Using a Cox proportional hazards regression model, this dissertation explores three factors currently not addressed in the literature on men who batter women and who are court ordered to participate in a battering intervention program. These factors are the cumulative effects of civil and criminal legal sanctions (dose-response of sanctions) for domestic violence related offenses on recidivism, reduced opportunities to recidivate, and whether the number of legal sanctions imposed has an effect on how long a man maintains his non-recidivism status. Because one domestic violence case may involve multiple sanctions, this study uses the Legal Sanction Dose-Response Index to gauge the cumulative impact of civil and criminal sanctions upon recidivism of domestic violence. The Cox proportional hazards model indicates that the risk of recidivism is 45% lower for men who experienced two legal sanctions (typically arrest and probation) in response to the index case, relative to men who experienced one legal sanction (typically civil protective order). In other words, those with two legal sanctions are able to maintain their non-recidivism status longer relative to those with one sanction. Men with prior criminal court involvement for domestic violence related offenses are more likely to recidivate. Additionally, rather than incarceration reducing opportunities to recidivate, this study finds that incarceration for any offenses committed during the follow-up period is a predictor of recidivism of domestic violence related offenses. It is possible that, rather than incarceration reducing opportunities, recidivists are persistent and use whatever opportunities are available to them to commit domestic violence, despite legal sanctions.

The primary goal of this study was to evaluate the effectiveness of a program being run by the Bedford Police Department’s Repeat Victimization Unit on domestic violence and mental health and mental retardation. The study sought to determine whether the program was effective in reducing instances of repeat victimization in domestic violence and MHMR victims. Additionally the program investigated whether or not the program was effective at reducing victimization severity, and which demographic could be identified as the most victimized. Participants consisted of 157 domestic violence and MHMR victims in the city of Bedford, Tx between November 11, 2012 to July 30, 2013. Findings indicate that levels of repeat victimization for domestic violence and MHMR are relatively low regardless of whether the victim received services through the repeat victimization program or not. Additionally the severity of these repeat victimizations remains relatively constant regardless of whether services were received through the program or not. Implications and findings are discussed.