Domestic abuse (DA) is a vexing problem for law enforcement. Often in a home with no neutral witnesses police must ascertain the facts of the case and decide on a course of action. The couple is romantically involved, so unsurprisingly, both couple members are unwilling to help the police. A substantial fraction of these couples will have the police return—13% in the data that we recently collected. With the large differences in physical strength and aggressiveness between women and men, women disproportionately suffer from domestic abuse. Their intimate partners are responsible for the deaths of one-third of all female homicide victims; and domestic abuse accounts for about 15% of all violent crimes.
Aware of these ominous statistics, the police and social service agencies often seek help assessing the risks associated with the incidence of DA. In the US, 42% of police departments use some form of risk assessment. Risk assessment is also used for DA cases in Canada and several EU nations. In England and Wales, almost all police agencies follow the Domestic Abuse, Stalking, Harassment, and Honor-Based (DASH) violence risk assessment model, which we evaluate in our study. DASH consists of a 27-item questionnaire and a risk grade assigned by the responding officer (one of three risk categories: standard, medium, and high). Victims designated as high-risk are provided with a customized package of services designed to keep them safe.
Police may also begin official proceedings against the perpetrator. In England and Wales, police may charge the perpetrator. Charges may or may not result in the perpetrator being taken into custody, what we in the US would call arrested. We explore the impact of being categorized as high-risk in the DASH system and the impact of being charged. Both being charged and designated high-risk are not random samples of domestic abuse calls: referred incidents are more likely to recidivate. To control these differences, we use matching on a rich set of other characteristics, such as previous DA offenses and other offenses, responses to the DASH questionnaire, and characteristics of the current offense, including alcohol and drug use.
The DASH program has no measurable impact on recidivism, with an estimated effect of no more than a 0.6 percentage point reduction (on a base of 13%). In contrast, charging the perpetrator results in a substantial, 5 percentage point reduction in the risk of recidivism.
Interpreting these results, however, is hard. For instance, the DASH result could arise because the DASH questions do a poor job of discerning who is at risk of re-offending or because protective services agencies fail to protect the victim. Similarly, the success of charging the perpetrator may result in an incapacitation effect, or it may reflect that the perpetrator, and perhaps even the victim, understand the gravity of the offenses. For whatever reason, however, the involvement of the judicial system appears more effective than solely relying on social services.
© Dan A. Black
Dan A. Black is professor in the Harris School of Public Policy at the University of Chicago and a Research Fellow of IZA.
Please note:
We recognize that IZA World of Labor articles may prompt discussion and possibly controversy. Opinion pieces, such as the one above, capture ideas and debates concisely, and anchor them with real-world examples. Opinions stated here do not necessarily reflect those of the IZA.