Abstract
The purpose of this study was to better understand young adults’ perceptions of what behaviors constitute intimate partner violence (IPV) and the correlates of these perceptions using a comprehensive measure of IPV perceptions and behaviors. Participants were undergraduates (aged 18-25), including 357 women and 346 men (N = 703) from the midwestern region of the United States, who completed surveys for course credit. Results demonstrated that young women and men on average reported that acts of physical, sexual, and psychological IPV were abusive. However, young women generally rated these behaviors as more abusive than young men, male-to-female (M-to-F) IPV was viewed as more abusive than female-to-male (F-to-M) IPV, and physical IPV was considered the most abusive form of IPV, followed by sexual IPV, which was rated as more abusive than psychological IPV. Furthermore, among men, a history of IPV perpetration and victimization generally predicted decreased perceptions that acts were abusive; however, among women, histories of IPV perpetration and victimization were generally unrelated to abuse perceptions. These data underscore the importance of the inclusion of psychoeducation about the seriousness of all forms of IPV in IPV prevention programming and the importance of situation-specific and targeted IPV prevention messages. Moreover, future research is needed to replicate and better understand the explanatory mechanisms underlying the relationships among a history of IPV, abuse perceptions, and gender.
Research consistently documents that intimate partner violence (IPV), which includes physical, psychological, and sexual aggression toward a current or former partner, is a major public health concern (Black et al., 2011). College students are an especially important group to study, given the high rates of IPV on college campuses (Katz, Kuffel, & Coblentz, 2002; Nabors, Dietz, & Jasinksi, 2006). Specifically, reviews of the literature indicate that 10% to 50% of college students experience some form of IPV (Kaukinen, 2014), and when psychological abuse (i.e., disparaging or hurtful comments) is considered, more than 80% of college women and men endorse experiencing such behaviors (e.g., Harned, 2002; Shook, Gerrity, Jurich, & Segrist, 2000). There is a growing body of research focusing on the correlates and predictors of IPV victimization and perpetration (see Dardis, Dixon, Edwards, & Turchik, 2015, for a review). However, less research has focused specifically on the extent to which college students consider specific acts of physical, psychological, or sexual violence to be abusive. Such research is critical to understand given that perceptions of and experiences with IPV are linked and provide an important point of prevention and intervention with adolescents and young adults. Grounded in gender schema theory (Bem, 1981), the purpose of the current study was to examine college students’ perceptions of what types of behaviors constitute IPV, including whether genders of respondent, victim, or perpetrator are associated with perceptions of abuse, and how a personal history of IPV perpetration or victimization is linked with perceptions of abuse. Given the focus of the current study on college students in the United States, all the studies cited herein are based on U.S. samples unless otherwise noted as there are great cross-cultural variations in perceptions of gender and IPV (Nayak, Byrne, Martin, & Abraham, 2003).
Theoretical Framework
Our understanding of people’s perceptions of gender and IPV can be understood through gender schema theory. Gender schema theory (Bem, 1981) stipulates that the male–female distinction forms a schema, an anticipatory cognitive structure that leads to the automatic invocation of sex-typing when processing and assimilating incoming information. Schematic processing leads to a general readiness to process information according to the schema-relevant category. When applied to gender schema theory, Bem (1981) asserts that this process leads to readiness to apply certain dimensions and attributes to one gender rather than the other. Schemas develop because of the cultural insistence on gender norms (Bem, 1983). The masculine gender role emphasized in gender schema theory includes qualities of toughness, independence, aggressiveness, and dominance, whereas girls and women are expected to be affiliative, submissive, emotionally compliant, and sexually passive (Herek, 1986; West & Zimmerman, 1987). Utilization of the gender schema is so automatic and powerful that entire dimensions (e.g., “strength” for men, “nurturance” for women) are essentially absent from application to the other gender (Bem, 1981).
Gender schemas, as they relate to violence and aggression, may affect perceptions of abuse based on a combination of (a) the gender of the individual whose perceptions are being assessed, (b) the genders of the perpetrator and victim, and (c) personal history of IPV. First, gender schemas include self-referential and self-evaluative information that are believed to become a “prescriptive standard or guide” (Bem, 1981), such that one’s own gender is likely to influence perceptions of gender-related constructs. Violence and aggression, which are more consistent with the masculine role, may be considered more socially acceptable by men than by women. Thus, men may perceive acts to be less problematic or abusive than do women. Second, because violence and aggression are more consistent with the masculine role, they are more easily processed as masculine traits than as feminine traits, and the “victim” role (passivity, compliance) is more easily processed as a feminine gender role. Women’s attributes of passivity, affiliativeness, and smaller size, are inconsistent with violent or threatening figures, which may lead to women’s use of violence being perceived as less problematic than men’s (Hamby & Jackson, 2010). These socially constructed notions of masculinity and femininity may influence perceptions of abuse based on the gender of the perpetrator and victim, such that IPV perpetrated by men and experienced by women may be perceived as more severe than IPV perpetrated by women and experienced by men (Hamby & Jackson, 2010). Third, considering the role of self-referential information in schematic processing, an individual’s own history of IPV perpetration and victimization may influence his or her perceptions of IPV. Previous research suggests that adolescents and young adults often minimize experiences of IPV (Chung, 2007; Edwards et al., 2012) and that IPV victims and perpetrators are generally more accepting of IPV than non-victims and perpetrators (Price & Byers, 1999).Thus, individuals who have perpetrated IPV or who have been victims of IPV may include this information in their schematic perceptions of abuse and perceive acts to be less abusive (e.g., Ehrensaft & Vivian, 1999).
