The Causal Influence of Perceived Social Norms on Intimate Partner Violence Perpetration: Converging Cross-Sectional,Longitudinal,and Experimental Support for a Social Disinhibition Model

Abstract

Across three studies, we develop a model of the direct and indirect paths through which the perceived prevalence (perceived descriptive norms [PDN]) of intimate partner violence (IPV) among peers may influence individuals’ likelihood of engaging in IPV. Study 1 replicated and extended previous cross-sectional research by demonstrating a positive longitudinal association between PDN and subsequent IPV perpetration. Study 2 further showed the influence of PDN on IPV perpetration to be mediated through its relation to perceived peer acceptance of IPV (perceived injunctive norms [PIN]), which in turn predicted personal IPV acceptance. Study 3 built on this model using an experimental paradigm to show that increasing PDN leads to corresponding increases in PIN and, in turn, personal IPV acceptance, which both predicted IPV perpetration. Furthermore, the effects of PIN on personal IPV acceptance and IPV propensity were strongest for dominance-oriented individuals. Results bear important implications for social norms–based interventions for IPV.

Keywords

intimate partner violence social norms social influence prevention

Intimate partner violence (IPV) has been recognized as a major public health issue on college campuses, with physical violence occurring in an estimated 20% to 30% of college dating relationships across the globe (Straus, 2004, 2008). Surveys of IPV perpetration and victimization among college students have found these rates to be fairly symmetrical across sexes, with nearly one third of both male and female students reporting that they either experienced or engaged in some form of physical IPV toward a romantic partner in the previous year (Gover, Kaukinen, & Fox, 2008; S. F. Lewis & Fremouw, 2001; Sellers & Bromley, 1996). While there are significant limitations in self-report assessment of violent acts in dating relationships (see Hamby, 2016), the equivalent rates of physical violence by both sexes in college and high school samples have been shown in dozens of studies and mirror that of young married couples entering marriage therapy clinics for relationship dissatisfaction (Stith, McCollum, Amanor-Boadu, & Smith, 2012).

For years, researchers were baffled by these seemingly mutually violent relationships, particularly when global depictions of domestic violence are clearly characterized by male dominance over women. Rates of IPV as recorded by police reports and shelter settings clearly indicate that this is a gendered problem. However, researchers are now beginning to demonstrate different patterns of IPV that may explain the discrepant findings: intimate terrorism, which represents the most severe form of domestic violence carried out predominantly by men toward women (Johnson, 1995), and situational couple violence, which typically involves bilateral aggression over a particular issue and is thought to be the most prevalent type of relationship violence among the general population (Simpson, Doss, Wheeler, & Christensen, 2007). Because of these distinctions, new theories have been established to understand situational couple violence. I³ (I-cubed) theory (Finkel et al., 2011) is consistent with the General Aggression Model (Bushman & Anderson, 2002) and suggests that Instigating factors (situational events) often precipitate aggressive urges, and individuals with Impelling factors (e.g., testosterone, dispositional hostility) with very few Inhibiting factors (e.g., sobriety, relationship commitment) will be likely to enact aggression in the situation. Finkel and colleagues (2011) outline several potential disinhibiting factors, one of which is the role of social norms, or the surrounding culture’s attitudes and behaviors with regard to IPV. The goal of this research is to test whether individual differences in perceptions of the social norm (i.e., how often individuals believe other, similar individuals engage in violence toward an intimate partner) play a role in decisions to enact violence.

The Role of Perceived Social Norms in IPV

Growing public awareness of the prevalence and consequences of physical aggression in young, dating relationships has led to a proliferation of school-based IPV prevention programs, which have been implemented in educational settings across the developmental path from middle school (Cornelius & Resseguie, 2007) into college (Rosen & Bezold, 1996; Schwartz, Griffin, Russell, & Frontaura-Duck, 2006). To date, these programs have focused largely on changing individual-level factors associated with IPV perpetration, including personal attitudes condoning violence (Avery-Leaf, Cascardi, O’Leary, & Cano, 1997; Foshee, 1998), patriarchal views (Wolfe, Wekerle, Scott, Straatman, & Reitzel-Jaffe, 2003), and sexist attitudes (Jaffe, Sudermann, Reitzel, & Killip, 1992). However, studies evaluating these programs have found little evidence that they have been effective in reducing IPV (Cornelius & Resseguie, 2007; Whitaker et al., 2006). These findings have led to growing consensus among researchers regarding the need to expand the scope of existing IPV prevention research and programming to address the role of social and community-level influences (DeGue et al., 2012; Whitaker et al., 2006). To that end, newer models of prevention have endorsed more socially inclusive strategies, such as bystander interventions, which aim to increase community members’ awareness of IPV and foster a sense of social responsibility to reduce its perpetration (Banyard, 2011).

The shift toward more ecological approaches to IPV prevention has sparked an increase in scientific attention toward the ways in which social norms may contribute to the attitudes and beliefs that precipitate the use of violence between romantic partners (Banyard, Moynihan, & Plante, 2007; Gidycz, Orchowski, & Berkowitz, 2011). Social norms generally refer to the implicit or explicit standards for appropriate conduct held by a particular group, and may encompass both injunctive norms, which reflect the acceptability of a given behavior, and descriptive norms, which describe the prevalence of a behavior (Cialdini, Reno, & Kallgren, 1990; Sherif, 1936).

Collective Versus Perceived Social Norms

An important conceptual distinction in the study of social norms is that between collective norms, which refer to the prevailing attitudes, behaviors, and social rules that exist at a group, community, or cultural level, and perceived norms, which refer to the subjective interpretations of these norms at the individual level, and are the focus of this investigation. A major development in research on normative influence has been the recognition that these two types of norms frequently diverge; because perceived norms are the product of individuals’ psychological construal processes, the extent to which they accurately reflect actual, collective norms may vary widely (Lapinski & Rimal, 2005). For example, numerous studies have shown that many college students reliably misperceive the majority of their peers to be significantly heavier and more frequent drinkers than they actually are (Borsari & Carey, 2001; Perkins, 2002; Perkins et al., 1999).

Whereas accurate perceptions of group norms typically encourage individuals to engage in socially advantageous behaviors, inaccurate perceptions of those norms may lead to behaviors that are harmful or maladaptive (Berkowitz, 2004; Sherif, 1936). Studies have consistently shown that individuals who overestimate the acceptability (injunctive norm) or prevalence (descriptive norm) of problematic behaviors among their peers are more likely to personally engage in them at a higher than average frequency (Berkowitz, 2004). This effect has been demonstrated by a substantial body of literature examining the implications of normative misperceptions for a wide variety of harmful or risky behaviors, including alcohol use (M. A. Lewis & Neighbors, 2004), sexual risk-taking (Miner, Peterson, Welles, Jacoby, & Rosser, 2009), smoking (Hancock & Henry, 2003), and gambling (Neighbors & Larimer, 2004; Neighbors et al., 2015).

Recent research suggests that normative misperceptions may also predict individuals’ likelihood of engaging in physical aggression in an intimate relationship (Neighbors, Walker, et al., 2010; Witte & Mulla, 2013; Witte, Mulla, & Weaver, 2015). Neighbors, Walker, et al. (2010) were the first to demonstrate a positive association between perceptions of descriptive IPV norms and IPV perpetration in a community sample of male perpetrators. Results of their study showed that IPV perpetrators significantly overestimated the prevalence of physical perpetration among other men, and that their estimates were positively correlated with their own rates of physical partner abuse over the past 90 days (Neighbors, Walker, et al., 2010). This association has been replicated in several studies of college students (Witte & Mulla, 2012, 2013; Witte et al., 2015), which have shown in agreement that students who have personally engaged in IPV are more likely to overestimate descriptive norms for IPV on their campus, relative to those with no history of perpetration.

