Borderline and Antisocial Personality Scores as Mediators Between Attachment and Intimate Partner Violence

Borderline and Antisocial Personality Scores as Mediators Between Attachment and Intimate Partner Violence


Anne Marie Mauricio, PhD Jenn-Yun Tein, PhD

Arizana Prevention Research Center Arizana  State  University

Frederick G. Lopez, PhD

University of Houston,  TX


Court-mandated male batterers (n = 192) attending an intervention program completed measures examining adult attachment orientations (anxious and avoidant), personality disorders (borderline and antisocial), type of violence (psychological and physical), and social desirability. Structural equation modeling was used to determine whether there were significant relationships between anxious attachment and physical and psychological violence that are mediated by either borderline or antisocial personality disorders. Social desirability was included in both models as a covariate. Results indicated that personality disorders fully mediated the relationship between avoidant attachment and physical as well as psychological violence. Personality disorders only partially mediated the relationship between anxious attachment and psychological violence. Implications for intervention are discussed.

The recognition of intimate partner violence (IPV) as a serious and pervasive problem  has prompted  research  exploring  factors contributing  to such violence.

The urgency to attend to IPV is further highlighted by findings that it permeates every race, religion, social class, and educational level (Bureau of  Justice  Statistics, 1995; National Institute of Justice, 1998, 2000; Straus & Gelles, 1986). Research exploring the origins of IPV  has  identified  a  wide  range  of  predictors  that  account for IPV against women (Schumacher, Feldbau-Kohn, Smith Slep, & Heyman, 2001; Schumacher, Smith Slep, & Heyman, 2001). Adult attachment characteristics and per­ sonality disorders are two predictors of IPV that have received increasing attention in recent years.



Attachment theory provides a rich conceptual framework for understanding IPV. Attachment theory (Bowlby, 1973, 1980, 1982) posits  that,  for the  purpose  of  safety and survival, humans are innately driven to seek attachments or close enduring emo­  tional bonds with others. Moreover, through attachment processes, individuals develop  an internalized set of beliefs about self and other, known as “internal working models” (Bowlby, 1973, 1988). The internal working model of self influences one’s perceptions about his or her  self-worth,  competence,  and  Jovability,  whereas  the working  model of other is responsible for expectations about the availability and trustworthiness  of others.

When caregivers are available and consistently responsive during childhood, secure attachments and corresponding positive internal working models of self and other  develop and promote healthy developmental trajectories and relationship patterns (Ainsworth, Blehar, Waters, & Wall, 1978). By contrast, disruptions in the parent-child bond are the precursors to insecure attachment and corresponding negative  models  of  self and/or others, thus promoting maladaptive relationship patterns that can continue to regulate relationship behavior into adulthood. Specifically, attachment theory maintains that insecurely attached individuals, compared with securely attached individuals, dem­ onstrate more anxiety, avoidance, or both characteristics in their intimate relationships (Fraley & Waller, 1998).

Research supports the conceptualization of attachment as a two-dimensional structure (Bartholomew & Horowitz, 1991; Brennan, Clark, & Shaver, 1998; Brennan & Shaver, 1995; Lopez & Brennan, 2000). The first dimension (i.e., attachment avoidance) is charac­ terized by a pervasive discomfort with intimate closeness and a strong orientation toward self-reliant and counterdependent relationship behavior. The second dimension (i.e., attachment anxiety) is represented by low self-esteem, pervasive fears of partner rejection and abandonment, and dependent relationship behavior. Whereas the avoidant dimension is closely related with a negative model of others, a negative model of self is associated with anxious attachment.

Attachment theory has been proposed  as  a  conceptual  framework  for  understand­ ing interpersonal dynamics in adult relationships (Hazan & Shaver, 1987), including relationship violence (Bowlby, 1984; Dutton, 1995; Fonagy,  1999; Mayseless,  1991). For example, given their low self-esteem and fears of rejection, adults with an anxious attachment orientation who perceive their intimate partner as unavailable may respond to this distress by engaging in overt expressions of anger and manipulation targeted toward keeping their partner engaged in the relationship. On the other hand, given their gener­ ally hostile interpersonal orientation and negative models of others, avoidantly attached adults may be more distressed by fears of engulfment and partner assertion and more disposed to use violence and other abusive strategies to control and intimidate their part­ ners (Mayseless, 199I ). Research exploring associations between adult insecure attach­ ment and male-perpetrated IPV against women has provided support for the hypothesis that insecurely attached men are at risk for committing IPV (Tweed & Dutton, 1998) and that attachment dynamics corresponding with insecure attachment (e.g., fear of aban­ donment) may explain  IPV  (Dutton  & Browning,  1988; Dutton, Saunders, Starzomski, & Bartholomew, 1994; Dutton, Starzomski, & Ryan, 1996; Woike, Osier, & Candela, 1996).



As noted earlier, personality disorders have also recently gained prominence in under­ standing the etiology of IPV. Specifically, research has demonstrated that personality disorders related to the need to control others (e.g., narcissistic and antisocial) and related to self-concept and identity (e.g., borderline) are particularly prominent among batterers (Hamberger & Hastings, 1988). Additionally, research on violent husbands (Holtzworth­ Munroe, Bates, Smutzler, & Sandin, 1997) suggests that violent husbands compared with nonviolent men are more likely to evidence borderline and antisocial personality disor­ ders. For example, Hastings and Hamberger (1988) found that batterers demonstrated more borderline symptomatology than nonbatterers, and Dutton et al. ( 1996) found that descriptions of batterer personalities and behaviors fit the clinical classifications of bor­ derline personality disorder. Murphy, Meyer, and O’Leary (1993) found that violent men, as compared to nonviolent men, consistently demonstrated higher scores on a measure of antisocial personality disorder after controlling for social desirability.

Studies demonstrating the prevalence of personality dynamics descriptive of bor­ derline and antisocial personality disorders among batterers provide  additional  support for the hypothesis that the presence of a personality disorder may be a risk marker for committing IPV. For example, a hallmark of borderline personality disorder is fear of abandonment (Gunderson, 1984; Sack, Sperling, Fagen, & Foelsch, 1996), and Dutton  and Browning (1988) found that, compared with nonabusers, physical abusers reported the most anger to scenarios depicting abandonment of a male by his female partner. Additionally, research suggests a positive relationship between excessive dependency needs, which are also a defining feature of the borderline personality (Gunderson, 1984; Sack et al., 1996) and relationship violence (Murphy et al., 1993). Deficiencies in empa­ thy as well as a lack of remorse are often associated with antisocial behaviors (American Psychiatric Association, 1994), such as violence. Supporting this association, Mehrabian (1997) found that measures of aggression and violence were negatively correlated with measures of emotional empathy, and Bovasso, Alterman, Cacciola, and  Rutherford (2002) found that deficient empathy predicted violent crime. Research with samples of men perpetrating  IPV suggests that these men exhibit antisocial characteristics,  such as a tendency to externalize responsibility for behavior (Flourny & Wilson, 1991; Hale, Zimostrad, Duckworth,  & Nicholas, 1988).



