The Open Collections website will be unavailable July 27 from 2100-2200 PST ahead of planned usability and performance enhancements on July 28. More information here.

Open Collections

UBC Undergraduate Research

A Consideration of Gender in University Students’ Perceptions of Intimate Partner Abuse Cormier, Nicole 2006-07

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


52966-Cormier _Woodworth_ THESIS 073006.pdf [ 257.18kB ]
JSON: 52966-1.0086059.json
JSON-LD: 52966-1.0086059-ld.json
RDF/XML (Pretty): 52966-1.0086059-rdf.xml
RDF/JSON: 52966-1.0086059-rdf.json
Turtle: 52966-1.0086059-turtle.txt
N-Triples: 52966-1.0086059-rdf-ntriples.txt
Original Record: 52966-1.0086059-source.json
Full Text

Full Text

     Partner Abuse     1 Running head: PERCEPTIONS OF INTIMATE PARTNER ABUSE           A Consideration of Gender in University Students’ Perceptions of Intimate Partner Abuse Nicole Cormier University of British Columbia – Okanagan Supervisor: Dr. Mike Woodworth In partial fulfilment of the requirements of PSYC 480 July 30, 2006             Partner Abuse     2 Abstract Forty-one men and 67 women undergraduate students from the University of British Columbia’s Okanagan Campus were presented with vignettes depicting a physical abuse scenario. The vignettes were identical, except that gender pronouns were manipulated to portray all possible gender combinations of victim and perpetrator (e.g., male-female, male-male, female-male, and female-female). After reading each vignette, participants rated them by responding to a number of questions designed to measure perceptions of the depicted abuse (Harris & Cook, 1994), as well as ranking the four vignettes in order of severity.                                                                                                                                                            A mixed factorial design was utilized to examine potential differences in perception according to victim and perpetrator gender, and between men and women participants. Analyses supported the prediction that participants’ ratings of the vignettes would indicate an overall view that female perpetrators are not as capable of inflicting harm as male perpetrators. The prediction that women would react more strongly to any battering incident than men was partially supported. A third hypothesis expecting that men would show greater reluctance in reporting vignette situations to the police was not supported. It is hoped the proposed study will enhance academic and social understanding of partner abuse, and provoke increased research and education concerning the topic.       Partner Abuse     3 A Consideration of Gender in University Students’ Perceptions of Intimate Partner Abuse Physical violence in intimate relationships is a disquieting issue, constituting a major public health crisis in North American culture. The costs to the criminal justice system, health care, and law enforcement may pale in comparison to the personal suffering and even potential fatalities among the victims (see Campbell, Harris, & Lee, 1995).  The phenomenon of domestic violence against women is well documented in the research literature (see Archer, 2000; Dobash & Dobash, 1978; Walker, 1984; etc.). An estimated two million women each year are physically assaulted by a male partner (Burgess et al, 1997; Hamberger, 1994). The population estimates for incidence rates range from 3% (Murty et al., 2003) to as high as 51% (Magdol et al., 1997). The variation in these figures is believed to depend on what criteria are used, as well as whether clinical or population samples are drawn (Hegarty & Roberts, 1998).  Clinical samples often produce findings that women are predominantly the victims of partner abuse, and men the perpetrators (Archer, 2000; Capaldi & Owen, 2001; Magdol et al., 1997). These clinical samples consist of victim’s reports, law enforcement data, and crime surveys – all of these deal primarily with self-selected battered women. Therefore, they are not likely to be representative samples, as men are unlikely to report incidences of abuse to the police or others (Archer, 2000; McNeely &  Robinson-Simpson, 1987). The women in these samples also likely differ significantly from the larger population of battered women due to volunteer effects as well as possible differences in circumstances (Capaldi & Owen, 2001). George (1999) notes that reliance on criminal justice statistics is also likely to lead to problems due to a failure of victims of women to report, as well as because of policing and legal policies that do not      Partner Abuse     4 recognize men as victims or women as perpetrators of violent assaults. Archer suggests that studies based on such clinical samples characterize partner abuse as a female health issue, stemming from a patriarchal society that encourages the political and personal subjugation and control of women. In this conceptualization of partner abuse, the man who batters his wife is believed to be acting normally within a patriarchal framework; his violence is not seen as an abnormal act, but one encouraged as a tactic to subvert women (see Dobash & Dobash, 1979; Dutton, 1994b). These studies and theories have formed the basis of a feminist activist movement in the past couple of decades that has been aimed at altering social policy, laws, and education to reflect this side of the issue (McNeely & Robinson-Simpson, 1987). The existence of this movement has been crucial historically in demanding attention to the seriousness of domestic violence against women, but because feminist theory implicates patriarchal attitudes in the etiology of abuse, victims other than women assaulted by men have not been a focus (see Dutton, 1994b; George, 2003; Steinmetz, 1978).     Noting the politically-charged nature of many partner abuse studies, Straus (1979; 1999) developed an objective measure, the Conflict Tactics Scale (CTS), a standardized assessment tool that has been used extensively with national survey data to estimate levels of abusive behaviors in the population (see McNeely & Robinson-Simpson,  1987; Straus, 1999). The CTS has been administered to over 70,000 participants in some 400 published studies to measure partner abuse (Straus, Hamby, Boney-McCoy, & Sugarman, 1996). Using the CTS, Straus and his colleagues have found that men are as likely to be assaulted by their wives as women are by their husbands (Straus, 1999). Further, Straus, Gelles, and Steinmetz’s 1980 national survey data (as cited in O’Leary,      Partner Abuse     5 2000) indicated that 12% of both men and women admitted physically aggressing against their partner in the past year. Of the total sample collected, 3.8% of women indicated having been beaten by a partner, as did an even higher number (4.6%) of men. Straus and Gelles (1986, as cited in McNeely & Robinson-Simpson, 1987) also compared national survey data from 1975 with data from 1985 and found that while overall victimization had declined in frequency for women (from 12.1% to 11.3 %), it had slightly increased for men (from 11.6% to 12.1%). More recently, Milardo (1998) outlined research indicating that one in three college students (male and female) experienced physical violence, and that the frequency of violence in university students has increased from 1982 to 1992. Dutton, Kwong, and Bartholomew (1999) found equivalent one-year prevalence rates of spousal assault among both men and women in a large-sample study in Alberta. Researchers in Australia and Britain have found slightly higher female perpetration rates in large-scale studies. Magdol et al. (1997) conducted interviews as part of a large-scale longitudinal study in New Zealand, and found that women were more likely to report aggressing against their partners than men on all levels of abuse measured. Magdol et al.’s background research suggested these differences might be a result of men’s overreporting their victimization and underreporting their perpetration, and women the reverse. However, a concurrent study by several of the authors (Moffitt et al., 1997) revealed that both men and women tended to overreport their victimization and underreport their perpetration. Therefore, it may be reasonable to argue against  “gender-determined” tendencies towards overreporting or underreporting abuse as explanations for differences in reported violent experiences. Carrado, George, Loxam,      Partner Abuse     6 Jones, and Templar (1996)’s study of 1,978 heterosexual men and women in Britain also indicated slightly higher victimization rates for men than women using a 12-item (shortened) version of the CTS. Archer’s (2000) meta-analytic review of partner abuse research from 1976 to 1997 found that while women were actually more likely to physically aggress against their partners, and with greater frequency, men were more likely to cause serious injuries, particularly those requiring medical attention. However, although the proportions were lower, there were men in Archer’s study who sustained serious injuries and required medical attention.  The CTS studies have received widespread criticism from feminist theorists. Many of these theorists maintain – often backed by little or dated evidence – that the increased physical capacity of men to cause injury (as well as the “inherently violent” quality of the masculine gender role) makes male batterers a more relevant issue than female batterers (see Berliner, 1990; Bograd, 1999; Burke, Stets, & Pirog-Good, 1988; Hamberger & Potente, 1996; Steinmetz, 1980; Vivian & Langhinrichsen-Rohling, 1996). One large-scale (but somewhat dated) study found that women were more often victims than men, and more likely to sustain moderate and severe injuries than men (Makepeace, 1986). However, there is evidence to refute this position. George (1999) found when examining female-perpetrated assaults, male victims of women were slightly more likely to report sustaining injuries from their assaults than the women victims were. Capaldi and Owen (2001) also found that men were slightly more likely to have sustained infrequent injuries from a female partner, and equally likely to have sustained frequent injuries, with men were almost twice as likely to have experienced non-mutual aggressive victimization      Partner Abuse     7 as women. Although the three most severe cases of injury were sustained by women in their sample, men also reported injuries of considerable severity. McNeely and  Robinson-Simpson (1987) presented evidence suggesting women equalize any physical inequities by utilizing weaponry at much higher rates than men. In summary, although considerable variation exists in the data, there is currently a growing amount of literature demonstrating that men are being victimized by their spouses as well as women. In recent years, the absence of attention paid to male victims and female perpetrators of partner abuse in the literature has been pointed out by increasing numbers of both researchers and theorists (e.g., Carney and Buttell, 2004; George, 2003; Harris & Cook, 1994; McNeely and Robinson-Simpson, 1987; and Steinmetz, 1978). For example, Straus and Gelles (as cited in McNeely & Robinson-Simpson, 1987) noted that violence committed by women has not been given the public attention necessary to educate the population about this phenomenon. In an attempt to call attention to the presence of male victims of intimate partner violence, George (2003) and Steinmetz (1978) both presented evidence of abused men, including historical accounts of men being abused by their wives. George noted that such occurrences have been documented in Europe as early as the 1500’s. According to George and Steinmetz, there are also records of these victims being subjected to additional humiliation and abuse at the hands of their neighbors. For example, in France, beaten husbands were forced to ride through the town backward on a donkey, dressed in garish outfits. In Britain, they were tied to carts and driven around town so onlookers could mock them. These public procedures were traditional mechanisms used by the community to punish those who deviated from their established gender roles. Perhaps as a result of the historical shame attached to male victimization, it      Partner Abuse     8 is noted in a number of sources (DeMaris, Pugh, & Harman, 1992; George, 1994, 1999, 2003; Harris & Cook, 1994; McNeely & Robinson-Simpson, 1987; Steinmetz, 1978; Straus, 1999; Straus & Sweet, 1992) that men are less willing to divulge information about violent incidents, and are less likely to report violent acts against themselves than women. An unwillingness to report abuse to the appropriate authorities for fear of ridicule may partially explain the discrepancies between clinical samples and national survey findings mentioned earlier. McNeely and Robinson-Simpson (1987) and Letellier (1996b) argue that academic research and literature focused exclusively on female victimization has the consequence of affecting the understanding and views of policy makers, health care personnel, and the legal system, to the detriment of male victims who have nowhere to turn for help. Illustrating this bias in social policy further, Carney and Buttell (2004) considered the implications of U.S. legislation in the 1980’s mandating warantless arrests of batterers, which removed the reporting burden from the victims. An unexpected consequence of this policy was the appearance of substantial numbers of convicted female batterers being treated along with male batterers. Traditionally, batterer treatment programs have been designed according to feminist principles focused on ameliorating the patriarchal attitudes assumed to cause partner abuse (Carney & Buttell, 2004; McNeely & Robinson-Simpson, 1987). The effectiveness of this mode of treatment has been questionable at best for male batterers; its focus on breaking down male patriarchal attitudes makes it a ridiculous treatment to apply to female batterers. Carney and Buttell point out that despite the obvious presence of women who are violent in relationships, there are no social policies or treatments designed specifically to address them. Such a      Partner Abuse     9 social blind spot could be another major detriment to male victims (and female batterers) getting the help (and treatment) they require. In summary, there is a growing amount of literature establishing the presence of battered males, and all of them have emphasized the lack of attention given to these victims in terms of research and social policy. The paucity of data concerning heterosexual male victims of partner abuse is mirrored by the lack of attention given to partner abuse among same-sex couples, a phenomenon that appears to occur in similar proportions as opposite gender relationships (e.g., Coleman, 1994; Cruz & Firestone, 1998; Letellier, 1996b; Lie & Gentlewarrior, 1991; McClennan, Summers, & Vaughan, 2002; Renzetti, 1996; Wise and Bowman, 1997).  