Bystander Responses to Intimate Partner Misconduct: Barriers to Responding to Physical and Sexual Violations

Abstract

All forms of unchecked acts of violence against women may harm individual women while also normalizing the ways in which women are routinely violated. Violence against women manifests across a continuum of linked behaviors, yet few studies have investigated bystander responses to less extreme forms of intimate partner violence. We examined bystander responses to different forms of misconduct: physical (grabbing and imminent slapping) or sexual (groping and unwanted kissing). Undergraduates (N = 402) read and responded to dating conflict scenarios in which they witnessed a young man verbally insult a young woman while perpetrating either sexual or physical misconduct. Across conditions, 42% of participants described misconduct as abusive, although this was significantly more common among those assigned to the physical (52%) than sexual (32%) conditions. Compared with those in the sexual misconduct condition, participants in the physical misconduct condition reported greater intent to directly intervene. Furthermore, participants in the physical misconduct condition also reported more barriers to intervention, including less awareness/attention to misconduct, less perceived danger to the victim, and less personal responsibility to intervene. In multivariate analyses, less awareness/attention to misconduct and less personal responsibility uniquely predicted lower intent to intervene; these same barriers also explained the tendency for bystanders to report lower intent to intervene in response to sexual than physical misconduct. These results suggest the need for education to promote awareness of the continuum of violence against women. Education also is needed to increase feelings of personal responsibility to challenge the normalization of less extreme violent acts.

Keywords

dating violence sexual assault sexual harassment

Intimate partner violence is a common problem. About 35% of women worldwide report lifetime partner physical or sexual victimization (World Health Organization [WHO], 2017). Similarly, in a nationally representative sample of U.S. adolescents, 18.5% of young women reported partner physical victimization and 13.6% reported sexual victimization (Ybarra, Espelage, Langhinrichsen-Rohling, Korchmaros, & Boyd, 2016). Prevalence rates across studies vary based on many factors, including which discrete behaviors “count” in operational definitions of partner violence.

Feminist scholars have conceptualized violence against women across a continuum of linked behaviors (e.g., Kelly, 1988, 2012; Stout, 1991). On one side of this continuum are more extreme acts such as battering, rape, and murder. On the other side of the continuum are less extreme acts such as sexist comments and threats. Located between these two extremes are intermediate forms of physical or sexual violence, such as grabbing and groping. In the current article, for clarity, we refer to these intermediate forms of violence as misconduct. Importantly, more and less extreme forms of violence may overlap and are not clearly discrete; in addition, rather than implying a hierarchy of harm, the continuum refers to prevalence (see Kelly, 1998, 2012). For example, Ybarra et al. (2016) found that the most common types of partner physical violence reported by young women were “threw something at you” (7.6%) and “pushed, grabbed, kicked, shoved, or hit you” (6.4%), and the most common type of sexual violence was “your partner kissed, touched, or did something sexual when he or she knew you didn’t want to” (11%) (p. 1092). Fewer young women reported more extreme acts such as choking (2%) or forced sex (4%).

Bystanders who witness any type of potential or actual violence against women may intervene to interrupt and promote the woman’s safety. Yet, overall, the available research suggests that bystanders believe that more extreme experiences on the continuum of violence warrant bystander intervention (Ermer, Roach, Coleman, & Ganong, 2017). Bystanders also report greater willingness to intervene in response to more extreme than less extreme forms of violence. For example, bystanders reported greater intent to intervene in response to physical than verbal aggression enacted by a man against a female partner (Chabot, Tracy, Manning, & Poisson, 2009). Likewise, bystanders reported greater intent to intervene in response to risk for a potential incapacitated rape than in response to actual nonconsensual sexual groping (Bennett, Banyard, & Edwards, 2017). In general, researchers tend to investigate bystander responses to extreme forms of violence against women (e.g., Burn, 2009; Fischer, Greitemeyer, Pollozek, & Frey, 2006; Katz & Nguyen, 2016); fewer studies have focused specifically on understanding bystander responses to less extreme forms of violence.

This is an important gap in the literature. Given the higher prevalence of less extreme forms of violence, there may be many more opportunities for bystanders to respond to misconduct than to rape or battering. Likewise, given that acts of misconduct are perceived as more “normal” than extreme violence (Kuijpers, Blokland, & Mercer, 2017), misconduct may be more often enacted in public where bystanders are present. Understanding responses to intimate partner misconduct in particular is important given that onlookers tend to minimize the harm that partners or ex-partners inflict on women via physical and sexual violence (e.g., Ewoldt, Monson, & Langhinrichsen-Rohling, 2000; Katz & Rich, 2017; Shotland & Straw, 1976). Yet regardless of the perpetrator’s identity, all forms of violence against women function to constrain women’s autonomy while imposing risk for emotional harm such as feelings of fear, vulnerability, intimidation, or humiliation. In public settings, unchecked acts of violence against women may harm individual victims while also normalizing the ways in which women are routinely violated (Kelly, 1988; Manne, 2018). Bystanders can challenge this normalization.