In addition to providing a rich conceptual framework for understanding IPV, attachment theory also provides a theoretical framework for understanding personality disorder ori­ gins, especially borderline and antisocial personality disorders. Bowlby ( 1988) argued that the quality of one’s early attachments determine internal representations of self and other. In turn, these mental representations influence interpersonal functioning and impact later psychological health. Secure attachments early in life contribute to healthy psychological development, whereas insecure attachments make one vulnerable to psychopathology. Specifically, anxious attachment produces a vulnerability to disorders that exaggerate negative affect and distress in an effort to get the attention of an attachment figure. Fear of abandonment, clinging behaviors, and an uncertainty as to whether the attachment figure will respond accompany overwhelming negative affects. In contrast, avoidant attachment makes one prone to behaviors that minimize distress and vulnerable  to disorders that  deny the presence of negative affect. Because avoidantly  attached individuals expect to be rebuffed, they become compulsively self-reliant and deny intimacy needs. Thus, the psychopathology of anxious attachment is more likely to reflect borderline traits, whereas the psychopathology of avoidant attachment is more likely to reflect antisocial personality traits.

Others have followed Bowlby in positing the etiological role of attachment and disturbed working models of self and other in theoretical conceptualizations of personality disorders (Fonagy, 2003; Sable, 1997). Consistent with theoretical suppositions, many researchers have underscored the importance of attachment in understanding personality disorders (Brennan & Shaver, 1998; Livesley, Schroeder, & Jackson, 1990; Sheldon & West, 1990), providing evidence of relationships between anxious attachment and borderline personal­ ity disorder (Fonagy et al., 1996; Patrick, Hobson, Castle, Howard, & Maughan, 1994; Rosenstein & Horowitz, 1996; West, Keller, Links, & Patrick, 1993) as well as between avoidant attachment and antisocial personality disorder (Gacano & Meloy, 1991, 1992; Gacano, Meloy, & Berg, 1992).



 Relationships between anxious  attachment  and borderline  personality  disorder  as well as relationships between avoidant attachment and antisocial personality disorder raise questions about whether these adult attachment orientations and personality disorders operate independently in predicting IPV or whether the presence or absence of personal­ ity disorders functionally mediates the relationship between insecure adult attachment  and IPV. While previous research has demonstrated the contributions of adult attachment orientations and personality disorders in generally predicting IPV as well as relationships between adult attachment and personality disorders, no study has tested whether personal­ ity disorders mediate the effects of adult attachment on violence. Answering this question has important implications for batterer intervention, as models of intervention are related to current understandings of risk factors for committing IPV. It also remains unclear whether any observed interrelationships among adult attachment orientations, personality disorders, and IPV are a function of the form of IPV that is being assessed. A review docu­ menting different risk markers for psychological and physical violence (see Schumacher, Feldbau-Kohn, et al., 2001; Schumacher, Smith Slep, et al., 2001) supports the inclusion of psychological and physical violence as distinct outcomes in IPV research. Schumacher et al.’s reviews identified insecure attachment and personality disorders as risk markers for both psychological and physical violence. It is thus possible that direct and mediational effects of adult attachment and personality disorders on physical and psychological abuse differ depending on the form of IPV that is being predicted. The current study will address these questions by posing and testing mediational models that, respectively, predict scores on indicators of psychological abuse and physical violence.

The proposed effort to consider how attachment and personality disorders function together to explain IPV may also contribute to the discussion on batterer typologies. A significant contribution to the field of IPV has been the advancement of batterer typology research  premised  on the assumption  that batterers  are a heterogeneous  group that may respond differentially to treatment (Langhinrichsen-Rohling, 2005). In general, typology research suggests that there are subgroups of batterers that are behaviorally, emotionally, and cognitively diverse. Moreover, typology research has indicated that anxious and avoid­ ant adult attachment orientations as well as borderline and antisocial personality disorders each play central roles in distinguishing batterer subgroups. Specifically, one of the pro­ posed subgroups is avoidantly attached with antisocial personality disorder characteristics, and a second proposed subgroup is anxiously attached with borderline personality disorder characteristics (Hamberger, Lohr, Bonge, & Tolin, 1996; Holtzworth-Munore, Meehan, Herron, Rehman, & Stuart, 2000; Tweed & Dutton, 1998; Waltz, Babcock, Jacobson, & Gottman, 2000). The conceptualization outlined in this article, which juxtaposes anxious attachment and borderline personality disorder as well as avoidant attachment and antiso­ cial personality disorder, is consistent with the typology literature. However, understand­ ing mediational processes between attachment and personality dysfunction will answer questions regarding the dynamic by which attachment and personality may operate to influence IPV for a particular subgroup of batterers.



Based on research demonstrating associations between insecure attachment and personality disorders, we hypothesized that relationships between adult attachment orientations and IPV would be mediated by personality disorders. More specifically, we hypothesized that borderline personality disorder would mediate the relationship between anxious attachment and IPV, whereas antisocial personality disorder would mediate the relationship between avoidant attachment and violence. In this study, IPV was operationalized by measures of psychological and physical violence against intimate female partners.






Participants were 192 heterosexual men with at least an eighth-grade reading level that were court mandated to attend a community batterer intervention program. Three percent of the sample had a graduate degree, 6% had a college degree, 27% had at least 1 year of college, 40% had a high school diploma or graduate equivalent, 17% had at least 1 year  of high school, and 6% had only a junior high school education. Thirty-six percent of the sample reported that they were married, 22% were single and not living with a girlfriend, 22% were single and living with a girlfriend, 11% were divorced, and 9% were separated. The age of participants ranged from 17 to 56 years old, with a mean of 33 (SD = 8.83). The racial ethnic breakdown of the sample was 67% Caucasian, 16% African American, 8% Hispanic, 3% Native American,  1% Asian, and 5% self-identified as “other.”



Study participants were invited to participate in this study by the first author during one  of the participants’ court-mandated weekly group sessions. These men were informed that participation in this study was voluntary and anonymous. If they declined to participate, they would complete an alternate activity with the group facilitator. Participants were also informed that they would receive $15 for their participation. All solicited men agreed to participate. Participants then completed informed consent forms and a packet of self-report questionnaires. The order in which the questionnaires were administered was varied from participant to participant. Participation time was approximately 45 minutes.



Demographic and Participant Background Questionnaire. This questionnaire was developed to gather information regarding the participants’ ethnicity, age, educational level, and relationship status for the purpose of describing the sample.

Reading Level. The Basic Reading Inventory: Graded Passage-Form B (BRI; Johns, 1997) was individually administered to participants as an informal reading test to ensure that respondents had an eighth-grade reading level. Published initially in 1978, the BRI has been used extensively by various types of professionals in education and psychology. Form B is a silent reading test, where participants read and subsequently respond in writing to 10 corresponding comprehension questions (i.e., “What is this passage about?”). If the participant responds inaccurately to more than three questions, his comprehension of the material is below the eighth-grade level. At least an eighth-grade reading level was needed to understand the self-report inventories administered. All participants demonstrated an eighth-grade reading level.

