This study, which used data on targeted court cases from the third quarter of 1998, was conducted after New York City had adopted policies and practices for DV cases intended to increase the likelihood of arrest, prosecution, and conviction in DV cases. Data were obtained on the prosecution, conviction, and sentencing of defendants in New York City criminal courts for DV and non-DV cases. The data analysis sought to determine how case outcomes for DV cases differed from those of comparable non-DV cases; how defendants and case characteristics in DV cases differed from those in comparable non-DV cases; and whether differences in case outcomes between DV and non-DV cases was due to differences in defendant and case characteristics. As expected, the study found significant differences in case outcomes for DV and non-DV cases. DV cases were less likely to result in conviction, and convicted defendants in DV cases were less likely than those in non-DV cases to be sentenced to jail. Once sentenced to jail, however, defendants in DV cases received longer sentences, on average, than defendants in non-DV cases. Defendants in DV cases tended to be older than those in non-DV cases, and they were slightly more likely to be male. Defendants in DV cases generally had less serious criminal records and stronger community ties. Overall, the findings indicate that defendants in DV and non-DV cases had different backgrounds, and their cases were processed differently in the criminal courts of New York City. The study concludes that the conviction rate in DV cases compared with non-DV cases indicates that weak evidence continues to be a major problem in DV cases even after efforts to address it. Weak evidence in DV cases typically results from lack of victim cooperation in the prosecution of the case. When prosecutors prosecute cases without the cooperation of the victim, the likelihood of an acquittal increases. Recommendations for addressing problems of weak evidence in DV cases include decreasing prosecutors' caseloads to allow more time and resources for counseling victims and gaining their cooperation and reconsidering whether it is cost-effective to prosecute DV cases without the cooperation of the victim. Suggestions are offered for future research. 19 tables, 51 references, and appended coding of variables for regression models