Individuals – Berk et al. (1992)

Study Reference:

Berk, R. A., Campbell, A., Klap, R., & Western, B. (1992). A Bayesian Analysis of the Colorado Springs Spouse Abuse Experiment. The Journal of Criminal Law and Criminology, 83 (1), 170-200.


Location in the Matrix; Methodological Rigor; Outcome:

Individuals; Focused; Reactive; Very Rigorous; Effective


What police practice or strategy was examined?

The evaluation examined the Colorado Springs Spouse Abuse Experiment. In the Colorado Springs experiment, suspects apprehended for misdemeanor spouse abuse were assigned randomly to one of four treatments: (1) an emergency order of protection for the victim coupled with arrest of the suspect; (2) an emergency order coupled with immediate crisis counseling for the suspect; (3) an emergency order only; or (4) restoring order at the scene only. The current study focused specifically on the impact of arrest on recidivism for different kinds of offenders.


How was the intervention evaluated?

The authors focused primarily on the effects of arrest on different types of offenders, given that the arrest effect was found to differ based on the employment status of offenders in research in Milwaukee (see Sherman et al., 1992). In this study, offenders who were employed or were in the military were categorized as “good risks” and those who were not as “bad risks.” The outcome measure was the failure of the suspect, which was defined as either a new reported offense by the suspect involving the same victim (from police data), or an incident in which the same suspect struck or caused injury to the same victim (from victim report data) within approximately six months of the original offense.


What were the key findings?

Results demonstrate a deterrent effect for arrest among “good risk” offenders, who presumably have more to lose by being arrested. The chance for new violence by this category of offenders was reduced by 15%. Though the evidence is not conclusive, the authors suggest that the likelihood of future violence could be increased for those in the “bad risk” suspect category.


What were the implications for law enforcement?

The authors advise that more replication data and analysis are required before definitive policy recommendations can be made on police policy and procedures for responding to domestic assault. Even though mandatory arrest seems to be a deterrent for “good risk” suspects, one must consider the balance of beneficial and harmful effects upon the whole set of individuals for whom the intervention is relevant. It is clear that the impact of mandatory arrest for domestic abusers is complex.


Where can I find more information about this intervention, similar types of intervention, or related studies?