Antisocial and Borderline Personality Disorder Characteristics. Antisocial and bor­ derline personality disorder characteristics were assessed using the Personality Diagnostic Questionnaire- Revised (PDQ-R; Hyler et al., 1988), a self-report measure that  screens for personality disorders described in the Diagnostic and Statistical Manual of Mental Disorders (American Psychological Association, 1987). Scores can be computed as con­ tinuous or categorical variables. Categorical scores can be compared with cutoff scores and used to determine whether a person meets criteria for a specific personality disorder. Participants in this study completed only borderline and antisocial subscales. Because the PDQ-R was used in this study to assess dimensions of antisocial and borderline person­ ality disorders rather than to diagnose participants, continuous scores were used for the purpose of this study. The borderline subscale has 12 items, and the antisocial subscale  has 11 items.

“Some people consider me a drifter” and  “lying comes easily to  me  and  I often  do it” are representative of items on the PDQ-R antisocial subscale. “I feel empty or bored much of the time” and “my feelings toward another person can often change drastically” are representative of items on the PDQ-R borderline subscale. The PDQ-R is valid  for  use with both clinical (Hyler, Skodol, Oldham, Kellman, & Doidge, 1992; Hyler et al., 1990) and nonclinical samples (Johnson & Bornstein, 1992; Wierzbicki & Gorman, 1995). Correlations between borderline and antisocial subscales of the Millon Clinical Multiaxial Inventory-II (MCMl-11)-a well-validated measure of personality disorders-and the PDQ-R borderline subscales and antisocial subscales have been as high as .57 and .37, respectively, suggesting construct validity of the PDQ-R borderline and antisocial sub­ scales (Wierzbicki & Gorman, 1995). In the present study, Cronbach alpha coefficients of

.66 and .62 were obtained for scores on the borderline and antisocial subscales, respec­ tively. However, dropping the item “I drive pretty well even if I have been drinking” from the antisocial subscale increased this subscale’s score reliability to .64. As such, this item was not included in the antisocial subscale used for analyses in this study. Fossati et al. (1998) found somewhat comparable internal consistency coefficients, with an alpha equal to .70 for the borderline  scale and .63 for the antisocial scale.

Anxious and Avoidant Adult Attachment Orientations. Adult attachment orientations were  assessed  using  the Experiences  in  Close Relationships  (ECR)  short form. The ECR  is a 36-item measure of adult romantic attachment orientation comprised of two subscales: orientation toward anxious adult attachment and orientation toward avoidant adult attach­  ment (Brennan et al., 1998). Each subscale is comprised of 18 items. “I worry about being abandoned” and “I worry a lot about my relationships”  are representative  of  items  making  up the anxious attachment orientation subscale. “I prefer not to show a partner  how  I feel  deep down” and  “I get uncomfortable  when  a romantic  partner  wants  to be very  close” are representative of items making up the avoidant attachment orientation subscale. Respondents indicate  their  level  of  agreement  with  each  item  on  a 7-point  rating  scale  (l = Disagree Strongly; 7 = Agree Strongly), and, following recoding of some reverse-keyed items, item ratings on each subscale are summed to produce total scores, with higher scores indicating higher attachment anxiety and avoidance, respectively.  Alpha  coefficients for the anxious  and avoidant scales reported by Brennan et al. are .94 and .91, respectively, indicating good internal consistency. In the present study, Cronbach alpha coefficients were .90 and .86 for scores on the anxious and avoidant subscales,  respectively.

Intimate Partner Physical Violence. Physical violence was assessed using the Conflict Tactics Scale (CTS; Straus, 1979, 1990). The CTS is an 18-item self-report inventory con­ sisting of a list of actions the respondent might take in a conflict with his partner. Participants indicated their use of violent tactics in the 12-month period prior to their arrest. The modes of conflict are scored on three dimensions: reasoning, verbal aggression, and physical violence. For the purpose of this study, only the nine-item  physical  violence subscale was   used.

Straus (1979) reported that scores on the CTS physical violence subscale obtained a Cronbach alpha coefficient of .83 and demonstrated both construct and concurrent validity. Additionally, evidence of the CTS’s concurrent validity is demonstrated by anticipated relationships between CTS violence subscales and variables predictive of male violent behavior toward female partners (i.e., the lower a husband’s economic  and  prestige  resources, the greater his tendency to use violence to maintain a male-dominant power position; Straus, 1979, 1990). In the present study, the Cronbach alpha coefficient for the physical  violence  subscale  was .71.

Intimate Partner Psychological Violence. Psychological  violence  was  assessed  using the short form of the Psychological  Maltreatment  of Women  Scale (PMWS; Tolman,   1989, 1999). The PMWS short form contains 14 items describing various acts of coercion, verbal abuse, and intimidation of intimate partners. Using a 5-point rating scale (1 = Never; 5 = Very  Frequently),  participants  indicated  the  frequency  with  which  they  engaged  in these acts of psychological violence in the 6 months prior to their arrest. The PMWS contains two factor-analytically-derived subscales: emotional/verbal abuse and dominance/isola­ tion (Tolman, 1989, 1999). Emotional/verbal abuse items include withholding emotional support, verbal attacks, and humiliation. “I called my partner names” and “I treated my partner like an inferior” are representative of items in the emotional/verbal abuse domain. Dominance/isolation items tap the respondent’s demands for subservience.  “I interfered  in my partner’s relationships with other family members” and “I restricted my partner’s use of the telephone” are representative of dominance/isolation items. Tolman reported Cronbach alpha coefficients for scores on the dominance/isolation and emotional/verbal short subscales of .88 and .92, respectively (Tolman, 1999). In the present study, the two PMWS subscales were strongly intercorrelated ( r = .67); therefore, scores on these scales were aggregated to create a composite psychological violence variable that obtained a Cronbach alpha coefficient of .87.

Social Desirability. When conducting domestic violence research, it is important to control for socially desirable response bias (Arias & Beach, 1987). In this study, the short form of the Marlowe-Crowne Social Desirability Scale (SDS; Crowne & Marlowe, 1964) was used. The SDS contains 10 items. “I am quick to admit making a mistake” is repre­ sentative of  items on the short form of the SDS. Participants  were asked to respond  on   a 7-point Likert scale the extent to which they agreed or disagreed with inventory items. Previous studies using the SDS have yielded Cronbach alpha coefficients ranging from .72 (Loo & Thorpe, 2000) to .77 (Crino, Svoboda, Rubenfeld, & White, 1983) and a I-month interval test-retest reliability coefficient of .86 (Crino et al. 1983). In the present study, a Cronbach alpha coefficient of .74 was obtained.



