regulation?

Method

Participants

Participants were recruited from a midsized, private university in the Midwestern United States. Toward the end of the fall semester, 650 students were randomly selected from the first-year class (N = 1,547) and invited via email to take part in an online study investigating how students handle risky situations. Of the 297 (46%) who agreed to participate, 281 (43%) completed the study and received a US$10 e-gift card for participating. The participants (N = 281) ranged in age from 18 to 21 years (mean age = 19 years), and the majority were female (65%) and identified their ethnicity as White (77.9%), with smaller numbers identifying as Asian (8.5%), Latino/a (5.7%), Multiracial (5.3%), Black or African American (2.1%), and American Indian (.40%). Differences between students invited to participate and the final sample could not be determined because demographic information was not available for those who chose not to participate. However, the sample was comparable to the first-year class as a whole in terms of gender, race, and ethnicity. Chi-square goodness-of-fit tests indicated that the sample included somewhat larger percentages of female (65% vs. 57%; χ² = 7.568, p = .01) and White (77.9% vs. 69%; χ² = 10.49, p = .01) students than the first-year class as a whole, but did not differ on other racial or ethnic categories.

Procedure

The study was administered online using Qualtrics survey software. After electronically signing the informed consent form, participants were directed to a survey consisting of a series of self-report measures (described below) that they completed anonymously. The university’s Institutional Review Board approved all procedures.

Measures

Bystander situations

Participants completed the Bystander Situation Questionnaire, which was developed to assess situations experienced by college students that present the threat or occurrence of sexual or physical assault. The situations were chosen from two existing questionnaires, the Bystander Behavior Scale (BBS; Banyard, 2008) and the Conceptual Framework for the Prevention of Sexual Violence through Bystander Intervention (McMahon & Banyard, 2012), and using input from a group of seven female undergraduate research assistants (five, who identify as White; two, who identify as Latina; mean age = 20 years) about the kinds of risky situations that they viewed as most common on campus. The 10 situations selected represent a range of interactions that could precede, occurring during, or follow a sexual or physical assault and include the following: (a) Offensive joke: heard someone make sexist, racist, or homophobic jokes, or catcalls; (b) Drunk and vulnerable: saw an intoxicated person who appeared to be left alone at a party, going home alone, or passed out at a party; (c) Uncomfortable woman: saw a woman at a party or bar looking very uncomfortable with a man or group of men around her; (d) Drugging: saw a drink get roofied or someone appear to be trying to get another person drunk; (e) Questionable intentions: saw an intoxicated person being led away by someone with questionable intentions; (f) Sexual coercion: saw someone acting in a harassing or sexually aggressive manner toward someone else; (g) Physical violence: saw someone act aggressively (i.e., shoving, yelling, controlling) toward a person they were dating or involved with; (h) Bruises: saw someone with unexplained bruises that may be signs of an abusive relationship; (i) Learned of sexual assault: learned or suspected that a friend was sexually assaulted; (j) Victim blaming: heard someone imply or say, “she was asking for it” in reference to a person who had been sexually assaulted.

Students reported whether or not they had encountered each situation within the last 3 months (i.e., since the start of the semester) by checking “yes” or “no.” If they chose “yes,” the next screen asked how many times they had experienced each situation. If participants had been in a situation more than once, they were asked to answer the questions that followed in relation to their most recent experience. They then were asked to indicate whether they had intervened or not. If they indicated that they had, they were presented with a list of behaviors and asked to indicate which they had engaged in; if they indicated that they had not intervened, they were asked why they chose not to and presented with a list of reasons (i.e., barriers) for not responding. The barriers were chosen to reflect the steps in Burn’s (2009) model and worded to fit each situation. For instance, barriers for the “Drunk and Vulnerable” situation included, “I didn’t think anything needed to be done” (Perceived Risk), “I thought someone else would do something” (Perceived Responsibility), “I didn’t know what to do” (Skill Deficit), and “I was afraid that other people would make fun or criticize me if I did something” (Audience Inhibition). Participants also had the option of writing in a response if there was a reason they had failed to intervene that was not listed. Few participants wrote in responses (6%), and those that fit into one of the four types of barriers described above were coded accordingly. Those that did not correspond to any of the categories (many of which consisted of descriptions of the situation or stated that someone else intervened instead) were coded as “Other.” Because the number of “Other” responses is small, they were not analyzed for this study.

