Preventing Sexual Violence Through Bystander Intervention: Attitudes,Behaviors,Missed Opportunities,and Barriers to Intervention Among Australian University Students

Abstract

The concept of bystander intervention is gaining popularity in universities as a mechanism to prevent sexual violence. Prior research has focused on correlates of bystanders’ intentions to intervene and intervention behaviors in situations where there is a risk of sexual violence. The current study builds on this literature by exploring the nature of missed opportunities, including perceived barriers to intervention. In all, 380 Australian undergraduate university students completed an online survey. Measures included a rape myth acceptance scale, bystander intentions to intervene, actual intervention behaviors, missed opportunities for intervention, and perceived barriers for missed opportunities. Promisingly, students reported high levels of intentions to intervene in situations where there was a risk of sexual violence and reported relatively few missed opportunities to do so when these situations did occur. Intervention behaviors varied by important demographic characteristics such as gender, age, attitudes toward sexual violence, and the nature of the situation. Younger female students, with lower levels of rape myth acceptance, who had previously engaged in bystander intervention behaviors were more likely to report intentions to intervene in future risky situations, and female international students reported fewer missed opportunities for intervention. The most common barrier to intervention for identified missed opportunities was a failure to recognize situations as having a potential risk for sexual violence, and students were most likely to intervene in situations when the opportunity to help a friend in distress arose. This study provides some preliminary empirical evidence about bystander intervention against sexual violence among Australian university students, and identifies unique contexts for intervention and what current barriers to intervention may be.

Keywords

barriers bystander intervention sexual assault sexual victimization missed opportunities

Sexual violence among university students in Australia is receiving increased attention in the research literature. For decades, research internationally has shown that the risk of sexual violence is higher for university students, particularly females, compared with their counterparts in the general population (e.g., Fisher, Cullen, & Turner, 2000; Krebs, Lindquist, Warner, Fisher, & Martin, 2007). According to the Australian Public Safety Survey, approximately 17% of women aged 18 years and over have experienced sexual assault since the age of 15 and women aged 18 to 24 were most likely to have experienced sexual assault in the previous 12 months (Australian Bureau of Statistics [ABS], 2012). In terms of university students, Patton and Mannison (1995) found up to 25% of female undergraduate students experienced unwanted intercourse with their partner. In a more recent study, Cale, Tzoumakis, Leclerc, and Breckenridge (2017) found that approximately one third of female university students experienced some form of sexual coercion by their partners. Importantly, these figures are possibly underestimates given the sampling strategies of these studies and underreporting patterns. Perpetrators are most often known to the victim, with the most likely type of known perpetrator being a partner or date. However, a vast majority of these experiences, whether committed by partners, acquaintances, or strangers, are not reported to police or other authorities (e.g., ABS, 2016). In Australia, rates of sexual assault reported to police are consistently lower than rates of sexual assault reported in corresponding victimization surveys (e.g., ABS, 2005a, 2005b, 2012, 2016).

Sexual violence among Australian university students is also gaining attention in the public sphere where universities are increasingly being called upon to address the problem of sexual violence. This is in spite of limited research on the topic in the Australian context in general, and more specifically, research into prevention. This study aims to take one step forward by providing initial empirical evidence about contemporary prevention and intervention practices involving bystander intervention in the context of sexual violence among university students in Australia.

Sexual Violence Prevention Among University Students

One of the most prominent avenues for sexual violence prevention has involved generally raising awareness of the issue with a focus on women to manage their risk of rape or dangerous situations. Other prominent programs focus on challenging cultural norms of male violence emphasizing the responsibility of, typically, men (e.g., Berkowitz, 1992, 1994; Carmody, 2006). A key critique of these programs has been the narrow focus on gender specific individual-level prevention. The basis for sexual violence stems in part from broader cultural norms and values of societies in which universities are embedded, cultural and organizational characteristics of universities in these contexts, and individual characteristics of victims and perpetrators that converge in situations that can result in sexual violence (e.g., Banyard, Plante, & Moynihan, 2004; Koss & Dinero, 1989; Hall & Barongan, 1997). Banyard and colleagues have advocated for adopting an ecological approach that involves engaging with the broader community across multiple levels of prevention. In the university context, one way this has emerged is through promoting bystander intervention aimed at creating a sense of community and individual responsibility for preventing sexual violence.

Bystander Intervention in Preventing Sexual Violence Among University Students

In the past decade or so, bystander intervention models on university campuses have gained increasing prominence (Banyard et al., 2004). In terms of sexual violence, bystander intervention refers to engaging third parties who are not directly involved in incidents as victims and/or perpetrators but have an opportunity to intervene. Banyard and Moynihan (2011) outlined four key subtypes of bystander intervention behaviors that included the following: (a) dealing with sexual and intimate partner violence specific incidents, where there is an identified victim, such as, telling a friend if you think their drink has been spiked or providing support for a victim after an unwanted sexual experience or physical violence; (b) general party safety behaviors, such as, making sure you leave the party with the same people you came with; (c) general helping friends in distress behaviors, such as, walking a friend home from a party who has had too much to drink; and (d) addressing confronting language, such as, indicating displeasure when you hear a sexist joke or catcalls.

There are obviously a multitude of factors that affect the likelihood that an individual would intervene across any of these contexts. A key focus in the literature along these lines has been the relationship between whether individuals carry intentions to intervene in risky situations in the first place, attitudes toward sexual violence, and how these relate to actual intervention behaviors. Bystander intentions are associated with bystander intervention behaviors; individuals who report higher intentions to intervene in risky situations typically are also more likely to do so (Banyard, Moynihan, & Plante, 2007; Brown, Banyard, & Moynihan, 2014; McMahon, 2010). This association tends to be small, and many other factors play a role in whether an individual actually engages in helpful bystander behavior (e.g., see Moschella, Bennett, & Banyard, 2018).

