The sample was drawn from all criminal defendants arrested for a domestic-violence offense in Shelby County, TN, over the 12-month period that began in August 2000. All of the cases were assigned to a specialized Domestic Violence Court presided over by a single judge. A rotating team of four prosecutors and a continuing lead prosecutor handled these cases. All of the cases in the sample were disposed in the Domestic Violence Court by April 2002. Case outcomes were coded as dismissed/not guilty or pled guilty/found guilty/diversion. The length of sentence was the actual number of days the defendant was ordered to serve in jail or prison after taking into account the proportion of the sentence suspended by the judge. A variety of legal and extralegal variables were collected for each case, including offender characteristics (e.g., age, race, gender, socioeconomic status, employment status, and prior criminal record) and the offense (e.g., victim/offender relationship, charge type, dual arrest, weapon usage, victim injury, and defendant's use of substances). The defendant's race was unrelated or inconsistently related to court outcome; however, compared to males, females were more likely to be released on recognizance prior to trial and to have their charges dropped by prosecutors. Female defendants were also less likely than male defendants to be convicted; and when convicted, they received lighter sentences. Few of the offense characteristics were consistently associated with decisions during the adjudication process. 4 tables and 49 references