Efficacy for intervening

Participants completed 14 items from the Bystander Efficacy Scale (BES; Banyard et al., 2007) to indicate their feelings of efficacy for intervening when witnessing actual or potential sexual assault situations. For each item, students rated how confident they felt performing each particular behavior (0 = can’t do, 100 = very certain can do). Example items include, “Ask a friend if they need to be walked home from a party” and “Do something to help a very drunk person who is being brought upstairs to a bedroom by a group of people at a party.” Item responses were averaged to provide an index of feelings of efficacy with higher scores indicating a greater degree of confidence intervening in potential situations of risk and violence. Only one participant failed to complete all the items, but as only two responses were missing, this participant’s score was computed as the average of the 12 items they answered and included in the analyses. The internal consistency of this measure in the current sample was α = .90.

Emotion regulation

Participants completed the Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004) to measure emotion regulation ability. The DERS is a 36-item self-report questionnaire that assesses characteristic patterns of emotion regulation and includes items such as, “I experience my emotions as overwhelming and out of control” and “I pay attention to how I feel.” Respondents rated their level of agreement with each statement on a 5-point scale (1 = almost never, 5 = almost always). Item responses were summed to provide an index of emotion regulation. The scores on items were reversed, so that higher scores reflect a greater emotion regulation difficulty. The internal consistency of this measure in the present sample was α = .93.

Results

Rates of Experiencing and Responding in Risky Situations

Nearly all of the participants (n = 260, 93%) reported experiencing at least one situation in their first semester on campus in which responsive bystander behavior could have been helpful. The average number of situations encountered was two (M = 2.14, SD = 1.50) but the range was large, from zero to nine. As Table 1 shows, the most common situation was hearing someone make a sexist, racist, homophobic joke, or catcall (n = 239, 85%), and the least common was seeing someone with unexplained bruises that could be signs of an abusive relationship (n = 7, 3%). Many participants also reported being in situations in which someone was at heightened risk for sexual assault; nearly half reported seeing an intoxicated person alone at a party, going home alone, or passed out at a party, and about 20% saw an intoxicated person being led away by someone who appeared to have questionable intentions or witnessed a woman at a party or bar looking uncomfortable with a man or group of men. Participants also reported fairly high rates of bystander behavior (see Table 1). In all but two situations, over half of the participants reported intervening, and for many situations, 75% or more of participants indicated that they did something. The situations that led to the highest rate of bystander behavior were learning or suspecting that a friend had been sexually assaulted (88%), seeing an intoxicated person left alone at or after a party (84%), and seeing a woman who looked uncomfortable with one or more men (82%). The situations in which the fewest participants reported intervening were seeing someone with suspicious bruises (43%; also the least commonly encountered situation) and hearing an offensive joke (56%; the most frequently encountered situation).

Table 1.

Situations of Violence Experienced and Action Response by First-Year Students (N = 281).

Situation(Since the start of the fall semester, have you . . .)	Situation Experiencedn (%)	FrequencyM	Number of Responsen (%)	Actively Respondedn (%)	Othern (%)
1. Offensive joke	239 (85)	5.21	133 (56)	65 (27)	41 (17)
2. Drunk and vulnerable	136 (48)	3.32	18 (13)	115 (84)	3 (2)
3. Questionable intentions	56 (20)	2.14	15 (27)	41(73)	0 (0)
4. Uncomfortable woman	50 (18)	2.37	6 (12)	41 (82)	3 (6)
5. Learned of a sexual assault	34 (12)	1.74	3 (9)	30 (88)	1 (3)
6. Drugging	25 (9)	2.12	5 (20)	19 (76)	1 (4)
7. Victim blaming	21 (8)	1.67	3 (14)	16 (76)	2 (10)
8. Physical violence	19 (7)	1.74	7 (37)	10 (53)	2 (10)
9. Sexual coercion	10 (4)	2.2	4 (40)	6 (60)	0 (0)
10. Bruises	7 (2)	1.57	4 (57)	3 (43)	0 (0)

Note. Percentages in the response columns are calculated using the total number of students who had experienced each situation.