Rape myth acceptance (RMA; that is, prejudicial, stereotyped, or false beliefs about rape, rape victims, and rapists; Burt, 1980) has been shown to be consistently associated with lower willingness to intervene as a bystander (Banyard, 2008; Banyard et al., 2007; McMahon, 2010) and lower rates of actual intervention behavior (Banyard, 2008; Banyard et al., 2007), particularly in the context of interpersonal violence among university students. Compared with females, males tend to be more endorsing of rape myths, are less likely to carry intentions to, and, actually, intervene in sexual violence situations (Aronowitz, Lambert, & Davidoff, 2012; Banyard, 2008; Banyard & Moynihan, 2011; Banyard et al., 2007; McMahon, 2010). Banyard and Moynihan (2011) also found that students who held attitudes that diminished violence as an important issue, and those with lower levels of awareness around the issue of sexual violence were also less likely to intervene in the context of sexual and intimate partner violence. Increased knowledge around sexual violence increases the likelihood of bystander intervention. Students who report knowing someone who has been sexually assaulted, or who have participated in a course discussing sexual violence have shown higher intentions to intervene (Banyard, 2008; McMahon, 2010) as well as more experience doing so (Banyard, 2008).

When examining the nature and extent of bystander intervention with sexual violence, it is also necessary to distinguish whether or not a lack of intervention behaviors is actually due to the lack of an opportunity to do so. For example, a recent study by Brown et al. (2014) found that, overall, students reported relatively low levels of missed opportunities to intervene in situations of sexual violence. At the same time, they also reported that males were more likely to report missed opportunities compared with females and that participants with lower intentions of engaging in bystander behaviors tend to report more missed opportunities. Therefore, a critical question that warrants exploration is what potential barriers are associated with missed opportunities for intervention in situations of sexual violence.

Social psychologists have long described a sequence of cognitive and behavioral processes that precede intervention behaviors in emergency situations (Latane & Darley, 1970). Burn (2009) identified five barriers to intervention along these lines that included the following: (a) failure to notice, where bystander distraction may lead to a failure to notice the event, for example, being distracted by noise and their own social activities at an event; (b) failure to identify a situation as high risk, where a bystander may not interpret the event as intervention appropriate due to (i) ignorance of sexual assault risk markers, such as, what constitutes sexual consent, (ii) ambiguity of the situation, such as ambiguity regarding the relationship between the potential victim and potential perpetrator, or (iii) pluralistic ignorance, that is, relying on others’ reactions to decide on what to do; (c) failure to take responsibility for intervening, that is, failing to intervene due to not believing it is their responsibility to do so. This may include perceptions of the victim’s “worthiness” or feeling a decreased sense of responsibility due to many other bystanders; (d) skills deficit, where a bystander may not feel confident in knowing what to say or do; and (e) audience inhibition, where a failure to act may be due to fears of embarrassment, awkwardness, and social concerns. This evaluation apprehension may prevent action if the bystander fears making an error by intervening. Therefore, in instances where an individual has the opportunity to intervene in a potentially risky situation, but chooses not to, it is of interest to determine why they chose not to engage.

Study Aims

The aim of the current study is to explore perceived barriers for specific missed opportunities for intervention in situations where there was a risk of sexual violence among a sample of Australian university students. Although previous research has explored barriers to intervention with participants’ intentions to intervene or behaviors, few have controlled for the extent of actual opportunities. Therefore, the current study explores barriers to intervention for actual missed opportunities. A further exploratory aim was to investigate experiences with, and attitudes toward, intervening in situations where there is a real or perceived risk of sexual violence.

Method

Participants

The sample for the current study initially included a total of 397 students enrolled in an undergraduate introductory psychology course at a large university in Australia. The current study specifically focused on 18- to 24-year-old students. Over two thirds of participants were in their first year of university (67%), from a range of different academic majors, including all faculties at the university. A similar proportion were female (68%) and the vast majority were domestic students (90%). The mean age among participants was 19.2 years old (SD = 1.5). Slightly over half of the sample (53%) reported Australian as their ethnicity, just below one fifth (19%) reported Chinese as their primary ethnicity, and the remaining respondents were distributed across other ethnic backgrounds. Approximately 30% of the respondents indicated they were born overseas and the vast majority of respondents indicated they resided in off campus accommodation (89%).

Procedures

Ethics approval was obtained from the Human Research Ethics Advisory Panel at the host university and participants were recruited using the participant pool from the School of Psychology. All students enrolled in this introductory psychology course can participate in studies for partial course credit. Participants aged 18 and over were invited to complete an online survey that lasted approximately 30 minutes. Approximately 1,000 students were able to view the invitation to participate, with 400 spaces being allocated. In all, 16 participants were omitted from the sample because they were older than 25 years of age and four were removed due to erratic responses and/or significant missing data. In total, the final sample used for the current analysis was 380.

Measures

Demographic characteristics

Various demographic characteristics were collected including age, sex, student type, accommodation type, ethnicity, whether the participant knew anyone who had been a victim of sexual assault, and whether the participant had participated in any course or training that discussed sexual violence.

Illinois Rape Myth Acceptance Scale–Short Form (IRMAS-SF)

The IRMAS-SF (Payne, Lonsway, & Fitzgerald, 1999) is a 20 item measure,¹ where participants indicate their level of agreement with rape myth statements on a 7-point Likert-type scale. The items were presented in a randomized order and higher scores indicate greater endorsement of rape myths. The scale has demonstrated adequate reliability and validity by the original authors (Payne et al., 1999), and in the current study, internal reliability for the scale was high (Cronbach’s α = .92).