Descriptive statistics and zero-order correlations for all variables included in the model can be found in Table 1. Scores on the PDQ-R antisocial and borderline personality disorder subscales were significantly and positively correlated with adult attachment orientations. As expected, anxious attachment and borderline personality disorder scores were highly correlated, and avoidant attachment and antisocial personality disorder scores were correlated. Unexpectedly, PDQ-R antisocial scores were also correlated with anxious attachment, and PDQ-R borderline scores were also correlated with avoidant attachment. Psychological violence was correlated with both anxious and avoidant attachment as well as with PDQ-R antisocial and borderline scores. Physical violence was correlated with anxious attachment as well as with antisocial and borderline scores but not with avoidant attachment. Social desirability was negatively correlated with measures of attachment, personality disorders, and violence, thus justifying its statistical control in our primary analyses (Arias & Beach,  1987).


TABLE 1. Descriptives and Zero-Order Correlations for Covariates, Predictors, and Outcomes in Models Presented in Figure 1and Figure 2

Scale                                  X            SD                           2           3           4          5         6


1.  Social


2.   Psychological











3.  Physical violence









4.  Borderline PD 3.67 2.09 .50** .57** .36**
5.  Antisocial PD 2.3 1.94 -.52** .47** .36** .60**

6.  Avoidant















7. Anxious 3.68 1.28 -.40** .50** .24** .57** .41** .23**

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


Path modeling with EQS 5.1 (Bentler,  1995) was used  to test  the  hypotheses  that the relationship between anxious attachment and violence (physical and psychological) behaviors is mediated by borderline personality disorder and that the relationship between avoidant attachment and violence is mediated by antisocial personality disorder. Because of significant correlations between borderline and antisocial personality scores, anxious attachment and antisocial personality, and avoidant attachment and borderline person­ ality, we also explored the possibility of borderline personality disorder mediating the relationship between avoidant attachment and violence and the possibility of antisocial personality disorder mediating the relationship between anxious attachment and violence. Preliminary analyses indicated that the data met assumptions regarding normality, linear­ ity, and homoscedasticity.

To establish potential mediational pathways, two hypotheses should be supported (see Judd & Kenney, 1981; MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002): (a) the predictor variable should be significantly related to the mediator, and (b) the mediator should significantly predict the outcome after adjusting for the effect of  the predictor.  For example, in the mediation model with avoidant attachment, antisocial personality, and violence, avoidant attachment should have a significant path to antisocial personal­ ity, and antisocial personality should have a significant path to physical or psychological violence after controlling for avoidant attachment. Given the support of these hypotheses, the effect of antisocial personality in mediating the relation between avoidant attachment and physical or psychological violence or the effect of borderline personality in mediating the relation between anxious attachment and physical or psychological violence can be tested for statistical significance using the multivariate-delta method (MacKinnon et al., 2002; Sobel, 1986). This method calculates the standard error of the mediation effect (the multiplication of the path from the predictor variable to the mediator [path a] and the path from the mediator to the outcome [path b], ab), which is then used to test the significance of the mediation effects (i.e., z = ab/SEab; z 2: 1.65 for p s; .05, one-tail test).

Figure 1 depicts the results of the mediation analysis for the model with avoidant attach­ ment as the predictor, antisocial and borderline personality disorders as the mediators, and physical and psychological violence as the outcomes. Figure 2 depicts the results of the mediation analysis for the model with anxious attachment as the predictor, antisocial and borderline personality disorders as the mediators, and physical and psychological violence


Figure 1. Mediational model with avoidant attachment, antisocial and borderline personality disorder scores, and physical and psychological violence. Social Desirability entered as covariate but path to dependent variables not shown for simplicity of presentation.

Figure 2. Mediational model with anxious attachment, antisocial and borderline personality disorder scores, and physical and psychological violence. Social Desirability entered  as covariate but path to dependent variables not shown for simplicity of presentation.


as the outcomes. Because of the saturated nature of these two models, the model fit the data perfectly  (i.e., x2  [0, N = 192] = 0, RMSEA  = 1.00, SRMR = 1.00) for both  models.  For the model shown in Figure 1, avoidant attachment was significantly related to both  antiso­  cial personality (P = .17, p < .01) and borderline personality (P = .21, p < .01). Antisocial personality was significantly related to both physical violence (P = .23, p < .01) and psy­ chological violence (P = .16, p < .05). In addition, borderline personality was significantly related  to both  physical  violence  (P = .24, p < .01) and psychological  violence  (P = .42, p < .01). The direct path from avoidant attachment to physical violence or to psychological violence  was  not  significant  after taking  the  mediator  variables,  antisocial  and borderline personality  disorders,  into account  (P = .02, ns; p = -.02, ns, respectively).  Results  of  the multivariate delta method (Sobel, 1982, 1986) indicated that both antisocial and borderline personality  disorders  were  significant  mediators  of  the  relation  between  avoidant  attach­ment and physical violence (z = 1.89,p = .05; z = 2.17, p = .01, respectively) and the relation between  avoidant  attachment  and  psychological  violence  (z  =  1.65, = .05; z =  3.02, p = .01). Antisocial and borderline personality disorders together served as a full mediator of the relation between  avoidant attachment  and physical  violence and psychological  violence. A full mediation process is characterized as having a significant mediation pathway with a nonsignificant direct path. A partial mediation process is characterized as having a signifi­  cant mediation pathway as well as a significant  direct path.  The model  accounted  for  16% of the variance  for physical  violence  and 36% of variance  for psychological   violence.

For the model shown in Figure 2, anxious attachment was significantly related to borderline personality (P = .44, p < .01) and antisocial personality (P = .24, p < .01). Borderline personality was significantly related to both physical violence (P = .23, p < .05) and psychological violence (P = .30, p < .01). Antisocial personality was significantly related to physical violence (P = .23, p < .05) but not psychological violence. After taking borderline and antisocial  personality  into account, the direct path  from anxious attachment  to psychological violence remained significant  (P =  .24, < .01).  However,  the  direct  path  from  anxious attachment  to physical  violence  was  no longer  significant. Multivariate delta method results indicated that borderline personality was a significant mediator of the relationship between anxious attachment and physical violence (z = 2.32, p = .01) and the relationship between anxious attachment and psychological violence (z = 3.39, p = .01). In addition, antisocial personality  was a significant mediator of the relationship   between anxious attachment and physical violence (z = 2.13, p = .01). Antisocial and borderline personality disorders together served as a full mediator of the relationship between anxious attachment and physical violence but a partial mediator  of the relationship  between  anx­ ious attachment and psychological violence. The model accounted for  16% of the variance for physical  violence  and  39% of  variance  for psychological violence.




 This study’s findings indicated that personality disorders significantly mediated the rela­ tionships between adult attachment orientations and indices of physical and psychological violence. More specifically, our results support our hypotheses that antisocial personality dis­ order functions as a mechanism through which avoidant adult attachment is related to both physical and psychological violence and that borderline personality disorder functions as a mechanism through which anxious adult attachment is related to both physical and psycho­ logical violence. Although not hypothesized, borderline personality disorder in addition to antisocial personality disorder mediated relationships between adult avoidant attachment and violence, and antisocial personality disorder in addition to borderline person mediated the relationship between anxious attachment and physical violence. However, the relationship between anxious attachment and psychological violence was only partially mediated by per­ sonality disorders. After controlling for personality disorders, anxious attachment continued  to have a direct effect on psychological violence. Taken together, this study’s findings sug­ gest that, after controlling for the batterer’s personality disorder, avoidant  attachment  does  not have a direct effect on physical or psychological violence, whereas anxious attachment does have a direct effect on psychological  violence but not physical   violence.

