The question of how to reduce the recidivism rates among IPV offenders is one that plagues criminologists to this day. Though a difficult issue to address, educational treatment programs have started to gain popularity as one idea to achieve this reduction. By examining the dataset from the “Domestic Violence Experiment in King's County (Brooklyn), New York, 1995-1997,” conducted by Robert C. Davis et al. (2000), it was found that the results of the educational program showed a great promise in reducing recidivism rates. Though it is important to focus on and analyze the results from this study, it is also important to extrapolate from them by running and examining specific models and variables with the dataset. Focusing on specific variables within the dataset allows researchers to find different themes and results in smaller ideologies of research, versus trying to find one overall answer on how to reduce recidivism.

By examining specific variables such as length of relationship, I wonder how length of relationship between an IPV offender and victim impact recidivism rates? This thesis will discuss IPV history and theoretical perspectives, history of educational programs, length of relationship, and the dataset conducted by Davis et al. (2000).

This thesis examines how the likelihood of IPV recidivism is effected by length of relationship, the different length of treatment programs (overall, eight-week, or twenty-six-weeks), and the interaction between length of relationship and the different treatment programs. The results show overall that length of relationship slightly decreases the rate of recidivism for IPV. When length of relationship is ran in the models with the separate treatment programs, it is found that the overall treatment and twenty-six-week programs have drastic and significant reduction results on recidivism, but that the eight-week program actually increases recidivism rates slightly. The results also indicate that when examining the interaction between length of relationship and the different treatment programs, length of relationship slightly moderates the reduction of the recidivism rates for the individuals enrolled in the overall treatment and eight-week programs, but slightly increases the rates for those in the twenty-six-week program.

Mentally ill offenders continue to contribute to mass-incarceration within the United States. The cost, both social and economic, of housing a large number of mentally ill inmates in our prison system has reached a breaking point. The need for empirically founded correctional research, with an emphasis on individuals who suffer from a mental illness, is crucial to reducing the number of incarcerated individuals in the United States. The current study analyzes whether mentally ill inmates reported statistically significant differences in levels of perceived reentry social support, when compared to their non-mentally ill counterparts. The current study utilized data from the APVP. The APVP contained a sample of 231 individuals, 121 female and 110 male, from two Arizona Department of Corrections facilities. The majority of respondents were white (44.58%), medium security (40.26%), non-married (77.49%), and had a mean age of 36.04 years (SD=11.74). The current study conducted both bivariate and multivariate analyses to determine whether mentally ill inmates perceived differences in the reentry social support available to them as compared to non-mentally ill inmates. Further multivariate analyses were conducted to determine whether there were any significant differences the key independent variable and the dependent variables across gender. Mentally ill female inmates reported significantly lower rates of perceived reentry social support in a number of emotional support factors. The findings of this study are a crucial first step for future empirical research on inmate perceptions of social support—perceptions that may directly affect successful reentry.