Intervention scenarios

The current study included 29 situations that were characterized by a potential risk of sexual violence (adapted from Brown et al., 2014). Each situation was presented in a block of questions, in the following order, measuring: (a) intentions to intervene, (b) opportunities to intervene, (c) intervention behaviors or missed opportunities, and (d) perceived barriers to intervention. The 29 situations were presented in a randomized order, with the behaviors and barriers questions only being displayed to participants depending on their answer to the previous question (see Figure 1).

Figure 1.

Example flow of bystander questions.

Bystander intentions

This measure was designed to assess the likelihood as to whether or not an individual would intervene in 29 different situations characterized with a potential risk for sexual violence (see Appendix A). For each item the participants were asked how likely they would be to intervene at some point in the future on a 5-point scale. Reponses regarding intentions to intervene to all 29 items were averaged to create an overall bystander intentions score, where higher scores indicated a higher intention to intervene overall. Internal reliability for the scale was high (Cronbach’s α = .91).

Bystander behaviors

Next, for each of the 29 situations, participants were asked whether they ever had the opportunity to intervene in the specific type of situation during the past 6 months, and if they had the opportunity, whether or not they actually did intervene. Responses were summed to create an overall bystander behavior score, with higher scores indicating more experience with intervening in the context of potential risk of sexual violence.

Missed opportunities

Missed opportunities were measured by calculating each time a participant indicated they had the opportunity to intervene in one of the 29 situations, but subsequently indicated that they did not intervene. The number of missed opportunities across items was summed to create a missed opportunities variable, with higher scores indicating more occasions where a participant indicated they had the opportunity to intervene, but did not.

Barriers to intervention

For each situation where a participant identified a missed opportunity, a follow-up question was presented asking what the main perceived barrier for not intervening was (see Appendix B). These barriers were adapted from Burn (2009) and coded as the following (see Table 1).

Table 1.

Barriers to Intervention.

Code	Barrier
1	Failure to notice
2	Failure to identify situation as high risk—Ignorance of sexual assault risk markers
3	Failure to identify situation as high risk—Ambiguity regarding consent or danger/pluralistic ignorance
4	Failure to take intervention responsibility
5	Failure to intervene due to a skills deficit
6	Failure to intervene due to audience inhibition
7	Other

Social desirability

A 13 item limited disclosure scale (Straus, Hamby, Boney-McCoy, & Sugarman, 2010) was included to control for the possibility of socially desirable responding given the nature of the survey questions. Higher scores indicate a tendency to minimize disclosure of socially undesirable behavior. The internal reliability for the scale was acceptable (α = .71). The scale has demonstrated adequate reliability and validity by the original authors (Straus et al., 2010).

Results

Table 2 shows the descriptive statistics and correlations among the study variables. More than one third of the students (36.6%, Table 2) stated they knew someone who has been a victim of sexual assault, additionally, just over 5% (n = 21) stated they knew someone who had been a victim of sexual assault while the victim was a student at university. Scores on the bystander measures indicated students had overall high levels of intentions to intervene in situations that posed a risk of sexual violence, M = 4.10 (SD = 0.52), and low levels of missed opportunities, M = 3.05 (SD = 2.49), with an average of 11.09 (SD = 4.93) opportunities out of a possible 29 over the past 6 months.

Table 2.

Descriptive Statistics and Correlations Between Variables.

Variable	M (SD) or %	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. Intentions	4.10 (.52)	—
2. Behaviors	8.01 (4.15)	.45**	—
3. Opportunities	11.09 (4.93)	.24**	.86**	—
4. Missed opportunities	3.05 (2.49)	−.28**	.05	.55**	—
5. RMA	1.82 (0.83)	−.42**	−.31**	−.31**	−.09	—
6. Social desirability	2.60 (0.36)	.03	.02	−.01	−.06	−.09	—
7. Know victim	36.6%	.19**	.23**	.26**	.13*	−.26**	−.04	—
8. Training	32.1%	.08	.12*	.11*	.02	−.12*	−.07	.11*	—
9. Gender (female)	67.8%	.37**	.23**	.13*	−.13*	−.30**	−.04	.13*	.11*	—
10. Year (2nd or above)	33.4%	−.12*	−.04	−.01	.05	.11*	−.07	.04	−.06	−.10	—
11. Age	19.21 (1.49)	−.17**	−.10	−.04	.09	.08	−.07	.15**	−.08	−.09	.44**	—
12. Student type (International)	10.0%	−.18*	−.20**	−.24**	−.14**	.34**	−.09	−.11	−.00	−.03	.08	.25**	—
13. Born overseas	29.5%	−.07	−.12**	−.20**	−.19**	.29**	−.08	−.06	.00	−.04	.03	.12*	.52**	—
14. On campus accommodation	10.8%	−.03	−.05	−.06	−.03	−.01	.08	−.03	−.00	.00	−.03	.04	.34**	.24**	—

Note. Some variables were skewed; year in university was recoded as a dichotomous variable and a log transformation was applied to RMA. Gender: male = 0, female = 1; year: first year student = 0, second year or above = 1; student type: domestic = 0, international = 1. Correlations for dichotomous variables are point-biserial. RMA = rape myth acceptance.