Despite unique diagnostic criteria distinguishing antisocial and borderline personality disorder diagnoses, some research suggests that the two diagnoses may not be mutually exclusive. In fact, borderline and antisocial personality disorders represent  a common clus­  ter of personality disorders (e.g., cluster B) characterized by  behaviors  that  are impulsive and violate  social  norms  (American  Psychiatric  Association,  1994). Moreover,  measures of antisocial and borderline personality, including the PDQ-R,  often  have  overlapping  items. Illustratively, Gacono and colleagues (1992) found support for the presence of borderline symptomatology among a group of offenders meeting criteria for antisocial personality disorder and psychopathy. Overlap between borderline  and antisocial personal­ ity disorder symptoms may explain the high correlation between antisocial and borderline scores  in  this  study’s sample.

Batterer typology research has also prompted discussion regarding the distinctiveness of the antisocial and borderline batterer subgroups. For example, Holtzworth-Munroe, Meehan, Herron, Rehman, and Stuart (2003) found that antisocial and borderline batterers did not always differ significantly on measures of antisociality, and Tweed and Dutton’s (1998) “impulsive” batterer also produced high mean scores on both antisocial and bor­ derline scales. Waltz et al. (2000) identified a “pathological” batterer subtype that scored highest on measures of both antisocial and borderline personality disorders and also found a strong correlation ( r = .64) between antisocial and borderline measures among their com­ munity sample of self-reported batterers. Holtzworth-Munroe et al. (2003) also found that borderline batterers did not differ significantly from antisocial batterers on a composite measure of preoccupied and fearful attachment, suggesting that preoccupied and fearful attachment may be characteristic of both antisocial and borderline batterers.

Given relationships between antisocial and borderline personality disorders, relation­ ships between the two personality disorders’ corresponding attachment orientations are understandable. Attachment literature provides additional insight regarding the relationship between avoidant and anxious adult attachment orientations. Specifically, Bartholomew and Horowitz ( 1991) tested a fourfold attachment typology that distinguished between two types of avoidant attachment. The first type, dismissive, is characterized by a negative internal working model of others but a positive internal working model of self. The second type, fearful attachment, is characterized by a negative internal working  model of both self and other and combines abandonment anxiety with avoidant behavior (Brennan et al., 1998). This study’s unexpected finding that avoidant attachment functioned as a stronger predictor of borderline than antisocial personality disorder may be explained by the pres­ ence of batterers with fearful attachment in the study’s sample. Support for a batterer char­ acterized by fearful attachment has been provided by Dutton and colleagues’ ( 1994) work suggesting that early trauma can lead to fearful attachment and, in conjunction with real or perceived unmet attachment needs, violence against his partner. Research investigating relationships between antisocial symptomatology and Dutton’s fearful batterer is needed before assuming that the fearful batterer who may combine avoidant attachment with abandonment anxiety also evidences both antisocial and borderline characteristics.

The results of this study contribute to the dialogue on batterer typologies. Extant typol­ ogy research associates anxious attachment and borderline personality disorder with a specific subgroup of batterers while associating avoidant attachment and antisocial per­ sonality disorder with another subgroup of batterers. A third subgroup without personality disorder symptomatology that exhibits less severe violence than either the anxious border­ line or the avoidant antisocial batterer has also been implicated in typology literature. This study provides information regarding the dynamic by which adult attachment orientations and personality disorders operate conjointly to influence the IPV enacted by batterer sub­ types. Many of this study’s findings are consistent with typology research. For example, our results indicate that, for at least for some batterers, psychological violence may be directly related to one’s anxious attachment and not a function of personality pathology. Next, this study’s results support relationships between anxious adult attachment and borderline personality disorder as well as relationships between adult avoidant attachment and antisocial personality disorder. However, in contrast to typology research, anxious attachment also predicted antisocial personality, and avoidant attachment also predicted borderline personality disorder. Mediational analyses suggest that (a) both borderline and antisocial personality disorders are mechanisms influencing avoidantly attached batterers’ physical and psychological violence and (b) both borderline and antisocial personality disorders are mechanisms influencing anxiously attached batterers’ physical violence. However, whereas borderline personality disorder also functioned as a mediating influence of the anxiously attached batterers’ psychological violence, antisocial personality disorder did not.

An emergent clinical implication of this study’s findings is that batterer intervention pro­ grams (BIPs) serving batterers presenting with antisocial and/or borderline personality dis­ orders may need to address the batterer’s personality disorders to effectively end his violence against women. Research suggesting that tailoring interventions to batterer characteristics resulted in improved outcomes (Saunders, 1996) as well as findings indicating that men scoring in the clinical range on borderline or antisocial scales were less likely to benefit from treatment compared to men scoring in the normative range (Dutton, Bodnarchuk,  Kropp,  Hart, & Ogloff,  1997) are consistent  with the assertion that addressing personality disorders in batterer intervention is important. Moreover, practitioners (Murphy & Dienemann, 1999) have expressed concern that batterers with personality disorders impede the progress of less pathological batterers, suggesting that BIPs combining nonpathological batterers with bat­ terers evidencing personality disorders may be counterproductive.

If BIPs are extended to address batterers’ pathology, drawing from the psychotherapy literature on treating borderline and antisocial personality  disorders  should prove use­ ful. For example, Dialectical Behavior Therapy (DBT) is being advanced by Linehan (1993) as a model for treating borderline patients, and empirical findings suggest that it  is effective in eliminating borderline behaviors (Bohus et al., 2004). In fact, a model of intervention extending DBT principles to batterer treatment has already been introduced (Fruzetti & Levensky, 2000), although it has yet to be implemented. Research on effective treatments with persons diagnosed with antisocial personality disorder is less promising. However, in a review of attempted treatments with antisocial persons, Reid and Gacono (2000) indicated that “strict social management, with consistent, predictable, even rigid consequences for one’s [antisocial] acts is most likely to extinguish the behavior” (p. 657). Reid and Gacono’s recommendation emphasizes the importance of following  through with sanctions and consequences imposed on batterers by the criminal justice system. Future research should examine whether incorporating tenets of interventions demon­ strated as effective with borderline personality persons in general clinical samples results in more effective treatment of batterers presenting with borderline personality disorder. Additionally, mediational models that test whether criminal sanctions are an especially effective mechanism to explain decreases in IPV among antisocial batterers could be an important line of inquiry.