The current study investigated bystanders’ responses to witnessing male intimate partner misconduct against a woman. In past research, young adult bystanders reported being reluctant to act in response to dating or sexual assault unless an assault meets a certain threshold of observable harm involving physical injury or intense victim distress (Edwards, Rodenhizer-Stämpfli, & Eckstein, 2015). This finding suggests the need to identify specific barriers to responding to physical and sexual misconduct in the absence of observable physical or emotional injury. Based on Latané and Darley’s (1970) classic model of bystander behavior, the first three barriers to intervention are a failure to notice or attend to the event, a failure to perceive the event as dangerous or severe enough to warrant bystander intervention, and a failure to feel personally responsible to address the event. In past studies of responses to extreme violence against women, bystanders intervened more when they showed greater awareness/attention to the violence (Fischer et al., 2006), when the violence was perceived as more severe or harmful to the victim (Bennett et al., 2017), and when they felt greater personal responsibility to intervene (Burn, 2009; Gracia, Garcia, & Lila, 2009). These same factors also may be associated with behavioral responses to misconduct, although to our knowledge, no studies have investigated the independent effects of awareness/attention, perceived danger, and personal responsibility on intent to intervene in misconduct situations.

The current study also compared bystander responses to physical versus sexual misconduct. Multiple studies have assessed bystander responses to both physical and sexual violence without directly comparing responses to each type (e.g., Bennett, Banyard, & Garnhart, 2014; Edwards et al., 2015; Palmer, 2016; Palmer, Nicksa, & McMahon, 2018). We hypothesized that, when compared with sexual misconduct, physical misconduct may be more recognizable as abusive/violent and more likely to elicit direct bystander intervention. According to Harding’s (2015) description of rape culture in contemporary society, people generally tend to trivialize sexual violations, viewing victims with skepticism while offering offenders an automatic benefit of the doubt. For example, bystanders may assume a couple has past history of consensual sexual activity, and this assumption may appear to imply consent for future sexual activity (Ewoldt et al., 2000; Monson, Langhinrichsen-Rohling, & Binderup, 2000; Shotland & Goodstein, 1992), which may decrease intervention responses to sexual misconduct. In contrast, most bystanders are unlikely to assume that victims have implicitly consented to partner physical misconduct.

We also hypothesized that potential barriers to misconduct in general (i.e., less awareness/attention, less perceived danger to the victim, and less personal responsibility) would be more common in response to sexual than physical misconduct. Sexual misconduct may attract less attention than physical misconduct, in part, because sexual misconduct involves acts that often occur in private and that could appear to be public displays of affection. Sexual misconduct may also be seen as less dangerous than physical misconduct because sexual acts are unlikely to result in physical injury (McGregor, 2012). In fact, if enacted with consent, these same acts might be expected to elicit feelings of pleasure. Bystanders also may feel less personally responsible to intervene in response to sexual than physical misconduct because sexual attention from men may seem complimentary and perhaps “wanted” or “invited” by the victim (Harding, 2015; Vera-Gray, 2017), and when women are seen as behaving provocatively, bystanders feel less personal responsibility to intervene to help them (Burn, 2009).

This study investigated college students’ responses to misconduct perpetrated against a woman by a male partner. First, we explored how often bystanders characterized misconduct as reflecting abuse; it was expected that abuse would be more often identified in response to physical than sexual misconduct (Hypothesis 1). Second, across conditions, we hypothesized that bystanders who reported more awareness/attentiveness to misconduct (Hypothesis 2), who perceived misconduct as more dangerous to the victim (Hypothesis 3), and who felt more personally responsible to intervene (Hypothesis 4) would report greater intent to directly intervene. Given that these perceptions may be interrelated, we sought to explore which perceptions uniquely predict intervention. Next, given the general trivialization of sexual violations specifically, we hypothesized that bystanders would be more likely to directly intervene in response to physical than sexual misconduct (Hypothesis 5) and that bystanders would report more awareness/attention to the physical misconduct (Hypothesis 6), perceive the physical misconduct as more dangerous to the victim (Hypothesis 7), and feel more personally responsibility to intervene (Hypothesis 8). We also planned to test which specific barriers to intervention associated with different types of misconduct might uniquely explain the expected tendency for bystanders to intervene more in response to physical than sexual misconduct.

Method

Participants

Data were collected from 402 undergraduates (77.6% women) at a small public college in the Northeastern United States. Although the majority (79.3%) identified as White/European American, participants also identified as Asian or Asian American (7.8%), Hispanic/Latinx (7.1%), Black/African American (4.3%), or Other (1.6%). The mean age of participants was 18.85 years (SD = 1.36, range = 17-22). Just under half of the participants were freshman (46.6%; n = 185), about a third were sophomores (33.2%; n = 132), and others were juniors (12.3%; n = 49) or seniors (7.9%; n = 31).