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

Intentions to intervene and actual bystander behaviors were moderately correlated (r = .45, p < .01), higher intentions, but not actual behaviors, were associated with fewer missed opportunities (r = –.28, p < .01). Lower RMA was associated with higher intentions (r = –.42, p < .01) and actual behaviors (r = –.31, p < .01). Female gender was positively associated with intentions (r = .37, p < .01), behaviors (r = .23, p < .01), opportunities (r = .13, p < .05), and negatively associated with missed opportunities (r = –.13, p < .05) and RMA (r = –.30, p < .01). Age was negatively correlated with intentions (r = –.17, p < .01), as was year in university (r = –.12, p < .05).

International students tended to report lower bystander intentions (r = –.18, p < .05), lower intervention behaviors (r = –.20, p < .01), fewer opportunities for intervention (r = –.24, p < .01), and higher levels of RMA (r = .34, p < .01). International students also reported lower levels of missed opportunities (r = –.14, p < .01). Student type and being born overseas were also strongly correlated (r = .52, p < .01); therefore, only student type was used in the subsequent regression models as students were not asked how long they had resided in Australia.

Students who indicated they knew someone who had been sexually assaulted reported higher levels of intentions to intervene (r = .19, p < .01), higher intervention behaviors (r = .23, p < .01), more opportunities to intervene (r = .26, p < .01), and lower levels of RMA (r = –.26, p < .01). These students also tended to report having more missed opportunities for intervention (r = .13, p < .05).

Associations Among Bystander Intentions, Behaviors, and Missed Opportunities

Next, regression analyses were used to explore the relative importance of the study variables in terms of associations with bystander intentions, behaviors, and missed opportunities (see Table 3). Demographic variables were entered in the first model, followed by RMA and social desirability in the second model. For bystander intentions, a third model introduced previous bystander behaviors. Due to missing data, the regression analyses were based on 367 cases.

Table 3.

Regression Analyses for Variables Predicting Bystander Outcomes.

	Bystander Behaviors(n = 367)		Bystander Intentions(n = 367)			Missed Opportunities(n = 367)
	Model 1	Model 2	Model 1	Model 2	Model 3	Model 1	Model 2
Age	−0.19 (–.07)	−0.20 (–.07)	−0.04* (–.12)	−0.04* (–.13)	−0.04* (–.11)	0.18 (.11)	0.18 (.11)
Gender (female)	1.71*** (.19)	1.28** (.14)	0.37*** (.33)	0.28*** (.25)	0.23*** (.21)	−0.79** (–.15)	−0.91** (–.17)
Year (2nd or above)	0.15 (.02)	0.25 (.03)	−0.04 (–.04)	−0.02 (–.02)	−0.03 (–.03)	0.05 (.01)	0.06 (.01)
Student type (International)	−2.29** (–.17)	−1.55* (–.11)	−0.20* (–.12)	−0.05 (–.03)	0.01 (.01)	−1.40** (–.17)	−1.26** (–.15)
Know victim	1.62*** (.19)	1.31** (.15)	0.17** (.16)	0.11* (.10)	0.06 (.05)	0.50 (.10)	0.41 (.08)
Training	0.65 (.07)	0.57 (.06)	0.01 (.01)	−0.01 (–.01)	−0.03 (–.03)	0.19 (.04)	0.14 (.03)
RMA		−1.94** (–.18)		−0.40*** (–.30)	−0.33*** (–.25)		−0.48 (–.08)
Social desirability		−0.03 (–.00)		−0.01 (–.01)	−0.01 (–.01)		−0.46 (–.07)
Bystander behaviors					0.04*** (.30)
Constant	9.93	11.49	4.64	4.98	4.55	−.09	1.61
F	9.38***	8.53***	14.89***	16.45***	20.91***	4.49***	3.76***
(df)	(6, 360)	(8, 358)	(6, 360)	(8, 358)	(9, 357)	(6, 360)	(8, 358)
R ²	.14	.16	.20	.27	.35	.07	.08

Note. The first line of data represent unstandardized beta coefficients, the second line represents standardized coefficients. Gender: male = 0, female = 1; year: first year student = 0, second year or above = 1; student type: domestic = 0, international = 1. RMA = rape myth acceptance.

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

The first series of regression models examined the number of actual bystander behaviors engaged in by participants in the previous 6 months (Table 3). Here, being female (β = .19, p < .001), knowing a victim of sexual assault (β = .19, p < .001), and being a domestic student (β = –.17, p < .01), were significantly associated with higher levels of bystander behaviors. In the second model, which introduced RMA and controlled for social desirability, female gender (β = .14, p < .01), knowing a victim (β = .15, p < .01), and being a domestic student (β = –.11, p < .05) still remained significant, while lower RMA was also associated with increased bystander behaviors (β = –.18, p < .01).

The second series of regression models examined intentions to intervene in potential bystander behaviors. In the first model, knowing a victim of sexual assault (β = .16, p < .01), being female (β = .33, p < .001), age (β = –.12, p < .05), and domestic student status (β = –.12, p < .05) were significantly associated with higher levels of intentions to intervene. In the second model controlling for RMA and social desirability, student type was no longer a significant predictor of intentions to intervene. Higher RMA was associated with lower levels of bystander intentions (β = –.30, p < .001), over and above student type. In the final model, age (β= –.11, p < .05), being female (β = .21, p < .001) and lower RMA (β = –.25, p < .001) were associated with intentions to intervene, in addition to prior bystander behaviors (i.e., in the previous 6 months; β= .30, p < .001).

The third series of regression models examined the number of missed opportunities for bystander intervention in the previous 6 months. In the first model, being female (β = –.15 p < .01) and an international student (β = –.17, p < .01) was significantly associated with fewer missed opportunities. In the second model, which introduced RMA and controlled for social desirability, again only being female (β = –.17, p < .01) and international student status (β = –.15, p < .01) were associated with fewer missed opportunities.

