NCJ Number
139735
Date Published
Unknown
Length
134 pages
Annotation
The Colorado Springs Spouse Assault Replication Project was conducted by the Colorado Springs Police Department and involved a multisite experiment in Wisconsin, Nebraska, Florida, Georgia, and North Carolina, as well as Colorado Springs.
Abstract
The project design at each site was tailored to local conditions and questions of interest but was constructed around common data elements that could be compared across sites. At all sites, the core of the project was to test effects of arrests and other dispositions on the frequency of subsequent violent domestic incidents. A pretest was conducted between March and May 1987, and the full experiment was launched in June 1987. A total of 1,658 cases of spouse assault had a randomized disposition imposed. One of four randomly selected dispositions was selected: restoring order only; issuing an emergency protection order only; issuing an emergency protection order and having the suspect go immediately to a counseling session; or issuing an emergency protection order and arresting the suspect. Threat or injury to the victim or hostile behavior toward police officers were associated with larger number of arrests imposed than assigned, especially when the original assignments were to counseling or to restore order. Using victim reports of subsequent incidents that involved hitting, slapping, hurting, or detaining, couples assigned to the arrest disposition had lower recidivism rates. When data from official arrest records were used to measure failure, however, no differences were found among the alternative dispositions assigned. Victim reports of new incidents and official arrest records did not match well.