The Causal Order of Perceived Social Norms and IPV Behaviors

The evidence of the association between perceived IPV norms and IPV perpetration described above suggests that correcting normative misperceptions of IPV may be a useful component in interventions to reduce its perpetration. Specifically, some researchers have found success in providing personalized normative feedback (i.e., corrective information for individuals who overestimate the prevalence of a particular behavior in their social environment) for reducing problematic drinking (Neighbors, Larimer, & Lewis, 2004). However, further research explicating the nature of the relationship between perceptions of IPV norms and IPV behaviors is needed to establish an adequate empirical basis for such programs. A particularly important issue that has yet to be systematically examined is the directionality of this relationship. To date, it remains unclear whether individuals’ perceptions of IPV norms among peers can have a causal influence on their subsequent likelihood of personally engaging in physical aggression toward a romantic partner. Although this possibility is congruent with theories of social influence (Asch, 1956; Latane, 1981; Sherif, 1936), only a few studies thus far have produced longitudinal evidence that beliefs about violence precede its perpetration (Connolly, Friedlander, Pepler, Craig, & Laporte, 2010; Foshee, Linder, MacDougall, & Bangdiwala, 2001), and none of these studies focused specifically on beliefs about IPV prevalence. Results from Witte & Mulla (2012) suggest that perceived IPV norms may influence the self-reported likelihood of subsequent engagement in violent behavior toward intimate partners. In their study, college students who overestimated descriptive norms for situational violence (i.e., violence in response to partner infidelity) reported greater likelihood of becoming violent themselves in a similar situation in the future, compared with those who made lower estimates of the norm. These findings are consistent with the broader body of research showing normative misperceptions to influence later behavior (Borsari & Carey, 2001).

The Relation Between Descriptive and Injunctive Norms for IPV

Previous research suggests that individuals’ perceptions of both injunctive and descriptive IPV norms may influence their own risk of IPV perpetration (Witte & Mulla, 2012, 2013). However, no studies thus far have directly examined how these types of perceived norms are related to one another and how they may interactively contribute to individuals’ attitudes toward IPV and/or propensity to engage in IPV. Theories of normative influence (Deutsch & Gerard, 1955) suggest that individuals’ perceptions of descriptive norms for certain behaviors may often serve as a basis for their beliefs about the injunctive norms for those behaviors. Whereas the perception that many others engage in a particular behavior may lead individuals to conclude that it is socially acceptable and adaptive, the perception that few others engage in that behavior may lead them to construe it as unacceptable or maladaptive (Lapinski & Rimal, 2005). Accordingly, individuals who perceive the descriptive norm for IPV among their peers to be high may also be more prone to believe that it is considered more acceptable by their in-group, and thus feel less at risk of incurring social sanctions for engaging in IPV themselves (Kallgren, Reno, & Cialdini, 2000). Furthermore, individuals who believe IPV is common and unlikely to result in social sanctions may adopt their own personal favorable attitudes toward IPV, which then may lead to poor behavioral choices. There is a large body of literature suggesting that favorable attitudes toward IPV are related to perpetration of IPV (Lichter & McCloskey, 2004; McDonnel, Ott, & Michell, 2010; O’Keefe, 1998; Slep, Cascardi, Avery-Leaf, & O’Leary, 2001).

The Moderating Role of Social Dominance Orientation (SDO)

Another important issue that has yet to be addressed is the extent to which personality characteristics interact with perceived norms and IPV perpetration. Social norms theory suggests that perceived norms influence behavior by providing information on the potential social consequences of refraining from, or engaging in, certain behaviors. For instance, individuals who perceive drinking norms to be particularly high among peers may be motivated by the fear of not fitting with their in-group to drink more frequently themselves (Rimal & Real, 2003). In this case, perceived norms would promote participation in a behavior by providing information about its social benefits. Another possible way perceived norms may increase participation in a behavior is by providing social information that disinhibits individuals from acting on preexisting motivations or dispositions. People with a propensity to use violence against romantic partners might often curtail such urges out of fear of being labeled deviant and then ostracized. For such individuals, the perception that IPV is more acceptable and prevalent among in-group members may act to disinhibit them from acting on their violent tendencies. As such, we propose that it is primarily through this latter function (disinhibition) that perceived norms affect behavior in the context of IPV.

Based on the I³ theory of dating violence, aggressive predispositions interact with factors that disinhibit individuals to act aggressively when provoked (Finkel et al., 2011). In other words, it is likely that individuals with a predisposition to use violence may be socially disinhibited from engaging in IPV by the perception that it is common, and therefore acceptable, among peers. This possibility suggests that identifying background characteristics associated with a propensity for antisocial behavior may be critical to understanding which individuals may be at increased risk of acting on the misperception that IPV is highly prevalent or accepted by one’s in-group members. Furthermore, predispositions that make someone more likely to be influenced by his or her in-group would be important to discover as well.

Previous research investigating the associations between personality factors and social attitudes suggests that certain traits may augment the relationship between individuals’ perceptions of peer norms for IPV and their personal engagement in it. For example, SDO (Pratto, Sidanius, Stallworth, & Malle, 1994) is a personality disposition that reflects one’s propensity to prefer and promote the superiority of one’s in-group through exploitation of power dynamics inherent to social hierarchies. Research has shown SDO to be associated with lower levels of prosocial characteristics, such as empathy and tolerance, and higher levels of prejudiced social attitudes, including negative attitudes toward women (Perry & Sibley, 2011; Pratto et al., 1994; Sidanius, Levin, Liu and Pratto, 2000). Given that many sociocultural models of IPV recognize perpetration as instrumental in promoting social order and maintaining power within the dyad, it is likely that SDO plays an important role in IPV propensity. Furthermore, given that SDO individuals hold their in-groups in high regard, the biases and attitudes held by their in-groups may be more salient and highly valued to them.

The Present Investigation

The research described above elucidates three important issues we sought to address in the present investigation: (a) the directionality of the association between perceived descriptive IPV norms and IPV perpetration, (b) whether this association is mediated through injunctive norms and personal attitudes toward IPV, and (c) whether it is moderated by SDO. We present three studies in which we systematically examined each of these issues, and in doing so developed a conceptual model of the interrelations between perceived norms, personal attitudes, SDO, and IPV propensity.

In Study 1, we used longitudinal methods to assess support for the normative influence hypothesis that college students’ perceptions of descriptive IPV norms on campus would predict their likelihood of subsequent IPV perpetration. In Study 2, we extended our examination of this association using a larger cross-sectional sample to test the prediction that descriptive norms influence IPV perpetration through their effects on perceived injunctive norms (PIN) and personal attitudes, respectively. In Study 3, we employed an experimental design to examine the causal influence of perceived descriptive IPV norms on perceived injunctive IPV norms, personal IPV acceptance, and IPV propensity and tested the moderating role of SDO. Sample sizes in all three studies surpassed the minimum requirement for sufficient power (⩾80) to detect the medium-sized effects observed in previous research on the relation between perceived descriptive norms (PDN) and IPV perpetration (Neighbors, Walker, et al., 2010; Witte & Mulla, 2013).

Study 1

The ability of previous studies to determine the directionality of the association between perceived IPV norms and IPV perpetration has been limited by their use of cross-sectional methods. Thus, in Study 1, we sought to address this limitation by using a longitudinal design to test whether individuals became more likely to perpetrate IPV after perceiving the descriptive norm to be high among their in-group members (normative influence hypothesis). This hypothesis was examined by comparing baseline and follow-up measures of perceived IPV norms and IPV behaviors in a sample of college students from a large public university. Participants were queried about their previous IPV behaviors and their estimates of the prevalence of IPV (descriptive norm) among same-gender friends at the beginning of the semester and again 12 weeks later. Because aggression perpetrated by men is often viewed differently than aggression perpetrated by women (Hamby & Jackson, 2010) and because students tend to be more influenced by perceived norms for proximal referent groups that share similar traits to their own (i.e., gender; Neighbors, LaBrie, et al., 2010; Neighbors, Walker, et al., 2010), we assessed perceptions of gender-specific norms and limited our focus to same-gender friends. In accordance with social norms theory, we predicted that PDN for IPV in same-gender peers at baseline would predict the likelihood of engaging in IPV over the next 3 months.