Although this study’s findings suggest that interventions for batterers presenting with antisocial and/or borderline personality disorders may require addressing the batterer’s per­ sonality disorders to effectively end his violence against women, it is important to note that personality disorders are stable characteristics with a poor prognosis (American Psychiatric Association, 1994) even with many years of intensive psychotherapy (Gunderson, 1984, 1996). As such, BIPs targeting personality disorder pathology should be lengthy and intensive. Many current interventions are based on a 12-week model with once-weekly sessions. For batterers presenting with personality disorders, year-long programs with twice weekly sessions may be more appropriate. Additionally, there is no evidence that treating the personality disorder alone is likely to result in the cessation of violence, and many other variables are important etiological contributors to IPV. As such, until evidence suggests otherwise, interventions targeting batterer pathology should supplement rather than substitute extant violence-cessation curriculum components already in use. However, some research has already demonstrated that tailoring interventions to batterer character­ istics has positive outcomes (Chang & Saunders, 2002; Saunders, 1996). Future research should continue to explore the effectiveness of BIPs that specifically address batterer pathology and examine if separate interventions for batterers with and without personality disorders will result in improved outcomes for each of these groups.

Identification of batterers presenting with antisocial and/or borderline personality disor­ ders could be accomplished via assessments that inform intervention providers regarding a batterer’s underlying pathology. The purpose of this assessment should be twofold. First, knowledge of this underlying pathology can assist practitioners in making appropriate referrals for batterers to increase the likelihood of violence cessation. However, a second and more important purpose would be to provide practitioners and victims with insight regarding the batterer’s potential threat to the victim, given that research has demonstrated that severity of violence among batterers is positively associated with pathology (Edwards, Scott, Yarvis, Paizis, & Panizzon, 2003; Lawson et al., 2003).

This study’s findings that, above and beyond indirect effects, anxious attachment has a direct effect on psychological violence suggest that targeting the attachment dysfunction of an anxiously attached batterer who does not evidence borderline pathology and enacts only psychological violence may have merit. Some researchers have already initiated the development of treatment models premised on the tenets of attachment theory (Sonkin & Dutton, 2003; Stosny, 1995). Inherent in these treatment models is the notion that once bat­ terers’ dysfunctional attachment  dynamics  desist, battering  will also desist. Specifically, if the anxiously attached batterer who uses psychological violence to control his partner’s proximity learns that his hostile and violent tactics to prevent abandonment may  actu­ ally cause his partner to seek greater emotional distance from him, he may be motivated toward compliance with an intervention teaching nonviolent communication and relation­ ship skills that allow for heightened emotional intimacy and that consequently minimize abandonment fears. However, this study’s findings that physical violence is fully mediated by personality disorders suggest that targeting attachment dysfunction is not appropriate as an intervention for physical violence.

Some limitations should be considered when evaluating results of this study. First, recent etiological models of IPV have been multidimensional and have included both  individual­  and system-level predictors of IPV (O’Neil & Harway, 1997). The models in this study included only individual-level predictors representing psychological theories of violence. Thus, researchers were unable to test hypotheses regarding moderating and/or mediating effects of variables representative of other theoretical perspectives (i.e., sociopolitical) on relationships between individual-level  predictors  and physical  and psychological  violence. A second limitation associated with the study’s design is the cross-sectional nature  of the  data. Although our findings support expected relationships between attachment and per­ sonality disorders, use of cross-sectional data prohibits conclusions regarding the temporal precedence of  attachment. Additionally,  use of cross-sectional  data does not allow for tests  of cause-and-effect  relationships  between  attachment,  personality  disorders, and  violence.

This study also has some sampling and measurement limitations. Participants included in this study were involved with  the criminal justice  system and on probation for a domestic assault misdemeanor. As  such, this  study’s findings may  not  generalize to batterers who enact minor violence and exist under the radar of the criminal justice system. This study’s findings are also not generalizable to batterers who have  been arrested for domestic assault felonies and are in jail. Because  pathology  is  associated with severity of IPV (Edwards et al., 2003; Lawson et al., 2003), antisocial and bor­ derline personality disorders are probably most prevalent among incarcerated batterers. Because men in this sample committed only misdemeanor violence, PDQ-R borderline and antisocial scores for men in this sample may not be representative of borderline and antisocial personality traits for all batterers. Furthermore, the modest internal  consis­ tency reliabilities of PDQ-R scores observed within our sample may be a contributing factor to the high correlation between antisocial and borderline scores and would sup­  port the use of alternative measures of personality disorders in future studies. Relatedly, this study was limited by monomeasure and monomethod bias because measures in this study were exclusively self-report  and data were collected  from only one source (Cook & Campbell, 1979). This may be especially problematic for self-reports  of psychologi­  cal and physical violence, given batterers’ tendencies to minimize self-report of acts of violence  (Arias  & Beach, 1987).



Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological   study  of  the strange  situation.  Hillsdale,  NJ: Erlbaum.

American  Psychiatric  Association.  (1987).  Diagnostic  and  statistical   manual  of  mental    disorders

(3rd ed.). Washington, DC: Author.

American   Psychiatric  Association.   (1994).  Diagnostic  and  statistical   manual  of  mental  disorders

(4th ed.). Washington, DC: Author.

Arias, I., & Beach, S. R. H. (1987). Validity of self-reports of marital violence. Journal of Family Violence, 2, 139-149.

Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61,  226-224.

Bentler, P. M. (1995). EQS program manual. Los Angeles: Multivariate Software Inc.

Bohus, M., Haaf, B., Simms, T., Limberger, M. F., Schmahl, C., Unckel, C., et al. (2004) Effectiveness of inpatient dialectical behavioral therapy for borderline personality disorder: A controlled trial. Behavior, Research, and Therapy, 42, 487-499.

Bovasso, G. B., Alterman, A. I., Cacciola, J. S., & Rutherford, M. J. (2002). The prediction of violent and nonviolent criminal behavior in a methadone maintenance population. Journal of Personality Disorders, 16, 360-373.

Bowlby, J. (1973). Attachment and loss. Vol. II: Separation. New York: Basic Books. Bowlby, J. (I980). Attachment and loss. Vol. III: Loss. New York: Basic Books.

Bowlby, J. ( 1982). Attachment and loss. Vol. /: Attachment. New York: Basic Books.

Bowlby, J. (1984). Violence  in the family as a disorder of the attachment and caregiving    systems.

American Journal of Psychoanalysis, 44,  9-27.

Bowlby, J. (1988). A secure base: Parent-child attachments and healthy human development. New York: Basic Books.

Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measurement of adult attachment: An integrative overview. In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory and close relationships (pp. 46-76). New York: Guilford.

Brennan, K. A., & Shaver, P. R. (1995). Dimensions of adult attachment, affect regulation, and romantic relationship functioning. Personality and Social Psychology Bulletin, 21,  267-283.

Brennan, K. A., & Shaver, P. R. (1998). Attachment styles and personality disorders: Their connec­ tions to each other and to parental divorce, parental death, and perceptions of parental caregiv­ ing. Journal of Personality, 66, 835-878.

Bureau  of  Justice  Statistics.  (1995).  Violence against  women:  Estimates from   a  redesigned  survey.

Washington,  DC:  U.S.  Department  of Justice.

Chang, H., & Saunders, D. G. (2002). Predictors of attrition in two types of group programs for men who batter. Journal of Family Violence, 17, 273-292.

Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design and analysis issuesfor field settings. Chicago: McNally College Publishing.