Manipulation

Participants were randomly assigned to read about witnessing a conflict in which a man is verbally aggressive toward a woman while also perpetrating acts of either physical or sexual misconduct. The acts of physical misconduct (grabbing her arm, raising a hand to slap her) were based on the “minor” Violence subscale of the Revised Conflict Tactics Scales (Straus, Hamby, Boney-McCoy, & Sugarman, 1996). The acts of sexual misconduct (grabbing her butt, kissing her neck) are similar to behaviors studied by Bennett and Banyard (2016) in their low severity sexual violence condition. Phrases included only in the sexual misconduct condition are underlined; the one phrase included only in the physical misconduct condition is italicized.

It is a Saturday afternoon. You’re at the mall waiting for a friend. You sit at a table in the nearly empty food court when you notice a young couple about your age (Steve and Sarah) arguing nearby. You can’t see or hear everything, but you see the guy (Steve) try to grab his girlfriend’s (Sarah) butt and call her “a whore.” You see Sarah brush his hand away and turn around, as if to leave. In response, Steve continues to grab Sarah’s butt and then starts to kiss Sarah’s neck/grabs her arm to pull her back and then raises his hand as if he’s about to slap Sarah. Sarah winces, notices you, and the two of you make eye contact.

Measures

Two undergraduate women coded participants’ qualitative responses to a single open-ended question: “Please briefly write what you think is happening, what you would do, and why. Please be as honest as you can. There are no right or wrong answers.” Responses were coded for perceptions of either the behaviors or relationship as abusive, assaultive, or involving domestic/dating violence. Eight participants provided no response. A sample response reflecting abuse is “Steve is abusing Sarah. I would try to find a security guard because I would want to help but don’t know how.” There was good interrater agreement across coders (κ = .86).

For exploratory purposes, responses also were coded for three types of constructive behavioral intent based on McMahon, Hoffman, McMahon, Zucker, and Koenick (2013): to help the victim, to confront the perpetrator, and to delegate/recruit others to intervene. A sample response for helping the victim is “It looks like Steve is assaulting Sarah. I would pretend to know Sarah and try to help her exit the situation and stay with her in a safer place.” A sample response for confronting the perpetrator is “I think Steve and Sarah are in an abusive relationship. I would tell Steve to back off because hitting a woman is awful and sucks.” A sample response coded as reflecting intent to delegate is “I think Sarah is being abused. I would notify a security guard that a man is abusing a woman and that she is visibly in distress.” In a few cases, multiple intentions were described and coded; for example, “Steve and Sarah are arguing and it got out of hand and Steve is about to hit Sarah. I go over there and say something or call for help” was coded as reflecting both helping the victim and delegating. There was moderate to good interrater agreement for codes for helping the victim (κ = .77), confronting the perpetrator (κ = .75), and delegating (κ = .82).

Intent to intervene directly to help the victim of misconduct was assessed with three items from Katz and Nguyen (2016): (a) “try to talk to the girl,” (b) “ask the girl if she is okay,” and (c) “offer to walk the girl away from the situation.” Participants rated how likely they would be to enact each behavior on a 7-point scale (1 = strongly disagree, 7 = strongly agree). Responses were averaged such that higher scores indicated greater intent to intervene. Katz and Nguyen (2016) reported evidence for reliability and convergent validity given significant associations with theoretically related constructs in a similar sample. In the current sample, the estimate of internal consistency was good (Cronbach’s α = .95).

Awareness/attention toward the misconduct was assessed with three items from Fischer et al. (2006): (a) “I recognized very quickly that something was going wrong there,” (b) “I attended to the situation very quickly,” and (c) “My attention toward the situation was very high.” Each item was rated on an 11-point scale (0 = not at all, 10 = definitely) and averaged; higher scores reflect greater awareness/attention. The scale authors showed an association between greater awareness and reaction time in response to male violence against a woman. The estimate of internal consistency in the current sample was good (Cronbach’s α = .92).

Perceived danger was assessed with three items adapted from Levine, Cassidy, Brazier, and Reicher (2002): (a) “How serious was this incident?” (b) “How violent was this incident?” and (c) “In how much danger was the girl?” Responses were made on a 7-point scale ranging from 1 (not at all) to 7 (extremely) and averaged such that higher scores reflected greater perceived danger to the victim. In the current sample, the estimate of internal consistency was acceptable (Cronbach’s α = .86).

Personal responsibility to actively intervene was assessed with three items from Fischer et al. (2006): (a) “I felt personally responsible for helping in that situation,” (b) “The girl was in need and therefore I was responsible to help,” and (c) “I recognized injustice and therefore felt responsible to help.” Each item was rated on an 11-point scale (0 = not at all, 10 = definitely) and averaged such that higher scores reflected greater personal responsibility. In the current sample, the estimate of internal consistency was good (Cronbach’s α = .94).