Method

Participants

Baseline (Time 1 [T1]) measures were completed by a total of 239 undergraduate students (M_years = 18.97, SD = 2.18, 75% female) recruited from an introductory social science class at a public university in the Southeastern United States. The study sample was primarily White (71.1%) and consisted mostly of freshmen (64%) and sophomore (25.1%) students. The majority of participants who completed the baseline measures (n = 136, 57%) provided follow-up data (Time 2 [T2]) approximately 12 weeks later.

Procedure and materials

Institutional review board approval was granted prior to data collection. During the first week of classes in fall semester of 2014, research assistants entered an introductory social/behavioral science classroom in a large lecture hall. The enrollment size was 292 at the start of the semester. Students were given an opportunity to fill out questionnaires during class time to earn extra credit for class. Research assistants entered the same classroom 12 weeks later, and students were, once again, given an opportunity to fill out the study questionnaires for extra credit. The students were unaware of the posttest date to reduce the potential of selective attrition. Each participant was tracked from T1 to T2 with a personal identification code to protect confidentiality. The measures completed by participants at each time point are described in detail below.

Demographic form

Participants completed a short demographic form which asked them to report their age, gender, sexual orientation, relationship status, height and weight, and year in school.

Physical IPV perpetration

The Revised Conflict Tactics Scale (CTS2; Straus, Hamby, Boney-McCoy, & Sugarman, 1996) was used to measure participants’ history of IPV perpetration. The CTS2 is a widely used instrument for assessing multiple forms of IPV, including physical, psychological, and sexual as well as rates of injury and positive negotiation tactics. The CTS2 physical violence subscale queries participants about their use of “mild” and “severe” forms of violence (see Straus et al., 1996). Only questions from the mild physical abuse subscale of the CTS2 were used in the present study, as these behaviors are the most often endorsed among male and female college students (Archer, 2000). For the T1 assessment, participants were asked whether they had thrown something at their partner, twisted their partner’s arm or hair, pushed or shoved their partner, grabbed their partner, or slapped their partner in the past 12 months. Response options ranged from 0 (never) to 6 (20 or more times). The response scale also included a seventh answer choice (not in the past 12 months but it did happen before), but this response was coded as 0 because it referred to behaviors prior to the time period of interest in this study. For the T2 assessment, participants were given the same measure and asked to indicate how often they had perpetrated each of the aforementioned behaviors “since the beginning of the semester.”

Because the time frames differed for T1 and T2 assessment of IPV (12 months vs. one semester) and because scores on the CTS2 tend to be positively skewed, responses for each item were converted to dichotomous scores (0 = 0, >0 = 1) and then summed to yield a variety score, which represented the variety of IPV behaviors perpetrated by participants during the specified time frame. Variety scores for IPV have been shown to have stronger psychometric properties than frequency scores (Shorey, Brasfield, Febres, Cornelius, & Stuart, 2012).

PDN for physical IPV

For T1 and T2, participants were instructed to estimate the percentage (from 0% to 100%) of same-gender friends that had ever engaged in each of the five behaviors (previously mentioned) from the mild physical abuse subscale of the CTS2 (Straus et al., 1996). This method of assessment is consistent with that used in previous studies by Lewis and Neighbors (2004), Neighbors, Walker, and colleagues (2010), and Witte & Mulla (2013). Participants’ estimates for the five behaviors at each time point were averaged together to create a PDN score for T1 and T2.

Results and Discussion

The means, standard deviations, and bivariate correlations of the Study 1 variables are reported in Table 1. All variables were significantly correlated with one another with the exception of gender, which was not significantly associated with perceived norms or IPV perpetration at either time point and thus was not included as a covariate in our subsequent analyses. Although there was significant attrition in the study sample between T1 and T2 (43%), it is unlikely this was due to systematic differences between participants, as students had no prior knowledge that they would be asked to complete the baseline measures again until they arrived in class on the day the T2 survey was administered. The homogeneity of variance between participants who completed the study and those who dropped out after T1 was further confirmed by the results of a one-way analysis of variance (ANOVA), which showed no significant differences between the two groups on baseline measures of PDN, F(1, 220) = 2.420, p = .121, or IPV perpetration, F(1, 219) = 1.702, p = .193.

Table 1.

Means, Standard Deviations, and Correlations of Study 1 Variables.

Variable	M	SD	1	2	3	4	5
1. Gender	—	—	1.00	—	—	—	—
2. T1 perpetration	0.560	1.130	.051	1.00	—	—	—
3. T1 perceived norms	13.017	14.954	.063	.513**	1.00	—	—
4. T2 perpetration	0.360	1.100	−.024	.537**	.435**	1.00	—
5. T2 perceived norms	11.150	13.819	.023	.232**	.487**	.358**	1.00

Note. Columns 1 to 5 correspond with numbers of measures listed in the variable column. T1 = Time 1; T2 = Time 2.

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

Longitudinal effects

The directionality of the relation between PDN and IPV perpetration was assessed using autoregressive cross-lagged panel analyses. This approach allowed us to test both the normative influence hypothesis and the cognitive dissonance hypothesis simultaneously be examining (a) the stability of IPV perpetration and perceived IPV norms over time, (b) the synchronicity of IPV perpetration and perceived IPV norms within each time point, and (c) the cross-lagged effects of these variables over time. Results of the full model are summarized in Figure 1.

Figure 1.

Study 1 path model showing standardized coefficients for direct paths between variables.

PDN and IPV perpetration were both stable across the two time points (β = .546, p < .001 for perceived norms and β = .429, p < .001 for IPV perpetration). PDN and IPV perpetration were significantly positively correlated at both T1 (β = .515, p < .001) and T2 (β = .182, p = .035). At T2, the association between PDN and IPV perpetration remained significant even when both variables were accounted for at T1 through autoregressive paths. Cross-lagged paths indicated a significant positive relationship between perceived norms at T1 and IPV perpetration at T2 (β = .315, p < .001). The cross-lagged path between IPV perpetration at T1 and perceived norms at T2 was not significant. Therefore, both participants’ history of IPV perpetration and their perceptions of descriptive IPV norms at the beginning of the semester predicted their likelihood of subsequently engaging in IPV.

Consistent with the normative influence hypothesis, these results suggest that individuals are more likely to perpetrate IPV when they perceive higher rates of IPV perpetration among their same-gender close friends. In our sample, this association remained significant even when previous IPV perpetration was controlled for, suggesting a unique relation between perceived IPV norms and subsequent IPV perpetration.

Study 2

The results of Study 1 provided evidence of the longitudinal influence of PDN on subsequent IPV perpetration. Building on these findings, Study 2 aimed to examine the potential mediating roles of PIN and personal IPV acceptance in the relation between perceived descriptive IPV norms and IPV perpetration. Theories of social influence suggest that perceiving a behavior to be more common among in-group members may lead individuals to also perceive that behavior as more accepted among in-group members (Deutsch & Gerard, 1955) and, in turn, adjust their own acceptance of that behavior to be more in line with the norm. In Study 2, we examined this theoretical set of relationships in the context of IPV using a multiple mediation model that allowed us to assess both the hypothesized indirect effects of perceived descriptive IPV norms and IPV perpetration through perceived injunctive IPV norms and personal IPV acceptance, as well as the direct relations between these variables. As a basis for comparison, we also tested a modified exploratory model where personal attitudes predicted perceptions of descriptive and injunctive norms, which in turn predicted IPV perpetration.

Method

Participants

Study 2 included data from 538 undergraduate students at a public university in the Southeastern United States who completed a survey study on health, stress, and problematic dating behaviors; 422 provided complete data for Study 2 analyses (M_years = 20.29, SD = 1.53; 64.3% females). The majority of the sample was White (81.2%) and comprised roughly equal proportions of freshmen (26.1%), sophomores (22.4%), juniors (24.3%), and seniors (27.2%). Only participants’ demographic information and their responses to measures of perceived descriptive and injunctive norms for IPV, personal acceptance of IPV, and personal history of IPV perpetration were included in the analyses for Study 2.

Procedure and materials

Institutional review board approval was granted prior to data collection. Participants were recruited from undergraduate social science classes to complete the full paper-and-pencil survey, which included several measures of stress and health in addition to the measures of demographics, perceived descriptive and injunctive IPV norms, personal IPV acceptance, and IPV perpetration history. The average survey completion time was approximately 15 min.

Demographic form

Participants were asked to report age, gender, race, sexual orientation, and year in school on a standard demographic form.

Physical IPV perpetration

Consistent with Study 1, IPV perpetration history was assessed using the mild physical abuse subscale of the CTS2 (Straus et al., 1996) with Participants’ responses to each item scored dichotomously (0 = 0, >0 = 1) and then summed to create an index of the variety of physically abusive behaviors they had perpetrated in the past.

PDN for physical IPV

Students were instructed to estimate the percentage (from 0% to 100%) of their same-gender friends that had ever engaged in the five IPV behaviors from the CTS2: thrown something at their partner, twisted their partner’s arm or hair, pushed or shoved their partner, grabbed their partner, and slapped their partner. Participants’ estimates were then averaged across all five behaviors to calculate their PDN score.

PIN for physical IPV

Students were instructed to estimate the degree to which students and their same-gender friends believed it was acceptable to engage in each of the five IPV behaviors measured by the CTS2. Participants responded to each item using a scale from 1 (highly unacceptable) to 5 (highly acceptable). Responses were then averaged across all five behaviors to calculate their PIN score.

Personal IPV acceptance

Participants’ personal attitudes regarding the acceptability of engaging in IPV were assessed by asking them to indicate their level of personal acceptance for each of the five IPV behaviors included in the measures of IPV perpetration and perceived norms using the same scale (1 = highly unacceptable to 5 = highly acceptable).

Results and Discussion

The means, standard deviations, and bivariate correlations of the Study 2 variables are reported in Table 2. All predictor variables were significantly correlated with each other, as well as with the outcome variable of IPV perpetration, suggesting that conditions were appropriate for further testing the indirect relations between them.

Table 2.

Means, Standard Deviations, and Correlations of Study 2 Variables.

Variable	M	SD	1	2	3	4	5
1. Gender	—	—	1.00	—	—	—	—
2. PDN	29.06	19.17	.164**	1.00	—	—	—
3. PIN	1.63	0.67	.074	.417*	1.00	—	—
4. PA	5.78	2.26	−.121**	.281**	.749**	1.00	—
5. IPV perpetration	0.46	1.10	−.114*	.365 **	.303**	.394**	1.00

Note. Columns 1 to 5 correspond with numbers of measures listed in the variable column. PDN = perceived descriptive norms; PIN = perceived injunctive norms; PA = personal IPV acceptance.

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

Hypothesized model

A series of multiple regression path analyses were conducted to examine the hypothesized indirect effect of perceived descriptive IPV norms on IPV perpetration through PIN and personal IPV acceptance, respectively. Standardized scores for the study variables were entered in a multiple mediation path model constructed using Preacher and Hayes’s (2008) SPSS PROCESS macro; IPV perpetration was regressed on the predictor variables of personal acceptance, PIN, and PDN in that order, with gender entered as a covariate. This approach allowed us to examine (a) the direct relation of PDN to PIN, (b) their direct relation to personal acceptance after controlling for PIN, (c) their direct relation to the outcome variable of IPV perpetration after controlling for both PIN and personal acceptance, and (d) the indirect path(s) through which the effects of PDN on IPV perpetration were mediated. Bias-corrected 95% confidence intervals (CIs) based on 10,000 bootstrapped samples were used as a criterion for identifying interpretable relations among the terms of the model.

Direct effects

The hypothesized multiple mediation model and standardized coefficients for significant direct paths are presented in Figure 2a. The predicted positive associations between PDN and PIN, and the predicted positive association between PIN and personal acceptance were significant. Both PDN and personal acceptance were also significantly directly related to IPV perpetration.

Figure 2.

Study 2 (a) hypothesized path model and (b) exploratory path model with corresponding standardized coefficients for significant direct paths between variables.

Mediation effects

The standardized coefficients and 95% CIs for all indirect paths in the hypothesized model are reported in Table 3 (Model 1). Our hypothesis that PDN influence IPV perpetration through their effects on PIN and in turn personal acceptance was supported. However, the direct effect of PDN on IPV perpetration remained significant even after accounting for PIN and personal acceptance, suggesting that it was only partially mediated through these variables.

Table 3.

Standardized Coefficients, Bootstrapped CIs, and Effect Sizes (P_M) for Indirect Effects in Study 2 Models.

Path	β	SE	95% CI	P _M
Model 1
PDN → PIN → Perp	−0.007	0.037	[–0.073, 0.072]	−0.027
PDN → PIN → PA → Perp	0.090*	0.044	[0.030, 0.199]	0.348
PDN → PA → Perp	0.017	0.021	[–0.068, 0.018]	−0.065
Total indirect effect	0.055	0.034	[–0.005, 0.130]	0.255
Model 2
PA → PDN → Perp	0.072*	0.036	[0.025, 0.160]	0.182
PA → PDN → PIN → Perp	−0.001	0.008	[–0.017, 0.011]	−0.003
PA → PIN → Perp	−0.012	0.067	[–0.137, 0.123]	−0.037
Total indirect effect	0.058	0.062	[–0.056, 0.193]	0.148

Note. CI = confidence interval; P_M = percentage mediated; PDN = perceived descriptive norms; PIN = perceived injunctive norms; Perp = IPV perpetration; PA = personal IPV acceptance; IPV = intimate partner violence.

Exploratory model

Next, we constructed an alternative model in which we reordered the sequence of variables to explore the possibility that personal IPV acceptance indirectly influenced IPV perpetration by increasing perceptions of descriptive IPV norms and injunctive IPV norms, respectively. IPV perpetration was regressed on the predictor variables of PIN, PDN, and personal acceptance in that order, with gender entered as a covariate. The direct paths and corresponding standardized coefficients for this model are presented in Figure 2b. Consistent with the results of our hypothesized model, personal IPV acceptance was directly positively associated with PIN and IPV perpetration, as well as with PDN. Furthermore, perceived descriptive were significantly positively associated with IPV perpetration, but PIN were not. Standardized coefficients and 95% CIs for all indirect paths in the exploratory model are reported in Table 3 (Model 2). Results indicated that the relation of personal IPV acceptance to IPV perpetration was significantly partially mediated through perceived descriptive IPV norms, but not through perceived injunctive IPV norms.

Gender differences

Finally, we examined whether the series of associations between the study variables observed in our hypothesized model differed between male and female participants in our sample. Standardized scores for the study variables were entered in separate multiple regression models for women (Figure 3a) and men (Figure 3b); IPV perpetration was regressed on the predictor variables of personal acceptance, PIN, and PDN in that order. Results showed that the pattern of significant relationships between PDN, PIN, personal IPV acceptance, and IPV perpetration did not differ across genders, although the correlations between these variables were stronger for men.

Figure 3.

Study 2 path models for (a) women and (b) men with corresponding standardized coefficients for significant direct paths between variables.

Study 3

Study 2 provided evidence that PDN may increase IPV perpetration risk in part through their influence on PIN. Individuals who perceive the prevalence of IPV to be high among peers tend to perceive its acceptability to be high as well, and they, in turn, tend to have favorable personal attitudes toward IPV. Thus, Study 3 aimed to provide experimental support for this presumed causal chain by using a between-subjects experimental paradigm in which individuals’ perceptions of descriptive norms were manipulated. Study 3 further sought to examine the role of SDO (Pratto et al., 1994) as a potential moderator in the relationship between individuals’ perceptions of IPV norms and the propensity to personally engage in IPV.

Method

Participants

Study 3 included data from 293 undergraduate students at a public university in the Southeastern United States. Participants were recruited from a subject pool of students enrolled in an introductory psychology course and were offered partial course credit in exchange for completing the study. We aimed to enroll approximately 300 students to obtain a sample well above the minimum size of 200 cases recommended for structural equation modeling by research (Kline, 2011; Wolf, Harrington, Clark, & Miller, 2013). Of the 293 students who took part in the study, 35 either spent inadequate (under 10 min) or excessive (over 60 min) time completing it, or did not complete all of the study measures, and were excluded from analyses. The final sample included a total of 258 participants (M_years = 18.64, SD = 0.854; 69% females), the majority of whom were White (82.2 %) and in their freshmen (72.1%) or sophomore (20.9%) year of college.