Crino, M. D., Svoboda, M., Rubenfeld, S., & White, M. C. (1983). Data on the Marlowe Crowne and Edwards Social Desirability  Scales. Psychological  Reports, 53, 963-968.

Crowne, D. P., & Marlowe, D. (1964). The approval motive. New York: Wiley.

Dutton,  D. G. (1995). Intimate  abusiveness.  Clinical Psychology:  Science  and  Practice,  2,  207-224.

Dutton, D. G., Bodnarchuk, M., Kropp, R., Hart, S. D., & Ogloff, J. P. (1997). Client personality disorders affecting wife assault post-treatment recidivism.  Violence and Victims, 12, 37-50.

Dutton, D. G., & Browning, J. J. (1988).  Power  struggles  and  intimacy  anxieties  as causative factors of wife assault. In G. W. Russell (Ed.), Violence in intimate relationships  (pp.  163-176). New York:  PMA Publishing.

Dutton, D. G., Saunders, K., Starzomski, A., & Bartholomew, K. (1994). Intimacy-anger and inse­ cure attachment as precursors of abuse in intimate relationships. Journal of Applied Social Psychology,  24, 1367-1386.

Dutton, D. G., Starzomski, A. J., & Ryan, L. (1996). Antecedents of abusive personality and abusive behavior in wife assaulters. Journal  of Family  Violence, 11, 113-132.

Edwards, D. W., Scott, C. L., Yarvis, R. M., Paizis, C. L., & Panizzon, M. S. (2003). Impulsiveness,

impulsive   aggression,  personality   disorder,  and  spousal  violence.   Violence  and   Victims, 18, 3-14.

Flournoy, P. S., & Wilson, G. L. (1991). Assessment of MMPI profiles of .male batterers. Violence and  Victims, 6, 309-320.

Fonagy, P. ( 1999). Male perpetrators of violence against women: An attachment theory perspective. Journal of Applied  Psychoanalytic  Studies, 1, 7-27.

Fonagy, P., Leigh, T., Steele, M., Steele, H., Kennedy,  R., Mattoon,  G., et al. (1996). The relation  of attachment status, psychiatric classification, and response to psychotherapy. Journal of Consulting and  Clinical Psychology,  64, 22-31.

Fonagy,  P., Target,  M.,  Gergely,  G., Allen,  J.  G., & Bateman, A. W. (2003). The  developmental

roots of borderline personality disorder in early attachment relationships: A theory and some evidence. Psychoanalytic  Inquiry, 23, 412-459.

Fossati, A., Maffei, C., Bagnato, M., Donati, D., Donini, M., Fiorilli, M., et al. (1998). Brief communication: Criterion validity of the personality diagnostic questionnaire in a mixed psy­ chiatric sample. Journal of Personality Disorders,  12,  172-178.

Fraley, R. C., & Waller, N. G. (1998). Adult attachment patterns: A test of the typological model.  In

  1. A. Simpson & W. S. Rholes (Eds.), Attachment theory and close relationships (pp. 77-114). New York: Guilford.

Fruzetti, A. E., & Levensky, E. R. (2000). Dialectical behavior therapy for domestic violence: Rationale and procedures. Cognitive and Behavioral  Practice,  7,  435-447.

Gacano, C. B., & Meloy, J. R. (1991). A Rorschach investigation of attachment and anxiety in anti­ social personality disorder. Journal  of Nervous and Mental Disease,  179, 546-552.

Gacano, C. B., & Meloy, J.  R. (1992). The Rorschach  and  the DSM-III-R  antisocial  personality: A tribute to Robert Lidner. Journal  of Clinical Psychology, 48,  393-406.

Gacano, C. B., Meloy, J. R., & Berg, J. L. (1992). Object relations, defensive operations, and affective states in narcissistic, borderline, and antisocial personality disorder. Journal of Personality Assessment,  59, 32-49.

Gunderson, J. G. (1984). Borderline personality disorder. Washington, DC: American Psychiatric Press.

Gunderson, J. G. (1996). The borderline patient’s intolerance of aloneness: Insecure attachments and therapist availability. American Journal of Psychiatry,  153, 752-758.

Hale, G., Zimostrad, S., Duckworth, J., & Nicholas, D. (1988). Abusive partners: MMPI profiles of male batterers. Journal  of Mental Health  Counseling, 10, 214-224.

Hamberger, L. K., & Hastings, J. (1988). Characteristics of male spouse abusers consistent with personality  disorders. Hospital  and Community Psychiatry, 39, 763-770.

Hamberger, L. K., Lohr, J. M., Bonge, D., & Tolin, D. F. (1996). A large sample empirical typology of male spouse abusers and its relationship  to dimensions  of  abuse.  Violence and  Victims, 11, 277-292.

Hastings, J. E., & Hamberger, L. K. (1988). Personality characteristics of spouse abusers: A controlled comparison. Violence and Victims, 3, 31-48.

Hazan, C., & Shaver, P. ( 1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52,  511-524.

Holtzworth-Munroe, A., Bates, L., Smutzler, N., & Sandin, E. (1997). A brief review of the research on husband violence. Part I: Maritally violent versus nonviolent men. Aggression and Violent Behavior,  2, 65-99.

Holtzworth-Munroe, A., Meehan, J. C., Herron, K., Rehman, U.,  & Stuart, G. L.  (2000). Testing  the Holtzworth-Munroe and Stuart batterer typology. Journal of Consulting and Clinical Psychology,  68, 1000-1019.

Holtzworth-Munroe, A., Meehan, J. C., Herron, K., Rehman, U., & Stuart, G. L. (2003). Do subtypes of maritally violent men continue to differ over time? Journal of Consulting and Clinical Psychology,   71, 728-740.

Hyler, S. E., Rieder,  R. 0., Williams,  J. B. W.,  Spitzer, R. L., Hendler, J., & Lyons, M. (1988).  The Personality Diagnostic Questionnaire: Development and preliminary results. Journal of Personality  Disorders,  2, 229-237.

Hyler, S. E., Skodol, A. E., Oldham,  J. M.,  Kellman,  H. D.,  & Doidge,  N.  (1992). Validity  of the Personality Diagnostic Questionnaire-Revised: A replication in an outpatient sample. Comprehensive  Psychiatry,  33,  73-77.

Hyler, S. E., Skodol, A. E., Oldham, J. M., Kellman, H. D., Oldham, J. M., & Rasnick, L. (1990). Validity of the Personality Diagnostic Questionnaire-Revised: Comparison with two structured interviews. American  Journal  of  Psychiatry,  147, 1043-1048.

Johns, J. L. (1997). Basic Reading Inventory: Pre-primer through grade twelve and early literacy assessments. Dubuque, IA: Kendall/Hunt.

Johnson, J. G., & Bornstein, R. F. (1992). Utility of the Personality Diagnostic Questionnaire­ Revised  in  a non-clinical  sample. Journal  of Personality  Disorders,  6(4), 4557.

Judd, C. M., & Kenny, D. A. (1981). Process analysis: Estimating mediation in treatment evaluations. Evaluation Review, 5, 602-619.