Social desirability was assessed with the Marlowe–Crown Social Desirability Scale (Crowne & Marlowe, 1960), a 33-item true/false measure of a respondent’s need for social approval. A representative item is “I have never intensely disliked anyone.” Evidence for convergent and discriminant validity has been reported (Marlowe & Crowne, 1961). Higher scores reflect a greater motivation to present oneself in a manner consistent with perceived social and cultural expectations. The estimate of internal consistency in the present sample was adequate (Cronbach’s α = .70).

Procedure

Undergraduate students were recruited through a voluntary human participants pool for “College Students’ Responses to Stressful Situations.” Data collection sessions were held in classrooms and lasted no more than 1 hr. Participants were seated in alternating rows to ensure privacy. After providing informed consent, participants responded to a self-report measure of social desirability. They also were randomly assigned to read about a scenario involving either physical or sexual misconduct; in all scenarios, the perpetrator, a young man, enacts verbal aggression by calling a young woman a “whore.” Participants were asked an open-ended question about their response and then were asked to complete measures of intent to intervene and perceptions of the situation. Finally, participants submitted their materials to a slotted box for privacy and were fully debriefed. Participants earned course credit for their time. The Institutional Review Board approved all study materials and procedures.

Results

Table 1 shows the frequencies for each of the codes for the open-ended responses across the entire sample and separately for the physical and sexual misconduct conditions. Across conditions, just less than half of the sample characterized the situation as involving abuse/assault/violence. Hypothesis 1 was that physical misconduct would be described as abusive more often than sexual misconduct. In support of this hypothesis, about half of participants in the physical misconduct condition but only about a third of those in the sexual misconduct condition described having witnessed relationship abuse, χ²(1) = 14.62, p < .001, ϕ = .19. Also as shown in Table 1, the modal behavioral intent that participants described was to intervene to try to help the victim, which was reported significantly more often in response to physical than sexual misconduct, χ²(1) = 17.26, p < .001, ϕ = 21. Just under a quarter of the sample reported intent to either delegate or confront the perpetrator. Although more participants reported an intent to delegate in response to misconduct that was physical than sexual, χ²(1) = 5.40, p = .02, ϕ = .12, there was no difference between conditions in reported intent to confront, χ²(1) = 2.55, p =.11, ϕ = .08.

Table 1.

Coded Open-Ended Responses to Witnessing Physical and Sexual Misconduct (N = 394).

Code	Total(N = 394)		Physical(n = 200)		Sexual(n = 194)
Code	n	%	n	%	n	%
Abusive behavior/relationship	166	42.1	103	51.5	63	32.5
Try to help victim	186	47.2	115	57.5	71	36.6
Try to confront perpetrator	95	24.1	55	27.5	40	20.6
Try to delegate	93	23.5	57	28.5	36	18.6

Note. Because individual responses may have contained multiple behavioral intentions, the total percentages across types of behavioral intent were greater than 100%.

Descriptive statistics and zero-order correlations (Pearson’s r for continuous variables, Spearman’s rho for categorical) among all the other study variables as well as social desirability are reported in Table 2. The three barriers to intervention (low awareness/attention, low perceived danger to the victim, and low personal responsibility) were positively intercorrelated. Although these associations were not strong enough as to suggest redundancy, awareness/attention and personal responsibility were the most strongly interrelated potential barriers.

Table 2.

Descriptive Statistics and Bivariate Correlations.

	M (SD)	1	2	3	4	5
1. Type of misconduct^a	—	—
2. Intent to intervene^b	5.40 (1.64)	.10*	—
3. Awareness/attention^c	7.49 (1.82)	.12*	.53***	—
4. Perceived danger^b	4.71 (1.97)	.35***	.30***	.45***	—
5. Personal responsibility^c	6.96 (2.45)	.14*	.64***	.70***	.42***	—
6. Social desirability	13.99 (4.46)	−.04	.14**	.12*	−.01	.14**

Type of assault was coded as 0 = sexual, 1 = physical.

Possible scores ranged from 1 to 7.

Possible scores ranged from 0 to 10.

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

For exploratory purposes, potential gender differences in responses to misconduct were investigated. On average, women reported greater overall awareness (M = 7.80, SD = 1.73) than men (M = 7.07, SD = 2.08), t(395) = 2.42, p = .016, and women also perceived greater danger to the victim (M = 4.78, SD = 1.20) than men (M = 4.48, SD = 1.15), t(395) = 2.08, p = .04. However, there were no gender differences in either intent to intervene, t(395) = 0.89, p = .38, or personal responsibility, t(395) = 0.14, p = .88.

It was expected that, across conditions, intent to intervene in response to misconduct would be positively associated with awareness of/attention to the misconduct (Hypothesis 2), perceptions of the misconduct as dangerous to the victim (Hypothesis 3), and personal responsibility to intervene (Hypothesis 4). As shown in Table 2, all of these hypotheses were supported.