Exploring Barriers to Bystander Intervention

Across the sample, a total of 1,159 missed opportunities were reported, 1,151 of which had reported barriers (Table 4). Across all of the barriers, a failure to identify situation as high risk/ignorance of sexual assault risk markers (ignorance barrier) was the most common barrier (40.6%). This was followed by failure to notice the situation (i.e., respondent did not think about the risk for sexual violence; 20.4%) and “other” barriers where participants provided a free response. A failure to take intervention responsibilitywas the least reported barrier (1.9%).

Table 4.

Comparison of Barriers to Intervention Across Type of Bystander Behavior.

Barrier	Helping Friends in Distress	Confronting Language	Party Safety	Dealing With Specific Sexual Violence Incidents	Total(n = 1,151)
1. Notice	6 (10%)	55 (15.7%)	132 (30.8%)	42 (13.5%)	235 (20.4%)
2. Ignorance	19 (31.7%)	152 (43.3%)	174 (40.6%)	122 (39.2%)	467 (40.6%)
3. Ambiguity	0	47 (13.4%)	25 (5.8%)	26 (8.4%)	98 (8.5%)
4. Responsibility	8 (13.3%)	4 (1.1%)	5 (1.2%)	5 (1.6%)	22 (1.9%)
5. Skills	0	31 (8.8%)	9 (2.1%)	19 (6.1%)	59 (5.1%)
6. Audience	1 (1.7%)	23 (6.6%)	3 (0.7%)	10 (3.2%)	37 (3.2%)
7. Other	26 (43.3%)	39 (11.1%)	81 (18.9%)	87 (28%)	233 (20.2%)
Proportion of missed opportunities/total opportunities	6.5%	49.6%	26.0%	33.2%	27.4%

Table 4 also compares the seven barriers to intervention across the four contexts identified by Banyard and Moynihan (2011). Students tended to intervene the most when there was an opportunity to help a friend in distress, with missed opportunities only making up 6.5% of total opportunities for this type of behavior. When a missed opportunity did occur, this was mostly associated with the ignorance barrier (31.7%). The type of behavior with the highest proportion of missed opportunities was confronting language (49.6%). Again, the most common barrier was a failure to identify the situation as high risk (43.3%).

Intervention across the categories of party safety and dealing with specific sexual violence incidents had a similar proportion of missed opportunities (26% and 33.3%, respectively). In the context of party safety, a failure to notice and ignorance/failure to identify a situation as high risk combined made up 71.4% of conveyed barriers. When dealing with specific sexual violence incidents, the most common barrier for students was again, failure to identify the situation as high risk and ignorance (39.2%). Failing to intervene due to failure to take intervention responsibility was most common in helping friends in distress (13.3%) compared with the other types of behaviors. Failing to intervene due to ambiguity or pluralistic ignorance was more common in confronting language opportunities (13.4%), along with skills deficit (8.8%) and audience inhibition (6.6%).

Table 5 outlines the proportion of responses to each barrier within gender and student type. Overall, males reported a higher proportion of missed opportunities (out of total opportunities) as compared with females. A higher proportion of males compared with females reported a failure to identify a situation as high risk, with 45.8% of males and 37.6% of females reporting an ignorance of sexual assault risk markers, and 12.9% of males and 5.9% of females reporting ambiguity or pluralistic ignorance. In contrast, a higher proportion of females seemed to report the barriers of skills deficits (2.8% of males, 6.4% of females) and audience inhibition (1.9% of males, 4.0% of females). In terms of student type, the largest proportional difference in reported barriers was a failure to intervene due to a skills deficit, with 14.5% of international students and 4.5% of domestic students reporting this barrier.

Table 5.

Comparison of Barriers to Intervention Across Gender and Student Type.

Barrier	Gender		Student Type
Barrier	Male	Female	Domestic	International
1. Notice	18.1%	21.9%	20.3%	21.1%
2. Ignorance	45.8%	37.6%	41.1%	31.6%
3. Ambiguity	12.9%	5.9%	8.1%	14.5%
4. Responsibility	1.6%	2.1%	1.8%	3.9%
5. Skills	2.8%	6.4%	4.5%	14.5%
6. Audience	1.9%	4.0%	3.1%	5.3%
7. Other	16.9%	22.1%	21.1%	9.2%
Proportion of missed opportunities/total opportunities	34.7%	24.5%	27.5%	26.8%

A three-way mixed-method ANOVA was used to explore patterns in perceived barriers of reported missed opportunities, within variables that were significantly associated with missed opportunities in the regression analysis, specifically, gender and student type. Mauchly’s test of sphericity indicated violation of the sphericity assumption, χ²(20) = 1,238.7, p < .001, therefore Greenhouse-Geisser correction was used. A main effect of type of barrier was found to be significant, F(3.33, 1246) = 18.38, p < .001. Total missed opportunities varied across participants by barrier type, where a failure to identify a situation as high risk—ignorance of sexual assault risk markers was the most reported barrier. As expected based on the results of the regression analysis, main effects were found to be significant for student type, F(1, 374) = 6.97, p < .01, and marginally significant for gender, F(1, 374) = 3.45, p < .10. Here international students, on average, reported fewer missed opportunities than domestic students and females reported fewer missed opportunities than males.

