All men (n=447) convicted of misdemeanor domestic violence in Broward County (Florida) during a 5-month period in 1997 were randomly assigned to either the experimental or control group. The purpose of the evaluation was to test whether court-mandated counseling reduced the likelihood of repeat violence by the men. The evaluation tested two hypotheses. First, men who are mandated into counseling will have a lower likelihood of repeat violence compared with men assigned to the control (no treatment) group. Second, men who have a high stake in conformity will have a lower likelihood of recidivism than those with a low stake in conformity. The control group was assigned to probation instead of the treatment program. There were no significant differences between the groups in offender demographics, stake in conformity, criminal record, and instant incident; however, the average age of the control group was 2 years younger than that of the experimental group. This could be significant, since younger men are more likely to abuse their partners and to recidivate. Each batterer was interviewed at the time of adjudication and again 6 months after adjudication. The victim was also interviewed at adjudication and 6 and 12 months after adjudication. Standardized measures with known reliability were used when possible. Probation records and computer checks with the local police were conducted to determine new arrests for 1 year after adjudication. The research found no significant changes between the experimental and control group on beliefs about the legitimacy of wife beating, their sense of responsibility for these incidents, and their attitudes regarding the proper roles for women. Further, there were no significant differences between groups or within groups over time in men's self-reported likelihood of engaging in any of the five subscales listed in the Conflict Tactics Scale2 (negotiation, psychological coercion, physical abuse, sexual coercion, and injury). Victim reports on partner violence did not find any difference between or within the groups over time in the likelihood of their partners engaging in any of the five subscales listed in the Conflict Tactics Scale2. Regarding rearrests, 24 percent of men in both the experimental and control groups were rearrested on one or more occasions during their 1 year's probation. The authors note that although batterer intervention programs have been mandated by courts in jurisdictions across the country since the late 1980's, researchers still cannot definitively answer whether these program actually make things better for the victims of domestic violence. The current research suggests that one of the more popular treatment models being used does not achieve its goal of making victims safer. 2 exhibits and 29 references