A potential limitation in the current research study is that participants might be motivated to be perceived in a positive light, which might affect their responses to the dependent measures. That is, the greater the participants’ need for social approval, as indexed by the measure of social desirability, the more likely participants might be to report prosocial behavioral intent and perceptions. In fact, as shown in Table 2, greater social desirability was significantly associated with greater intent to intervene as well as greater awareness/attention and greater personal responsibility. To determine whether the expected patterns of association remained significant regardless of participants’ need for social approval, a linear regression analysis was conducted to examine the associations between each of the potential barriers to intervention while also controlling for social desirability. Both awareness/attention (b = .15, β = .16, t = 2.94, p = .003) and personal responsibility (b = .35, β = .52, t = 9.46, p < .001) were uniquely related to intent to intervene. In contrast, social desirability was not significantly related to intent to intervene (b = .02, β = .05, t = 1.18, p = .24), nor was perceived danger to the victim (b = .01, β = .00, t = 0.10, p = .92), full model F(4, 392) = 72.15, p < .001.

We also expected that bystanders would directly intervene more in response to physical than sexual misconduct (Hypothesis 5) and that they would report greater awareness/attention to physical misconduct (Hypothesis 6), perceive physical misconduct as more dangerous to the victim (Hypothesis 7), and feel more personally responsible to intervene in response to physical than sexual misconduct (Hypothesis 8). Hypothesis 5 was supported by the greater frequency of intent to help the victim within the physical versus sexual condition (Table 1) as well as by the positive association between type of misconduct (coded as 0 = sexual, 1 = physical) and intent to intervene (Table 2). Also as expected, compared with those assigned to the sexual misconduct situation, bystanders assigned to the physical misconduct situation reported greater awareness/attentiveness, perceived greater danger, and felt more personally responsible to intervene. These associations, as shown in Table 2, fully supported Hypotheses 6, 7, and 8.

Again, to determine whether the association between type of misconduct and bystander intent to intervene remained significant even after accounting for social desirability, a linear regression analysis was conducted in which social desirability and type of misconduct (0 = sexual, 1 = physical) were entered simultaneously as predictors of intent to intervene, F(2, 393) = 6.06, p = .003. Results of this analysis showed that both social desirability (b = .05, β = .14, t = 2.81, p = .005) and type of misconduct (b = .35, β = .11, t = 2.16, p = .03) were significantly, positively related to intent to intervene. These findings indicated that although participants who sought to present themselves in a positive light reported a greater intent to intervene, after controlling for this tendency, bystanders reported greater intent to intervene in response to physical than sexual misconduct.

Finally, to identify unique predictors of intent to intervene in response to different types of misconduct, controlling for the significant effect of social desirability, an additional regression analysis was conducted. More specifically, intent to intervene was regressed on type of misconduct, awareness/attention, perceived danger, and personal responsibility, along with social desirability as a control variable. Hayes’s (2013) PROCESS macro was used to test for the significance of awareness, danger, and personal responsibility as potential mediators (i.e., indirect effects) of the effect of type of misconduct on intent to intervene. Bootstrapping uses repeated random resampling with replacement from the available data to approximate a sampling distribution of the indirect effect. The significance of the indirect effect, based on the 95% confidence interval (CI) derived from 5,000 bootstrap resamples, is indicated when the bias-corrected CI values do not include 0. The results of this analysis are shown in Figure 1. There was no direct effect of type of misconduct on intent to intervene. In addition, there was no significant path linking type of misconduct to intent to intervene via perceived danger to the victim. However, there were significant indirect paths through both awareness/attention and personal responsibility, full model F(5, 391) = 57.58, p < .001, adjusted R² = .42, suggesting these barriers explained the tendency for bystanders to report less intervention in response to sexual than physical misconduct.

Figure 1.

Path analysis testing indirect effects of type of misconduct (coded as sexual = 0, physical =1) on intent to intervene to help the victim via awareness, severity/danger, personal responsibility, and blame.

Discussion

The current study investigated college students’ bystander responses to intimate partner misconduct. We found considerable variability in how likely bystanders were to identify a victim of misconduct as experiencing abuse as well as how likely they were to offer help. Barriers to intervention included being less aware/attentive, perceiving the victim was in less danger, and feeling less personally responsible to intervene. Different types of misconduct also elicited different responses. Fewer bystanders perceived sexual than physical misconduct as abusive. Bystanders also intended to intervene less in response to sexual than physical misconduct, and this difference could be explained by two barriers: less awareness/attention and less personal responsibility.

Based on their self-generated responses to misconduct, about 42% of the full sample identified misconduct as abusive, although abuse was more often identified by those assigned to the physical (52%) than sexual (32%) conditions. Furthermore, just less than half of the full sample reported that they would likely intervene to try to help the victim; helping the victim was significantly more common among those witnessing physical (58%) than sexual (37%) misconduct. These results suggest that many bystanders described themselves as likely to intervene to help a victim of intimate partner misconduct who was grabbed and threatened with slapping. Still, rates were variable. In contrast, relatively few bystanders described themselves as likely to intervene to help a victim of intimate partner misconduct who was groped and kissed without consent. Although intent to help the victim was the most common behavioral intent, some bystanders reported that they would intervene via confrontation or delegation. Rates of confrontation did not differ across conditions, but bystanders in the physical misconduct condition were more likely to delegate than those in the sexual misconduct condition. Overall, these findings suggest that public acts of intimate partner misconduct are not frequently or consistently likely to be interrupted by bystanders. In particular, sexual violence may often be ignored.