Literature Review
Gender of Respondent
In accordance with the notion from gender schema theory that men will view violence as less problematic than will women, research among high school students and undergraduates consistently finds that boys and men are more accepting of IPV than are girls and women (e.g., Bryant & Spencer, 2003; Carlson, 1999; Dardis, Edwards, Kelley, & Gidycz, 2013; Nabors et al., 2006). For example, Price and Byers (1999) found in their sample of Canadian high school students that boys, compared with girls, were significantly more accepting of male-to-female (M-to-F) physical, M-to-F psychological, and M-to-F sexual IPV as well as female-to-male (F-to-M) physical, F-to-M psychological, and F-to-M sexual IPV. It is likely that individuals who are more accepting of specific forms of violence would consider such acts to be less “abusive” than would individuals who are less accepting of such acts. However, the existing research largely examines the “acceptability” of violence rather than the extent to which acts are considered abusive. In the only known study among college students to examine gender differences in the extent to which acts are considered abusive, Beyers, Leonard, Mays, and Rosen (2000) found that college men’s and women’s ratings of abusiveness did not vary by severity, abuse type, or victim gender, with the exception that female participants rated the F-to-M sexual IPV scenario as significantly more abusive than male participants. Future research is needed on the gender of the respondent to determine whether boys and men, in addition to deeming IPV more acceptable than do women, also deem IPV acts to be more abusive than do women.
Gender of Perpetrator/Victim
In addition to varying by the gender of the respondent, and consistent with gender schema theory, research has also found that the gender of the perpetrator and victim impact perceptions of abuse. Survey and experimental vignette research across high school, undergraduate, and community samples consistently finds that violence perpetrated by boys and men against their female partners (i.e., M-to-F IPV) is perceived as more severe, illegal, frightening, harmful, in need of social intervention, and deserving of action than violence perpetrated by girls and women against their male partners (e.g., Carlson & Worden, 2005; Cormier & Woodsworth, 2008; Hamby & Jackson, 2010; Price & Byers, 1999; Seelau & Seelau, 2005; Sorenson & Thomas, 2009).
History of IPV Victimization/Perpetration
Furthermore, in line with the role of self-referential processing outlined in gender schema theory, the extent to which behaviors are considered abusive may also vary based on one’s own experience with IPV. For example, some researchers have asked individuals who report behavioral experiences of IPV if they actually label their own experiences as IPV. These self-labeling studies using college and community samples find that 15% to 60% of individuals who experience IPV victimization and/or perpetration actually label their own experiences as such (Hammond & Calhoun, 2007; Miller, 2011; Perry & Fromuth, 2005; Stets & Pirog-Good, 1989). Researchers suggest that individuals do not self-identify because they minimize or become desensitized to abuse (Perry & Fromuth, 2005). Based on this minimization conceptualization, it is possible that individuals with a personal history of IPV would be less likely to rate IPV acts as abusive. For example, Ehrensaft and Vivian (1999) found that college students with a history of physical IPV victimization and/or perpetration rated restrictive, domineering, and coercive behaviors as less controlling than individuals without a history of physical IPV victimization or perpetration. By contrast, Beyers et al. (2000) found college students’ abusiveness ratings of vignettes depicting emotional, physical, and sexual abuse of both female and male victims were not correlated with participants’ personal experiences with IPV perpetration and victimization; however, this study did not utilize a validated measure of IPV victimization and perpetration. Further research is needed to clarify the nature of the relationship between one’s own experiences with IPV victimization and perpetration and their perceptions of IPV using validated measures.
Type of Violence
Although not explicitly guided by gender schema theory, previous research has found that the type of violence affects perceptions of how abusive IPV is deemed to be. Specifically, physical and sexual abuse among married couples are more likely to be perceived to be “domestic violence” and considered abusive, illegal, and in need of social interventions than psychological marital/cohabiting IPV by community adult and undergraduate samples (Capezza & Arriaga, 2008; Carlson & Worden, 2005; Nabors et al., 2006; Sorenson & Thomas, 2009). In addition, psychological IPV is often overlooked by practitioners and policy makers more so than other types of IPV (Sullivan, McPartland, Armeli, Jaquier, & Tennen, 2012), perhaps because perceptions of abuse severity are based on the extent of injuries that the victim suffers (Lane & Knowles, 2000).Whereas the Nabors et al. (2006) and Capezza and Arriaga (2008) studies sampled college students, these studies assessed perceptions of husband–wife violence, and did not inquire about perceptions of IPV among dating couples. No known study has examined college students’ perceptions of the relative abusiveness of physical, sexual, and psychological IPV committed by college students. Examining perceptions of the abusiveness of these acts among dating couples may be important given that college dating relationships include younger partners, fewer economic and familial attachments, greater peer pressure, and potentially more pronounced gender roles than marital relationships (e.g., Shorey, Cornelius, & Bell, 2008).