Barriers to Intervention

Although bystander intervention was fairly common in this sample, there were many instances in which participants reported that they did not do anything in a potentially dangerous situation. Across the 10 situations, more than half of the participants (73%) reported that they experienced but did not take action as a bystander in at least one of the situations, and only 27% of participants reported actively responding in all of the high-risk situations that they witnessed. Participants who failed to intervene generally identified more than one factor as a salient barrier, which reflects the complexity of these situations. As Table 2 shows, the frequency of reporting each type of barrier varied across situations, but in nearly every situation the most frequently identified barriers to intervening were not feeling responsible for doing something and not knowing what to do. Failure to identify the circumstances as potentially risky was reported in several situations, but concern about how others in the situation would respond was rarely identified as a barrier to action.

Table 2.

Barriers to Intervention Perceived by Students for Each Situation.

Situation	Perceived Riskn (%)	Perceived Responsibilityn (%)	Skill Deficitn (%)	Audience Inhibitionn (%)	Othern (%)	Barriers ReportedM
1. Offensive joke (n = 133)	36 (27)	72 (54)	55 (41)	20 (15)	13 (10)	1.73
2. Drunk and vulnerable (n = 18)	6 (33)	10 (56)	8 (44)	0 (0)	3 (17)	2.13
3. Questionable intentions (n = 15)	3 (20)	13 (87)	6 (40)	1 (7)	1 (7)	1.92
4. Uncomfortable woman (n = 6)	1 (17)	5 (83)	3 (50)	0 (0)	0 (0)	1.83
5. Learned of a sexual assault (n = 3)	1 (33)	2 (67)	0 (0)	0 (0)	0 (0)	1.00
6. Drugging (n = 5)	1 (20)	4 (80)	4 (80)	0 (0)	0 (0)	1.80
7. Victim blaming (n = 3)	0 (0)	3 (100)	1 (33)	0 (0)	0 (0)	2.00
8. Physical violence (n = 7)	2 (28)	4 (57)	3 (43)	0 (0)	0 (0)	1.86
9. Sexual coercion (n = 4)	0 (0)	0 (0)	2 (50)	0 (0)	2 (50)	1.00
10. Bruises (n = 4)	0 (0)	4 (100)	1 (25)	1 (25)	0 (0)	2.50
Total across situations (n = 159)	46 (29)	93 (58)	72 (45)	21 (13)	19 (12)

Note. Percentages are based on total number that reported no action for each respective situation. They do not sum to 100 because participants could choose more than one barrier for each situation.

Because the number of participants reporting a particular barrier in a particular situation tended to be small, we could not conduct statistical tests to determine whether some barriers were significantly more likely to be reported than others in each situation. However, we did examine whether some barriers were reported more frequently than others across situations and whether men and women differed in the types of barriers they reported. A repeated measures ANOVA was conducted with barrier type as a within-subjects factor with four levels and gender as a between-subjects factor. Due to a sphericity violation (Mauchly’s W = .41, χ² = 230.55, p = .001), a Greenhouse-Geisser correction was used for the tests of within-subjects effects.

As shown in Table 3, the main effect for barrier type was significant, F(1.87, 481.74) = 45.30, p = .001, partial η² = .15, but the main effect for gender was not, F(1, 258) = .64, ns. Post hoc tests then were conducted to evaluate whether there were significant differences among barrier types. Pairwise comparisons utilizing the Bonferroni correction revealed that not feeling responsible for intervening (Perceived Responsibility) was reported significantly more often than failing to perceive the situation as high risk (Perceived Risk), not knowing what to do (Skill Deficit), and fearing that others would respond negatively (Audience Inhibition), while Audience Inhibition was reported significantly less than Perceived Risk and Skill Deficit barriers.

Table 3.

Repeated Measures ANOVA of Barrier Type × Gender.

Source	df	Sum of Squares	Mean Square	F	p	Partial η²
Between subjects
Gender	1	0.56	0.56	0.64	.42	.002
Error	258	224.00	0.87
Total	259	224.56
Within subjects
Barrier type	1.87	51.24	27.44	45.30	.001	.15
Barrier Type × Gender	1.87	9.83	5.26	8.69	.001	.03
Error	481.74	291.80	0.61
Total	485.48	352.87

This main effect was qualified by a significant interaction between barrier type and gender, which was explored by conducting simple main effects tests separately for females and males. The main effect of barrier type was significant for both males, F(1.51, 134.38) = 27.22, p = .001, partial η² = .23, and females, F(2.05, 346.54) = 19.61, p = .001, partial η² = .10. However, pairwise comparisons using the Bonferroni correction showed that the pattern of means across barriers differed by gender (see Table 4). Specifically, for males, failure to take responsibility was cited significantly more frequently than any other barrier (pairwise differences with other barriers all p < .001), and failure to perceive that the situation was dangerous was the second most-endorsed barrier. For females, failure to take responsibility and perceived skill deficits were the most frequently reported barriers (pairwise differences with other barriers all p < .01), and did not differ significantly from each other. Men also reported significantly more failure to take responsibility barriers than did women, t(145.51) = 2.16, p = .03, whereas women reported more skill deficit barriers to intervention than did men, t(228.08) = −2.61, p = .01.

