On categorizing intimate partner violence: A systematic review of exploratory clustering and classification studies.

Many theorists have proposed that intimate partner violence (IPV) is not one homogeneous phenomenon but instead has several distinct types. For example, Johnson (1995) typology described some perpetrators' violence as stemming from a desire to control and others' violence stemming from emotional dysregulation, whereas Holtzworth-Munroe and Stuart's (1994) typology classified perpetrators by severity of violence, whether violence was specific to intimate partners and perpetrators' psychopathological profiles. Other typologies are based on personality profiles, severity levels, and variety of violent acts. We conducted a systematic review of studies that tested these hypothesized IPV typologies, using exploratory clustering and classification methods to identify underlying groups. We used the databases such as PsycINFO, PsycARTICLES, MEDLINE, Social Sciences Full Text (H. W. Wilson), and Social Work Abstracts. We located 80 such studies that empirically tested IPV typologies. After reviewing the 34 studies that met our a priori inclusion criteria, we found the following: (a) the modal number of types identified was three, but there was substantial variance across studies and (b) although Holtzworth-Munroe and Johnson's models had mixed support, the inconsistency across studies calls into question the validity of existing typologies and the certainty with which typologies are described by researchers and practitioners. Therefore, we recommend caution in using a categorical approach to IPV. (PsycInfo Database Record (c) 2023 APA, all rights reserved).