Consistent with Latané and Darley’s (1970) model of bystander behavior, in univariate analyses, bystanders reported less intent to intervene to the degree that they also reported less awareness/attention to the misconduct, perceived the misconduct as less dangerous to the victim, and felt less personally responsible to intervene. In a multivariate analysis controlling for social desirability, when each of these barriers was entered as a predictor of intent to intervene, awareness and personal responsibility were unique predictors, but perceived danger was not. These findings match with and extend past studies of bystander responses to more extreme acts of violence against women in which intent to directly intervene was associated with awareness/attention (Fischer et al., 2006) and personal responsibility (Gracia et al., 2009).

In contrast, the null effect of perceived danger to the victim in the multivariate analysis showed that perceived danger did not independently predict direct intervention beyond the effects of either awareness/attention or personal responsibility. This was unexpected given past research in which young adults describe reluctance to intervene in dating or sexual assault situations unless they observe injury or intense victim distress (e.g., Edwards et al., 2015). However, Gracia et al. (2009) also found that personal responsibility was related to willingness to directly intervene in response to a range of types of physical violence, but perceived severity was not. A possible explanation for mixed findings across different studies is that perceived danger or severity may be important only to the degree that it functions to reduce ambiguity given that ambiguity is linked to bystander inaction (Latané & Nida, 1981). Although perceived danger to the victim may decrease ambiguity, other factors may also reduce ambiguity (such as awareness/attentiveness). In addition, other factors may reduce concerns about ambiguity (such as personal responsibility) that inhibit intervention. As another possibility, the importance of perceived danger in past studies could be related to the tendency to compare responses to less and more extreme forms of violence against women (e.g., Bennett et al., 2017; Chabot et al., 2009; Ermer et al., 2017) rather than a focus on a range of partner violence (Gracia et al., 2009) or, as in the current study, a focus on less extreme violence. Overall, our results suggest that low perceived danger to the victim is a barrier to intervention in response to misconduct, although this barrier is not independent of either low awareness/attention or low personal responsibility.

The current study also compared bystander responses to different types of intimate partner misconduct. Results indicated that participants assigned to the physical misconduct condition, compared with the sexual misconduct condition, reported more intent to intervene, more awareness/attention, more perceived danger to the victim, and more personal responsibility. Each of these perceptions was positively related to intent to intervene. These results directly extend past research on bystander behavior in which responses to physical and sexual violence were assessed but not statistically compared (e.g., Bennett et al., 2014; Edwards et al., 2015; Palmer, 2016; Palmer et al., 2018).

In path analyses, there were significant indirect paths linking type of misconduct to intent to intervene via both awareness/attention and responsibility. These results indicate that bystanders who witness sexual misconduct tend to report less awareness/attention and less responsibility to intervene; in turn, both tendencies explained the lower intent to intervene reported in response to sexual than physical misconduct. These results suggest that bystanders may be more alarmed and thus attentive to displays of physical than sexual possessiveness. Likewise, feelings of responsibility to intervene may be stronger for physical misconduct, perhaps because sexual misconduct may seem “wanted” or “invited” by the victim (Vera-Gray, 2017) or perhaps due to bystander assumptions about sexual consent within intimate relationships (e.g., Ewoldt et al., 2000; Shotland & Goodstein, 1992). These results match with the trivialization of sexual violations characteristic of rape culture more generally (Harding, 2015).

The current results suggest that the primary barriers to intervening in cases of intimate partner misconduct, and particularly in response to sexual misconduct, are low awareness/attention and low personal responsibility to intervene. The strong association between these two barriers suggests that those who are less aware/attentive feel less responsible to help victims; at the same time, those who feel less personally responsible may be less aware/attentive toward misconduct. Regardless, to the degree that less extreme forms of violence against women are seen as “normal” (Kuijpers et al., 2017), such forms of violence may attract less attention and elicit less personal responsibility to address misconduct when it occurs. The current results suggest the importance of educational efforts such as those described by Stout (1991) to raise awareness about how the various forms of violence against women, regardless of their place on the continuum, limit women’s freedom and feelings of personal safety. Reduced normalization may help break the cycle of misconduct being perceived as a non-event, one that is unworthy of attention or responsibility on the part of onlookers.