Procedure and materials

Institutional review board approval was granted prior to this online experimental study. Participants agreed to complete two ostensibly separate tasks online. The first of these purportedly examined the relationship between demographics, social relationships, and cognitive ability. After completing a demographics form and measures of SDO, attachment style, and sexist attitudes (presented in randomized order), participants completed a distractor task, followed by a second task purportedly examining their perceptions of social interactions. As part of this task, participants received fabricated information about “trends in relationship behavior on campus” (i.e., descriptive norms), which depicted prevalence rates of five nonviolent behaviors (neutral norm condition), low prevalence rates of five violent behaviors among dating partners (low norm condition), or high prevalence rates of the same five violent behaviors among dating partners (high norm condition). Participants then completed measures assessing their perceptions of IPV injunctive norms, their personal acceptance of IPV, and their personal propensity to engage in IPV in future situations. The full list of measures and materials used in Study 3 are described in detail below.

Demographics

Participants provided their sex, age, race, sexual orientation, year in college, and fraternity/sorority status.

SDO

Participants’ SDO was assessed via self-reported feelings (1 = very negative; 7 = very positive) toward 14 items from Pratto et al.’s (1994) SDO Scale (e.g., “Some people are just more worthy than others”; M = 2.58, SD = 1.15; α = .911).

Distractor task

Next, in line with our cover story intended to conceal the true purpose of the study, participants completed a Raven’s Progressive Matrix task (Raven, 1998), which comprised six matrices that each contained a sequence of eight shapes illustrating a pattern of change. For each matrix, participants were asked to examine the pattern and select a ninth shape (out of several options) that would best complete it.

Norm manipulation

Next, participants ostensibly began “Task 2” under the guise of investigating different ways people interact with close others. Participants were told that they would view some “background information” on trends in certain social interactions on their home campus and subsequently respond to some related questions. Importantly, the background information purportedly represented students with the same demographic profile of the participants. Here, participants experienced one of three conditions: In the neutral norm condition, participants viewed a histogram depicting percentages (ranging from 32% to 80%) of students who reported engaging in five nonviolent behaviors with their romantic partner (e.g., Talk to their romantic partner on the phone at least once a day); in the low norm condition, participants viewed a histogram depicting percentages (ranging from 1% to 3%) of students who reported engaging in five violent behaviors with their romantic partner (e.g., Slapped their girlfriend/boyfriend), which were taken from the mild physical abuse subscale of the CTS2, similar to Studies 1 and 2; in the high norm condition, participants viewed a histogram of the same behaviors from the low norm condition, but the percentage rates for the behaviors ranged from 49% to 68%. These percentages were selected as face valid indicators of “low” and “high” prevalence rates of behaviors and were pilot tested by several undergraduate research assistants to confirm their believability among members of the target population. On the same page of each graph, participants answered manipulation-check questions regarding the proclivity of each behavior depicted in the graph (one question per behavior) to ensure that participants viewed and understood the graph.

Perceived injunctive IPV norms and personal IPV acceptance

Immediately following the descriptive norm manipulation, participants completed measures of their perceived injunctive IPV norms and personal IPV acceptance that were nearly identical to those given in Study 2, but used a slightly different scale (1: never OK to 4 = always OK) to record participants’ estimates of peer acceptance and their personal acceptance of the same five IPV behaviors.

IPV propensity

Participants’ propensity to engage in IPV was assessed using the Proximal Antecedents to Violent Episodes (PAVE; Babcock, Costa, & Green, 2004) scale—a widely used measure of IPV propensity consisting of 30 items designed to assess participants’ likelihood of using violence in a variety of situations (e.g., Finkel, DeWall, Slotter, Oaten, & Foshee, 2009). Participants are instructed to rate how likely they would become physically aggressive in various situations using a scale of 1 (not at all likely) to 9 (extremely likely).

Results and Discussion

Main effects of norm condition on PIN

The descriptive statistics and bivariate correlations of the Study 3 variables are reported in Tables 4 and 5, respectively. We first compared the effects of the high, neutral, and low descriptive norm conditions on participants’ estimates of injunctive IPV norms using a one-way between-subjects ANOVA. There was a significant main effect of norm condition on PIN, F(2, 255) = 13.211, p < .001, $η_{p}^{2} = . 094$ . Post hoc analyses using Tukey’s HSD (honestly significant difference) indicated that estimates of injunctive IPV norms were significantly higher among participants in the high norm condition (M = 2.04, SD = 0.71) compared with those in the neutral (M = 1.59, SD = 0.61, $η_{p}^{2} = . 080$ ) and low norm (M = 1.64, SD = 0.59, $η_{p}^{2} = . 062$ ) conditions. However, the low norm condition and the neutral condition did not significantly differ in their perceptions of injunctive norms.

Table 4.

Means and Standard Deviations of Study 3 Dependent Variables by Condition.

Variable	Low norm condition		Neutral condition		High norm condition
Variable	M	SD	M	SD	M	SD
PIN	1.64	0.59	1.59	0.61	2.04	0.71
PA	1.16	0.26	1.24	0.39	1.29	0.41
SDO	2.47	1.06	2.61	1.18	2.64	1.21
PAVE	33.16	13.04	31.65	13.15	31.67	15.45

Note. PIN = perceived injunctive norms; PA = personal IPV acceptance; SDO = social dominance orientation; PAVE = Proximal Antecedents to Violent Episodes scale; IPV = intimate partner violence.

p < .05.

Table 5.

Bivariate Correlations of Study 3 Variables Across Conditions.

Variable	1	2	3	4	5
1. Gender	1.00	—	—	—	—
2. PIN	.107	1.00	—	—	—
3. PA	.028	.454**	1.00	—	—
4. SDO	−.167**	.030	.119	1.00	—
5. PAVE	.184**	.312**	.398**	.168**	1.00

Note. Columns 1 to 5 correspond with numbers of measures listed in the variable column. PIN = perceived injunctive norm; PA = personal IPV acceptance; SDO = social dominance orientation; PAVE = Proximal Antecedents to Violent Episodes scale; IPV = intimate partner violence.

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

Effects of SDO and PIN on personal IPV acceptance and IPV propensity

Our next set of analyses sought to determine whether SDO moderated the influence of PIN on individuals’ own acceptance of IPV and their propensity to personally engage in it. Multiple regression analyses were conducted to test the effects of SDO, PIN, and their interaction term on the dependent variables of personal IPV acceptance and IPV propensity in two separate moderation models constructed using the SPSS PROCESS macro (Preacher & Hayes, 2008). All variables were standardized prior to analysis.

The interaction effect of perceived norms by SDO on personal IPV acceptance is graphically presented in Figure 4. As predicted, individuals high in SDO who perceived injunctive IPV norms to be high reported significantly greater personal IPV acceptance compared with those with lower SDO (β = .141, p < .05). However, neither SDO nor PIN had a significant main effect on personal IPV acceptance. The interaction effect of PIN by SDO on IPV propensity is graphically presented in Figure 5. As with personal IPV acceptance, individuals high in SDO reported significantly higher IPV propensity than those who were lower in SDO (β = .229, p < .001). There was also a significant main effect of SDO on IPV propensity (β = .194, p < .001), as well as a significant main effect of PIN on IPV propensity (β = .394, p < .001).

Figure 4.

The interaction effect of social dominance orientation and PIN on personal IPV acceptance.

Figure 5.

The interaction effect of social dominance orientation and perceived injunctive norms on IPV propensity as measured by PAVE scores.

Final path model

Based on the results of the analyses above, we extended the conceptual model tested in Study 2 to include SDO as an exogenous moderator of the effects of PIN on personal IPV acceptance and IPV propensity. In addition to accounting for the role of SDO, this approach allowed us to assess the indirect paths through which our experimental manipulation of descriptive norms may have influenced participants’ IPV propensity.