Limitations of Existing Research
Although a growing body of research has enhanced our understandings of individuals’ perceptions of behaviors that constitute IPV, there remain several identified gaps in the literature. Of the studies that have examined abuse perceptions among college students, half of them have assessed perceptions of marital/cohabiting IPV (i.e., Cormier & Woodsworth, 2008; Nabors et al., 2006; Seelau & Seelau, 2005) as opposed to perceptions of IPV among dating partners (i.e., Beyers et al., 2000; Hamby & Jackson, 2010; Price & Byers, 1999), which is clearly more relevant to this age group and important for IPV prevention efforts. Furthermore, whereas a number of studies have examined whether perceptions of acceptability and severity of IPV vary as a function of the participant, perpetrator, and victim gender (e.g., Price & Byers, 1999), there is much less research that has concurrently examined whether the participant’s IPV history and the type of acts (e.g., physical, sexual) affect perceptions of IPV as well. Understanding which specific acts are perceived to be most and least abusive is critical for prevention efforts. For example, if college students believe some abusive acts are not abusive (e.g., sexual IPV), social norms approaches may need to focus more heavily on education and shifting norms about these behaviors. Beyers et al. (2000) is the only known study with college students that has compared the extent to which men and women perceive specific acts as abusive; although this study made an important contribution and is the only study to examine the role of past relationship violence in perceptions of abuse, Beyers and colleagues utilized a non-validated measure of IPV history, which may limit the validity of the findings and our understanding of the extent of students’ perceptions of specific acts of IPV. In addition, the Beyers et al. study was a fully crossed, between-subjects study assessing perceptions of vignettes, such that participants rated the abusiveness of just one form of violence and for a particular gender of perpetrator and victim. Consequently, comparisons cannot be made between ratings of abusiveness across forms of violence and gender of perpetrator and victim. Finally, no study to date has considered the role of social desirability in the study of perceptions, despite research documenting that there is a negative relationship between reporting of IPV and social desirability (e.g., Sugarman & Hotaling, 1997), which underscores the need for the inclusion of social desirability in perceptions of IPV research. In sum, this is the first study to our knowledge that has concurrently examined the extent to which college men and women consider M-to-F and F-to-M physical, sexual, and psychological acts to be abusive using standardized instruments as well as examined how IPV victimization and perpetration histories relate to abuse perceptions, while controlling for social desirability. In addition to providing information that could be important in tailoring IPV prevention and intervention efforts, the current study adds to the broader literature on gender and cultural issues as they relate to discourses about and perceptions of IPV among college students in U.S. society.
Research Aims and Hypotheses
The current study examined the extent to which perceptions of IPV varied as a function of (a) participant gender, (b) perpetrator/victim gender, (c) participant history of IPV victimization and perpetration, and (d) abuse type (i.e., psychological, sexual, physical). Based on previous research and gender schema theory, which holds that violence and aggression are more accepted by men due to the relevance of aggression to the male schema, the first hypothesis was that, across forms of IPV, young men will view IPV as less abusive than will young women. In addition, because aggression is less consistent with the female schema, the second hypothesis was that M-to-F IPV would be considered more abusive than F-to-M IPV; however, given the dearth of research, the extent to which these perceptions may vary by men and women was exploratory. Despite mixed empirical research regarding the role of IPV history and perceptions of abuse, gender schema theory suggests that self-referential information is critical to perceptions, and that victims and perpetrators of abuse are more likely to minimize its impact. Therefore, the third hypothesis was that a history of IPV perpetration and/or victimization would be related to lower abusiveness ratings; given the dearth of research, the extent to which these perceptions may vary by men and women was exploratory. Finally, drawing on existing research indicating that severity of violence is associated with the extent of injury (Lane & Knowles, 2000), the fourth hypothesis was that physical and sexual IPV would be considered more abusive than psychological IPV; we ventured no a priori hypotheses, however, about potential differences in abuse perception between physical and sexual IPV or how these perceptions may vary by gender.
Method
Participants
Participants were 357 college women and 346 college men (N = 703) who were between the ages of 18 and 25 (M = 18.89, SD = 1.06). The majority of participants identified as White (87.5%), followed by Black (5.1%), Other/Multiracial (3.6%), Asian (2.7%), and Hispanic (1.1%). A little more than half (62.1%) of the sample reported that their combined annual family incomes were greater than US$51,000 a year, 18.3% reported that they were less than US$51,000 per year, and 19.6% reported that they did not know their family income. The vast majority (93.2%) of the sample reported exclusively heterosexual experiences, whereas 6.8% of the sample reported more heterosexual experiences than homosexual experiences or equal heterosexual and homosexual experiences. Of note, individuals who reported exclusively or predominantly homosexual experiences were not included in the final sample, given that IPV perceptions among opposite-sex couples was the focus of the present study. Table 1 provides demographic data separately for women and men.