The phenomenon of lesbian batterers and gay male victims has been fundamentally difficult for feminist theorists to accept; their belief that patriarchy is the root cause of all partner violence is challenged by  same-sex abuse because it highlights the existence of victims and perpetrators not fitting the patriarchal mold (Coleman, 1994; Letellier, 1996a). As a result, there has been some reluctance in the academic field to accept the occurrence of same-sex partner violence,    possibly because feminist theory has historically contributed so much to our understanding of domestic violence. It is likely that theorists fear challenges to feminist theory will invalidate previous work and ultimately hurt their efforts to eliminate violence against women (e.g., Steinmetz, 1980; Straus, 1999). In addition to this ideological barrier, there has also been reluctance within the gay and lesbian communities to publicly address the issue of same-sex partner violence (Coleman, 1994; Merrill, 1998; Renzetti, 1996, etc.). A key reason for this reluctance stems from a wish of these already stigmatized groups to avoid further stigma; similarly, there are fears within the same-sex      Partner Abuse     10 communities that evidence of high rates of same-sex partner abuse could be used to suggest that homosexuality has an inherently pathological nature (Renzetti, 1996; Letellier, 1996a, 1996b).  Despite these attempts to avoid acknowledging the problem, research has demonstrated that same-sex partner abuse is not uncommon. Wise and Bowman (1997) presented estimates that rates of abuse among lesbian couples are likely comparable to rates of partner abuse in different gender couples. Lie and Gentlewarrior’s (1991) earlier study corroborates this assertion, finding in their large-sample study of lesbians at a women’s music festival that just over half of their lesbian participants reported having been victimized by a female intimate partner. Although they cautioned that this sample was not likely a fully representative sample of the battered lesbian population, this study does appear to have been the largest-scale study addressing abuse in same-sex relationships conducted to date. Similar to lesbian abuse rates, McClennan et al. (2002) reviewed studies demonstrating that abuse rates between gay male partners are comparable with those of different gender partners (see Gunther & Jennings, 1999; & Wallace, 1996, as cited in McClennan et al., 2002). Island and Letellier (1991, as cited in McClennan et al., 2002) estimated that 500,000 gay men are victimized by their partners each year in the United States. Cruz and Firestone (1998) and Letellier (1996b) pointed out that such figures may be underestimates, as gay male victims may hide their victimization due to fear of exacerbating the social stigma they already suffer due to homophobic attitudes. McClennan et al. (2002) reported that among gay men, partner abuse is now the third largest health issue faced, after AIDS and substance abuse.       Partner Abuse     11 Additional evidence has been presented to demonstrate that same-sex couples experience partner abuse in a similar manner to different-gender partners. For example, many of the abusive tactics and assaults are utilized, and the same “cycle of abuse” pattern identified in different-gender relationships is evident in abusive same-sex couples (Coleman, 1994; Merrill, 1998; Renzetti, 1998). Batterers of all genders and sexual orientations reportedly “tailor” abusive activities to the vulnerabilities of their victim (Letellier, 1996b). In addition to these similarities, there are some forms of abuse unique to same-sex relationships. Renzetti (1998) found in her research that “outing” was a commonly used abusive tactic among lesbians; that is, the abusive partner would threaten to expose the victim’s sexual identity to others as a means of coercion, control, or isolation. Letellier reports that HIV/AIDS is a common weapon used in gay male partner abuse. Not only may the perpetrator threaten to infect his partner as a coercive tactic, but he may also use his (or his victim’s) HIV/AIDS status to trap the victim in the relationship.  McClennen and her associates (2002) also found that gay male victims of partner abuse had a tendency not to report it, apparently to avoid additional stigmatization by mainstream society (as mentioned above). In many cases, such denial isolated the victims, increasing their vulnerability to their abusers and preventing them from seeking help. Gay men and lesbians have also expressed views that the traditional services sought by battered women (such as the police, social workers, counselors, and women’s shelters) were not very supportive or helpful for victims of same-sex partner abuse (Letellier, 1996b; Lie & Gentlewarrior, 1991; McClennen et al., 2002; Renzetti, 1992, 1996, 1998; Wise & Bowman, 1997). Letellier noted that gay men were often rejected by women’s      Partner Abuse     12 shelters and crisis lines and told instead to call batterer treatment programs, indicating a severe lack of acknowledgement for male victimization within the present assistance structure. Letellier also noted a tendency among police officers to label abuse incidents as “mutual combat” even when a victim and perpetrator were clearly distinguishable, due to an inability to believe that the gay victim did not act to defend himself. Despite the apparent lack of social attention to the victims of same-sex partner abuse, it is evident from the above studies that it is a phenomenon warranting the same serious consideration as partner abuse in different gender relationships.   This survey of the literature unfortunately suggests that there are prevailing attitudes in society that may be affecting social policy, legislation, reporting behaviors, and outreach services, barring male and same-sex oriented victims from receiving the assistance they require. In a cross-cultural study including only female participants,  Peek-Asa et al. (2002) compared the perceptions of Mexican and American women in rating the severity of several indicators of partner abuse, and found that the Mexican women in general regarded abusive situations as less severe. The authors point out that these lower perception ratings were well explained by the lack of sufficient education regarding partner abuse in Mexican society. This lack of education may mirror the state of perceptions in North America concerning gay, heterosexual male, and lesbian victimization, as the following studies illustrate: Archer (2000) analyzed a number of studies indicating that both men and women rated abusive situations more seriously and attributed more negative qualities to the perpetrator when a male abusing a woman was depicted. Bethke and DeJoy (1993) found that female perpetrators were rated as more acceptable overall, and male perpetration was      Partner Abuse     13 rated to be more physically and emotionally injurious, less acceptable, and more criminal than the identical case of female perpetration. Further, Harris and Cook (1994) discovered that vignettes depicting gay male partner abuse were perceived more seriously than were the vignettes depicting a woman abusing a man, but less seriously than the vignettes depicting a man abusing a woman. In Harris and Cook’s study, the male victim was taken more seriously than the female perpetrator, even in light of potentially homophobic reactions. Wise and Bowman’s (1997) results demonstrated that lesbian partner abuse was rated as less severe than male to female partner abuse. These ratings were particularly distressing in that they were assigned by graduate counseling students, many of whom did not even recommend the lesbian victim leave her abuser. These studies indicate that female perpetrators are not generally perceived as seriously as male perpetrators. Some of the studies mentioned above considered gender differences in perception (see Archer’s 2000 meta-analysis; Beyers, Leonard, Mays, & Rosen, 2000; Harris & Cook, 1994). Beyers et al.’s perception study hypothesized that women would react more strongly to depicted abusive scenarios than men would. Although they failed to support this hypothesis, this may have been due to methodological issues such as the decreased power of their between-subjects design. Harris and Cook (1994), on the other hand, did find significant gender differences on several of their perception measures. The women in their study rated all of the abuse scenarios more seriously than the men did. This difference was found for both female and male depicted victims. Therefore, the present study will also undertake to examine the potential effects of participant gender differences.         Partner Abuse     14 Personality-based theories relating to perpetration of abuse are often avoided in the literature due to the belief that a psychopathological etiology of abuse will lead to an excuse for violent behaviors and subsequent avoidance of the social inequities thought to underpin them (Coleman, 1994). Regardless of this avoidance, Dutton has undertaken research to examine the relationship between partner abusive behaviors and borderline personality organization (BPO) (1994a; 1998), noting that victim descriptions of abusive men are strikingly similar to BPO characteristics. His 1994(a) study found that BPO characteristics were positively corelated with perpetration of abuse in his two samples of male batterers, and perpetrators in the high-BPO group reported significantly more verbal and physical abuse than the low-BPO group. His analysis also established BPO as a central construct in partner abusive behaviors. Magdol et al.’s (1997) study discovered that several personality characteristics were associated with partner abuse perpetrations for men, but not for women. These characteristics included factors such as poor education, low socio-economic status, drug and alcohol abuse, mental health symptoms, and symptoms of antisocial personality disorder (APD). However, despite Magdol et al.’s (1997) findings that personality characteristics  were not associated with female perpetration of abuse, recent research has examined personality characteristics and trauma symptoms in female intimate partner assaulters and found significant relationships between them (Capaldi & Owen, 2001; Hernandez, Spidel, Nicholls, Dutton, & Kendrick, 2006; Spidel et al., 2006). Capaldi and Owen found that both men and women perpetrating frequent or bidirectional spousal assauls showed more antisocial personality characteristics than the men and women in their low abuse category. Further, Hernandez and her associates found that symptoms of trauma      Partner Abuse     15 were significantly correlated with the four subscales of the CTS, and trauma is often highly associated with development of BPO characteristics (see Dutton, 1994a; Coleman, 1994). Spidel and her associates found that those with Cluster “B,” or the “dramatic/erratic” personality disorders [as measured by the Structured Clinical Interview for DSM-IV Personality Disorders (SCID-II)] were more likely than  non-personality disordered participants to perpetrate minor and severe physical abuse as measured by the CTS.  Although no similar larger-sample research involving gay or lesbian participants was located, Coleman's (1994) article on lesbian battering reviews findings (largely case examples) that both male and lesbian batterers exhibit many characteristics of borderline, narcissistic, and/or antisocial personality disorder. It therefore appears that symptoms typical of a personality disorder may be a component of partner assaulters of any gender or sexual orientation.  BPO may relate to perceptions of abuse, as well as perpetration – the latter two phenomena are linked. Carney and Buttell (2004) note in their evaluation of a female batterer treatment program that the defense mechanisms of denial, minimization and blame are crucial issues to address in treatment because they allow the batterer to perceive abuse as less serious and continue their abusive behaviors – it follows from this that the perceptions of personality disordered individuals might differ from the rest of the population when judging abuse severity.  Based on the preceding research and theory, it appears that people are likely to perceive a situation in which a male is abused by a female, a male by a male, or a female by a female, differently (and less seriously) than a situation where a female is abused by a      Partner Abuse     16 male. These differences may also be mediated by the gender of the individual perceiving such a situation. Such differences could not only affect  perceptions of severity in a scenario depicting physical abuse, but could additionally tap into an educational and social blind spot that could be affecting reporting rates, help-seeking behaviors and social supports for many violence victims. We propose the following hypotheses:  1. Based on previous perception studies (Bethke & DeJoy, 1993; Harris & Cook, 1994; Wise & Bowman, 1997), we expect both male and female participants would order the severity (most to least) of the vignettes as follows: male-to-female, male-to-male, female-to-male, and female-to-female.  2. Further, consistent with Harris and Cook (1994), we anticipate that women will react more strongly to any battering incident, regardless of configuration, than will men.  This could reflect a higher level of education regarding abuse among women than among men, parallel to the cultural difference that Peek-Asa, Garcia, McArthur, & Castro (2002) found.  3. Based on previous research and theory described above (DeMaris, Pugh, & Harman, 1992; George, 2003; Harris & Cook, 1994; McNeely & Robinson-Simpson, 1987; Steinmetz, 1978; Straus, 1999; Straus & Sweet, 1992), we also expect male participants to show greater reluctance to report any of the hypothetical vignette situations to the police than women.   