The revised model was constructed and tested using AMOS 23 (Arbuckle, 2014). Given that participants in the neutral and low norm conditions did not significantly differ in their estimates of PIN, we combined them into one group. The two norm groups were then dummy coded (neutral + low norm = 0; high norm = 1) to allow us to treat descriptive norm condition as a continuous variable in our model. Norm condition was entered as a primary predictor with scores for PIN and personal IPV acceptance entered as mediators of the relation between norm condition and IPV propensity. SDO and the SDO by perceived injunctive norm interaction term were entered as exogenous predictors of personal acceptance and IPV propensity. All variables were standardized prior to analysis. We assessed model fit via the chi-square, χ²(5) = 1.80, p = .876, normed fit index (NFI = .99), comparative fit index (CFI = 1.00), and root mean squared error of approximation (RMSEA = .000 [.000, .043]) fit indices, which collectively suggested excellent model fit.

Direct effects

The final model with standardized coefficients of significant paths is presented in Figure 6. Our analyses revealed a series of direct paths that replicated the findings of Study 2; norm condition had a direct effect on injunctive norms (β = .64, p < .001), which had a direct effect on personal acceptance (β = .46, p < .001), which had a direct effect on IPV propensity (β = .27, p = .002). The path model also showed that the total effect of norm condition on IPB propensity was nonsignificant. However, norm condition did have an unanticipated negative direct effect on IPV propensity (β = –.26, p < .05) after controlling for all other predictor variables. This finding suggests that participants who were not influenced by their perceptions of injunctive norms, SDO, or personal IPV attitudes became less inclined to personally engage in IPV after hearing that it was highly prevalent among their peers.

Figure 6.

Study 3 final path model with standardized coefficients for significant direct paths between variables.

Mediated effects of norm condition on IPV propensity

Consistent with our predictions, norm condition had a positive indirect effect on IPV propensity (95% CI: [.111, .378]) through two mediated pathways. In one path, the effect of norm condition on IPV propensity was mediated through its positive influence on PIN, which in turn were positively related to IPV propensity (95% CI: [.054, .272]). In the second path, the effects of norm condition on IPV propensity were mediated through its influence on PIN, which were positively associated with personal acceptance, which in turn was positively related to IPV propensity (95% CI: [.025, .156]).

Effects of SDO on IPV propensity

SDO also had a significant total effect on IPV propensity (β = .20, p = .001) that could be separated into a direct effect (β = .16, p = .01) and an indirect effect via personal acceptance (95% CI: [.003, .076]). Furthermore, SDO moderated the effect of PIN on personal acceptance (β = .14, p = .028), wherein people higher in SDO were especially likely to personally accept an injunctive norm favoring IPV behaviors. SDO also moderated the effect of PIN on IPV propensity (β = .18, p = .010), wherein people higher in SDO who perceived injunctive IPV norms to be high were especially likely to report greater propensity to engage in IPV themselves; this moderation effect was mediated by personal acceptance (95% CI: [.003, .086]).

In sum, experimentally inducing an increased perception of IPV prevalence among peers increased participants’ perceived injunctive IPV norms, which increased their personal acceptance of IPV, which increased self-reported propensity to engage in IPV. Furthermore, people higher in SDO seemed more likely to engage in IPV and were especially likely to allow perceived injunctive IPV norms to increase their personal acceptance of IPV, as well as their propensity to engage in it.

General Discussion

Perceptions of social norms influence a variety of behaviors, from alcohol use (M. A. Lewis & Neighbors, 2004) to sexual practices (Miner et al., 2009). Recent investigations into perceived norms for IPV have shown correlations between PDN for IPV (i.e., estimates of IPV prevalence) and IPV perpetration (e.g., Witte & Mulla, 2013), but these cross-sectional studies were unable to demonstrate the direction of the association between social norms and IPV. Thus, until this point, it has been premature to use social norms–based intervention techniques (e. g., personalized normative feedback) in prevention programs for IPV given the lack of empirical evidence supporting the predictive and causal nature of perceived social norms for IPV. The present investigation addressed these gaps in the literature by demonstrating that PDN for IPV predict future IPV behaviors and influence IPV behavioral intentions; these associations were partially mediated by PIN and personal attitudes and moderated by SDO.

Results of Study 1 showed a prospective link between PDN for IPV and IPV perpetration 3 months later. Consistent with theories of social influence, male and female college students with elevated estimates of IPV prevalence (i.e., elevated perceived norms) among their peers at baseline were more likely to perpetrate IPV in the subsequent 12 weeks. This association remained significant even after controlling for past IPV behaviors. These findings justified the continued investigation of potential mediating mechanisms of this temporal relationship (Study 2) as well as potential causal associations (Study 3).

Study 2 investigated the mediating roles of PIN (i.e., perception of cultural acceptance of IPV) and personal attitudes accepting IPV in the relation between PDN and IPV perpetration. We expected perceptions of the frequency of IPV would be consistent with perceptions about the acceptability of IPV (Deutsch & Gerard, 1955) and further predicted that personal attitudes would develop based on these perceptions of the social ecology (Bronfenbrenner, 1979). Results showed evidence for partial mediation; PDN for IPV had a direct effect on IPV perpetration even after controlling for the mediators. Thus, perceived norms for IPV played an influential role in participants’ IPV behaviors, in part due to the anticipation of social approval (or disapproval) and personal beliefs about the acceptability of IPV. These findings were stable across the genders, suggesting that both men and women use PDN to gauge the culture’s approval (or disapproval) of IPV, and form their own attitudes toward IPV based on such perceptions. An alternative model was tested to measure the influence of personal acceptance of IPV on perceptions of descriptive and injunctive IPV norms. Although the results of this model replicated several of the direct paths observed in our hypothesized model (i.e., the direct relations between personal acceptance and both types of norms), our primary interest—the mediated pathway of personal acceptance to IPV propensity through perceived descriptive and injunctive norms, respectively—was nonsignificant.

In Study 3, we replicated and extended the findings from Studies 1 and 2 by using an experimental design to demonstrate a causal association between PDN and IPV behavioral intentions. Individuals who were led to believe that rates of IPV were elevated among demographically similar individuals believed IPV was accepted by their in-group. They, in turn, reported greater personal acceptance of IPV and greater tendencies toward engaging in IPV in the near future. Also in Study 3, individuals who reported elevated levels of SDO had a higher proclivity to engage in violence in future situations, in part due to their accepting attitudes toward IPV. Furthermore, those with higher social dominance showed a strong relation between their estimates of acceptability among peers and their own personal attitudes, which in turn led to greater behavioral intentions to become violent toward an intimate partner. This suggests that socially dominant individuals may be more influenced by peer acceptability and potential social disapproval compared with those low in social dominance, and this pressure may guide their own attitudes and behaviors in favor of IPV.

Taken together, the findings of these studies expand upon the prevailing I³ theory of situational couple violence (Finkel et al., 2011), which suggests that there are important factors that both propel (e.g., elevated levels of testosterone) and inhibit (e.g., ego strength) aggressive responses when an individual is provoked. In accordance with this theory, the current investigation provides evidence that perceived peer norms may be an important disinhibiting factor in situations of intimate partner conflict. Specifically, our findings suggest that individuals with violent predispositions may be socially disinhibited from engaging in IPV simply by the perception that it is common, and therefore acceptable, in their subculture. This cascade of social-cognitive processes seems to occur more robustly in individuals with higher levels of SDO, based on our results.

The current investigation also provides support for the use of social norms–based intervention techniques in prevention programs for IPV. Such programs would provide personalized normative feedback (i.e., presenting accurate norms to individuals who may be overestimating the norm) to create an inhibiting factor—the threat of social disapproval—which would ultimately decrease abusive behavior in dating relationships. Indeed, this approach has been successful with other behaviors, such as alcohol use. For example, the presentation of normative feedback in an alcohol intervention was shown to be effective in changing both individuals’ perceived drinking norms and drinking behavior at 3- and 6-month follow-up assessments (Neighbors et al., 2004).