Demographic Data for Female and Male Participants.
Measures
Adolescent/adult dating violence victimization and perpetration
The Revised Conflict Tactics Scale (CTS2; Straus, Hamby, Boney-McCoy, & Sugarman, 1996) was used to identify men and women who had experienced any IPV victimization and perpetration from a dating partner, to obtain a lifetime estimate of IPV victimization and perpetration (modified from original scope of past 12 months to obtain a more inclusive lifetime history of IPV). Participants were instructed to consider acts between themselves and a dating partner, defined on the survey as “someone you were/are in any type of relationship with ranging from casual and non-committed relationship to a very serious relationship.” Items assessed victimization and perpetration of physical (12 items; e.g., “pushed or shoved me [him/her]”), sexual (7 items; e.g., “made me [him/her] have sex without a condom”), and/or psychological (8 items; e.g., “called me [him/her] fat or ugly”) IPV. For each item, participants were asked, “how many times a dating partner has done these things to you” and “how many times you have done this to a dating partner,” to assess victimization and perpetration, respectively. Response options for all items range from Never (0) to More than 20 times (6). For each of the three victimization and three perpetration subscales, participants received a continuous score reflecting the number of times they experienced that specific type of IPV. Dichotomous scores were also computed, to enable comparison of rates with previous studies.
Perceptions of dating violence
The CTS2 (Straus et al., 1996), which was used to measure physical, sexual, and psychological IPV victimization and perpetration, was modified to assess participants’ perceptions of IPV. Specifically, participants were asked to rate the extent to which they considered each of the CTS2 items abusive for both M-to-F IPV and F-to-M IPV, using a 7-point scale, with anchors of 1 (Not at all abusive), 4 (Somewhat abusive) and 7 (Very Abusive). Mean scores were computed for each of the six subscales: M-to-F Physical, Sexual, and Psychological, and F-to-M Physical, Sexual, and Psychological. An example is, “a boyfriend pushing or shoving his girlfriend” (M-to-F Physical). Higher scores were indicative of greater perceptions that the acts were abusive. Internal consistencies for each of the subscales were acceptable: M-to-F Physical (Cronbach’s α = .83), M-to-F Sexual (α = .74), M-to-F Psychological (α = .85), F-to-M Physical (α = .92), F-to-M Sexual (α = .88), and F-to-M Psychological (α = .88).
Social desirability
Participants’ desire for social approval was assessed using the Marlowe-Crowne Social Desirability Scale (MCSDS; Reynolds, 1982) as a control variable, given that self-report measures are susceptible to social desirability. The MCSDS consists of 13 true–false items (e.g., “I am always courteous, even to people who are disagreeable”) assessing personal attitudes and traits relating to social desirability. Each participant received a summed total score, with higher scores representing greater social desirability. Internal consistency for the current sample was moderate (Cronbach’s α = .68).
Procedure
Men and women were recruited via an online database of all introductory psychology students at a midwestern university and received course credit for their participation. The solicitation message for the study was titled “Opinions About Social and Dating Experiences,” and all individuals 18 years and above were invited to participate, with no other requirements or information about the study provided in the recruitment message; detailed study information, however, was provided during the informed consent process. Although participants signed up for the study via the website, participants completed the identical surveys in person (paper and pencil) in gender-specific group testing environments administered by same-gender graduate students. Of note, the questionnaires assessing perceptions of abusive behavior were presented prior to the questions about IPV experiences, to attempt to minimize the personal relevance of the perceptions of abuse questions. Participants received and signed consent forms prior to participation and were provided with debriefing and referral information following participation. All study procedures were approved by the university’s institutional review board.
Results
Descriptive and Basic Inferential Statistics
Descriptive statistics for IPV victimization and perpetration are displayed in Table 2, with frequencies were provided to enable comparison with existing research. When examining IPV as a continuous variable, which reflected the sum of the number of times that participants experienced each type of victimization or engaged in each form of perpetration, women reported significantly higher physical and psychological perpetration, whereas men reported significantly higher sexual perpetration. There were no differences between men and women on any of the IPV victimization variables. Regarding the control variable, women (M = 6.86, SD = 2.80) reported significantly higher levels of social desirability than men (M = 6.86, SD = 2.80), t(681) = 3.21, p < .01.
Descriptive Statistics for IPV Victimization and Perpetration Variables.
Note. IPV = intimate partner violence.
p < .05. **p < .01. ***p < .001.
IPV Perceptions as a Function of Participant Gender (Hypothesis 1)
To test the hypothesis that men would consider physical, sexual, and psychological IPV to be less abusive than women (Hypothesis 1), a one-way MANCOVA was conducted with social desirability as a covariate. There was a statistically significant effect for gender on the combined abuse perception dependent variables, F(6, 620) = 22.34, p < .001, Wilks’s Lambda = .82, η2 = .18. Follow-up, tests of between-subject effects (Table 3) suggested that men rated all forms of M-to-F and F-to-M IPV as significantly less abusive than did women, although the magnitude of these differences varied from small (i.e., M-to-F physical, M-to-F psychological, F-to-M psychological) to medium (i.e., F-to-M physical, M-to-F sexual) to large (i.e., F-to-M sexual). Thus, results supported the hypothesis that men would consider all IPV acts as less abusive than women. For men and women, the average rating for all forms of violence were between 4.30 and 6.65, indicating that they considered the acts to be between “somewhat” and “very” abusive.
Perceptions of Dating Violence as a Function of Gender of Participants.
Note. F and Cohen’s d values reflect follow-up, tests of between-subjects effects as part of the MANCOVA examining perceptions of IPV as a function of gender. Perceptions scores are measured on a 1 to 7 scale, with higher scores indicating greater perception that a given act is abusive. Positive d values indicate that women endorsed higher perceptions that an act was abusive. Estimated marginal means displayed in the table apply to all MANCOVA analyses. IPV = intimate partner violence.
p < .001.
IPV Perceptions as a Function of Perpetrator/Victim Gender (Hypothesis 2), and Abuse Type (Hypothesis 4)
To test the hypotheses that M-to-F IPV would be viewed as more abusive than would F-to-M IPV (Hypothesis 2), and that physical and sexual IPV would be considered more abusive than psychological IPV (Hypothesis 4), a two-way mixed MANCOVA was conducted. The six different abuse perception subscales (i.e., M-to-F physical, M-to-F sexual, M-to-F psychological, F-to-M physical, F-to-M sexual, and F-to-M psychological) were the within-subject variables, gender of the participant was the between-subject variable, and social desirability was the covariate. Mauchley’s test indicated violation of the assumption of sphericity, and thus, values were corrected using the Greenhouse-Geisser estimates of sphericity (ε = .782). The multivariate main effect for the abuse perception subscales was significant, F(3.59, 2242.17) = 198.29, p < .001, η2 = .24, indicating that there were significant differences in abusiveness ratings across the different abuse perception subscales. However, as suggested by the significant multivariate abuse perception subscales by gender interaction, F(3.59, 2242.17) = 31.09, p < .001, η2 = .05, the degree of differences in abuse perceptions was not equal for men and women; post hoc comparisons of marginal means were thus conducted separately for men and women. Although Levene’s test indicated violation of the assumption of equality of error variances, this was not deemed to be problematic given the large sample size, and the ratio of the largest to smallest variance was no more than 4:1 (Meyers & Well, 1995).
Post hoc comparisons of marginal means (Table 3) generally supported the Hypothesis 2 that M-to-F IPV would be viewed as more abusive than would F-to-M IPV. This hypothesis was supported with regard to physical and psychological abuse (all ps < .001); both men and women rated M-to-F physical IPV and M-to-F psychological IPV as more abusive than F-to-M physical IPV, and F-to-M psychological IPV, respectively. However, with regard to perceptions of sexual IPV, whereas men (p < .001) rated M-to-F sexual IPV as more abusive than F-to-M sexual IPV, women (p = .05) rated sexual IPV regardless of the gender of the perpetrator as equally abusive. These results are graphically displayed in Figure 1.
Estimated marginal means for abuse perception subscales for Hypothesis 2.
Post hoc comparisons of marginal means (Table 3) generally support Hypothesis 4 that physical and sexual IPV would be considered more abusive than psychological IPV; these findings were consistent regardless of the participant and perpetrator gender. More specifically, across both participant and perpetrator gender, physical IPV was rated most abusive, and significantly more abusive than both sexual and psychological IPV, with psychological IPV rated the least abusive (significantly less abusive than sexual IPV). The pattern of results was such that, among both men and women (all ps < .001), M-to-F physical IPV was rated as more abusive than M-to-F sexual IPV, which were both rated as more abusive than M-to-F psychological IPV. A parallel pattern of significant differences (all ps < .001) emerged for the three F-to-M forms of IPV. Likewise, F-to-M physical IPV was rated as more abusive than M-to-F sexual IPV, which were both rated as more abusive than F-to-M psychological IPV.
Relationships Between IPV Victimization and Perpetration History and IPV Perceptions Among Men and Women (Hypothesis 3)
To test the hypothesis that individuals with a history of IPV perpetration or victimization would consider behaviors to be less abusive than individuals without such histories (Hypothesis 3), 12 gender-specific hierarchical, linear regression models were conducted (6 for men and 6 for women). The criterion variables were the six abuse perception variables: M-to-F physical IPV, M-to-F psychological IPV, M-to-F sexual IPV, F-to-M physical IPV, F-to-M psychological IPV, and F-to-M sexual IPV. In the first block of all regressions, we entered social desirability as a control variable. In the second block, we first conducted the analyses by entering all six subscales of the CTS2 into each regression. However, because of issues with multicollinearity due to the strong relationships among different CTS2 subscales (correlation coefficients ranging from .27 to .70, all significant at p < .001), these results are not presented and instead we present models in which we only included the one continuous IPV victimization or perpetration variable that corresponded to the criterion variable in the second block. Furthermore, we conducted analyses separately for men and women rather than conducting them together and including gender as an interaction term given that different independent variables (i.e., IPV subscales) were used for men and women for the same criterion variable. For example, to determine the effects of personal experience with each variable, when predicting M-to-F physical IPV abuse perceptions among men, physical IPV perpetration was entered in the second block (as men’s previous role in M-to-F physical IPV would be as the perpetrator), whereas for the M-to-F physical IPV abuse perception regression with women, physical IPV victimization was entered in the second block (as women’s previous role in M-to-F physical IPV would be as the victim).
For men (see Table 4), after controlling for social desirability, a history of the corresponding forms of IPV perpetration predicted lower levels of M-to-F physical, F(2, 319) = 12.50, p < .001, R2 = .073; sexual, F(2, 323) = 17.03, p < .001, R2 = .095; and psychological, F(2, 316) = 11.43, p < .001, R2 = .068, abuse perceptions; a history of the corresponding IPV victimization predicted lower levels of F-to-M physical, F(2, 321) = 7.55, p < .001, R2 = .045, and sexual F(2, 321) = 14.50, p < .001, R2 = .083, abuse perceptions. A history of psychological IPV victimization was unrelated to F-to-M psychological IPV abuse perceptions, yet the model, F(2, 319) = 3.81, p < .05, R2 = .023, was significant given that social desirability was a significant correlate of F-to-M psychological IPV abuse perceptions among men.
Hierarchical Linear Regressions Predicting Abuse Perceptions (Men).
Note. All IPV variables were continuous. IPV = intimate partner violence.
p < .05. **p < .01. ***p < .001.
For women (see Table 5), after controlling for social desirability, a history of physical IPV perpetration predicted lower levels of F-to-M physical abuse perceptions, F(2, 341) = 9.89, p < .001, R2 = .055. However, no other forms of IPV perpetration or victimization predicted perceptions of M-to-F or F-to-M abuse perceptions. More specifically, a history of the corresponding forms of IPV perpetration were unrelated to F-to-M sexual IPV abuse perceptions, F(2, 337) = 1.14, p = .321, R2 = .007, and F-to-M psychological IPV abuse perceptions, F(2, 336) = 3.24, p < .05, R2 = .019, and a history of the corresponding IPV victimization was unrelated to M-to-F psychological IPV, F(2, 333) = 0.20, p = .820, R2 = .001, M-to-F sexual IPV, F(2, 338) = 0.40, p = .961, R2 = .000, and M-to-F physical IPV abuse perceptions; M-to-F: F(2, 332) = 5.17, p < .01, R2 = .030. However, the models for both M-to-F physical IPV and F-to-M psychological IPV abuse perceptions were significant given that social desirability was a significant correlate of both types of abuse perceptions among women.
Hierarchical Linear Regressions Predicting Abuse Perceptions (Women).
Note. IPV = intimate partner violence.
p < .05. **p < .01. ***p < .001.
To summarize, the Hypothesis 3 that individuals with a history IPV would consider behaviors to be less abusive than individuals without such histories was only partially supported. Although among men, a history of IPV perpetration and victimization generally predicted lower levels of abuse perceptions, among women, histories of IPV perpetration and victimization were generally unrelated to abuse perceptions.
Discussion
The purpose of this study was to better understand college students’ perceptions of what behaviors constitute IPV and the correlates of these perceptions using a comprehensive measure of IPV perceptions and behaviors. Results were generally consistent with gender schema theory in implicating the role of gender and self-referential information in understanding perceptions of IPV. First, the gender of the respondent was associated with perceptions of abuse, as women reported that acts were more abusive than did men. In addition, the genders of the perpetrator and victim affected perceptions, as both women and men rated abuse perpetrated by men and experienced by women to be more abusive than abuse perpetrated by women and experienced by men. Furthermore, consistent with the concept of self-referential processing within gender schema theory, a history of IPV perpetration and victimization was generally related to lower perceptions that such acts were abusive, though this relationship was stronger among men than women. Finally, consistent with previous research (e.g., Capezza & Arriaga, 2008), psychological IPV was considered less abusive than were physical and sexual IPV.
As hypothesized, women consistently rated all forms of IPV (across gender of perpetrator and type of IPV) as more abusive than men. This is consistent with gender schema theory and with previous research with college and community samples documenting that men hold more accepting attitudes of IPV than women (e.g., Bryant & Spencer, 2003; Carlson, 1999; Nabors et al., 2006) and that women often perceive IPV to be more serious and severe than men (Hamby & Jackson, 2010; Little & Terrance, 2010). Whereas both men and women are harmed by violence, and experience injury, increased posttraumatic stress disorder (PTSD), depression, somatic and other negative consequences of IPV (e.g., Hines & Douglas, 2009; Próspero & Kim, 2009), women’s greater reported levels of fear, psychological consequences, and severe physical injury associated with IPV victimization (Archer, 2000; Foshee, 1996; Molidor & Tolman, 1998; Swan, Gambone, Caldwell, Sullivan, & Snow, 2008; Tjaden & Thoennes, 2000) could lead women to be more aware of the unacceptability and severe nature of these acts. Men, for whom aggression and violence are deemed heteronormative and gender role consistent (Bem, 1981), may not view these behaviors as problematic. Such findings should be integrated into programming efforts with men and efforts that focus on bystanders. Helping men and witnesses to such behavior to better understand the perceived coerciveness of such behaviors may lead to less acceptance of such behaviors on college campuses and challenge norms that support such behaviors.
As hypothesized, and consistent with previous research, all types of M-to-F IPV were viewed as more abusive than F-to-M with the exception that women rated sexual IPV perpetrated by men and women as equally abusive. This result is consistent with gender schema theory, in that hegemonic masculinity norms lead to a reluctance for men to be deemed victims of interpersonal violence (Turchik & Edwards, 2012). It is concerning that there is so much minimization of men’s experiences with IPV victimization, as this could decrease men’s help-seeking related to their own victimization. Indeed, the existing research has found that, relative to women with male partners, men with female partners are significantly less likely to report physical and sexual IPV to the police (Felson & Paré, 2005). Moreover, such minimization may deter peers from intervening in situations of IPV when a boy or young man is the victim. Furthermore, this minimization of men’s experiences with IPV victimization could increase women’s likelihood of perpetrating IPV against men.
Whereas M-to-F forms of IPV were generally rated as more abusive than F-to-M forms of IPV, the only exception to this was the finding that women rated sexual IPV perpetrated by men and women as equally abusive. It is possible that women rated F-to-M sexual IPV as abusive as M-to-F sexual IPV because the image of a woman being a sexual aggressor diverges greatly from social scripts depicting women as sexually passive (O’Sullivan & Byers, 1993). In fact, our data suggest that relative to both psychological and physical IPV perpetration, very few women in the present study engaged in sexual IPV perpetration. This is consistent with data that consistently shows that men are much more likely than women to perpetrate sexual aggression (Harned, 2001; Shorey, Meltzer, & Cornelius, 2010). Although speculative, such an unexpected divergence from norms regarding female sexuality could have influenced women’s ratings of abusiveness, and this may be especially salient for young women who are just transitioning to a college environment where attention to norms is particularly salient. However, across genders of respondent, victim, and perpetrator, physical IPV was rated as most abusive followed by sexual IPV, which was in turn followed by psychological IPV. These results are consistent with the cultural scripts regarding what is “legitimate” abuse, despite some research demonstrating that psychological abuse leads to psychological consequences just as severe if not more severe than physical and sexual abuse (e.g., Kessler, 2000). One reason that participants might minimize psychological abuse could be the lack of obvious or visible injury. In addition, psychological violence in the form of put-downs and disparaging remarks may be seen as relatively normative to many young adults, as approximately 80% of women and men endorsed experiencing psychological abuse, a rate consistent with other research (e.g., Harned, 2002; Shook et al., 2000).
In the final set of analyses, we tested the relationships between a history of IPV victimization and perpetration and IPV abuse perceptions. Results for men were consistent with gender schema theory, in that self-referential information appeared to affect men’s perceptions of violence. Among men, histories of IPV victimization and perpetration were generally related to lower abuse perceptions, whereas for women, histories of IPV victimization and perpetration were generally unrelated to abuse perceptions. However, effect sizes for the regressions were generally small. Previous research with college students has documented that there were generally null relationships between prior experiences with IPV and current abuse perceptions (e.g., Beyers et al., 2000). This previous research, however, neglected to examine how these relationships may differ for men and women, and did not include a validated measure of IPV victimization/perpetration. It is unclear why men’s, and not women’s, abuse histories were associated with perceptions of abuse. As there was somewhat greater variability in young men’s perceptions of abuse in the current study, personal experiences may have been more likely to exert an effect on perceptions for men than for women. Women, who, consistent with schemas for their gender, generally consider most acts to be more abusive than men, may be less affected by personal history. Regarding the relationship between IPV victimization and perceptions of acts as less abusive, whereas many men do experience physical or psychological injury as a result of IPV (Hines & Douglas, 2009, 2011), young men may minimize their IPV victimization as a result of cognitive dissonance (i.e., to match traditional gender schemas), or due to greater previous experience with aggression and violence (i.e., men are at the greatest risk of being victims of violence more generally; Felson, 2002). Alternatively, it may be that men’s minimization of the seriousness of such behavior is a result of cultural norms that condone violence. Future longitudinal investigations would lead to a better understanding of the temporal relationship between perceptions of IPV and engagement in abusive behavior.
In contrast, young women’s IPV perpetration and victimization experiences were unrelated to abuse perceptions in the current sample, with the exception of women’s physical IPV perpetration being related to lower perceptions of F-to-M physical abuse. Women may be most motivated to minimize their physical IPV perpetration (rather than psychological or sexual perpetration) because physical IPV is perceived to be the most abusive by both men and women. It is also possible that the form or function of women’s physical abuse differs from that of men; that is, women may be able to contextualize physically abusive acts they have committed as being in self-defense. Or, given that men are less likely to report negative consequences of abuse, women may minimize its impact on male partners. Although women also perceive F-to-M sexual violence to be very abusive, the low endorsement of sexual IPV perpetration (7.5%) could lead to lower power to detect an effect for perpetration. However, the general lack of significant relationship between personal experience and perceptions of abuse for both victimization and perpetration could simply be explained by women’s greater general perception of IPV as more abusive than that of men. Furthermore, women may have greater awareness of the negative consequences of violence, such that personal experiences of abuse do not add additional influence in explaining women’s perceptions of what they consider IPV.