In addition, based on the work and theory of Coleman (1994), Dutton (1994a; 1994b; 1998), Magdol et al. (1997), and Spidel et al. (2006), we plan to examine potential relationships between perceptions of abuse severity, borderline personality organization      Partner Abuse     17 [as measured by the Borderline Personality Organization scale: BPO scale, (Oldham et al.,  as provided by Dutton, 1998)], and prior abuse perpetration or victimization [as measured by the Revised Conflict Tactics Scale: CTS2, (Straus, Hamby, Boney-McCoy, & Sugarman, 1996)]. The rationale for conducting the present study is similar to the logic presented by Beyers et al., (2000): If men and women view partner abuse in a significantly different manner, it may have an effect of the frequency of their reporting of abuse and their  help-seeking strategies. Although perception studies have been conducted contrasting male to female and female to male partner violence (Bethke & DeJoy, 1993; Beyers et al., 2000) as well as opposite gender to gay male (Harris & Cook, 1994) and lesbian (Wise & Bowman, 1997) partner violence, we believe this study will be the first to address all four potential configurations together. In doing so, we intend to determine whether female perpetrators (and/or male victims) are perceived less seriously than male perpetrators (and/or female victims). If there is a bias in social attitudes toward  non-heterosexual and non-female victims, such a bias could have devastating consequences for the health and well-being of these victims. A bias toward  non-traditional victims and perpetrators may have consequences for treatment and rehabilitation (e.g., Carney and Buttell, 2004). It is hoped that this study will call attention to such potential biases, and serve to further education concerning this topic. Method Participants  One-hundred and eight undergraduate students, 41 male and 67 female, were recruited for participation from the UBC-Okanagan Psychology and Science departments.      Partner Abuse     18 All participants were 19 years of age or older in accordance with the requirements of UBC-Okanagan’s research ethics board (M = 23.52, SD = 5.62, Mdn = 22.00).  Design  The present study used a mixed factorial design, with the gender of the participants as a quasi-independent, between-subjects variable, and the different configurations of the vignettes comprising four levels of the manipulated independent variable. The initial idea to use this design was partially derived from Brown’s thesis proposal (n.d.). Participants were instructed to read four vignettes (which are described in more detail below), completed Likert-Type rating questions for each, and then ranked them in order of perceived severity. Presentation order of the vignettes was completely counterbalanced. The design also includes a correlational element; the relationships between responses to the vignettes, the personality measure (see below), and the measure of personal abuse experience (see below) will be examined using Pearson  product-moment correlations.  Materials  Participants, either in a group setting or individually, were provided with the survey package composed of an instruction letter, Harris & Cook’s (1994) vignettes and Likert-Type questions (as well as the final ranking question devised for this study), Oldham et al.’s [1985; as cited in Dutton (1998)] Borderline Personality Organization (BPO) Scale, the CTS2, and a demographic information sheet, including two open-ended questions allowing the participants to provide feedback to the researchers.  Cover Letter. The first page of the survey package was a cover/instruction letter (refer to Appendix A). This letter outlined the study, explaining its voluntary nature, the      Partner Abuse     19 procedures to ensure informed consent, and that responses were confidential and anonymous.  Vignettes and Likert-Type Questions. The primary materials used in this study were the vignettes and ratings questions presented by Harris and Cook (1994), with only the genders and names of the depicted individuals changed as appropriate, as the authors instructed. These measures effectively captured perception differences based on victim and perpetrator gender in Harris and Cook’s own study, and the vignettes were successfully adapted by Wise and Bowman (1997) for a similar study. For the current study, two alterations were made to the original text of Harris and Cook’s materials. The first was to change the term “batterer” to “perpetrator;” this was due to our concerns that a loaded term such as “batterer” could distort scores on the dependent measure. Secondly, instead of using Kansas State Police officers in the scenario, the Kelowna R.C.M.P were used. This was because we believed participants may have had a different opinion of abusive situations occurring in the United States as opposed to in their own city. The participants were instructed to complete the Likert-Type questions provided by Harris and Cook for each vignette configuration. There were 11 ratings questions per vignette, adapted to a 9-point scale for the present study to facilitate parametric analyses. For an example of the vignettes and Likert-Type questions, please refer to Appendix B. After reading all of the vignettes and completing the ratings questions, participants were instructed to rank order the vignettes from most severe to least severe. Combining this ranking item with Harris & Cook’s (1994) items rating severity of abuse was intended to yield a more sensitive measure of perceptions.       Partner Abuse     20  Borderline Personality Organization Scale. Due to Dutton’s (1994a; 1998) findings that borderline personality organization (BPO) is highly associated with abusiveness toward partners, Oldham et al.’s Borderline Personality Organization Scale (as provided by Dutton, 1998) will be administered to determine whether BPO scores are associated with perceptions of severity (refer to Appendix C). The BPO scale is a 30-item self-report measure of borderline personality characteristics, and it contains "3 subscales of identity diffusion, primitive defenses, and reality testing" (Dutton, 1994a, pp.268). Scores on these items indicate levels of different facets of borderline personality traits. Dutton (1994a) provided evidence of the scale's statistically tested reliability and validity drawn from Oldham et al.'s (1985, as cited in Dutton, 1994a) study, and noted that his own study provides concurrent evidence of reliability and validity for Oldham et al.'s scale. Importantly, Dutton was able to demonstrate that the BPO scale and the Millon Clinical Multiaxial Inventory (version 2) (MCMI-II) were significantly correlated for his sample, thereby increasing the validity of the BPO scale by demonstrating it measures the same construct as the MCMI-II, a measure validated by large normative samples.  The Revised Conflict Tactics Scale. The CTS2 (Straus et al., 1996) was included in the survey package to determine whether previous victimization or perpetration of partner abuse affected perceptions of severity (see Appendix D). The original CTS has been used extensively in the field of partner abuse (see Arias & Beach, 1986; Arias, Samois, & O’Leary, 1987; Bethke & DeJoy, 1993; Dutton, 1994; Foo & Margolin, 1995; Magdol et al., 1997; Straus, 1990; White & Humphrey, 1994; etc). The CTS2 was a revision created to further improve the psychometric integrity of the scale, and to enhance its clarity and sensitivity as a measure (Straus et al., 1996).       Partner Abuse     21 Additional Materials. A final sheet of items was included (see Appendix E), requesting that participants provide their gender, age, marital status, and sexual orientation for analytical purposes. Studies of abuse perceptions have previously included such demographic information in their survey packages (eg: Bethke & DeJoy, 1993) and we believed that an item regarding participants’ sexual orientation was directly relevant to our study, as it examined same-sex partner abuse as well as opposite-gender abuse. Two open-ended questions were included in order to allow the participant to reflect upon the study. One requested that participants add any additional comments or opinions regarding the vignettes and ratings questions, and the other examined what they thought of the overall study. These questions provided the researchers with additional insights into the phenomenon in question as well as feedback concerning the research design.  Procedure All participants received survey packages sealed in individual envelopes. Each survey package included the instruction/cover letter, the vignettes and ratings questions, the BPO scale, the CTS2, and demographics sheet. Along with these, we included an  up-to-date list of counseling contacts at Campus Health to ensure that any participants upset by the material were able to seek the appropriate professionals for assistance.  Participants were largely recruited through solicitation of volunteers from upper level classes at the beginning of scheduled lecture times, with advance permission of the professor in question. These participants were given two options. They could include (on a provided index card) their first name, email address and/or telephone number and their preferred date to participate in a group survey completion session. Alternatively, they could obtain an envelope containing a survey package from the researcher, to complete      Partner Abuse     22 and return to the primary supervisor’s office on their own time. The majority of participants opted for the second method. A small group of participants were additionally recruited through Experimetrix, a web-based research volunteer program allowing Introduction to Psychology students to sign up for and participate in research for academic credit. These participants obtained survey packages from the researcher, completed them at their convenience, and received credit for participation once the surveys were returned.  The group collection procedures were conducted as follows. Classrooms at  UBC-Okanagan were reserved in advance, and participants wishing to attend were provided with a schedule. Contact info provided on a voluntary basis was used to remind participants of their indicated sessions. At the assigned times, participants completed the questionnaires, and the researcher remained to answer any questions. None of the participants withdrew from the study during a group session. Following completion of their surveys, the participants were thanked for volunteering.    Following survey completion, all packages were taken to the primary supervisor’s office and locked in a secure file drawer until all participants had handed in their data. Once all data were collected, compilation for analysis was undertaken using SPSS version 11.5. At no point were any participant names connected in any way with the survey packages. The survey packages containing the raw data will be kept for five years in accordance with UBC-O policy, and then destroyed by the primary supervisor. Results Hypothesis 1  All analyses were conducted using SPSS version 11.5. In our data base, the       Partner Abuse     23 male-to-female vignette was labeled “Scenario A,” the male-to-male “Scenario B,” the female-to-male “Scenario C,” and the female-to-female “Scenario D.”    Descriptives. The marital status of the sample was varied, with the modal status being “dating one person” (See Figure 1). The sexual orientation of participants, on the other hand, was extremely homogenous, with 91.7% of respondents reporting a heterosexual orientation.  Refer to Table 1 to examine men and women’s means and standard deviations for the total for the Likert-Type questions for each different scenario. Participant scores on the BPO scale ranged from 32 to 115, M = 60.35, Mdn = 57. There were no significant differences in BPO scores between men (M = 61.70) and women (M = 59.53), F(1, 106) = .49, p = .487. For a frequency histogram of the entire sample, refer to Figure 2. For a comparison of the means and standard deviations for male and female participants on the CTS2 subscales, please refer to Table 2. It is important to note that a MANOVA comparing these means found no significant main effect of gender for any CTS2 subscale item.    ANOVAs. In order to examine the order of severity hypothesis, the 11 Likert-Type questions for each scenario were summed to create a total. We then examined the effects of the totals across the four possible gender combinations (A-B-C-D), with gender, age, and marital status as between-subjects factors. Sexual orientation was not included because the majority of the sample was heterosexual. The analysis found a significant main effect of scenario, Greenhouse-Geisser F(2.6, 205.3) = 11.96, p <.001, partial  η2 = .13 (see Figure 3).       Partner Abuse     24 There were no other significant main effects or interactions with any factor, although the expected gender differences approached significance, F(1, 79) = 3.44,  p = .062, partial η2 = .04.  To better understand the within-subjects differences between the scenarios, and to see if gender differences reached significance on any particular Likert-Type item, we conducted a number of follow-up mixed-factorial ANOVAs. Due to the increased chance of Type I error produced by multiple comparisons, we adjusted the alpha using a Holm’s sequential Bonferroni procedure. Using this adjustment, we found a number of significant within-subjects main effects on particular survey items.  As crimes go, how violent was the incident? There was a significant  within-subjects main effect of scenario, Greenhouse-Geisser F(2.7, 285.8) = 12.57,  p <.001, partial η2 = .11. Post hoc analysis shows that Scenario A was rated as significantly more violent than Scenarios B, C, and D (see Table 3). If you had witnessed this incident from the window next door, how likely would it have been that you would have called the police? There was a significant within-subjects main effect of scenario, Greenhouse-Geisser F(2.5, 270.2) = 33.17, p <.001, η2 = .24. Post hoc comparisons found that participants reported being significantly more likely to call the police upon witnessing Scenario A, than Scenarios B, C, and D (see Table 4). In this case should the perpetrator be convicted of assault? There was a significant within-subjects main effect of scenario, Greenhouse-Geisser F(2.7, 285.1) = 13.87, p <.001, η2 = .12. Post hoc tests demonstrate that participants were significantly more likely to believe the perpetrator in Scenario A should be convicted of assault than the perpetrator depicted in B, C, and D (see Table 5).      Partner Abuse     25 Should the victim leave the perpetrator for good? There was a significant  within-subjects main effect here, too, Greenhouse-Geisser F(2.8, 294.6) = 9.03, p <.001,  η2 = .08. Interestingly, the post hoc comparisons here indicated that participants were significantly more likely to recommend that the victim leave the perpetrator in Scenario A than in Scenarios C and D, but also more likely to recommend that the victim in Scenario B leave the perpetrator than in Scenario C (see Table 6). Do you think the perpetrator has probably acted this way in the past? A within subjects main effect was found for this item, Greenhouse-Geisser F(2.6, 282.7) = 6.15, p = .001, η2 = .05. Post hoc comparisons show that participants were more likely to consider the perpetrator a repeat offender for Scenario A than for Scenarios C and D, but NOT for Scenario B. This may demonstrate that female perpetrators were seen as acting in more of a situation-specific manner than the male perpetrators (see Table 7). How much do you like the perpetrator? A within subjects main effect was found for this item as well, Greenhouse-Geisser F(2.7, 288.6) = 10.63, p <.001, η2 = .09. Post hoc comparisons indicated that participants disliked the perpetrator in Scenario A significantly more than in Scenario B, C, or D (see Table 8).  