Although the results of the current investigation help to justify and inform the application of this approach to reducing IPV, they should also be evaluated in light of several limitations. First, the attrition rate in Study 1 was high. Attrition was most likely due to random absenteeism rather than study materials because the students were unaware of the date for T2 assessment, and most, if not all, students in the class at T2 completed the study materials. Second, although the longitudinal association between PDN and IPV perpetration in Study 1 justified the investigation of mediators in Study 2, the assessment of mediators in Study 2 was conducted using a cross-sectional design. Future research should test mediators using prospective designs. Third, the causal link between PDN for IPV and IPV behaviors in Study 3 was assessed through a measure of behavioral intentions rather than actual behaviors, which reduces the strength of this finding.

Further work is also needed to explicate the ways in which the individual-level perceptions of IPV norms examined in the current investigation are influenced by, and interact with, actual collective norms at the community and cultural levels. Just as social rules for dealing with relationship conflict vary across cultural contexts, individuals’ subjective construal of those rules can also be expected to vary accordingly. For example, in communities or countries that more strongly promote a patriarchal social order, people may be more likely to perceive the use of IPV by men against women as culturally sanctioned. The results of our studies suggest that such contexts may foster more salient, socially disinhibiting perceptions regarding the acceptability of male IPV perpetration against women relative to more egalitarian settings, and that these perceptions may be especially disinhibiting for individuals with a predisposition toward social dominance. Subsequent studies could directly test this hypothesis by examining the association between individual-level perceptions of gender-specific descriptive and injunctive IPV norms, and collective norms among men and women across different cultural groups, as well as the factors that moderate that association. Research operationalizing and assessing the influence of sociohistorical and cultural norms regarding gender roles and gender disparities may be particularly helpful in advancing the scientific understanding of how such contexts may interact with individual-level moderators (i.e., gender identity, gender role adherence) to influence the use of situational violence between romantic partners.

Overall, the results of the current studies provide novel empirical support for the potential utility of incorporating social norms–based approaches in interventions for reducing IPV. Our investigation is the first to demonstrate the predictive and causal role of perceived descriptive IPV norms in young adults’ intentions to engage in IPV perpetration, and show these effects to be partially mediated by perceived injunctive IPV norms and personal IPV acceptance, and moderated by SDO. These findings may eventually help to shape current efforts at IPV prevention programming across college campuses. Future research expanding on the current investigation by addressing the limitations and questions raised above could provide an important next step in this direction.

Supplemental Material

Mulla_OnlineAppendix – Supplemental material for The Causal Influence of Perceived Social Norms on Intimate Partner Violence Perpetration: Converging Cross-Sectional, Longitudinal, and Experimental Support for a Social Disinhibition Model

Supplemental material, Mulla_OnlineAppendix for The Causal Influence of Perceived Social Norms on Intimate Partner Violence Perpetration: Converging Cross-Sectional, Longitudinal, and Experimental Support for a Social Disinhibition Model by Mazheruddin M. Mulla, Tricia H. Witte, Kyle Richardson, William Hart, Francesca L. Kassing, Christopher A. Coffey, Christine L. Hackman and Ian M. Sherwood in Personality and Social Psychology Bulletin

Footnotes

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.

Supplemental Material

Supplemental material is available online with this article.

ORCID iD

Mazheruddin M. Mulla

References

Arbuckle

J. L.

(2014). Amos 23.0 User’s Guide. Chicago, IL: IBM SPSS.

Archer

(2000). Sex differences in aggression between heterosexual partners: A meta-analytic review. Psychological Bulletin, 126, 651-680.

Asch

S. E.

(1956). Studies of independence and conformity: I. A minority of one against a unanimous majority. Psychological Monographs: General and Applied, 70(9), 1-70.

Avery-Leaf

Cascardi

O’Leary

Cano

(1997). Efficacy of a dating violence prevention program on attitudes justifying aggression. Journal of Adolescent Health, 21, 11-17.

Babcock

J. C.

Costa

D. M.

Green

C. E.

(2004). What situations induce intimate partner violence? A reliability and validity study of the Proximal Antecedents to Violent Episodes (PAVE) scale. Journal of Family Psychology, 18, 433-442.

Banyard

V. L.

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

Banyard

V. L.

Moynihan

M. M.

Plante

E. G.

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

Berkowitz

A. D.

(2004). The social norms approach: Theory, research and annotated bibliography. Retrieved from http://www.edc.org/hec/socialnorms

Borsari

Carey

K. B.

(2001). Peer influences on college drinking: A review of the research. Journal of Substance Abuse, 13, 391-424.

10.

Bronfenbrenner

(1979). The ecology of human development. Cambridge, MA: Harvard University Press.

11.

Bushman

B. J.

Anderson

C. A.

(2002). Violent video games and hostile expectations: A test of the general aggression model. Personality and Social Psychology Bulletin, 28(12), 1679-1686.

12.

Cialdini

R. B.

Reno

R. R.

Kallgren

C. A.

(1990). A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology, 58, 1015-1026.

13.

Connolly

Friedlander

Pepler

Craig

Laporte

(2010). The ecology of adolescent dating aggression: Attitudes, relationships, media use, and socio-demographic risk factors. Journal of Aggression, Maltreatment & Trauma, 19, 469-491.

14.

Cornelius

T. L.

Resseguie

(2007). Primary and secondary prevention programs for dating violence: A review of the literature. Aggression and Violent Behavior, 12, 364-375.

15.

DeGue

Holt

M. K.

Massetti

G. M.

Matjasko

J. L.

Tharp

A. T.

Valle

L. A.

(2012). Looking ahead toward community-level strategies to prevent sexual violence. Journal of Women’s Health, 21(1), 1-3.

16.

Deutsch

Gerard

H. B.

(1955). A study of normative and informational social influences upon individual judgment. The Journal of Abnormal and Social Psychology, 51, 629-636.

17.

Finkel

E. J.

DeWall

C. N.

Slotter

E. B.

McNulty

J. K.

Pond

R. S.

Atkins

D. C.

(2011). Using I3 theory to clarify when dispositional aggressiveness predicts intimate partner violence perpetration. Journal of Personality and Social Psychology, 102, 533-549.

18.

Finkel

E. J.

DeWall

C. N.

Slotter

E. B.

Oaten

Foshee

V. A.

(2009). Self-regulatory failure and intimate partner violence perpetration. Journal of Personality and Social Psychology, 3, 483-499.

19.

Foshee

V. A.

(1998). Involving schools and communities in preventing adolescent dating abuse. In Arriaga

X. B.

Oskamp

Arriaga

X. B.

Oskamp

(Eds.), Addressing community problems: Psychological research and interventions (pp. 104-129). Thousand Oaks, CA: SAGE.

20.

Foshee

V. A.

Linder

MacDougall

J. E.

Bangdiwala

(2001). Regular article: Gender differences in the longitudinal predictors of adolescent dating violence. Preventive Medicine, 32, 128-141.

21.

Gidycz

C. A.

Orchowski

L. M.

Berkowitz

A. D.

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

22.

Gover

Kaukinen

Fox

(2008). The relationship between violence in the family of origin and dating violence among college students. Journal of Interpersonal Violence, 23, 1667-1693.

23.

Hamby

(2016). Self-report measures that do not produce gender parity in intimate partner violence: A multi-study investigation. Psychology of Violence, 6, 323-335.

24.

Hamby

Jackson

(2010). Size does matter: The effects of gender on perceptions of dating violence. Sex Roles, 63, 324-331.

25.

Hancock

L. C.

Henry

N. W.

(2003). Perceptions, norms, and tobacco use of college residence hall freshmen: Evaluation of a social norms marketing intervention. In Perkins

H. W.

Perkins

H. W.

(Eds.), The social norms approach to preventing school and college age substance abuse: A handbook for educators, counselors, and clinicians (pp. 135-153). San Francisco, CA: Jossey-Bass.

26.

Jaffe

P. G.

Sudermann

Reitzel

Killip

S. M.

(1992). An evaluation of a secondary school primary prevention program on violence in intimate relationships. Violence and Victims, 7, 129-146.