Table 4.

Reported Barriers by Gender (N = 260).

Variables	M	SD
Perceived risk
Male	0.26	.51
Female	0.16	.38
Total	0.19	.43
Perceived responsibility
Male	0.88^a	1.36
Female	0.52^a	1.04
Total	0.65	1.17
Skill deficit
Male	0.21^b	.51
Female	0.41^b	.68
Total	0.34	.63
Audience inhibition
Male	0.04	.21
Female	0.11	.33
Total	0.08	.29

Significant mean difference for males and females at p < .05.

Finally, we tested whether self-efficacy and emotion regulation were associated with participants’ tendency to intervene in the situations that they had encountered and the kinds of barriers to intervening that they reported. We created a variable (“No Intervention”) that represented the total number of times that participants reported experiencing a particular situation but not engaging in any bystander behavior, and four variables representing the number of times that participants who failed to intervene in particular situations identified each barrier as a reason they did not intervene. The correlations of these variables with scores on the BES and DERS are shown in Table 5. The results showed that participants reporting lower global perceived efficacy reported more instances in which they failed to intervene in risky situations that they had encountered. However, emotion regulation did not predict whether or not participants intervened. Global self-efficacy and emotion regulation each were correlated with at least one barrier: Participants reporting greater self-efficacy were less likely to report that they failed to intervene because it was not their responsibility and because they did not know how to intervene, and those reporting greater emotion regulation were less likely to report Audience Inhibition as a barrier to intervening. Finally, we examined whether any of the barriers were particularly strongly associated with participants’ tendency to not intervene in risky situations they had experienced. As would be expected, the number of times participants failed to intervene was significantly correlated with all of the perceived barriers, with Perceived Responsibility and Skill Deficits exhibiting the largest associations (.53, .52) and Perceived Threat (.29) and Audience Inhibition (.14) the smallest.

Table 5.

Correlations and Descriptive Statistics for Study Variables.

Variables	1	2	3	4	5	6	7
1. No intervention	—
2. Perceived risk	.29***	—
3. Perceived responsibility	.53***	.24***	—
4. Skill deficit	.52***	.13*	.57***	—
5. Audience inhibition	.14*	−.10	.31***	.28***	—
6. Bystander efficacy	−.26***	−.07	−.18**	−.14*	−.01	—
7. Emotion regulation	.07	−.09	−.04	.06	.15*	−.04	—
n	260	260	260	260	260	281	281
M	0.94	0.19	0.65	0.34	0.08	74.04	2.30
SD	0.78	0.43	1.17	0.63	0.29	16.97	0.59
Range	0-6	0-2	0-8	0-5	0-2	10-100	1.2-4.2

Note. No intervention, Perceived Risk, Perceived Responsibility, Skill Deficit, and Audience Inhibition are sums of participant responses across the 10 situations. Bystander Efficacy is the mean score on the BES and Emotion Regulation the mean score on the DERS. BES = Bystander Efficacy Scale; DERS = Difficulties in Emotion Regulation Scale.

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

Discussion

The goal of the current study was to better understand the barriers that inhibit college students from engaging in responsive bystander behavior when they witness situations that could present the risk of physical and sexual assault. It builds on prior research in a number of important ways: it assessed situations that students actually had experienced rather than asking about hypothetical situations or behavioral intentions; it addressed a wide range of bystander situations that students may encounter in college; it included broader assessment of the factors that may be barriers to intervening in each situation. The data showed it was common for students in the sample to encounter situations where there is risk for sexual assault. Nearly half had seen an intoxicated person left alone, passed out at a party or leaving a party alone, and approximately 20% reported that they had witnessed an intoxicated person being led away by someone else or a woman at a bar or party looking uncomfortable with a man or group of men. Fewer students reported seeing someone slip drugs into a drink and learned or suspected that a friend had been sexually assaulted, but given that the study was conducted during participants’ first semester on campus, even these situations occurred at a concerning rate.

