The Factor Structure of Intimate Partner Violence Risk

Abstract

Intimate partner violence (IPV) has elicited a great deal of attention from the research, clinical practice, and policy communities because of its widespread prevalence and harmful impact on the individuals and society. Effective risk assessment can help identify individuals who are in the highest risk category and are thus in most need of management strategies. Therefore, a better understanding of risk factors that underlie IPV risk is necessary to optimize available risk assessment measures and effectively manage risk. The purpose of this dissertation was to explore the factor structure of combined items from the Ontario Domestic Assault Risk Assessment, Spousal Assault Risk Assessment, and Brief Spousal Assault Form for the Evaluation of Risk and examine whether the underlying risk factors predict recidivism outcomes. Data were collected for 300 adult men who were charged or accused of violence against their past or current female intimate partners and whose files were referred for a comprehensive threat assessment between 2010 and 2016. Using exploratory factor analysis, I found that items from these measures assessed six underlying risk factors. Although two factors significantly predicted IPV, any violent (violence against anyone), and general (violent and non-violent) recidivism outcomes, only one factor containing mostly Central Eight risk factors (for general violent and criminal recidivism) independently predicted violent and IPV recidivism over time above and beyond other factors, suggesting that not all risk factors included in these measures are independently predictive of recidivism, particularly IPV recidivism. Additionally, I found that randomized selections of items predicted as well as the original measures, indicating that these risk measures are potentially missing pertinent risk-relevant information. Overall, findings from this dissertation suggest that IPV risk assessment measures are far from ideal, and researchers have suggested that the development of informative causal theories of IPV recidivism can help motivate risk factors for consideration in risk assessment measures, as current IPV theories fail to produce predictive models. In addition to this, more empirical work is also needed to identify and validate risk factors strongly related to IPV recidivism that can assist in risk assessment and ultimately risk management.