27.

Johnson

M. P.

(1995). Patriarchal terrorism and common couple violence: Two forms of violence against women. Journal of Marriage and the Family, 57, 283-294.

28.

Kallgren

C. A.

Reno

R. R.

Cialdini

R. B.

(2000). A focus theory of normative conduct: When norms do and do not affect behavior. Personality and Social Psychology Bulletin, 26, 1002-1012.

29.

Kline

R. B.

(2011). Principles and practice of structural equation modeling (3rd ed.). New York, NY: Guilford Press.

30.

Lapinski

M. K.

Rimal

R. N.

(2005). An explication of social norms. Communication Theory, 15, 127-147.

31.

Latane

(1981). Psychology of social impact. American Psychologist, 36, 343-356.

32.

Lewis

M. A.

Neighbors

(2004). Gender-specific misperceptions of college student drinking norms. Psychology of Addictive Behaviors, 18, 334-339.

33.

Lewis

S. F.

Fremouw

(2001). Dating violence: A critical review of the literature. Clinical Psychology Review, 21, 105-127.

34.

Lichter

E. L.

McCloskey

L. A.

(2004). The effects of childhood exposure to marital violence on adolescent gender-role beliefs and dating violence. Psychology of Women Quarterly, 28, 244-257.

35.

McDonnel

Ott

Michell

(2010). Predicting dating violence victimization and perpetration among middle and high school students in a rural southern community. Children and Youth Services Review, 32, 1458-1463.

36.

Miner

M. H.

Peterson

J. L.

Welles

S. L.

Jacoby

S. M.

Rosser

B. R. S.

(2009). How do social norms impact HIV sexual risk behavior in HIV-positive men who have sex with men? Multiple mediator effects. Journal of Health Psychology, 14, 761-770.

37.

Neighbors

LaBrie

J. W.

Hummer

J. F.

Lewis

M. A.

Lee

C. M.

Desai

Larimer

M. E.

(2010). Group identification as a moderator of the relationship between perceived social norms and alcohol consumption. Psychology of Addictive Behaviors, 24, 522-528.

38.

Neighbors

Larimer

M. E.

(2004). Self-determination and problem gambling among college students. Journal of Social and Clinical Psychology, 23, 565-583.

39.

Neighbors

Larimer

M. E.

Lewis

M. A.

(2004). Targeting misperceptions of descriptive drinking norms: Efficacy of a computer-delivered personalized normative feedback intervention. Journal of Consulting and Clinical Psychology, 72, 434-447.

40.

Neighbors

Rodriguez

L. M.

Rinker

D. V.

Gonzales

R. G.

Agana

Tackett

J. L.

Foster

D. W.

(2015). Efficacy of personalized normative feedback as a brief intervention for college student gambling: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 83, 500-511.

41.

Neighbors

Walker

D. D.

Mbilinyi

L. F.

O’Rourke

Edleson

J. L.

Zegree

Roffman

R. A.

(2010). Normative misperceptions of abuse among perpetrators of intimate partner violence. Violence Against Women, 16, 370-386.

42.

O’Keefe

(1998). Factors mediating the link between witnessing interparental violence and dating violence. Journal of Family Violence, 13, 39-57.

43.

Perkins

H. W.

(2002). Social norms and the prevention of alcohol misuse in collegiate contexts. Journal of Studies on Alcohol, Issue S14, pp. 164-172.

44.

Perkins

H. W.

Meilman

P. W.

Leichliter

J. S.

Cashin

J. S.

Presley

C. A.

(1999). Misperceptions of the norms for the frequency of alcohol and other drug use on college campuses. Journal of American College Health, 47, 253-258.

45.

Perry

Sibley

C. G.

(2011). Social dominance orientation: Mapping a baseline individual difference component across self-categorizations. Journal of Individual Differences, 32, 110-116.

46.

Pratto

Sidanius

Stallworth

L. M.

Malle

B. F.

(1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67, 741-763.

47.

Preacher

K. J.

Hayes

A. F.

(2008). Contemporary approaches to assessing mediation in communication research. In Hayes

A. F.

Slater

M. D.

Snyder

L. B.

Hayes

A. F.

Slater

M. D.

Snyder

L. B.

(Eds.), The SAGE sourcebook of advanced data analysis methods for communication research (pp. 13-54). Thousand Oaks, CA: SAGE.

48.

Raven

J. C.

(1998). Raven’s progressive matrices. Oxford, UK: Oxford Psychologists Press.

49.

Rimal

R. N.

Real

(2003). Understanding the influence of perceived norms on behaviors. Communication Theory, 13, 184-203.

50.

Rosen

K. H.

Bezold

(1996). Dating violence prevention: A didactic support group for young women. Journal of Counseling & Development, 74, 521-525.

51.

Schwartz

J. P.

Griffin

L. D.

Russell

M. M.

Frontaura-Duck

(2006). Prevention of dating violence on college campuses: An innovative program. Journal of College Counseling, 9, 90-96.

52.

Sellers

C. S.

Bromley

M. L.

(1996). Violent behavior in college student dating relationships: Implications for campus service providers. Journal of Contemporary Criminal Justice, 12, 2-27.

53.

Sherif

(1936). The psychology of social norms. Oxford, UK: Harper.

54.

Shorey

R. C.

Brasfield

Febres

Cornelius

T. L.

Stuart

G. L.

(2012). A comparison of three different scoring methods for self-report measures of psychological aggression in a sample of college females. Violence and Victims, 27, 973-990.

55.

Sidanius

Levin

Liu

Pratto

(2000). Social dominance orientation, anti-egalitarianism and the political psychology of gender: An extension and cross-cultural replication. European Journal of Social Psychology, 30, 41-67.

56.

Simpson

L. E.

Doss

B. D.

Wheeler

Christensen

(2007). Relationship violence among couples seeking therapy: Common couple violence or battering? Journal of Marital & Family Therapy, 33, 270-283.

57.

Slep

A. M.

Cascardi

Avery-Leaf

O’Leary

K. D.

(2001). Two new measures of attitudes about the acceptability of teen dating aggression. Psychological Assessment, 13, 306-318.

58.

Stith

S. M.

McCollum

E. E.

Amanor-Boadu

Smith

(2012). Systemic perspectives on intimate partner violence. Journal of Marital & Family Therapy, 38, 220-240. doi:10.1111/j.1752-0606.2011.00245.x

59.

Straus

M. A.

(2004). Prevalence of violence against dating partners by male and female university students worldwide. Violence Against Women, 10, 790-811.

60.

Straus

M. A.

(2008). Dominance and symmetry in partner violence by male and female university students in 32 nations. Children and Youth Services Review, 30, 252-275.

61.

Straus

M. A.

Hamby

S. L.

Boney-McCoy

Sugarman

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

62.

Whitaker

D. J.

Morrison

Lindquist

Hawkins

S. R.

O’Neil

J. A.

Nesius

A. M.

. . . Reese

(2006). A critical review of interventions for the primary prevention of perpetration of partner violence. Aggression and Violent Behavior, 11, 151-166.

63.

Witte

T. H.

Mulla

M. M.

(2012). Social norms for intimate partner violence in situations involving victim infidelity. Journal of Interpersonal Violence, 27, 3389-3404.

64.

Witte

T. H.

Mulla

M. M.

(2013). Social norms for intimate partner violence. Violence and Victims, 28, 959-967.

65.

Witte

T. H.

Mulla

M. M.

Weaver

A. A.

(2015). Perceived social norms for intimate partner violence in proximal and distal groups. Violence and Victims, 30, 691-698.

66.

Wolf

E. J.

Harrington

K. M.

Clark

S. L.

Miller

M. W.

(2013). Sample size requirements for structural equation models: An evaluation of power, bias, and solution propriety. Educational and Psychological Measurement, 73, 913-934.

67.

Wolfe

D. A.

Wekerle

Scott

Straatman

C. G.

Reitzel-Jaffe

(2003). Dating violence prevention with at-risk youth: A controlled outcome evaluation. Journal of Consulting and Clinical Psychology, 71, 279-291.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.22 MB