The project used a quasi-experimental design that incorporated two treatment groups. One group received a single session of traditional crisis counseling, and the other received a session of counseling coupled with crime prevention training. The names of victims of robbery, burglary, and nonsexual assault were collected from crime reports at selected high-crime police precincts in New York City. A total of 191 victims (evenly divided between males and females) were recruited for the study. The first 78 participants were assigned to the control treatment, and the next 113 participants were assigned to the treatment group. By all accounts, the crime prevention training program achieved its goals. Two-thirds of the victims who received the training requested and received a security survey of their residence. Even 1 month later, victims who had gone through the training knew more about crime prevention, were more likely to believe that crime is avoidable, and took more precautionary measures than victims who were not exposed to the training; still, there was no indication of a reduction in rates of new victimizations. The failure of the treatment to impact on future victimization was not due simply to the intervention not being sufficiently potent. Analysis showed that the intermediate program outcomes -- knowledge, belief in crime avoidability, and precautions -- were unrelated to the likelihood of future victimization. 39 references