Although the findings of the current study contribute to our understanding of perceptions of abuse among college women and men, there are several limitations that need to be mentioned. Our target population was U.S., middle-class, White, heterosexual college undergraduates. Thus, these findings cannot be generalized to non-college young adults or individuals from other age groups or other cultures, especially in light of research documenting age differences in perceptions of IPV. Furthermore, we used a cross-sectional design, and thus the temporal sequencing of relationships among IPV victimization and perpetration experiences and abuse perceptions are unknown. For example, perceptions of violence may lead to IPV perpetration. Future research would benefit from using longitudinal studies that would allow for a better understanding of the temporal sequencing of relationships under study as well as the development trajectory of abuse perceptions. Future research that includes qualitative methodologies could also benefit from better understanding, in young adults’ own words, how they perceive IPV and the factors associated with these perceptions. Along these lines, future research is needed to replicate and better understand the explanatory mechanisms underlying the relationships among a history of IPV, abuse perceptions, and gender (e.g., empirically why and how IPV history is related to abuse perceptions for men but not women).
The current study also did not assess whether IPV experiences were current or in the distant past; future research should investigate the impact of current or recent IPV on perceptions, as this may be more salient than past experiences and would minimize potential recall bias. We also did not include any relational (e.g., length of relationship) or contextual (e.g., motives for IPV perpetration) information when rating the extent to which participants considered CTS items to be abusive. This could be an important consideration in future research, particularly given that gender might have been more artificially salient in the ratings of abusiveness in the current study in the absence of any other contextual information (e.g., length, chronicity of abuse, consequences of abuse). Future research should examine in what ways individuals might perceive violence motivated by different reasons to be more or less abusive (e.g., whether abuse motivated by power/control are perceived to be more abusive than acts perpetrated in self-defense), as well as how other contextual factors (e.g., length of relationship, size of victim and perpetrator, inciting events), might predict perceptions of violence.
Despite the aforementioned limitations, data from the current study provide important guidance in understanding perceptions of violence among U.S. college students. To the extent that other cultures experience similar levels of hegemonic masculinity and traditional gender role expectations, results from the current study may be important to investigate for IPV prevention efforts. Within a college undergraduate population, the current study underscores the critical importance of the inclusion of psychoeducation about the seriousness of all forms of IPV in IPV prevention programs. Increasing awareness of the behaviors that constitute IPV and the legal repercussions of such violence may help in early IPV intervention efforts, which is particularly salient given evidence that IPV often escalates among college young adults (Smith, White, & Holland, 2003). However, the current findings suggest that programmers need to go beyond simply clarifying definitions of abuse as participants viewed the abusive incidents generally as abusive. Program efforts that seek to personalize the information and provide an atmosphere that will lead participants to process their own experiences and behavior more fully are likely to be more effective than the often brief psychoeducational programs that are offered on many college campuses (see Orchowski, Gidycz, & Murphy, 2010 review).
In the current study, both men and women minimized male victimization. This result may be reflective of broader societal minimization of men’s victimization, particularly in light of evidence from the IPV service utilization literature indicating that male victims of IPV may not be provided the same extent of shelter, outreach, therapy, and other services as are female victims of IPV (Hines & Douglas, 2011). Therefore, increasing psychoeducation and efforts toward legal interventions for men as victims is particularly important. At the same time, it is important to understand the likely complex differences (e.g., fear, psychological consequences, and physical injury) in what appear on the surface to be comparable types of victimizations between genders (Archer, 2000; Foshee, 1996; Hamby, 2009; Molidor & Tolman, 1998; Swan et al., 2008; Tjaden & Thoennes, 2000). Furthermore, results are consistent with recommendations from others (e.g., Gidycz, Orchowski, & Edwards, 2011) that suggest that IPV risk reduction and prevention programs may best be used in gender-specific format. Young men may specifically benefit from processing their own previous experiences of IPV in relation to their understanding of what behaviors constitute abuse from a social norms approach, including correcting men’s norms about abusive behavior (see Berkowitz, 2010 for a review of the social norms approach). Providing men with specific scenarios where contextual factors are manipulated is also likely important to challenge norms that support violence. Such work becomes more imperative in light of prior research indicating that men in greater denial about the abusiveness of acts are less likely than other men to show changes in abusive behavior (Murphy & Baxter, 1997; Scott & Wolfe, 2003). For young women, it appears that personal experiences with violence do not greatly affect their perceptions of what constitutes IPV, and thus, prevention programs for women may need to focus less on correction of norms about what constitutes abuse, focusing more on conflict management and problem-solving strategies to supplant abusive tactics (Dardis et al., 2013). By fostering and creating individual and community stances that are intolerant of all forms of IPV among young adults, we will be one step closer to eliminating this form of violence from our society.
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.