Langhinrichsen-Rohling, J. (2005). Top 10 greatest “hits”: Important findings and future directions for intimate partner  violence research. Journal  of Interpersonal  Violence, 20,   108-118.

Lawson, D. M., Weber, D., Beckner, H. M., Robinson, L., Marsh, N., & Cool, A. (2003). Men who use violence: Intimate violence versus non-intimate violence profiles.  Violence and  Victims, 18, 259-277.

Linehan, M. M. (1993). Cognitive-behavioral treatment of borderline personality disorder. New York: Guilford.

Livesley, W. J., Schroeder, M. L., & Jackson, N. J. (1990). Dependent personality disorder and attachment  problems. Journal  of Personality  Disorders,  4, 131-140.

Loo, R., & Thorpe, K. (2000). Confirmatory factor analyses of the full and short versions of the Marlowe-Crowne  Social Desirability  Scale. Journal  of Social Psychology,  140, 628-635.

Lopez, F. G., & Brennan, K. (2000). Dynamic processes underlying adult attachment organization: Toward an attachment theoretical perspective on the healthy and effective self. Journal of Counseling  Psychology,  47, 283-300.

MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S., & Sheets. V. (2002). A comparison of methods to test mediation  and other intervening  variable  effects. Psychological  Methods,  7,  83-104.

Mayseless,  0.  (1991).  Adult  attachment  patterns  and  courtship  violence.  Family  Relations,  40, 21-28.

Mehrabian, A. (1997). Relations among personality scales of aggression, violence, and empathy: Validational evidence bearing on the risk of eruptive violence scale. Aggressive Behavior, 23, 433-445.

Murphy, C. M., & Dienemann, J. A. (1999). Informing the research agenda on domestic abuser intervention through practitioner-researcher dialogues. Journal of Interpersonal Violence, 14, 1314-1326.

Murphy, C. M., Meyer, S. L., & O’Leary, K. D. (1993). Family of origin violence and MCMI-II psychopathology  among partner  assaultive  men.  Violence and  Victims, 8, 165-176.

National Institute of Justice. ( 1998). Batterer intervention: Program approaches and criminal justice strategies. Washington, DC: U.S. Department of Justice.

National Institute of Justice. (2000). Prevalence, incidence, and consequences of violence against women. Washington, DC: U.S. Department of Justice.

O’Neil, J. M., & Harway, M. (1997). A multivariate model explaining men’s violence toward women.  Violence Against  Women, 3,  182-203.

Patrick, M., Hobson, R. P., Castle, D., Howard, R., & Maughan, B.  (1994). Personality  disorder  and the mental representation of early social experience. Development  and Psychopathology,  6, 375-388.

Reid, W. H., & Gacano, C. (2000). Treatment of antisocial personality, psychopathy, and other characterologic  antisocial syndromes. Behavioral Sciences and Law, 18, 647-662.

Rosenstein, D. S., & Horowitz, H. A. (1996). Adolescent attachment and psychopathology. Journal of Consulting and Clinical Psychology, 2,  244-253.

Sable, P. (1997). Attachment, detachment, and  borderline  personality  disorder.  Psychotherapy, 34,  171-181.

Sack, A., Sperling, M. B., Fagen, G., & Foelsch, P. (1996). Attachment style, history, and behav­ ioral contrasts for a borderline and normal sample. Journal of Personality Disorders, 10, 88-102.

Saunders, D. G.  (1996).  Feminist-cognitive-behavioral  and  process-psychodynamic  treatments for men who batter: Interaction of abuser traits and treatment models. Violence and  Victims,  11, 393-413.

Schumacher, J. A., Feldbau-Kohn, S., Smith Slep, A. M., & Heyman, R. E. (2001). Risk factors for male-to-female partner physical abuse. Aggression and Violent Behavior, 6,  281-352.

Schumacher, J. A., Smith Slep, A. M., & Heyman, R. E. (2001). Risk factors for male-to-female partner psychological  abuse. Aggression and Violent Behavior, 6, 255-268.

Sheldon, A. E., & West, A. M. (1990). Attachment pathology and low social skills in avoidant personality disorder: An exploratory study. Canadian Journal  of Psychiatry, 35,  596-599.

Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects in structural equation models. In S. Leinhardt (Ed.), Sociological methodology 1982 (pp. 290-312). Washington, DC: American Sociological Association.

Sobel, M. E. ( 1986). Some new results on indirect effects and their standard errors in covariance structure models. In N. Tuma (Ed.), Sociological methodology 1986 (pp. 159-186). Washington, DC: American Sociological Association.

Sonkin, D., & Dutton, D. (2003). Treating assaultive men from an attachment perspective. Journal  of Aggression, Maltreatment, and Trauma, 7,   105-133.

Stosny, S., 1995. Treating attachment abuse: A compassion approach. New York: Springer.

Straus, M. A.  (1979).  Measuring  intrafamily  conflict  and  violence:  The  Conflict  Tactics Scales.

Journal  of Marriage and the Family, 41, 75-88.

Straus, M. A. (1990). Measuring  intrafamily conflict and violence: The Conflict Tactics    Scales. In

  1. A. Straus & R. J. Gelles (Eds.), Physical violence in American families (pp. 29-48). New Brunswick, NJ: Transaction.

Straus, M. A.,  & Gelles, R. J. (1986).  Societal change and change in family  violence  from  1975   to 1985 as revealed by two national surveys. Journal of Marriage and the  Family,  48,  465-479.

Tolman, R. M. ( 1989). The development of a measure of psychological maltreatment of women by their male partners.  Violence and Victims, 4, 159-177.

Tolman,   R.  M.  (1999).  The  validation   of   the  psychological   maltreatment   of   women  inventory.

Violence and Victims, 14, 25-37.

Tweed, R. G., & Dutton, D. (1998). A comparison of impulsive and instrumental subgroups of batterers.  Violence and Victims, 13, 217-230.

Waltz, J., Babcock, J. C., Jacobson, N. S., & Gottman, J. M. (2000). Testing a typology of  batterers.

Journal  of  Consulting  and  Clinical  Psychology,  68, 658-669.

West, M., Keller, A., Links, P., & Patrick, J. (1993). Borderline disorder and attachment   pathology.

Canadian  Journal  of  Psychiatry,  38(Suppl.  1),  16-21.

Wierzbicki, M., & Gorman, J. L.  (1995).  Correspondence  between  students’  scores  on  the  Millon Clinical Multiaxial Inventory-II and Personality Diagnostic Questionnaire-Revised. Psychological  Reports,  77, 1079-1082.

Woike, B. A., Osier, T. J., & Candela, K, (1996). Attachment styles and violent imagery in thematic stories about relationships. Personality and Social Psychology Bulletin, 10,   1030-1034.

Correspondence regarding this article should be directed to Anne Marie Mauricio, PhD, Arizona Prevention Research Center, Arizona State University, P.O. Box 876005, Tempe, AZ 85287-6005. E-mail:

Comments are closed.
%d bloggers like this: