The role of batterers’ treatment in recidivism by domestic violence offenders

Graduation Date

2016

Document Type

Project

Program

Other

Program

Project (M.S.W.)--Humboldt State University, Social Work, 2016

Committee Chair Name

Jennifer Maguire

Committee Chair Affiliation

HSU Faculty or Staff

Keywords

Domestic violence, Recidivism, Humboldt State University -- Projects -- Social Work, Batterers' intervention program

Abstract

In this study I examined the role of a batterers' treatment program relative to other factors that influence the likelihood that a domestic violence offender will be arrested for further violence. Then I looked at factors that best predicted attrition from the program. I sampled 101 program participants from the men who did intakes in one batterers' program in Alaska. Using Random Forest, I modeled recidivism (re-arrest for violent offences after last contact with the program) and attrition (dropping out during the program) using 27 factors known to be correlated with domestic violence. Of the men who completed the batterers' intervention program, 20% went on to be arrested for a violent crime, compared to 41% of those who did not complete. The recidivism model was able to correctly predict re-offense 60% of the time. The top predictors in this model were the number of days since last contact with the program, whether the participant had committed any violent offenses during the program, and whether the participant had completed the program. The fact that program completion and attendance variables out-performed the rest of the 27 predictors found in the literature to relate to recidivism provides evidence in support of the effectiveness of the program. The attrition model classified program completion correctly 68% of the time. The top predictor that someone would drop out was race (particularly AK Natives), followed by whether the client was abused as a child, the number of prior offences on their record, and the charges for which they were ordered to the program. This model can be used to inform program efforts to increase retention by targeting interventions targeted to those at highest risk of dropping out and by improving cultural responsiveness.

https://scholarworks.calstate.edu/concern/theses/zk51vk45z

Share

 
COinS