It is encouraging that the majority of participants reported that they did something to intervene when they witnessed a threatening situation. In eight of the 10 situations, over half of the participants reported that they engaged in bystander behavior, with rates of bystander behavior ranging from 27% (hearing a sexist, racist, or homophobic joke) to 88% (learning or suspecting that a friend had been sexually assaulted). However, there were many situations where participants did not intervene, and only a minority of participants (27%) reported that they intervened in every situation that they encountered. Thus, the findings also underscore the need to understand and address the factors that serve to inhibit responsive bystander behavior.

The situations in which participants were least likely to intervene included hearing an offensive joke and seeing someone with suspicious bruises. Although sexist, racist, and homophobic jokes and comments contribute to a culture that tacitly condones sexual coercion and aggression by devaluing women and sexual and ethnic minorities (Gartner & Sterzing, 2016; Paludi, 1990; Vaux, 1993), they may seem further removed or less threatening to an identifiable individual than most of the other situations. Indeed, over a quarter of the students who reported experiencing this situation did not perceive it as risky or threatening. However, considerably more students reported other barriers, suggesting that they did see the behavior as problematic but either did not feel that it was their responsibility to say or doing something (54%) or did not know what to say or do (41%). Given the ubiquity of this kind of situation, it is important for bystander programs to help students understand how such jokes and comments can create conditions that enable aggressive or assaultive behavior, underscore individuals’ role in changing these conditions, and providing specific guidance in how to intervene effectively. Seeing someone with bruises also is ambiguous, but given that very few students reported this experience, we are reluctant to draw any conclusions about whether the level of responding is representative of how students typically act in that situation.

The measure used in the present study offered participants a list of potential barriers in each situation, and they could endorse as many as they found to be salient. Students’ perceptions of factors that inhibited responding reflect the complexity of the circumstances that call for bystander behavior. Participants who did not intervene reported a range of barriers to responding and on average identified two barriers per situation, indicating that bystanders must overcome multiple potential barriers to intervene. This complexity rarely is reflected in self-report measures, which typically are comprised of items that ask about only a single response or barrier in a given situation. Responses on such measures can be misleading because participants can indicate that a particular barrier was not present in the situation but still have failed to intervene for other reasons.

The most common barrier reported by students was the belief that it was not their responsibility to intervene. This barrier was represented by items such as “I thought someone else would say or do something” and “I felt it wasn’t my place to say or do anything.” This is consistent with the original work on bystander intervention by Latane and Darley (1970), which cited the diffusion of responsibility as a primary obstacle to responsive behavior, and with prior work on bystander behavior on college campuses (Bennett et al., 2014). The context for many of the situations included in the present study involved other people, and for some, potentially large numbers of people (e.g., parties, bars). When there are many people present who could intervene, the expectation that someone else—likely a friend of the vulnerable individual—will do something is likely to be widespread.

Men were more likely to indicate that they did not feel responsible for intervening across situations than were women. This gender difference may be due to a number of factors. First, college students generally are more likely to intervene when their friends are at risk (Banyard, Moynihan, Cares, & Warner, 2014; Bennett et al., 2014; Katz & Nguyen, 2016; Katz, Pazienza, et al., 2015b; McMahon & Farmer, 2009); female participants likely have more female friends than male students do and so may have been more likely to be friends with the woman who was threatened. Second, because women are at greater risk for sexual assault than are men, they may feel more empathy for potential victims and more willing to take responsibility for helping them whether they are friends or not. These findings highlight the importance of expanding students’ sense of collective responsibility for their fellow students, encouraging them to care about the welfare of all and not just their friends. For men in particular, it may also be important to directly address perceptions of social norms and attitudes related to gender and sexuality and to highlight the role that men can play in preventing sexual assault (e.g., Stewart, 2014).

Gender differences also were found in barriers reflecting skill deficits, which included items such as “I didn’t know what to do.” Female participants were more likely than male participants to report that they didn’t intervene because they did not know what to say or do in the situation. Endorsing this skill deficit barrier underscores the idea that recognizing a situation as risky and feeling responsible to do something are not sufficient to motivate action, and argue for the benefit of providing students with specific, concrete skills for appropriately and safely intervening in risky situations. Strategies taught in a one-session program may not be remembered, however, especially when students are in the midst of a stressful situation. Supplementing training programs with apps that can be easily accessed by students on their phones may be more effective for providing students with specific ideas of what to say or do in the moment.

Failure to identify situations as high risk was the second most frequently reported barrier by men, although it did not differ significantly from the frequency of reporting skill deficits or audience inhibition or from women’s mean level on this barrier, and was represented by items such as “I didn’t think anything needed to be done.” Even though it was not among the more prominent barriers reported in the study, recognizing the seriousness and potential implications of situations that could lead to sexual assault is an essential first step for responsive bystander behavior and consequently is an important element to emphasize in prevention programs.

Finally, fear that others who are present in the situation would react negatively to intervening was the least frequently endorsed barrier; in fact, the only situation in which this barrier was common was hearing an offensive joke or comment. As noted above, this type of situation rarely presents an immediate threat of assault, and students who are striving for social acceptance may perceive the potential for criticism by their peers if they “call out” someone for an inappropriate joke or comment to be a greater threat. Students also may believe that not laughing at the joke or not agreeing with the comment is a sufficient way to express their disapproval. It is not clear why audience inhibition was not reported in more situations given that it has been identified as a significant barrier in other studies (e.g., Hoxmeier Hoxmeier, Flay, & Acock, 2015). One possible, if optimistic, explanation for this finding is that increased attention to sexual assault and promotion of bystander intervention on college campuses has reduced fears that attempts to intervene will lead to adverse consequences for the bystander. The format of the questions also may have influenced participants’ responses: Students were presented with a list of possible barriers for each situation, and although they could choose more than one, some barriers may make others less salient. For example, if participants did not feel responsible for intervening, the potential reaction of others in the situation may not be relevant.

We also examined whether bystander behavior and the salience of particular barriers could be predicted by two intrapersonal characteristics, global self-efficacy for intervening and emotion regulation. Individuals who reported more confidence in their ability to intervene in general were more likely to intervene in the specific situations that they had encountered, and were less likely to report that skill deficits and perceived responsibility were barriers in situations where they did not intervene. In contrast, emotion regulation predicted only one barrier, audience inhibition. Although the prospect of intervening may be anxiety provoking, the ability to manage affect was not related to whether participants intervened in the bystander situations that they had experienced or their perception of most barriers. It appears that when students perceive that someone is at risk, feel responsible for helping them, and have confidence in their ability to respond effectively, their emotion regulation capacities are not a bar to intervening.

Limitations and Future Directions

The limitations to the study should be considered when examining the results. First, the sample was predominantly White, and the findings may not generalize to a more ethnically diverse student population. Second, the current study included only students in their first semester on campus. This is an important time to study bystander intervention because students are at particularly high risk for assault during their first year (Krebs et al., 2007), and these students may differ from their older peers in ways that are relevant to both bystander behavior and their vulnerability to sexual coercion and assault. For example, they have had less time to develop strong social connections and less experience with campus social life than older students. Consequently, they may feel less responsibility for helping other students and know less about what kinds of behavior are normative or acceptable in social settings and how to respond in potentially threatening situations. It would be interesting to examine if college juniors and seniors differ in their tendency to intervene and in the barriers they perceive to intervening. Third, the study relied on self-report data, and on participants’ ability to recall specific situations and willingness to report the failure to respond in a risky situation. In an attempt to reduce social desirability, the study was conducted online and responses were collected anonymously, but social desirability still may have influenced the results. It also is possible that students who had encountered more risky situations may have been more inclined to participate in the study. Finally, presenting students with descriptions of the situations reduced the capacity of the study to assess the fifth barrier in Burn’s (2009) model, failure to notice an event. It is not possible to determine whether participants who indicated that they did not encounter a particular event actually were not exposed to the event, or whether they were present but did not realize or notice it.

In summary, this study documents barriers to bystander intervention that college students encounter when they are faced with situations that present elevated risk of sexual assault. The findings have implications for efforts to promote responsive bystander behavior on campus, and suggest that existing efforts can be enhanced by explicitly fostering a sense of collective responsibility for the health and safety of all members of the campus community and teaching students specific behaviors to enact in risky situations through training programs and apps that could be accessed while in the situation.

Footnotes

Acknowledgements

We thank Christina Caiozzo and our undergraduate research assistants for their assistance conducting this study, as well as their helpful comments on a draft of this article.

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.

Author Biographies

Kristen Yule is a clinical psychology doctoral student in the Department of Psychology at Marquette University. Her research focuses on the promotion of resilience and prevention of violence for at-risk youth. Her research interests also include the prevention of sexual violence and reduction of risk for sexual assault victimization among adolescents and young adults.

John Grych, PhD, is professor and chair of psychology at Marquette University. His research has investigated a range of questions related to aggression in close relationships, including the impact of family conflict on children’s development, the origins of physical and sexual aggression in adolescent romantic relationships, and resilience in children exposed to violence and other forms of adversity. He also has published work on intervention and prevention, including the prevention of sexual assault and interpersonal violence.

References

Banyard

V. L.

(2008). Measurement and correlates of prosocial bystander behavior: The case of interpersonal violence. Violence and Victims, 23, 83-97. doi:10.1891/0886-6708.23.1.83

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.

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/cds.0.0083

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. doi:10.1002/jcop.20159

Banyard

V. L.

Plante

E. G.

Moynihan

M. M.

(2005). Rape prevention through bystander education: Bringing a broader community perspective to sexual violence prevention. U.S. Department of Justice. Retrieved from http://www.mcrdpi.marines.mil/Portals/76/Docs/SAPR/SAPR_Bystander%20Research.pdf

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

Bentler

P. M.

Speckart

(1979). Models of attitude-behavior relations. Psychological Review, 86, 452-464. doi:10.1037//0033-295X.86.5.452

Borges

A. M.

Banyard

V. L.

Moynihan

M. M.

(2008). Clarifying consent: Primary prevention of sexual assault on a college campus. Journal of Prevention & Intervention in the Community, 36, 75-88. doi:10.1080/10852350802022324

10.

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

11.

Burn

S. M.

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

12.

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

13.

DeGue

Valle

L. A.

Holt

M. K.

Massetti

G. M.

Matjasko

J. L.

Tharp

A. T.

(2014). A systematic review of primary prevention strategies for sexual violence perpetration. Aggression and Violent Behavior, 19, 346-362. doi:10.1016/j.avb.2014.05.004

14.

Foubert

J. D.

Langhinrichsen-Rohling

Brasfield

Hill

(2010). Effects of a rape awareness program on college women: Increasing bystander efficacy and willingness to intervene. Journal of Community Psychology, 38, 813-827. doi:10.1002/jcop.20397

15.

Gartner

R. E.

Sterzing

P. R.

(2016). Gender microaggressions as a gateway to sexual harassment and sexual assault expanding the conceptualization of youth sexual violence. Affilia, 31, 491-503. doi:10.1177/0886109916654732

16.

Gilboa-Schechtman

Foa

E. B.

(2001). Patterns of recovery from trauma: The use of intraindividual analysis. Journal of Abnormal Psychology, 110, 392-400. doi:10.1037/0021-843X.110.3.392

17.

Gratz

K. L.

Roemer

(2004). Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the Difficulties in Emotion Regulation Scale. Journal of Psychopathology and Behavioral Assessment, 26, 41-54. doi:10.1023/B:JOBA.0000007455.08539.94

18.

Hollander

J. A.

(2014). Does self-defense training prevent sexual violence against women? Violence Against Women, 20, 252-269. doi:10.1177/1077801214526046

19.

Hoxmeier

J. C.

Flay

B. R.

Acock

A. C.

(2015). When Will Students Intervene? Differences in students’ intent to intervene in a spectrum of sexual assault situations. Violence and Gender, 2, 179-184. doi:10.1089/vio.2015.0015

20.

Hoxmeier

J. C.

Flay

B. R.

Acock

A. C.

(2018). Control, norms, and attitudes: Differences between students who do and do not intervene as bystanders to sexual assault. Journal of Interpersonal Violence, 33(15), 2379-2401.

21.

Katz

Colbert

Colangelo

(2015). Effects of group status and victim sex on female bystanders’ responses to a potential party rape. Violence and Victims, 30, 265-278. doi:10.1891/0886-6708.VV-D-13-00099

22.

Katz

Moore

(2013). Bystander education training for campus sexual assault prevention: An initial meta-analysis. Violence and Victims, 28, 1054-1067. doi:10.1891/0886-6708.VV-D-12-00113

23.

Katz

Nguyen

L. J.

(2016). Bystander responses to risk for rape perpetrated by a friend, acquaintance, or stranger. Journal of Aggression, Maltreatment & Trauma, 25, 652-667. doi:10.1080/10926771.2016.1185755

24.

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

25.

Kelley

Mirer

T. W.

(1974). The simple act of voting. The American Political Science Review, 68, 572-591. doi:10.2307/1959506

26.

Kleinsasser

Jouriles

E. N.

McDonald

Rosenfield

(2015). An online bystander intervention program for the prevention of sexual violence. Psychology of Violence, 5, 227-235. doi:10.1037/a0037393

27.

Krebs

C. P.

Lindquist

C. H.

Warner

T. D.

Fisher

B. S.

Martin

S. L.

(2007). The campus sexual assault (CSA) study, final report. Washington, DC: National Institute of Justice, U.S. Department of Justice. Retrieved from www.ncjrs.gov/App/abstractdb/AbstractDBDetails.aspx?id=243011

28.

Krebs

C. P.

Lindquist

C. H.

Warner

T. D.

Fisher

B. S.

Martin

S. L.

(2009). The differential risk factors of physically forced and alcohol- or other drug-enabled sexual assault among university women. Violence and Victims, 24, 302-321. doi:10.1891/0886-6708.24.3.302

29.

Latane

Darley

J. M.

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

30.

Lonsway

Banyard

Berkowitz

Gidycz

Katz

Ross

. . . Edwards

(2009). Rape prevention and risk reduction: Review of the research literature for practitioners. VAWnet. Retrieved from http://scholars.unh.edu/psych_facpub/162

31.

McMahon

(2015). Call for research on bystander intervention to prevent sexual violence: The role of campus environments. American Journal of Community Psychology, 55, 472-489. doi:10.1007/s10464-015-9724-0

32.

McMahon

Banyard

V. L.

(2012). 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

33.

McMahon

Farmer

G. L.

(2009). The bystander approach: Strengths-based sexual assault prevention with at-risk groups. Journal of Human Behavior in the Social Environment, 19, 1042-1065. doi:10.1080/10911350902990304

34.

Miller

Tancredi

D. J.

McCauley

H. L.

Decker

M. R.

Virata

M. C. D.

Anderson

H. A.

. . . Silverman

J. G.

(2013). One-year follow-up of a coach-delivered dating violence prevention program: A cluster randomized controlled trial. American Journal of Preventive Medicine, 45, 108-112. doi:10.1016/j.amepre.2013.03.007

35.

Moynihan

M. M.

Banyard

V. L.

(2008). Community responsibility for preventing sexual violence: A pilot study with campus Greeks and intercollegiate athletes. Journal of Prevention & Intervention in the Community, 36(1-2), 23-38. doi:10.1080/10852350802022274

36.

Najdowski

C. J.

Ullman

S. E.

(2009). Prospective effects of sexual victimization on PTSD and problem drinking. Addictive Behaviors, 34, 965-968. doi:10.1016/j.addbeh.2009.05.004

37.

Orchowski

L. M.

Gidycz

C. A.

Raffle

(2008). Evaluation of a sexual assault risk reduction and self-defense program: A prospective analysis of a revised protocol. Psychology of Women Quarterly, 32, 204-218. doi:10.1111/j.1471-6402.2008.00425.x

38.

Palmer

J. E.

(2016). Recognizing the continuum of opportunities for third parties to prevent and respond to sexual assault and dating violence on a college campus. Crime Prevention & Community Safety, 18, 1-18. doi:10.1057/cpcs.2015.18

39.

Paludi

M. A.

(Ed.). (1990). Ivory power: Sexual harassment on campus. Albany: State University of New York Press.

40.

Stewart

A. L.

(2014). The Men’s Project: A sexual assault prevention program targeting college men. Psychology of Men & Masculinity, 15, 481-485. doi:10.1037/a0033947

41.

Ullman

S. E.

Filipas

H. H.

(2001). Predictors of PTSD symptom severity and social reactions in sexual assault victims. Journal of Traumatic Stress, 14, 369-389. doi:10.1023/A:1011125220522

42.

Vaux

(1993). Paradigmatic assumptions in sexual harassment research: Being guided without being misled. Journal of Vocational Behavior, 42, 116-135. doi:10.1006/jvbe.1993.1008

43.

Webb

T. L.

Sheeran

(2006). Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychological Bulletin, 132, 249-268. doi:10.1037/0033-2909.132.2.249