The only significant interaction found was the two-way interaction between type of barrier and student type, F(3.33, 1246) = 4.73, p < .01. Statistical significance of a simple main effect was accepted at a Bonferroni-adjusted alpha level of .007. There was a statistically significant simple main effect of student type within the ignorance barrier, F(1, 374) = 9.24, p < .01. The mean score of the perceived barrier of ignorance was higher for domestic students than for international students, a mean difference of .76 (p < .01).

Discussion

The findings from the current study provide additional empirical support for key factors associated with bystander intervention behaviors, and also uncover some unique characteristics of bystander intervention against sexual violence among university students in Australia. First, higher intentions to intervene were associated with an increased likelihood of actual intervention behaviors (e.g., Banyard et al., 2007; Brown et al., 2014; McMahon, 2010) and lower missed opportunities (e.g., Brown et al., 2014). Second, younger age, being female, low RMA, and prior experiences intervening were associated with the likelihood of respondents to report intentions to intervene (e.g., Banyard, 2008; Banyard et al., 2007; McMahon, 2010). At the same time, however, age was not significantly associated with actual intervention behaviors or missed opportunities. These results suggest that the relationship between age and bystander outcomes is not straightforward. As others have suggested (e.g. Brown et al., 2014), it may be appropriate to measure a sense of community or bonds to the university culture along with age or year in university to gain a clearer picture of these relationships.

Interestingly, endorsement of rape myths was not associated with missed opportunities, suggesting RMA may not necessarily play a role in the decision to help when an opportunity is in fact present. Banyard and Moynihan (2011) found that RMA was uniquely associated with not only lower intentions to intervene but also higher helpful bystander behaviors, especially in terms of intimate partner violence incidents. Attitudes and behaviors often have complex interactions (see Glasman & Albarracin, 2006) and more research is needed to understand this in the context of sexual violence. For example, Bennett, Banyard, and Garnhart (2013) suggested that both facilitators and barriers to intervention can be better understood by recognizing how individual differences link with situational variables.

The current results support the robust relationship between gender and bystander outcomes (e.g., Banyard, 2008; Banyard et al., 2007; Brown et al., 2014; McMahon, 2010); being male was associated with lower willingness to intervene, lower levels of actual intervention, more missed opportunities and higher RMA compared with females. Gender was the only variable that remained a significant predictor of all bystander outcomes, including missed opportunities (i.e., being male).

Student type was associated with all bystander outcomes in zero-order correlations, where international students had lower intentions to intervene, fewer opportunities and fewer behaviors, however, when the opportunity did present, they tended to intervene, that is, they reported fewer missed opportunities. When controlling for the association between student type and RMA, however, student type was no longer predictive of bystander intentions. Perhaps unexpectedly, types of barriers reported varied between student type; specifically, the barrier of ignorance was higher for domestic students than international students. Some recent studies have begun to explore the role of race in bystander intervention against sexual violence (e.g., Brown et al., 2014). It is likely that cultural differences between international and domestic students play a role in explaining these different patterns of bystander outcomes. For example, the majority of the international students identified as Chinese. Zhong (2010) described an ingrained culture regarding altruism and found higher intentions of helping in middle class Chinese communities where fear of crime was low. In addition, the relationship between the bystander and victim/perpetrator also plays a role (e.g., Bennett & Banyard, 2016; Bennett, Banyard, & Edwards, 2017; Nicksa, 2014); in this context, international students may simply be acquainted with fewer people because of the comparably limited opportunity they have had to make friends, in turn, resulting in fewer barriers to intervention related to knowing the parties involved.

Reported missed opportunities were relatively low among students; more often than not, they reported intervening when they had the opportunity to do so (e.g., Brown et al., 2014). Following the suggestion by Brown et al. (2014), we then examined proportions of missed opportunities (out of total opportunities), as neither behavior nor missed opportunity scores tell the full picture on their own. Although international students reported fewer missed opportunities, they also reported fewer opportunities overall, leaving the proportion of missed opportunities at a similar level to domestic students. However, males had a higher proportion of missed opportunities than females, indicating that although they helped more often than not when the situation arose, they had a tendency to do so less, compared with females.

Overall, the most common barrier to intervention reported was a failure to identify a situation as high risk due to an ignorance of sexual assault risk markers, where participants felt there was no need to intervene in the given situation. In the study of Bennett et al. (2013), a failure to intervene due to perceived skills deficits and failure to take responsibility by far made up the majority of barriers reported. Although these differences may be due to methodological differences it is also possible differences between campus cultures in Australia and the United States played a role. For example, in Australia, there is a comparative lack of fraternities and a minimal numbers of students living on campus compared with the United States. Therefore, one possibility is that situations used in this study may actually carry less of a risk of sexual violence in the Australian context. However another possibility is that Australian university students may have less understanding of sexual assault risk markers, and tend to fail to identify these situations as needing intervention. Developing bystander measures through focus groups in the Australian university context may provide vital information to clarify these differences.

Burn (2009) suggests that in party situations students may be more likely to fail to notice the actual level of risk due to the distractions of surrounding social interactions. The current results were consistent with this idea, where a failure to notice and ignorance of risk markers were the most common barriers reported in intervening during party safety situations. Across the subtypes of intervention situations, participants were most likely to intervene when the opportunity to help a friend in distress arose, where this subtype had the lowest proportion of missed opportunities. Recent research on bystander intervention against sexual violence has begun to look at differences in situations that involve friends versus strangers (e.g., Nicksa, 2014). Katz, Pazienza, Olin, and Rich (2014) found higher intentions to intervene for risky situations involving friends due to a shared social group membership (see also Bennett et al., 2017). Future research should explore missed opportunities and barriers across risky situations involving friends or strangers, as this may have important implications for bystander intervention programs, for example, promoting social identification across all students on campus in an attempt to increase intervention across all types of situations.

