The study identified four distinct problems that have stemmed from New York State's mandatory arrest policy for domestic violence cases: "dual arrests" (both parties involved in the violent encounter are arrested); "retaliatory arrests" ( complaining caller falsely accuses a partner who previously caused him/her to be arrested for domestic violence); "unwanted arrest" (victim does not want his/her partner arrested for any number of reasons); and "no arrest" (a case in which an arrest should have been made but was not made). The analysis of each of these consequences of the mandatory arrest policy for domestic violence cases suggests that further training and better supervision is required for responding officers in order to have more effective implementation of the law in terms of protection and remedies for domestic violence victims. Between July 1, 1998, and June 30, 2000, data were obtained from the telephone helpline of the Family Violence Project (FVP) of New York City's Urban Justice Center. The final sample consisted of 183 cases in which the presenting problem involved a police response. Cases were classified according to the arrest outcome indicated by the caller at the first point of contact. This means the cases were classified according to the arrest outcome described in the initial call to the helpline rather than the ultimate outcome of the case once advocacy was begun. 2 tables and 39 references