The issues addressed in this research are the impact on case outcomes of prosecution policies for screening DV cases and the impact of the use of specialized court parts on case outcomes in DV cases. On the issue of prosecutorial screening policies, the study examined whether there were differences in DV case outcomes between the Bronx, which had a prosecutorial screening policy of proceeding with those cases in which the victim signed the complaint, and the Brooklyn and Manhattan boroughs, which routinely proceeded with cases that did not have the victim's cooperation. Regarding the second issue, the study considered whether there were differences in DV case outcomes between Brooklyn and the Bronx, where DV cases were processed in specialized DV court parts, and Manhattan, where DV cases were processed in court parts with mixed dockets. The data for the study were drawn primarily from the CJA database, which contains information on the arrest, case processing, and case outcomes of most New York City arrestees. The study used the dataset on DV cases from the third quarter of 1998, which encompassed data collected on a 3-month cohort of arrests made from July 1, 1998, through September 30, 1998. The study found that prosecutorial screening that relied on first-party complaints produced a relatively high conviction rate (64 percent) for DV cases in the Bronx. Conviction rates for DV cases in Brooklyn and Manhattan, which processed DV cases without victim cooperation, produced conviction rates that were significantly lower (18 percent and 29 percent, respectively). As expected, convicted DV offenders in the two boroughs with specialized DV court parts (Brooklyn and the Bronx) were less likely to be incarcerated than convicted DV defendants in Manhattan. This difference in sentencing reflected the stronger emphasis on the use of batterer intervention programs to monitor defendants in boroughs with specialized DV court parts. This report advises that it is beyond its scope to decide which of the three boroughs has the "best" approach to prosecuting and adjudicating DV cases; however, it does make some observations about the problems posed by each approach and issues that could be addressed through policy changes. Suggestions for future research are also offered. 10 tables, 1 figure, 64 references, and appended coding of variables for regression models