A Latent Classification of Male Batterers

Article excerpt

Regression latent class analysis was used to identify batterer subgroups with distinct violence patterns and to examine associations between class membership and adult attachment orientations as well as antisocial and borderline personality disorders. Results supported three batterer subgroups, with classes varying on frequency and severity of violence. The high-level violence class represented 40% of batterers, and both anxious and avoidant adult attachment orientations as well as borderline personality characteristics predicted membership in this class. The moderate-level violence class represented 35% of the batterers, and adult anxious attachment orientation was associated with membership in this class. The low-level violence class represented 25% of the sample and reported significantly less violence than other classes. Neither adult attachment orientations nor personality disorders predicted membership in this class.

Keywords: domestic violence; batterer typology; attachment; borderline personality disorder; antisocial personality disorder

Batterer typology research suggests that batterers are a heterogeneous group with diverse behavioral, cognitive, and emotional characteristics (Holtzworth-Munroe & Stuart, 1994). Although these studies show some variation in the number of identified subtypes due to differences in sampling and/or measurement, research generally supports two to four batterer subtypes each exhibiting distinct patterns of interpersonal violence (IPV) and each differing on important personality dimensions. Personality disorders related to the need to control others (e.g., narcissistic and antisocial) and related to self-concept and identity (e.g., borderline) as well as adult attachment orientations are two dimensions that have played an important role in differentiating batterer types (Gondolf, 1988; Gottman et al., 1995; Hamberger & Hastings, 1988; Hamberger, Lohr, Bonge, & Tolin, 1996; Holtzworth-Munroe, Meehan, Herron, Rehman, & Stuart, 2000; Tweed & Dutton, 1998; Waltz, Babcock, Jacobson, & Gottman, 2000).

ATTACHMENT THEORY: AN OVERVIEW

Originally understood as a framework for conceptualizing relational dynamics between a primary caregiver and child, attachment theory (Bowlby, 1973, 1980, 1982, 1988) was subsequently and productively extended to the study of the interpersonal dynamics in adult intimate relationships (Hazan & Shaver, 1987). According to theory, individuals develop an internalized set of beliefs about self and other (i.e., "internal working models") that regulate their relationships with intimate partners. Early adult attachment research identified four adult attachment prototypes or styles (i.e., secure, preoccupied, dismissive, and fearful) based on the relative valences of persons' self and other models (Bartholomew & Horowitz, 1991). The secure person has a positive sense of self-worth along with positive expectations that others will be accepting and trustworthy. The preoccupied person has a sense of unworthiness accompanied with a positive view of others. The fearful person experiences the self as unworthy and expects the other to be untrustworthy and not accepting. Thus, despite desire for connection with others, the fearful person avoids intimacy because of expectations of rejection. The dismissive person has a positive sense of self-worth yet expects that others are untrustworthy and unreliable.

Subsequent studies using multi-item measures of adult attachment security have confirmed that a two-dimensional structure adequately explains (Brennan, Clark, & Shaver, 1998; Brennan & Shaver, 1995; Lopez & Brennan, 2000) and more accurately characterizes (Fraley & Waller, 1998) variation in adult attachment organization. The first dimension (i.e., attachment avoidance) is represented by a pervasive discomfort with intimate closeness and a strong orientation toward excessively self-reliant and counterdependent relationship behavior (i. …