All remaining items were not significant. There were no interactions with gender, and only one between-subjects main effect of gender approaching significance, F(1, 106) = 7.86, p = .006, η2 = .07. In this instance, involving the item measuring perceptions of perpetrator responsibility, women were slightly more likely to ascribe responsibility to the perpetrator for the assault than were men. However, the significance value exceeded the alpha cutoff dictated by our Holm’s sequential Bonferroni (α = .005).      Partner Abuse     26 Friedman/Wilcoxin analyses. To examine the results for the ranking item in further support of our first hypothesis, we conducted Friedman and Wilcoxin tests to determine differences in median rankings between the four scenarios. As 35 participants failed to respond as instructed to the forced ranking question (but completed all other sections), they were excluded from the Friedman and Wilcoxin analyses. The possible ranks for each scenario (A, B, C, and D) ranged from 1 to 4, with 1 indicating most harmful and 4 indicating least harmful. There were significant median differences among severity ratings for Scenario A (Mdn = 1), B (Mdn = 2), C (Mdn = 4), and D (Mdn = 3), χ2(3, N = 73) = 120.96, p <.001, and Kendall’s W coefficient of .552 indicates that differences among the four categories are strong. To follow this up, we conducted six pairwise comparisons utilizing Wilcoxin signed ranks tests, adjusting the alpha again with a Holm’s sequential Bonferroni to guard against the occurrence of Type I errors. These analyses indicated that Scenario A was ranked as significantly more severe than B (z = -6.74, p <.001), C (z = -7.01, p <.001), and D (z = -7.38, p <.001). B was ranked as significantly more severe than C (z = -3.89, p <.001) and D (z = -.2.37, p = .018) and that D was ranked as significantly more severe than C (z = -2.30, p = .022). For visual representations of the ranking breakdown of each scenario, please refer to Figures 3-6. Hypothesis 2 The ANOVAs conducted for hypothesis 1 failed to find any significant main effects of gender, although gender differences approached significance for each scenario’s total. To further examine any gender differences, we generated global sums for each participant across all four scenarios, and conducted a one-way ANOVA to examine possible gender differences in overall perceptions of severity. The results of this analysis      Partner Abuse     27 indicate that women (M = 350.14, SD = 39.88) perceived the scenarios as being significantly more severe overall than did men (M = 332.84, SD = 33.95). F(1, 106)  = 5.78, p = .018, partial η2 = .052 (see Figure 7). Hypothesis 3 The ANOVAs used to explore hypothesis 1 also failed to find significant main effects of gender for the “police call” Likert-Type questionnaire item. To further explore hypothesis 3, we created a sum for each participant of all “police call” items across the four scenarios. We then conducted a one-way ANOVA to determine whether women  (M = 30.31, SD = 8.78) were more likely to report all of the incidents to the police than men (M = 28.14, SD = 9.69). Although the means suggest a slight difference, we failed to find a significant effect of gender, F(1, 106) = 1.43, p = .234.  Exploratory Analyses  Intercorrelations between the BPO, the CTS2 subscales, and the totals for the four abuse scenarios were explored using Pearson product-moment correlations. Due to the exploratory nature of this analysis, we did not use a Holm’s sequential Bonferroni adjustment to guard against Type I errors; consequently, the following results are most appropriate for future hypothesis generation, not for drawing definitive conclusions about the interrelationships between the three phenomena.  Examination of the relationships between the BPO scale and the CTS2 subscales produced results similar to those expected from Dutton’s (1994; 1998) research. Scores on the BPO were positively correlated with participants’ own CTS2 sexual coercion perpetration scores, r(97) =.220, p = .028, as well as with their partners’ sexual coercion      Partner Abuse     28 perpetration scores, r(97) = .225, p = .025. Additionally, BPO scores were positively correlated with participants’ own CTS2 psychological abuse perpetration scores,  r(97) = .363, p < .001, as well as with their partners’ psychological abuse perpetration scores, r(97) = .265, p = .008. Analysis of relationships between the CTS2 subscales and the Likert-Type questions used to rate vignette severity produced a few interesting results. Participants’ partners’ CTS2 negotiation subscale scores were positively correlated with severity ratings for Scenarios B, r(97) = .238, p = .018, C, r(97) = .272, p = .006, and D,  r(97) = .235, p = .015, as well as with the total severity score across all four scenarios, r(97) = .236, p = .019. Alternatively, participants’ partners’ CTS2 psychological abuse perpetration scores were negatively correlated with severity ratings for Scenario A,  r(97) = -.199, p = .048 and D, r(97) = -.216, p = .032.  Finally, examination of the interrelationships between the BPO scale and the vignette severity ratings also produced some interesting results. The BPO was negatively correlated with severity perceptions for Scenarios B, r(106) = -.209, p = .030), C,  r(106) = -.281, p = .003, and D, r(106) = -.268, p = .005, as well as with overall severity perceptions across all 4 scenarios, r(106) = -.250, p = .009.  Discussion Findings Hypothesis 1. Our study revealed that the male-to-female scenario was consistently perceived as more severe than the other three scenarios; the totals for the male-to-female scenario were significantly higher. This supported the Harris and Cook (1994) and Wise and Bowman (1997) studies where participants rated male-to-female      Partner Abuse     29 violence as significantly more severe than other gender combinations of victim and perpetrator. However, this finding only informed us on the overall differences. We were also able to discern where these differences tended to come from by studying each of the 11 survey items separately. Participants found the male-to-female scenario more violent, were more likely to report it to the police, were more likely to believe the perpetrator should be convicted of assault, and disliked the perpetrator more than in any of the other scenarios.  Particularly startling was the difference found for the police reporting item. This question produced the largest effect size, and more than any other, this question inquired about how participants would act if observing such a situation, rather than simply gauging perceptions. Such gender-mediated differences in willingness to act could strongly influence participants’ reporting tendencies in a real-life situation. Participants’ minimizing of the non-traditional perpetrators on the other three items could also reflect social stereotypes affecting how perpetrators are dealt with by the legal system, and how their victims are received and supported by others. The results for two of the questions did not match our other results. When deciding whether the victim should leave the relationship for each scenario, participants most strongly believed the heterosexual female victim should leave. However, they were also more likely to recommend that the gay male victim leave than the heterosexual male victim. This was a very interesting result. The most parsimonious explanation is that the gay male victim was thought to be at considerable physical risk, similar to the risk faced by the female victim of the male, but not as severe, perhaps due to a stereotypical belief      Partner Abuse     30 that the male victim is “tougher” than the female victim. Alternately, the male victim of the female may not have been considered at risk of comparable future attacks. Both possibilities suggest that female violence perpetration is considered more incidental or infrequent.       This leads us to the second unexpected result. When participants examined whether the perpetrator had likely been abusive before, the heterosexual male perpetrator was seen as more likely to be a repeat offender than both female assaulters, but not more likely than the gay male. This suggested female perpetrated assaults were considered more situation-specific than male perpetrated assaults. Perhaps female violence was viewed as more incidental, and male violence more systematic (or “expressive” as opposed to “instrumental” – see Magdol et al., 1997).  In addition to these results for our Likert-type questions, our examinination of the forced-choice ranking item indicated that the participants were significantly more likely to rank the male victims and female perpetrators below the traditional male-to-female scenario. Participants overall tended to rank the scenarios in the following order:  Male-to-female, male-to-male, female-to-female, and female-to-male. Although this was not the order we had expected to find, it still demonstrated the bias we had predicted against male victims and female perpetrators. This corroborated Harris and Cook’s (1994) finding that male-to-female violence was considered more serious than  male-to-male, and female-to-male was considered least serious. This finding also supported Wise and Bowman’s (1997) study which demonstrated that female-to-female partner assault was taken less seriously than male-to-female. In addition, our examination      Partner Abuse     31 of all four gender configurations added a new finding that the weight given to victims and perpetrators may inhabit a hierarchy of priority according to their gender.     Hypothesis 2. The perceptual differences we had expected between men and women did not emerge as clearly as anticipated.  Harris and Cook (1994) found differences on several individual Likert-type questions, but we failed to find similar gender differences. We did find a significant gender difference when adding the scenarios all together. In general, women tended to consider the scenarios more severe than men. Nonetheless, although significant, this difference did not necessarily indicate that men minimized the violent scenarios in general. In fact, the means for both genders were very high, suggesting considerable sensitivity to partner abuse regardless of gender (see Table 1 and Figure 7).  Our failure to find prevalent gender differences may reflect greater education about partner abuse and human rights among Canadian university students. Students in our country might be less likely to differ according to gender than the general population in Canada or students in the United States (i.e., those in Harris and Cook’s study). Also, the study by Harris and Cook was conducted 12 years prior to this study; and it is possible that gender differences, which have arguably become increasingly less salient in modern times, are continuing to diminish.  Nonetheless, it is possible that men’s viewing abuse less seriously overall might affect both their tendencies to perpetrate abuse AND their willingness to report abuse against themselves. This would further contribute to the difficulties in identifying, preventing, and treating male perpetration and victimization.  Hypothesis 3. Our third hypothesis was not met; men and women both reported being highly likely to call the police if witnessing the depicted scenarios. This is in      Partner Abuse     32 contradiction to previous findings by Harris and Cook (1994) that men were less likely to report scenarios to the police than women. Although it is tempting to hope that our discovered lack of difference reflects a real willingness to report abuse to the authorities for both men and women, we recognize that failure to find a significant difference does not prove there is no difference. It is possible that our smaller sample size merely did not provide sufficient power to detect the between-subjects differences in our design as easily as the within-subjects differences.    Also, although heartening, the willingness to report demonstrated by our participants may not generalize to their own relationships. It is possible that both sexes would readily call the police when others are in danger, but fail to report abuse against themselves. In the future, it may be relevant to include a question examining whether the participants would call the police if they were the victim.   Exploratory Data Analyses. Our examination of differences between men and women on the subscales of the CTS2 indicated no significant differences according to gender in either victimization or perpetration of the various (psychological, physical, sexual, etc.) subcategories. This matches many of the large-scale national surveys indicating that men and women perpetrate and suffer abuse in roughly equal proportions (e.g., Straus, 1999); including Canadian data (see Dutton et al., 1999).  Our exploratory data analyses offered some support for Dutton (1994a; 1998) and Spidel et al’s (2006) research, as well as identifying new relationships worthy of exploration in future research. It is important that these findings be understood as exploratory, as we did not adjust the alpha level as with our other analyses. We were more interested in discovering relationships for future studies with larger samples.        Partner Abuse     33 Our analysis of the BPO scale and CTS2 subscales found a positive relationship between BPO and both CTS2 sexual coercion subscales. Specifically, the higher participants’ BPO scores, the higher sexual coercion perpetration scores were for themselves and their partners. This potentially demonstrates a link between borderline personality organization and membership in a sexually coercive relationship. Higher BPO scores were also positively related to psychological abuse perpetration for both participants and their partners, possibly indicating that borderline personality organization and membership in a psychologically abusive relationship are linked phenomena. Such relationships between BPO and partner assaultive behaviors may have implications for treatment. For example, it is possible that partner assaulters with borderline characteristics may require additional interventions to ameliorate their personality disordered symptoms. These correlations offered support for Dutton’s (1994a; 1998) research. We also discovered potential new relationships between BPO and sexual coercion by using the CTS2 (rather than the original CTS).  