Limitations of the current study include the use of written scenarios rather than a more naturalistic exposure to partner misconduct. Alternative methodologies with more naturalistic exposure to acts of violence against women, perhaps using videotapes of actors (e.g., Fischer et al., 2006; Levine et al., 2002), are needed. Another limitation is the recruitment of a convenience sample of mostly White women undergraduates in the Northeastern United States; it cannot be guaranteed that the present findings generalize to different populations or geographic regions. Furthermore, given past research suggesting that White bystanders are less likely to intervene in response to Black women at risk for incapacitated rape (Katz, Merrilees, Hoxmeier, & Motisi, 2017), additional research on responses to victims of different racial/ethnic backgrounds is needed. Women of color are often stereotyped in highly sexualized ways; as such, it is likely that bystanders may be even more likely to trivialize and avoid involvement in sexual violations enacted against women of color. This type of research is a clear priority for future investigators.

Future research should also explore bystander responses to partners as well as strangers who enact less extreme forms of violence against women, especially given that even extreme physical and sexual forms of violence by partners are seen as less harmful and less aberrant than violence by strangers (e.g., Ewoldt et al., 2000; Katz & Rich, 2017; Shotland & Straw, 1976). Future research also is needed to investigate bystander responses to intrusions and street harassment (Vera-Gray, 2017). Being told to smile, catcalled, or photographed without consent may be trivialized as hassles or even as compliments; as such, these experiences may garner even less attention, be perceived as less dangerous, and elicit less personal responsibility for intervention than misconduct.

More generally, the current results also suggest the importance of additional scholarly work focused on less extreme forms of violence. The tendency for researchers to primarily study criminal forms of violence may contribute to the more general normalization of violent acts that most commonly affect women’s daily lives (Kelly, 1988, 2012). This normalization takes many forms. For example, in the United States, the #metoo movement began in late 2017 as a protest against widespread harassment against girls and women. Backlash quickly followed, including backlash about a women-centered view of intrusive experiences and the ostensibly broad range of different types of acts deemed to be unacceptable. We speculate that some of the same barriers that prevent bystander intervention in response to sexual misconduct also function to inhibit compassion for victims of sexual assault who avoided rape (e.g., Dr. Christine Blasey Ford). Likewise, these same barriers may function to promote fears about men being in constant danger of false or unfair accusations.

In conclusion, additional research on bystander responses to women subjected to the full continuum of violence is needed. Violence of all types restricts women’s freedom and sense of safety. When bystanders fail to intervene in response to less extreme acts of violence perpetrated against women, bystanders appear to condone such behaviors. That is, bystander inaction in response to misconduct implies that victims do not warrant respect, dignity, or bodily autonomy. The current results suggest considerable variability in the likelihood of bystander responses to intimate partner misconduct; in addition, low awareness/attention to misconduct and low personal responsibility to intervene are important barriers to intervention. Future research is necessary to further understand when and how bystanders respond to less extreme acts of violence, and particularly sexual violence, against women. The findings from these studies may be used to encourage bystanders to promote the freedom and safety of individual woman as well as to shape societal norms by de-normalizing the common forms of violence that many women face.

Footnotes

Acknowledgements

The authors gratefully acknowledge Tess Ramos-Dries for her assistance with data management and coding.

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.

ORCID iD

Jennifer Katz

Author Biographies

Jennifer Katz, PhD, is a licensed clinical psychologist and professor of psychology at The State University of New York at Geneseo (SUNY Geneseo) in upstate New York. Her program of research focuses on understanding and preventing different forms of intimate partner violence.

Claire Edgington is an undergraduate student at SUNY Geneseo. She is majoring in psychology with a minor in women’s and gender studies. She is interested in bystander behavior, the prevention of intimate violence, and the promotion of healthy sexual attitudes, behaviors, and functioning.

Elizabeth McCabe is an undergraduate student at SUNY Geneseo. She is majoring in both English and psychology. She is interested in social justice and peer support as well as in the specific ways in which bystanders can help prevent interpersonal violence.

References

Bennett

Banyard

V. L.

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

Bennett

Banyard

V. L.

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, 5, 682-702. doi:10.1177/0886260515586373

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

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

Chabot

H. F.

Tracy

T. L.

Manning

C. A.

Poisson

C. A.

(2009). Sex, attribution, and severity influence intervention decisions of informal helpers in domestic violence. Journal of Interpersonal Violence, 24, 1696-1713. doi:10.1177/0886260509331514

Crowne

D. P.

Marlowe

(1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349-354. doi:10.1037/h0047358

Edwards

K. M.

Rodenhizer-Stämpfli

K. A.

Eckstein

R. P.

(2015). Bystander action in situations of dating and sexual aggression: A mixed methodological study of high school youth. Journal of Youth and Adolescence, 44, 2321-2336. doi:10.1007/s10964-015-0307-z

Ermer

Roach

A. L.

Coleman

Ganong

L. H.

(2017). Deconstructing attitudes about intimate partner violence and bystander intervention: The roles of perpetrator gender and severity of aggression. Journal of Interpersonal Violence. Advance online publication. doi:10.1177/0886260517737556

Ewoldt

Monson

C. M.

Langhinrichsen-Rohling

(2000). Attributions about rape in continuum of dissolving marital relationships. Journal of Interpersonal Violence, 15, 1175-1174. doi:10.1177/088626000015011004

10.