Bystander intentions and bystander behaviors are two different measures, often with unique correlates (e.g., Brown et al., 2014). Although we were able to explain 35% of variance associated with bystander intentions, only 16% of behaviors was explained, and even lower, 8% of missed opportunities. Therefore, it is necessary to explore what contributes to intentions as well as actual decisions to intervene. Promising areas of research have begun to look at social norms (e.g., Brown et al., 2014) and relational factors of friends or strangers (e.g., Banyard, Moynihan, Cares, & Warner, 2014; Katz et al., 2014). This is critical because in the current study, sexual assault training was positively associated with bystander behaviors and opportunities, but these relationships were not evident when controlling for gender and student type. Nonetheless, sexual assault training was associated with lower RMA, and lower RMA was a predictor of increased intentions to intervene. In effect, the results suggest that training does have an association with bystander intentions, but it is clear that broader cultural factors also play an important role in actual bystander behavior.

Limitations

The current study surveyed students who were all enrolled in a first year psychology course and therefore is not representative of the general university population. Due to the relatively small sample, comparisons of student type should also be considered preliminary. Furthermore, this study used bystander measures created in an American university social context; Australian university contexts are not characterized by fraternity and sorority systems, and comparably fewer students live on campus grounds in Australia. Following from this, the current study did not specify the context to conceptualize bystander situations (i.e., on campus vs. off campus) and this may have affected the way participants responded. Another limitation is that the sample in the current study may lack diversity as the majority of respondents were Caucasian Australian. As noted by Banyard (2015), bystander actions may vary according to social status, among other dimensions, and we echo the need for future research along these lines in the context of bystander intervention in sexual violence. For example, future research should conduct focus groups and develop bystander measures in the context of sexual violence with diverse groups of university students. Finally, bystander behaviors and missed opportunities were measured as a dichotomous variable (i.e., yes or no). Taking into account the amount of times each behavior had been performed in the previous 6 months may have resulted in different patterns in the findings. McMahon, Palmer, Banyard, Murphy, and Gidycz (2017) have recently suggested this as a main consideration for future research in this area.

Conclusion

The present study adds to the bystander intervention literature by examining missed opportunities and perceived barriers for intervention, while also being the only known study to examine bystander intervention against sexual violence in an Australian university student sample. Many Australian universities do not have any policies, or avenues for reporting, in place for sexual violence on campus. Mikhailovich and Colbran (1999) argued that some Australian universities may be reluctant to investigate or identify the occurrences of sexual assault for fear of negative publicity, uncertainties of legal responsibility, or inadequacies of campus-based safety initiatives. Consequently, this study also serves the purpose of starting a conversation about sexual violence and prevention within Australian university student populations.

Unique contexts for intervention were identified that can inform the continual development of bystander programs. The current findings provide some evidence that simply adapting bystander programs from other international settings without considering unique university/cultural contexts of Australia may hinder effectiveness. Awareness campaigns to increase students’ abilities to identify risky situations should precede bystander programs that focus on other barriers such as specific skills deficits. Given differences in perceived barriers to intervention across different types of students, further cultural considerations should be considered for creation and implementation of any bystander interventions on campuses. Furthermore, participants were most likely to intervene to help a friend in distress, therefore, promoting collective responsibility for preventing sexual violence may also enhance the effectiveness of intervention programs and campaigns. Importantly, bystander intervention should not be considered a panacea to the problem of sexual violence and prevention and intervention initiatives need to be based on empirical evidence and demonstrated effectiveness.

Footnotes

Appendix A

Appendix B

Barriers to Intervention.

Code	Barrier	Recode	Barrier
1	At the time, I did not really notice the situation (I did not think about the risk of sexual violence)	1	Failure to notice
2	I did not think the situation required intervention (I did not think there was a risk of sexual violence)	2	Failure to identify situation as high risk—Ignorance of sexual assault risk markers
3	I found it hard to tell whether there was a risk of sexual violence	3	Failure to identify situation as high risk—Ambiguity regarding consent or danger/pluralistic ignorance
4	Other people around me did not seem concerned	3	Failure to identify situation as high risk—Ambiguity regarding consent or danger/pluralistic ignorance
5	I did not know what to say or do (I did not feel I had adequate skills to intervene)	5	Failure to intervene due to a skills deficit
6	I left it for others to intervene (I thought someone else would intervene)	4	Failure to take intervention responsibility
7	I felt less responsible to intervene because the person put themselves at risk (e.g., intoxicated, dressed, or acted provocatively etc.)	4	Failure to take intervention responsibility
8	I was worried about the reaction or thoughts of others around me	6	Failure to intervene due to audience inhibition
9	Other	7	Other

Acknowledgements

We gratefully acknowledge the assistance of our colleagues, Professor Richard Kemp and Associate ProfessorKristy Martire, University of New South Wales, who provided expertise that assisted the research. We also would like to thank the annonymous reviewers for their constructive comments.

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

ORCID iD

Rachel Kania

Author Biographies

Rachel Kania completed a Master of Psychology (Forensic) at the University of New South Wales in Sydney, Australia, in 2015. Rachel has since been employed as a practicing psychologist working with young people involved in the criminal justice system. She has a particular interest in prevention and intervention in the area of sexual violence.

Jesse Cale is currently a senior lecturer in Criminology at the University of New South Wales in Sydney, Australia. He completed his PhD in Criminology from Simon Fraser University in Vancouver, Canada in 2010 and also previously worked as a research and policy analyst for the provincial government of British Columbia, Canada. His main areas of research interest are sexual offenders and offences, developmental and life-course criminology, victimisation and victimology, and crime prevention and social policy.

References

Aronowitz

Lambert

C. A.

Davidoff

(2012). The role of rape myth acceptance in the social norms regarding sexual behavior among college students. Journal of Community Health Nursing, 29, 173-182. doi:10.1080/07370016.2012.697852

Australian Bureau of Statistics (ABS). (2005a). Crime and safety, Australia. ABS cat no. 4509.0. Canberra: ABS.

Australian Bureau of Statistics (ABS). (2005b). Recorded crime—Victims, Australia. ABS cat no. 4510.0. Canberra: ABS.

Australian Bureau of Statistics (ABS). (2012). Personal safety, Australia. ABS cat no. 4906.0. Canberra: ABS.

Australian Bureau of Statistics (ABS). (2016). Personal safety, Australia. ABS cat no. 4906.0. Canberra: ABS.

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

(2015). Toward the next generation of bystander prevention of sexual and relationship violence: Action coils to engage communities. New York, NY: Springer International Publishing. doi:10.1007/978-3-319-23171-6

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

10.

Banyard

V. L.

Moynihan

M. M.

Plante

E. G.

(2007). Sexual violence prevention through bystander education: An experimental evaluation. Journal of Community Psychology, 35, 463-481. doi:10.1002/jcop.20159

11.

Banyard

V. L.

Plante

E. G.

Moynihan

M. M.

(2004). Bystander education: Bringing a broader community perspective to sexual violence prevention. Journal of Community Psychology, 32, 61-79. doi:10.1002/jcop.10078

12.

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

13.

Bennett

Banyard

V. L.

Edwards

K. M.

(2017). The impact of the bystander’s relationship with the victim and the perpetrator on intent to help in situations involving sexual violence. Journal of Interpersonal Violence, 32, 682-702. doi:10.1177/0886260515586373

14.

Bennett

Banyard

V. L.

Garnhart

(2013). 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

15.

Berkowitz

A. D.

(1992). College men as perpetrators of acquaintance rape and sexual assault: A review of recent research. Journal of American College Health, 40, 175-181. doi:10.1080/07448481.1992.9936279

16.

Berkowitz

A. D.

(1994). A model acquaintance rape prevention program for men. New Directions for Student Services, 1994(65), 35-42. doi:10.1002/ss.37119946505

17.

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

18.

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

19.

Burt

M. R.

(1980). Cultural myths and supports for rape. Journal of Personality and Social Psychology, 38, 217-230. doi:10.1037/0022-3514.38.2.217

20.

Cale

Tzoumakis

Leclerc

Breckenridge

(2017). Patterns of intimate partner violence victimization among Australia and New Zealand female university students: An initial examination of child maltreatment and self-reported depressive symptoms across profiles. Australia & New Zealand Journal of Criminology, 50, 582-601. doi:10.1177/0004865816666615

21.

Carmody

(2006). Preventing adult sexual violence through education. Current Issues in Criminal Justice, 18, 342-356.

22.

Fisher

B. S.

Cullen

F. T.

Turner

M. G.

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

23.

Glasman

L. R.

Albarracin

(2006). Forming attitudes that predict future behavior: A meta-analysis of the attitude-behavior relation. Psychological Bulletin, 132, 778-822. doi:10.1037/0033-2909.132.5.778

24.

Hall

G. C. N.

Barongan

(1997). Prevention of sexual aggression: Sociocultural risk and protective factors. American Psychologist, 52, 5-14. doi: 10.1037/0003-066X.52.1.5

25.

Katz

Pazienza

Olin

Rich

(2014). 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

26.

Koss

M. P.

Dinero

T. E.

(1989). Discriminant analysis of risk factors for sexual victimization among a national sample of college women. Journal of consulting and clinical psychology, 57, 242. doi:10.1037/0022-006X.57.2.242

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 (Document No. 221153). Washington, DC: U.S. Department of Justice.

28.

Latane

Darley

J. M.

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

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

Palmer

J. E.

Banyard

Murphy

Gidycz

C. A.

(2017). Measuring bystander behavior in the context of sexual violence prevention: Lessons learned and new directions. Journal of Interpersonal Violence, 32, 2396-2418.doi:10.1177/0886260515591979

31.

Mikhailovich

Colbran

(1999). Changing the rules: Responding to sexual assault in an Australian university. Journal of Higher Education Policy and Management, 21, 71-80. doi:10.1080/1360080990210106

32.

Moschella

E. A.

Bennett

Banyard

V. L.

(2018). Beyond the situational model: Bystander action consequences to intervening in situations involving sexual violence. Journal of Interpersonal Violence, 33(20), 3211-3231.

33.

Nicksa

S. C.

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

34.

Patton

Mannison

(1995). Sexual coercion in dating situations among university students: Preliminary Australian data. Australian Journal of Psychology, 47, 66-72. doi:10.1080/00049539508257502

35.

Payne

D. L.

Lonsway

K. A.

Fitzgerald

L. F.

(1999). Rape myth acceptance: Exploration of its structure and its measurement using the Illinois rape myth acceptance scale. Journal of Research in Personality, 33, 27-68. doi:10.1006/jrpe.1998.2238

36.

Straus

M. A.

Hamby

S. L.

Boney-McCoy

Sugarman

(2010). Manual for the personal and relationships profile (PRP). Durham: Family Research Laboratory, University of New Hampshire.

37.

Zhong

L. Y.

(2010). Bystander intervention and fear of crime: Evidence from two Chinese communities. International Journal of Offender Therapy and Comparative Criminology, 54, 250-263. doi:10.1177/0306624X08331214