The Judicial Oversight Demonstration (JOD) was designed to test the feasibility and impact of a coordinated response to intimate partner violence (IPV) that involved the courts and justice agencies in a central role. The primary goals were to protect victim safety, hold offenders accountable, and reduce repeat offending. The two primary evaluation objectives were: (1) to test the impact of JOD interventions on victim safety, offender accountability, and recidivism, and (2) to learn from the experiences of well-qualified sites who were given resources and challenged to build a collaboration between the courts and community agencies to respond to intimate partner violence. Dorchester, Massachusetts, and Washtenaw County, Michigan, participated in a quasi-experimental evaluation of the impact of the program. IPV cases reaching disposition during the JOD were compared to similar cases reaching disposition in Lowell, Massachusetts, and Ingham County, Michigan. All IPV cases reaching disposition from approximately January 2003 to November 2004 (see Study Time Periods and Time Frames) were reviewed and included in the sample if appropriate. To be eligible for the sample, cases had to involve: (1) criminal IPV charges; (2) victims and offenders age 18 or older; and (3) victims and offenders who lived in the target jurisdiction at the time of case disposition. Cases that reached disposition more than a year after the incident were excluded to limit loss of data due to poor recall of the facts of the incident and police response. The evaluation design of JOD in Milwaukee differed from that of the other two sites. The evaluation in Milwaukee was based on a quasi-experimental comparison of offenders convicted of IPV and ordered to probation during JOD (January 1, 2001, to May 21, 2002) and before JOD (October 8, 1997, to December 21, 1999). This design was selected when early plans for an experimental design had to be abandoned and no comparable contemporaneous comparison group could be identified. Data for this evaluation were collected from court and prosecutors' records of case and defendant characteristics, probation files on offender supervision practices, and official records of rearrest, but do not include interviews with victims or offenders. This data collection has 20 data files containing 3,578 cases and 4,092 variables. The data files contain information related to each site's Batterer Intervention Programs (Parts 1, 8, and 15), court data (Parts 2, 12, 13, 14, 16, and 18), law enforcement (Parts 3, 11, and 17), and victim data (Parts 4, 5, 6, 9, 10, and 19). The Dorchester, Massachusetts, and Washtenaw County, Michigan, Impact Evaluation Data (Part 7) include baseline and follow-up information for the offender and the victim. The data file also contains Probation Supervision Performance Reports, Victim Services Logs, and Case Incident Fact Sheet information. The Milwaukee, Wisconsin, Impact Evaluation Data (Part 20) include information related to the offender and the victim such as age, race, and sex, as well as arrest records including charges filed.