Fischer

Greitemeyer

Pollozek

Frey

(2006). The unresponsive bystander: Are bystanders more responsive in dangerous emergencies? European Journal of Social Psychology, 36, 267-278. doi:10.1002/ejsp.297

11.

Gracia

Garcia

Lila

(2009). Public responses to intimate partner violence against women: The influence of perceived severity and personal responsibility. The Spanish Journal of Psychology, 12, 648-656.

12.

Harding

(2015). Asking for it: The alarming rise of rape culture—And what we can do about it. Philadelphia, PA: Da Capo Press.

13.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.

14.

Katz

Merrilees

Hoxmeier

J. C.

Motisi

(2017). White female bystanders’ responses to a black woman at risk for incapacitated sexual assault. Psychology of Women Quarterly, 41, 273-285. doi:10.1177/0361684316689367

15.

Katz

Nguyen

L. J.

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

16.

Katz

Rich

(2017). Intimate partner violence, neoliberal ideologies, and controversies about victimhood. In Robertson

(Ed.), Handbook on victims issues in criminal justice (pp. 75-87). New York, NY: Routledge.

17.

Kelly

(1988). Surviving sexual violence. Minneapolis: University of Minnesota Press.

18.

Kelly

(2012). Preface. In Brown

J. M.

Walklate

S. L.

(Eds.), Handbook on sexual violence (pp. xvii-xxvi). New York, NY: Routledge.

19.

Kuijpers

K. F.

Blokland

A. A. J.

Mercer

N. C.

(2017). Gendered perceptions of intimate partner violence normality: An experimental study. Journal of Interpersonal Violence. Advance online publication. doi:10.1177/0886260517746945

20.

Latané

Darley

J. M.

(1970). The unresponsive bystander: Why doesn’t he help? Englewood Cliffs, NJ: Prentice Hall.

21.

Latané

Nida

(1981). Ten years of research on group size and helping. Psychological Bulletin, 89, 308-324. doi:10.1037/0033-2909.89.2.308

22.

Levine

Cassidy

Brazier

Reicher

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

23.

Manne

(2018). Down girl: The logic of misogyny. Oxford, UK: Oxford University Press.

24.

Marlowe

Crowne

D. P.

(1961). Social desirability and response to situational demands. Journal of Consulting Psychology, 25, 109-115. doi:10.1037/h0041627

25.

McGregor

(2012). The legal heritage of the crime of rape. In Brown

J. M.

Walklate

S. L.

(Eds.), Handbook on sexual violence (pp. 69-89). New York, NY: Routledge.

26.

McMahon

Hoffman

M. L.

McMahon

S. M.

Zucker

Koenick

R. A.

(2013). What would you do? Strategies for bystander intervention to prevent sexual violence by college students. Journal of College and Character, 14, 141-151. doi:10.1515/jcc-2013-0019

27.

Monson

C. M.

Langhinrichsen-Rohling

Binderup

(2000). Does no really mean no after you say yes? Attributions about date and marital rape. Journal of Interpersonal Violence, 15, 1156-1174. doi:10.1177/088626000015011003

28.

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 and Community Safety, 18, 1-18. doi:10.1057/cpcs.2015.18

29.

Palmer

J. E.

Nicksa

S. C.

McMahon

(2018). Does who you know affect how you act? The impact of relationships on bystander intervention in interpersonal violence situations. Journal of Interpersonal Violence, 33, 2623-2642. doi:10.1177/0886260516628292

30.

Shotland

R. L.

Goodstein

(1992). Sexual precedence reduces the perceived legitimacy of sexual refusal: An examination of attributions concerning date rape and consensual sex. Personality and Social Psychology Bulletin, 18, 756-764. doi:10.1177/0146167292186012

31.

Shotland

R. L.

Straw

M. K.

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

32.

Stout

K. D.

(1991). A continuum of male controls and violence against women: A teaching model. Journal of Social Work, 27, 305-320. Retrieved from https://www-jstor-org-443.web.bisu.edu.cn/stable/23042667

33.

Straus

M. A.

Hamby

S. L.

Boney-McCoy

Sugarman

D. B.

(1996). The Revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17, 283-316. doi:10.1177/019251396017003001

34.

Vera-Gray

(2017). Men’s intrusion, women’s embodiment: A critical analysis of street harassment. Abingdon, UK: Routledge.

35.

World Health Organization. (2017). Violence against women: Intimate partner and sexual violence against women. Retrieved from http://www.who.int/news-room/fact-sheets/detail/violence-against-women /

36.

Ybarra

M. L.

Espelage

D. L.

Langhinrichsen-Rohling

Korchmaros

J. D.

Boyd

(2016). Lifetime prevalence rates and overlap of physical, psychological, and sexual dating abuse perpetration and victimization in a national sample of youth. Archives of Sexual Behavior, 45, 1083-1099. doi:10.1007/s10508-016-0748-9