Our analysis of the CTS2 and severity ratings on the Likert-Type questions revealed interesting relationships. Participants with partners who often used negotiation tactics were more likely to rate the non-traditional (i.e., not male-to-female) scenarios more seriously, as well as all four scenarios overall. If partner negotiation is related to relationship security and satisfaction, it is possible that those in more supportive and secure relationships are more likely to view all physical violence as unacceptable. Perhaps negotiation as a problem-solving skill is a protective factor against violence in relationships. This could offer important empirical support for clinical practice. Carney      Partner Abuse     34 and Buttell (2004) indicated that a major focus in batterer treatment programs is enhancing social skills such as negotiation; if our findings are replicated, it may indicate that present initiatives are correct in emphasizing problem solving and social skills.  In contrast, participants with psychologically abusive partners rated scenarios with female victims less seriously. It would have been interesting to see if this tendency was shared by both genders. However, given our modest sample size, separate analyses by gender were untenable. If the relationship is examined by larger-scale research, it may be that women experiencing greater amounts of psychological abuse in their relationships minimize the abuse of victims of their own gender. Alternatively, men who are psychologically abused by female partners may exhibit less sympathy for females than other men. This is a potentially important relationship to address clinically, in that lowered sympathy for victims associated with psychological abuse could affect propensity for perpetrating assaults, as well as reduce the likelihood of help-seeking behaviors. Finally, we discovered that the higher a participant's score on the BPO scale, the lower their ratings of severity were for the non-traditional partner abuse scenarios. This may suggest that borderline personality characteristics in general are related to minimization of perceived abuse severity in intimate relationships. The failure to find a negative relationship for the male-to-female scenario may be due to education and social norms leading the majority of people to believe, or at least superficially endorse, that men hitting women is wrong (see Peek-Asa et al., 2002). Therefore, social desirability may be obscuring a relationship between BPO and male-to-female abuse severity. On the other hand, minimization of the non-traditional scenarios may reflect an inability of high BPO      Partner Abuse     35 scorers to empathize with depicted victims or perpetrators unlike themselves. For example, if our high-BPO participants indeed have higher rates of abuse victimization as suggested above, the majority of our participants being heterosexual females, it could follow that these participants identified most strongly with the heterosexual female victim.  These relationships between BPO and severity perceptions offer some new connections between personality and partner abuse, further supporting the findings of researchers such as Dutton (1994a; 1998) and Spidel and colleagues (2006). Links between BPO and perception could shed light on cognitive features characteristic of batterers; perhaps the lowered level of empathy suggested by these results could increase the likelihood of involvement in partner assaultive relationships, making this a potentially key factor to address in treatment and intervention attempts.  Additional Interpretations as well as Implications Homophobia. A main interpretation offered by Harris and Cook (1994) for a number of their study’s findings was that homophobic attitudes could be leading participants to discriminate against the depicted gay male victim. However, only the response to one of our questions raised some suspicion of homophobia – our item examining whether the victim should leave participants indicated that both the female and male victims of the male perpetrators should leave more urgently than the victims of the females. In contrast to our belief that this response reflected participants’ acknowledgement of the greater danger faced by the victims of the males, it is possible that participants were expressing homophobic opinions. For instance, they might be      Partner Abuse     36 likely to believe that any man should leave a relationship with another man. It also may be that participants have suggested the gay male victim leave due to a belief that  male-male relationships are less permanent than other relationships, as gay men often have many partners over a lifetime (see Rathus, Nevid, Fichner-Rathus, & Herold, 2003). However, these explanations seem less tenable because the analyses for the other questions did not indicate homophobic attitudes; indeed, the male-to-male scenario was taken relatively seriously.    In a related vein, we had anticipated that the female-to-female scenario would be ranked as least severe due to an expectancy that homophobia would influence participants’ perceptions of severity. However, the majority of our sample rated the female-to-female scenario above the female-to-male scenario, a finding that does not support the notion of homophobia playing a significant role. The lack of an apparent effect of homophobia is at odds with Harris and Cook’s (1994) study, where homophobia seemed to bias many of the participants’ responses. It is possible that their study, conducted several years ago in the Midwestern United States, involved a more homophobic sample than our Canadian undergraduates. It could be argued that the Canadian government’s legal sanction of same-sex unions reflects a more liberal social attitude towards same-sex partnerships among Canadians (and university students particularly) than seen in the U.S. However, it would be useful to include a measure of homophobia in future studies to evaluate whether it is correlated with perceptions of severity. It is likely, as demonstrated by Harris and Cook (1994), that homophobic attitudes affect perceptions for at least a subset of the population. If so, this certainly has implications for intervention and assistance programs, whose staff are presumably not      Partner Abuse     37 invulnerable to homophobia and may require sensitivity training to ensure that same-sex victims are afforded the same effort as heterosexual women (as suggested by Lie & Gentlewarrior, 1991).   With homophobia not appearing to play a key role in explaining perceptual differences in the current study, we propose alternate explanations for these differences, and explore their implications.: Minimizing Female Violence.  It can be theorized that our participants viewed female perpetrators as less capable of assaultive acts because they are stereotypically assumed to be physically weaker than men (e.g., Steinmetz, 1980). Indeed, violence by women is often downplayed or even considered humorous (Dutton et al., 1999; George, 2001; Steinmetz, 1978). The “men strong, women weak” stereotype is a common argument against the existence of female partner assaulters in the literature (Dutton, 1994b; George, 1994; McNeely & Mann, 1990). However, although studies may support mean differences in size and strength between men and women (Straus, 1999), other factors can equalize such differences. Weaponry, intoxicants, surprise tactics, physical combat training (such as martial arts), speed, co-ordination, and endurance all provide combat advantages – and none of these are inaccessible to women (see McNeely & Mann, 1990). Additionally, the type of assault can compensate for women’s physical inequities. George (1999) found a preferred female tactic was “kicking.” Kicking is combined with “punching” on the CTS, but was separated out by George for his study. He theorized that while men’s punches may indeed be more powerful due to greater upper body strength, women’s kicks are potentially powerful and injury-causing. He also notes (1994) that      Partner Abuse     38 head injuries do not require excessive force or impact to result in unconsciousness or brain trauma, allowing women to disable men quickly using blows to the head. The idea that men (or the larger individual) are the automatic victors in physical combat is an overly simplistic defense of gender-driven theory (see Dutton, 1994b. For reviews of female equalization tactics, refer to George, 1994; and McNeely & Mann, 1990).    The participants in this study may have also believed that women do not assault partners because they possess fundamentally non-violent natures, a fallacious assumption sometimes present in the literature as well (e.g., Hamberger & Potente, 1996; O’Leary, 2000). Despite theoretical arguments (see George, 1994; Hamberger & Potente, 1996; McNeely & Mann, 1990; Vivian & Langhinrichsen-Rohling, 1996) that abuse by women is exclusively self-defensive, a study by Carrado and his associates (1996) found that women cited self-defensive reasons for abusing male partners only 20% of the time. The remaining reasons for the women’s aggression were instrumental and control-oriented, similar to the motives behind male aggression. George (1994) also presented reports indicating that women use violence for instrumental and controlling purposes. Foo and Margolin (1995) found that humiliation by a dating partner was a better predictor of partner assaultive behavior than was self-defense for both men and women. Fiebert and Gonzales (1997) found that getting the attention of an emotionally distant partner was the primary motivation for women’s physical attacks. Renzetti (1992; 1996; 1998) has found that lesbian partner abuse typically involves one partner utilizing violent tactics to exert power over the other, not mutually-incited, infrequent “scraps.” Not only have women overwhelmingly reported non-defensive motivations for partner abuse, they have also been implicated in generic criminal assaults on both men      Partner Abuse     39 and women according to victims’ self-reports (George, 1999). Additionally, the very existence of lesbian partner abuse challenges the notion that women are inherently  non-violent victims (see Renzetti, 1992; 1996; 1998). In fact, a study by Lie, Schilit, Bush, Montagne, and Reyes (1991) found that lesbians with previous relationships with both men and women reported higher rates of victimization for their lesbian partners than their male partners, a finding that Dutton (1994b) characterized as particularly notable because each woman acted as her own control condition. Women are also widely documented perpetrators of assaults on children (Coney & Mackey, 1999; McNeely & Mann, 1990; Renzetti, 1992; Steinmetz, 1980; Straus & Mouradian, 1998). Minimization of female violence in these situations is evident in that women murdering their children are often charged with negligence, while fathers in similar situations are charged with homicide.  In addition to minimizing the physiological damage inflicted by female perpetrators, some theorists have argued (eg Hamberger & Potente, 1996; Jacobson et al., 1994, as cited in Capaldi & Owen, 2001) that male to female violence is more salient because the male batterer, but not the female batterer, can produce fear and psychological trauma in his victim. This assertion was tested by Capaldi and Owen, who instead failed to find any differences in fear levels experienced by male and female victims. George (1994) additionally outlined studies indicating that male victims suffer significant psychological trauma due to assaults by female partners, and argued that theories claiming the reverse are based on stereotypical conceptualizations of masculinity. There is therefore some evidence that the assumption that only men possess the power to incite fear and psychological trauma in a victim is false.          Partner Abuse     40 Our finding that female perpetrators were taken less seriously may have some implications in terms of attempts at intervention and treatment. Milardo’s (1998) study indicated that men were more (and highly) likely to anticipate or expect violence from women than vice versa in a number of conflict situations, demonstrating that violence by women is viewed as normative in many dating conflicts. George (1999) and Milardo (1998) presented findings that women consider abusing their partners to be more acceptable and less risky than abusing others outside the relationships, and use various rationalizations for their violence. Fiebert and Gonzales (1997) found that abusive women did not believe their victims were seriously hurt, or fear retaliation. This evidence of rationalization and acceptance of female violence may help to explain our college sample’s minimization of female perpetrated assaults. Such cognitive rationalizations are common occurrences among male batterers as well - a phenomenon that interferes greatly with treatment efforts (Carney & Buttell, 2004; Dutton, 1994b; Merrill, 1998).  Alarmingly, despite these cognitive similarities between male and female batterers, there are theoretical arguments suggesting differential treatment for females. A fundamental aspect of male batterer treatment is to break down cognitive rationalizations and focus on guiding the abuser to accept responsibility for his violent actions. The man’s prior victimization history is not addressed until close to the end of the program, as it is necessary to prevent the batterer from using his past to justify his actions (see Carney & Buttell, 2004). However, at least one proposed treatment for assaultive females suggested a primary focus on the woman’s past victimization history (Hamberger & Potente, 1996), assuming that her violent behaviors must be self-defensive, and expressive of past abuse – or of internalized misogyny (e.g., Dutton, 1994b; Hamberger & Potente, 1996).       Partner Abuse     41 Focusing on victimization is known to feed a batterer’s rationalizations by giving excuses for abusive behaviors. There is no reason to believe that this is not the case for women. Carney and Buttell noted that the majority of empirical research on female batterers has focused on abuse context, offering justifications for their actions rather than solid evidence concerning how to intervene and address them. Fortunately, Carney and Buttell’s evaluation of a women’s treatment program uncovered common treatment elements useful in treating men and women perpetrators, such as addressing cognitive distortions and erroneous beliefs, and enhancing social and problem-solving skills. Developing useful therapies for female partner assaulters is an important step in acknowledging the seriousness of female perpetrated violence. Minimizing Male & Lesbian Victims. We have seen how perceptual differences of perpetrators according to gender can effectively obscure perpetrators who are not heterosexual males. These gender differences not only serve to keep female and gay male perpetrators hidden, they also produce disastrous consequences for non-traditional victims, seriously impairing their ability to seek help and escape their violent relationships. Bethke and DeJoy (1993) found, similar to the present study, that male victimization was considered less serious than that of women. It is possible that part of the reason for this perception involves a stereotypical belief that women cannot hurt men due to their greater size and strength.  However, it is unfair to assume all men are natural physical combatants. It is likely that many are not, perhaps having little or no combat experience to derive fighting skills or pain endurance from. It is also likely that [similar to Merrill’s (1998) point] many men are unwilling to physically fight back when attacked by a woman, as they are      Partner Abuse     42 taught from an early age that it is extremely wrong to do so (George, 1999; Straus, 1999). Magdol et al. (1997) theorized that men may restrain themselves from taking defensive actions against their partners because the costs of physical combat are higher for them (i.e., they may fear causing injury or prosecution) than for abusive women - these theoretical constraints mirror the ones described by Letellier (1996b) involving same-sex victims. Conditions such as these could easily allow a woman to physically injure a man, despite her possessing lower physical strength or smaller stature.  The assumption that the larger, stronger person is naturally the victor in a partner abuse situation also causes problems for victims of same-sex partners. In many of these relationships, the abuser is smaller in stature than the victim, leading the victim to have difficulties convincing others that abuse is occurring. Renzetti (1998) indicated that many lesbian victims reported their battering partners as smaller and as more feminine than themselves. Letellier (1996b) illustrated that this phenomenon is especially salient to male victims of same-sex partners with AIDS – the sicker, weaker partner is assumed to be incapable of inflicting abuse on the healthy partner, and the victim therefore loses credibility and is unable to obtain assistance.  Letellier and Renzetti both contest the assumption that greater size is the key determinant, because the victim may be constrained by a number of factors besides physical capabilities. For example, same-sex victims (like heterosexual male victims) may not wish to hurt their attackers out of love, or for fear that the police will mistake them for the perpetrator if they defend themselves. Letellier notes that gay male victims may also fear harming their partners if they are physically weakened due to AIDS. Merrill (1998) noted in his review that it is not simply a person’s physical, financial, or      Partner Abuse     43 social power that explains one person’s abuse of another, but instead a person’s willingness to use whatever power they possess to control others.      There are additional barriers faced by non-traditional victims. Letellier (1996b) pointed out that male victims of same-sex partners are often criticized and ridiculed by friends and family, turned away by shelters and crisis lines, and mislabeled as perpetrators by the police and other officials. He also noted (1996a) that only six official organizations exist in all of the United States to assist male victims of same-sex partners, and that in California, victims of same-sex partners must legally sustain more severe injuries than battered women before the violence is considered “criminal.” Renzetti (1992, 1996) found that lesbians attempting to seek help face similar barriers to  help-seeking: Although the women’s shelter staff claimed to support lesbian victims, there were no formal protocols or initiatives in place to actually accommodate them. Even trained counselors and psychologists have been found to employ incorrect and even dangerous methods to assist lesbian (as compared to heterosexual female) victims: A particularly prevalent mistake involves initiating couples’ therapy (a tactic noted for its propensity to incite batterers to increasingly severe violence as the victim attempts to assert his/her independence) instead of assisting the victim’s escape (Renzetti, 1992; Wise & Bowman, 1997).  If victims are aware of biases such as those apparent in the present study, they may fear the humiliation and ridicule that could result from their attempts to seek help. Our research demonstrates that gender role expectations appear to govern peoples’ attitudes toward partner abuse, and may prevent them from evaluating abuse on a       Partner Abuse     44 case-by-case basis. As a result, victims may be ignored simply because they are male, or because their batterer is female (Coney & Mackey, 1999; George, 1994, 1999, 2003; McNeely & Robinson-Simpson, 1987). The following is a good example of such bias: One article proposing treatment goals for female batterers indicated that helping them learn to access women’s shelters to escape their presumably abusive victims should be a main priority (Hamberger & Potente, 1996). The implications of this suggestion were disturbing – it is possible that female batterers, as a consequence of gender alone, may actually be more welcome at shelters than are battered men. It is evident that gender role stereotypes common in society and even in feminist theories may be a serious detriment to providing necessary aid to non-traditional victims. It is therefore appropriate to challenge theories propagating such stereotypes.  A Challenge of Feminist Partner Abuse Theory and Gender Role Stereotypes. The presumptions that women are physically incapable of harming their partners and that men are impervious to harm indicate gaps in reasoning that must be eliminated if we are to extend aid to victims in an unbiased and humanitarian manner. It surprised this author that the fiercest citations of women’s physical inferiority and consequent inability to harm men have originated from feminist theorists (see Berliner, 1990; Bograd, 1990; Dutton, 1994b; George, 1994; Hamberger & Potente, 1996; Vivian &  Langhinrichsen-Rohling, 1996). For feminists to assert that women are weak individuals lacking sufficient personal power to harm others might lead one to question their devotion to their own principles, as this argument is one redolent of the very patriarchal ideals feminists oppose (see George, 1994). The patently incorrect stereotype that women are less powerful than men could, in fact, be a force that has truly victimized women for      Partner Abuse     45 generations, leading women to identify with the victim role more strongly than men, and possibly stripping them of the confidence and empowerment necessary to escape hostile relationships (see Lamb, 1999 for more on victims and empowerment).   It is important to challenge the patriarchal and feminist assumptions that women are inherently passive, non-violent, and weak, and that men in comparison are dominant, aggressive, and strong (see Burke et al., 1988; Letellier, 1996a; Milardo, 1998; O’Leary, 2000). This may be one of the most important steps toward successfully intervening in violent relationships and assisting victims of each gender, as both are harmed by these stereotypes. For men, credibility as victims is negated through stereotype (see Letellier, 1996b), even preventing them from recognizing their own victimization (Letellier, 1996a). For women, stereotypes lead to disempowerment and even entrapment in the victim role (Lamb, 1999; Letellier, 1996a).  A More Inclusive Direction. Based on the evidence provided, it would appear that patriarchal attitudes and male domination may no longer be a useful theory for explaining the origins of partner abuse. Issues of dominance and power may not be as central to the issue of partner abuse as is assumed. Dutton (1994b) offered a strong theoretical deconstruction of patriarchy and gender as sufficient explanations of partner abuse. He argued that patriarchy as a system of power and dominance does not explain abuse when tested in North America or cross-culturally, and that it is only a factor among many others implicated in abusive relationships. Additionally, Burke et al. (1988) found that stereotypically feminine, rather than masculine, qualities were significantly associated with perpetration of abuse in both men and women. These researchers refute the notion that patriarchal, hyper-masculine attitudes are the cause of partner assaultive behaviors.      Partner Abuse     46 In light of findings by Coleman (1994), Dutton (1994a; 1998), and Spidel et al (2006), and the present study, the influence of personality disorders in partner abuse may prove a more promising avenue of future theory and research, and provide new directions to improve treatment. Carney and Buttell (2004) noted that traditional batterer treatment programs have been demonstrated to have little or no impact on recidivism rates, and they believe this ineffectiveness is likely a result of the homogenous approach used to treat batterers. They suggest different subtypes of batterers may require different interventions. This may be particularly relevant if personality disorders are highly related to partner assaults. For example, an initiative combining the effective elements of traditional batterer treatment with an empirically based, multi-level treatment for borderline personality disorder might lead to greater decreases in partner assaultive behaviors in those batterers with borderline characteristics. For example, dialectical behavior therapy, developed by Linehan, appears to offer a very promising treatment approach for borderline personality disorder (see McMain, Korman, & Dimeff, 2001; Swales, Heard, & Williams, 2000; Swenson, Sanderson, Dulit, & Linehan, 2001), and has even demonstrated effectiveness in reducing anger, hostility, and violence in a small sample of male prison inmates with borderline personalities (Evershed et al., 2003). For batterers with narcissistic or psychopathic characteristics such as those noted by Coleman (1994), other interventions might be more appropriate.    In light of the constellation of research examined, including the findings of the present study, it would seem that a gender-neutral, multidimensional theory inclusive of all relationship configurations, personality characteristics, sociocultural influences, and intergenerational transmission of violence such as the ones suggested by Coleman (1994)      Partner Abuse     47 and Dutton (1994b), may be more useful in conceptualizing abuse and developing appropriate treatment and intervention strategies (see also Dutton, 1994a). Strengths/Limitations  Strengths. Our research offered solid support for previous research (Bethke & DeJoy, 1993; Harris & Cook, 1994; Wise & Bowman, 1997), though some subtle differences existed, partly because of the differences in research designs. Our use of a within-subjects design nonetheless offered solid convergent support for pre-existing findings; we were able to control for variance due to individual differences in a manner that the previous researchers’ between-subjects designs did not allow for.    Our study also provided a new dimension to these studies, in that all four of these configurations regarded together allowed us to identify a particular trend in participants’ rank-ordering of severity, a trend indicating that male victims’ and female perpetrators’ injury sustaining and inflicting potential are minimized in a systematic way according to gender. These findings are particularly powerful when one considers the within-subjects experimental design. All four of the scenarios were identical in all aspects except for names and gender pronouns. The injuries described for all of the victims were identical; the violent acts committed by the perpetrators were constant across all scenarios. Nonetheless, we were able to identify significant differences in perceived severity. The within-subjects design’s control of variance allows us rule out alternate explanations for these differences, such as depicted size and strength differentials (all were equal), individual differences between participants (each participant acted as his or her own control), or other differences between the four scenarios in addition to the manipulated IV      Partner Abuse     48 (there were none). Therefore, it is reasonable to argue that the perceptual differences we discovered originated from the participants’ stereotypical gender expectancies.    Limitations. The study used a convenience sample of university students. Such samples lack representativeness because university students tend to differ from the general population in a number of ways. For example, their higher mean education levels could affect their responses to the survey; it is possible that other community members would respond differently. Additionally, the study relied on self-selected, volunteer participants, and it is well-known that volunteers possess different characteristics than non-volunteers, such as greater needs for approval and social desirability. As shown by the gender split in our sample, volunteers are also more likely to be female (for a complete discussion of volunteer characteristics, see Rosenthal & Rosnow, 1975). Some might argue that the relative youth of the sample might truncate the results. However, there is evidence (e.g., Magdol et al., 1997) that partner abuse is most prevalent among young adults, making them well-suited to this study. Nonetheless, the results cannot be generalized with confidence until further research improves the external validity of our findings.  Our sample size poses another limitation. It is probable that our relatively small sample made some of our comparisons weaker than anticipated. This may explain why we were unable to replicate the gender differences Harris and Cook (1994) found. The smaller sample was also unable to support strong correlational analyses; hence the reason these analyses are considered exploratory. With a larger-scale sample, it is possible that stronger gender differences will emerge, and solid support for our exploratory correlations may be found.       Partner Abuse     49 There are some limitations of self-report data common to all survey research. The primary difficulty with such data is accuracy. Self-report measures rely on accuracy of participant recall, and human recall has proven somewhat unreliable (see Graziano & Raulin, 2004). Inaccurate recall may have distorted participant responses to the BPO scale and CTS2. Additionally, social desirability biases likely affected our dependent variables. For example, more participants refused the CTS2 questions involving forcible sexual coercion than any other items. This may indicate intentional underreporting of these behaviors, weakening our ability to find correlations between participants’ partner abuse experience and our other measures. However, Arias and Beach (1987) conducted a study examining the effects of social desirability on responses to the original CTS, and found that social desirability led to decreases in self-reported perpetration of violence in both men and women. We can therefore at least be confident that male and female perpetration rates were not affected by different reporting tendencies between men and women.  Social desirability also may have strongly affected our abuse perception measures. The within-subjects experimental design used to present the vignettes increased the risk that participants would try to guess the purpose of our study, and respond accordingly. This was likely to have led participants to think that, because all of the scenarios were the same except for gender, it would be socially desirable to rate them all equally and not appear biased. This may have affected our results. For example, the ranking item was missing for 35 of our participants, and 23 indicated instead that they considered all of the scenarios to be equal. However, we feel that this may have lent particular strength to our findings because it indicated the direction participants’ social      Partner Abuse     50 desirability tended to take. This tendency was echoed in many of the responses to the open-ended questions. We were therefore able to detect significant differences in perception despite, rather than because of, our participant’s impulses to respond in a socially desirable and unbiased fashion. The missing data for the ranking question also did not prohibit use of the nonparametric tests, as enough participants answered the question to support these analyses, so our ability to support our hypothesis remained. Nonetheless, in the future it would be prudent to include a measure of social desirability in order to more systematically determine its impact on our data.  We failed to find differences for certain items measuring the vignettes. The item “did the perpetrator have the right to use physical force,” contained very little variation between the scenarios or even between participants; nearly everyone indicated very strongly that the perpetrator did NOT have the right. This may be a reflection of the advanced education that university students typically receive concerning human rights - no human has the “right” to use violence. The question’s lack of sensitivity could therefore be attributed to the term “right” having a very specific meaning to the participants. Perhaps in the future, less leading wording such as “was the perpetrator’s use of force justified (or understandable)?” would be more appropriate. Additionally, the item “did the victim fight back when beaten” did not indicate differences between scenarios. Some of our participants indicated that this particular question was confusing because the victim is clearly not depicted as fighting back. It may be advisable to drop this item in future studies, or revise it. For example, the wording “do you think the victim was capable of fighting back” might be less confusing. Harris and Cook’s (1994) study      Partner Abuse     51 found no differences for either of these two particular items, so it is not surprising that we were unable to detect differences.     Future Directions  As mentioned above, it would be useful to externally validate the findings of the present study using a large, random sample from the greater population of Canada. Additionally, further iterations of the study including measures of homophobia and social desirability will identify subgroups within the samples and different perceptions they may have of partner abuse. Knowledge of such attitudinal subgroups could help to tailor educational and intervention initiatives. Minor modifications to the vignettes and  Likert-Type questions designed by Harris and Cook (1994) may increase the sensitivity of these measures, and additional research will further establish their psychometric soundness.  The most important future direction, however, will be to examine the perceptions of subgroups in the community who come in direct contact with victims and perpetrators, such as police officers, judges, lawyers, social workers, emergency room staff, general practitioners, and nurses. People in these professions are on the front lines of intimate partner violence; as a result, their perceptions directly affect whether victims can access the assistance they require to escape their abusers.  A Final Note We would like to point out that, as a whole, the men and women in our study took the abusive scenarios seriously; we do not mean to imply by highlighting the differences that our participants are insensitive to partner abuse. On the contrary, many written responses indicated thoughtfulness and careful consideration of this issue. Our      Partner Abuse     52 examination of perceptual differences according to victim and perpetrator gender was undertaken with the sole intention of highlighting the importance of all victims and perpetrators of intimate partner assault, and must not be taken as an attempt to minimize or destroy credibility for any victims or any perpetrators. Use of this research to justify undermining or diverting funding for women’s shelters, or to negate the severity of  male-to-female partner abuse, would demonstrate a comlete misunderstanding of the theoretical underpinnings and results herein. Hopefully, this research will emphasize that all violence by humans against other humans is serious, and that all victims are deserving of assistance and freedom from ridicule, stigma, and marginalization.        Partner Abuse     53 References Archer, J. (2000). Sex differences in aggression between heterosexual partners: A  meta-analytic review. Psychological Bulletin, 126, 651-680. Arias, I., & Beach, S. R. H. (1987). Validity of self-reports of marital violence. Journal of Family Violence, 2, 139-148.  Arias, I., Samois, M., & O’Leary, K. D. (1987). Prevalence and correlates of physical aggression during courtship. Journal of Interpersonal Violence, 2, 82-90.   Berliner, L. (1990). Domestic violence. Journal of Interpersonal Violence, 5, 128-129.   Bethke, T. M., & DeJoy, D. M. (1993). An experimental study of factors influencing the acceptability of dating violence. Journal of Interpersonal Violence, 8, 36-51. Beyers, J. M., Leonard, J. M., Mays, V. K., & Rosen, L. A. (2000). Gender differences in the perception of courtship abuse. Journal of Interpersonal Violence, 15, 451-466. Bograd, M. (1990). Why we need gender to understand human violence. Journal of Interpersonal Violence, 5, 132-135.  Brown, J. (n.d.). Assessing the role of individual differences in accepting rape myth: Rating blame in sexual assault scenarios. Retrieved February 3, 2005, from Burgess, A. W., Baker, T., Greening, D., Hartman, C. R., Burgess, A. G., Douglas, J. E., et al. (1997). Stalking behaviors within domestic violence. Journal of Family Violence, 2, 389-403. Burke, P. J., Stets, J. E., & Pirog-Good, M. A. (1988). Gender identity, self-esteem, and physical and sexual abuse in dating relationships. Social Psychology Quarterly, 51, 272-285.       Partner Abuse     54 Campbell, J. C., Harris, M. J., & Lee, R. K. (1995). Violence research: An overview. Scholarly Inquiry for Nursing Practice: An International Journal, 9, 105-126. Capaldi, D. M., & Owen, L. D. (2001). Physical aggression in a community sample of  at-risk young couples: Gender comparisons for high frequency, injury, and fear. Journal of Family Psychology, 15, 425-440.  Carney, M. M., & Buttell, F. P. (2004). A multidimensional evaluation of a treatment program for female batterers: A pilot study. Research on Social Work Practice, 14, 249-258. Carrado, M., George, M. J., Loxam, E., Jones, L., & Templar, D. Aggression in British heterosexual relationships: A descriptive analysis. Aggressive Behavior, 22,  401-415.  Coleman, V. E. (1994). Lesbian battering: The relationship between personality and the perception of violence. Violence and Victims, 9, 139-152.  Coney, N. S., & Mackey, W. C. (1999). The feminization of domestic violence in America: The woozle effect goes beyond rhetoric. The Journal of Men’s Studies, 8, 45-58.  Cruz, J. M., & Firestone, J. M. (1998). Exploring violence and abuse in gay male relationships. Violence and Victims, 13, 159-172. DeMaris, A., Pugh, M. D., & Harman, E. (1992). Sex differences in the accuracy of recall of witnesses of portrayed dyadic violence. Journal of Marriage and the Family, 54, 335-345. Dobash, R. E., & Dobash, R. P. (1978). Violence against wives. A case against the patriarchy. New York: Free Press.       Partner Abuse     55 Dutton, D. G. (1994a). Behavioral and affective correlates of borderline personality organization in wife assaulters. International Journal of Law and Psychiatry, 17, 265-277. Dutton, D. G. (1994b). Patriarchy and wife assault: The ecological fallacy. Violence and Victims, 9, 167-182.  Dutton, D. G. (1998). The abusive personality: Violence and control in intimate relationships. New York: The Guilford Press.  Evershead, S., Tennant, A., Boomer, D., Rees, A., Barkham, M., & Watson, A. (2003). Practice-based outcomes of dialectical behavior therapy (DBT) targeting anger and violence, with male forensic patients: A pragmatic and non-contemporaneous comparison. Criminal Behavior and Mental Health, 13, 198-213.  Fiebert, M. S., & Gonzalez, D. M. (1997). College women who initiate assaults on their male partners and the reasons offered for such behavior. Psychological Reports, 80, 583-590.  Foo, L. & Margolin, G. (1995). A multivariate investigation of dating aggression. Journal of Family Violence, 10, 351-377.  George, M. J. (1994). Riding the donkey backwards: Men as the unacceptable victims of marital violence. The Journal of Men’s Studies, 3, 137-159.  George, M. J. (1999). A victimization survey of female-perpetrated assaults in the United Kingdom. Aggressive Behavior, 25, 67-79.  George, M. J. (2003). Invisible touch. Aggression and Violent Behavior, 8, 23-60. Graziano, A. M., & Raulin, M. L. (2004). Research methods: A process of inquiry. Pearson Education Group, Inc.       Partner Abuse     56 Hamberger, L. K. (1994). Domestic partner abuse: Expanding paradigms for understanding and intervention. Violence and Victims, 9, 91-94. Hamberger, L. K., & Potente, T. P. (1996). Counseling heterosexual women arrested for domestic violence: Implications for theory and practice. In C. M. Renzetti & L. K Hamberger (Eds.), Domestic partner abuse. Springer Publishing Company, Inc.  Harris, R. J., & Cook, C. A. (1994). Attributions about spouse abuse: It matters who the batterers and victims are. Sex Roles, 30, 553-565. Hegarty, K., & Roberts, G. (1998). How common is domestic violence against women? The definition of partner violence in prevalence studies. Australian and New Zealand Journal of Public Health, 22, 49-54. Hernandez, A., Spidel, A., Nicholls, T., Dutton, D. G., & Kendrick, K. (2006). Relationship between traumatic symptoms and rates of abusive behaviors perpetrated by female spousal assaulters. Poster presentation at the 3rd Annual Forensic Psychiatry Conference, Vancouver, B.C. Kwong, M. J., Bartholomew, K., & Dutton, D. G. (1999). Gender differences in patterns of relationship violence in Alberta. Canadian Journal of Behavioral Science, 31, 150-160. Lamb, S. (1999). Constructing the victim: Popular images and lasting labels. In: New visions of vicims: Feminists struggle with the concept, (pp. 108-138). New York, NY, US: New York University Press.  Letellier, P. (1996a). Gay and bisexual male domestic violence victimization: Challenges to feminist theory and responses to violence. In C. M. Renzetti & L. K Hamberger (Eds.), Domestic partner abuse. Springer Publishing Company, Inc.      Partner Abuse     57 Letellier, P. (1996b). Twin epidemics: Domestic violence and HIV infection among gay and bisexual men. Journal of Gay and Lesbian Social Services, 4, 69-81.  Lie, G-Y., & Gentlewarrior, S. (1991). Intimate violence in lesbian relationships: Discussion of survey findings and practice implications. Journal of Social Service Research, 15, 41-59. Lie, G-Y., Schilit, R., Bush, J., Montagne, M., & Reyes, L. (1991). Lesbians in currently aggressive relationships: How frequently do they report aggressive past relationships? Violence and Victims, 6, 121-135.  Magdol, L., Moffitt, T. E., Caspi, A., Newman, D. L., Fagan, J., & Silva, P. A. (1997). Gender differences in partner violence in a birth cohort of 21-year-olds: Bridging the gap between clinical and epidemiological approaches. Journal of Consulting and Clinical Psychology, 65, 68-78. Makepeace, J. M. (1986). Gender differences in courtship violence victimization. Family Relations, 35, 383-388. McClennan, J. C., Summers, A. B., & Vaughan, C. (2002). Gay men’s domestic violence: Dynamics, help-seeking behaviours, and correlates. Journal of Gay and Lesbian Social Services, 14, 23-49. McMain, S., Korman, L. M., & Dimeff, L. (2001). Dialectical behavior therapy and the treatment of emotion dysregulation. Psychotherapy in Practice, 57, 183-196.    McNeely, R. L., & Mann, C. R. (1990). Domestic violence is a human issue. Journal of Interpersonal Violence, 5, 129-131. McNeely, R. L., & Robinson-Simpson, G. (1987). The truth about domestic violence: A falsely framed issue. Social Work, 32, 485-490.      Partner Abuse     58 Milardo, R. M. (1998). Gender asymmetry in common couple violence. Personal Relationships, 5, 423-438.  Moffitt, T. E., Caspi, A., Krueger, R. F., Magdol, L., Margolin, G., Silva, P. A., et al. (1997). Do partners agree about abuse in their relationship? A psychometric evaluation of interpersonal agreement. Psychological Assessment, 9, 47-56. Murty, S. A., Peek-Asa, C., Ewerling, C., Stromquist, A. M., Burmeister, L. F., & Merchant, J. A. (2003). Physical and emotional partner abuse reported by men and women in a rural community. American Journal of Public Health, 93,  1073-1075. O'Leary, K. D. (2000). Are women really more aggressive than men in intimate relationships? Comment on Archer (2000). Psychological Bulletin, 126, 385-389. Peek-Asa, C., Garcia, L., McArthur, D., & Castro, R. (2002). Severity of intimate partner abuse indicators as perceived by women in Mexico and the United States. Women & Health, 35, 165-180.  Rathus, S. A., Nevid, J. S., Fichner-Rathus, L., & Herold, E. S. (2003). Human sexuality in a world of diversity: Canadian edition. Toronto, Ontario: Pearson Education Canada, Inc.  Renzetti, C. M. (1992). Violence in lesbian relationships. In M. Hansen & M. Harway (Eds.), Battering and family therapy: A feminist perspective (pp. 188-199). Thousand Oaks: Sage Publications, Inc.   Renzetti, C. M. (1996). The poverty of services for battered lesbians. Journal of Gay & Lesbian Social Services, 4, 61-68.      Partner Abuse     59 Rosenthal, R., & Rosnow, R. L. (1975). The volunteer subject. Oxford, England: John Wiley & Sons.  Spidel, A., Nicholls, T. L., Greaves, C., & Dutton, D. G. (2006). Personality disorder profiles of female intimate partner assaulters. Poster presentation at the Annual American Psychology and Law Society (AP-LS) Conference, Jacksonville, Florida. Steinmetz, S. K. (1978). The battered husband syndrome. Victimology: An International Journal, 2, 499-509.  Steinmetz, S. K. (1980). Women and violence: Victims and perpetrators. American Journal of Psychotherapy, 34, 334-350. Straus, M. A. (1979). Measuring intrafamily conflict and violence: The conflict tactics (CT) scales. Journal of Marriage and the Family, 41, 75-86. Straus, M. A. (1990). The conflict tactics scales and its critics: An evaluation and new data on validity and reliability. In M. A. Straus & R. J. Gelles (Eds.), Physical violence in American families: Risk factors and adaptations to violence in 8, 145 families. Straus, M. A. (1999). The controversy over domestic violence by women: A methodological, theoretical, and sociology of science analysis. In X. B. Arriaga & S. Oskamp (Eds.), Violence in intimate relationships (pp.17-44). Thousand  Oaks: Sage Publications, Inc. Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. (1996). The revised conflict tactics scale (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17, 283-316.      Partner Abuse     60 Straus, M. A., & Mouradian, V. E. (1998). Impulsive corporal punishment by mothers and antisocial behavior and impulsiveness of children. Behavioral Sciences and the Law, 16, 353-374.  Straus, M. A., & Sweet, S. (1992). Verbal/symbolic aggression in couples: Incidence rates and relationships to personal characteristics. Journal of Marriage and the Family, 54, 346-352. Swales, M., Heard, H. L., & Williams, J. M. G. (2000). Linehan’s dialectical behavior therapy (DBT) for borderline personality disorder: Overview and adaptation. Journal of Mental Health, 9, 7-23. Swenson, C. R., Sanderson, C., Dulit, R. A., & Linehan, M. M. (2001). The application of dialectical behavior therapy for patients with borderline personality disorder on inpatient units. Psychiatric Quarterly, 72, 307-324. Vivian, D., & Langhinrichsen-Rohling, J. (1996). Are bi-directionally violent couples  mutually victimized? In C. M. Renzetti & L. K Hamberger (Eds.), Domestic partner abuse. Springer Publishing Company, Inc. Walker, L. E. A. (1984). The battered woman syndrome. New York: Springer. White, J. W., & Humphrey, J. A. (1994). Women’s aggression in heterosexual conflicts. Aggressive Behavior, 20, 195-202.  Wise, A. L., & Bowman, S. L. (1997). Comparison of beginning counselors’ responses to lesbian vs. heterosexual partner abuse. Violence and Victims, 12, 127-135.      Partner Abuse     61 Table 1 Means and Standard Deviations for Men and Women’s Perceptions of Each Scenario  Men Women Scenario M SD M SD A (M-F) 87.37 8.68 91.25 7.86 B (M-M) 82.02 11.47 86.51 9.46 C (F-M) 80.89 12.53 85.42 10.11 D (F-F) 82.56 9.90 86.96 9.54  Table 2  Means and Standard Deviations for Men and Women’s CTS2 subscale responses   Men Women  M SD M SD Negotiation (Self) 73.42 39.92 75.25 37.31 Negotiation (Partner) 67.26 40.67  71.05 38.38 Psych. Abuse Perpetration 16.00 27.45 17.95 18.05 Psych. Abuse Victimization 17.78 22.78 16.40 18.61 Phys. Abuse Perpetration 1.91 5.10 2.34 5.96      Partner Abuse     62 Phys. Abuse Victimization 2.83 7.10 1.35 3.13 Sex. Coercion Perpetration 3.98 8.80 2.60 8.29  Sex. Coercion Victimization 4.13 9.28 3.86 9.71  Injuries Sustained .09 .28 .48 3.14 Injuries Inflicted .11 .40 .47 3.14  Table 3 Means and Standard Deviations for “How Violent…”  M SD Scenario A (M-F) 7.08 1.79 Scenario B (M-M) 6.60 1.63 Scenario C (F-M) 6.51 1.73 Scenario D (F-F) 6.71 1.65  Table 4  Means and Standard Deviations for “Police call…”  M SD Scenario A 8.38 2.04 Scenario B 7.07 2.61      Partner Abuse     63 Scenario C 6.82 2.78 Scenario D 7.20 2.64  Table 5  Means and Standard Deviations for “Perpetrator/assault…”  M SD Scenario A 8.69 1.98 Scenario B 7.75 2.45 Scenario C 7.47 2.86 Scenario D 7.70 2.60  Table 6  Means and Standard Deviations for “Should victim leave…”  M SD Scenario A 8.39 1.92 Scenario B 7.93 2.10 Scenario C 7.55 2.17 Scenario D 7.87 2.12  Table 7  Means and Standard Deviations for “Perp past acts…”  M SD Scenario A 7.89 1.85 Scenario B 7.64 1.96      Partner Abuse     64 Scenario C 7.38 2.20 Scenario D 7.37 2.12   Table 8  Means and Standard Deviations for “Liking perp…”  M SD Scenario A 8.99 1.46 Scenario B 8.55 1.68 Scenario C 8.50 1.75 Scenario D 8.57 1.64       Partner Abuse     65 Figure 1 Participant Marital Status  5.6%6.5%36.1%20.4%.9%3.7%26.9%OtherMarriedDatingCohabitingCommunal livingM ultiple partnersSingle      Partner Abuse     66 Figure 2 Frequency Histogram: Scores on Borderline Personality Organization (BPO) Scale Total score on BPO Scale110.0100. Personality Organization (BPO)Frequency20100Std. Dev = 15.65  M ean = 60.4N = 108.00      Partner Abuse     67 Figure 3   Scenario A Rank Frequencies  2.7%1.4%2.7%93.2%Least HarmfulThird Most HarmfulSecond M ost HarmfulMost Harmful      Partner Abuse     68 Figure 4  Scenario B Rank Frequencies  16.4%30.1%52.1%1.4%Least HarmfulThird Most...Second Most...Most Harmful      Partner Abuse     69 Figure 5  Scenario C Rank Frequencies  60.3%16.4%17.8%5.5%Least HarmfulThird Most...Second Most...Most Harmful      Partner Abuse     70 Figure 6  Scenario D Rank Frequencies  20.5%52.1%27.4%Least HarmfulThird Most...Second Most...      Partner Abuse     71 Figure 7  Men and Women’s Overall Mean Abuse Severity Ratings  6741N =Biological sex of respondentFemaleMale95% CI Grand total across all scenarios400350300250200150100500  6741N =Biological sex of respondentFemaleMale95% CI Grand total across all scenarios380370360350340330320310300       Partner Abuse     72 Appendix A    A Consideration of the Effects of Gender on Perceptions of Conflict in Relationships  PLEASE READ THE FOLLOWING VERY CAREFULLY!!! Hello, My name is Nicole Cormier. I am a psychology honours student at UBC Okanagan, and I am conducting this research under the supervision of Dr. Michael Woodworth. If you have any questions before or during your participation, please feel free to ask me at any time or contact Dr. Woodworth via office phone or email if I am not present or available.   You are invited to participate in a study of people’s perceptions of conflict in relationships. You must be at least 19 years of age to participate. This task will take approximately 30-60 minutes of your time. You may complete this questionnaire during one of the group sessions I have scheduled on campus. Alternatively, you may take this questionnaire if you are interested and complete it when convenient. If you choose this option, we request that you complete the questions in private, in a quiet setting, and without consulting with others on your answers.  Please complete all material in the order presented. The first part of the questionnaire includes four scenarios to read. After reading each one, you will be asked to answer 11 questions examining your opinions about the conflict. After finishing all four, you’ll rank the scenarios in order of how serious you feel they are. Then, there are some questions about your personality characteristics, and about your own experiences with conflict in relationships. You will be asked to provide your age, sex, relationship status, and sexual identity. Last, you have a chance to voice your own opinions with two open-ended questions. If you are completing the questionnaire in a group session, please deposit the questionnaire in the box provided when you are done. If you are taking a questionnaire away to complete in your own time, when you have completed it, please seal it in the provided envelope and slide it under Dr. Woodworth’s office door (ART 331).  You may find some of the questions are sensitive or personal, particularly parts D, E, & F of the survey. To protect your privacy, you will not be asked to provide your name or contact information, and these things will not be included in the data or published. Your responses are completely confidential and anonymous, and only the principal investigator and I will see your answers, which will be locked in a file cabinet and only removed for data entry and analysis. Do not write your name anywhere on the questionnaire.  Please answer all questions as truthfully as you can. Do not worry about what answers I might want, whether your opinion is right or wrong, or how politically correct your answer is. This study is intended to look at different people’s perceptions and values, as well as their personal experiences. These things are unique for every individual – there are no wrong answers!       Partner Abuse     73 Participation in this study is strictly voluntary. You are free to decide not to participate at any point up until you hand the data back to me, even if you have filled out the questionnaire. You can also indicate that you’ve changed your mind about participating at the end of the questionnaire by checking the box at the very end. If you hand in a completed questionnaire without checking the box, you will be consenting to have your data included in the study. The data cannot be removed after you hand in the questionnaire because it will not be connected in any way to your name (I will not be able to find your data to remove it).  These scenarios may contain physical abuse and same-sex relationships, which might disturb some readers. The questions about your own experiences also cover some physical, emotional and sexual abuse and sexual identity questions. Remember, you are not obligated to participate; if you feel offended by the content, you are free to leave with no negative consequences. You can also leave any question blank if it makes you feel uncomfortable or upset.  This study will enhance your understanding of how research can be conducted, which will be valuable in your future academic career. If you are a first-year student, you will also receive bonus credit toward completion of Intro PSYC 121 through Experimatrix. The answers you provide in this questionnaire will help psychologists and others to better understand conflict in relationships – a very important topic both academically and socially. I, the researcher, will also directly benefit from the experience of conducting research. The data from this project will be analyzed and presented in an undergraduate thesis, and submitted for publication in academic journals. If you have any questions after you have finished the study, please feel free to contact the principal investigator, Dr. Mike Woodworth (phone: (250) 807-8913; email:; office #: ART 331). At the end of Semester 2, we invite you to contact us to find out what the results of the study are. If you have any questions or concerns about how you are treated or what your rights are as a participant in research, you can contact the Chair of the Research Ethics Board (REB) (Phone: 250-807-8150; or contact through the UBC Okanagan office, or through Research Services). There is contact information for the researchers and the Chair of the REB, as well as a list of Crisis and Health contacts, provided along with your questionnaire. Please keep them.      Partner Abuse     74 Appendix B  Sample Vignette & Likert-Type Questions  Mike Jones, a 28-year-old white male, was arrested last night on charges of domestic abuse. Two police officers arrived at the location of the dispute at 7:05 pm. Police conducted interviews with Mr. Jones (a sales representative for a local furniture store) and his wife, Mary Jones (an interior designer). According to Officer Kevin Smith, of the Kelowna R.C.M.P detachment, he and another officer found Mrs. Jones on the living room couch bleeding with a black eye. Mrs. Jones, a 28-year-old white female, told the officers that she had arrived home late from work, turned on the TV and then made some phone calls. Approximately ten minutes later, Mr. Jones arrived home and became angry because his wife was on the phone. He then yelled at her that she had things to do and should make sure that she gets home on time.  Mrs. Jones became upset, began yelling at her husband, and, as her anger heightened, she began to shout various obscenities at him, calling him a “nagging bastard” and a “miserable excuse for a man.” She threatened to leave him if he didn’t shape up. Mrs. Jones then went into the kitchen to prepare dinner. Mr. Jones followed her, grabbed her by the arm and slapped her, knocked her to the floor, and kicked her. As Mrs. Jones lay there in stunned surprise, Mr. Jones left the house. Upon his return, he was informed by the police that his wife was charging him with assault.  Please answer the following ratings questions to the best of your ability. Put a circle or dash on the line indicating your answer:  1. As crimes go, how violent was the incident? (1: not violent; 10: very violent) 1      2      3      4      5      6      7      8      9      10  2. How responsible was the perpetrator for the incident? (1: not responsible; 10: very responsible) 1      2      3      4      5      6      7      8      9      10  3. How responsible for the incident was the person who was beaten? (1: not responsible; 10: very responsible) 1      2      3      4      5      6      7      8      9      10  4. If you had witnessed this incident from the window next door, how likely would it have been that you would have called the police? (1: not likely; 10: very likely) 1      2      3      4      5      6      7      8      9      10  5. Did the perpetrator have the right to use physical force? (1: definitely; 10: definitely not) 1      2      3      4      5      6      7      8      9      10       Partner Abuse     75 6. In this case should the perpetrator be convicted of assault? (1: definitely; 10: definitely not) 1      2      3      4      5      6      7      8      9      10  7. Did the victim fight back when beaten? (1: definitely; 10: definitely not) 1      2      3      4      5      6      7      8      9      10  8. Should the victim leave the perpetrator for good? (1: definitely; 10: definitely not) 1      2      3      4      5      6      7      8      9      10  9. Do you think the perpetrator has probably acted this way in the past? (1: definitely; 10: definitely not) 1      2      3      4      5      6      7      8      9      10  10. Overall, how much do you like the perpetrator? (1: very much; 10: not very much) 1      2      3      4      5      6      7      8      9      10  11. Overall, how much do you like the victim of the beating? (1: very much; 10: not very much) 1      2      3      4      5      6      7      8      9      10              Partner Abuse     76   


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items