The effects of a group batterer treatment program: a randomized experiment in Brooklyn

Study of the effectiveness of Duluth style perpetrator programmes in New York, in which 376 men judged appropriate for treatment by the prosecutor, defendant and judge, were randomly assigned to a six-month Duluth-style perpetrator programme or to 6 months community service (control group). During the evaluation, some men were assigned to an alternative treatment programme of two months' duration, with two sessions a week rather than one. Outcomes were measured by interviewing the women partners of the men at sentencing, and again six and twelve months later, and by police record search twelve months after sentencing. Completion rates for victim surveys, however, were low. Found that the men in the longer (six-month programme) were significantly less likely to be arrested for domestic violence than the men in the control (community service) group and than men in the shorter treatment group. This treatment effect did not diminish significantly between the six- and twelve-month follow-up periods. However, when the researchers looked at victim reports of new incidents, there were no statistically significant differences between the groups.