(Submit to Charlotte Gill at gillce@sas.upenn.edu
)
1.
Title of the review
"A Systematic Review of Research on the Effects of Arrest for Misdemeanor
Domestic Violence”
2.
Background and objective of this review (briefly describe the problem and the intervention).
In 1984, the Minneapolis domestic violence arrest experiment (Sherman and Berk, 1984), reported that arrest in misdemeanor domestic assault cases reduced reoffending substantially in comparison to two informal alternatives (mediation or separation). Subsequent trials funded by the
National Institute of Justice, however, failed to replicate the Minneapolis findings, reporting inconsistencies in the direction and effect size of the results (Hirschel et al., 1992; Berk et al., 1992a; Pate and Hamilton, 1992;
Sherman, 1992a; Dunford et al., 1990a: 1990b).
Despite the importance of the research question these experiments addressed, no formal meta-analysis or systematic review of this evidence has been conducted. Some reanalyzes of these studies have been published (e.g., Sherman, 1992b; Garner et al., 1995; Maxwell et al., 2002;
Berk et al., 1992b). Yet some of the analyses were conducted by methods other than by intention-to-treat (ITT), while others are based largely on victims’ interviews reports with very low response rates. These methods as well as the data sources are less than optimal compared, for example, to
ITT analyses using official police records. In addition, existing reanalyzes did not try to calculate effect sizes, which may have led to some confusion about the comparisons of effects across trials. Policymakers and researchers alike could benefit from an assessment of the existing body of
1

research using Campbell Collaboration review procedures, particularly on the question of whether a policy of mandatory arrest is better than informal dispositions.
This review would be especially useful in relation to the moderator analysis suggested by Sherman (1992) in the absence of formal metaanalytic methods. In this analysis, Sherman suggested that the effects of arrest were conditional upon at least two “stakes in conformity” (Toby,
1957): employment and marriage. Sherman and others (Sherman and
Smith, 1992; Pate and Hamilton, 1992, and Berk et al., 1992b) have all reported a differential and interactive effect of arrest with the employment status of the suspect (as measured by victim interviews). Sherman (1992a) also reports similar but weaker interactions with marriage.
Based on these earlier, less rigorous moderator analyses, we propose to conduct subgroup analyses of the available evidence in order to test these hypotheses. We hypothesize that a meta-analysis will show that arrest deters employed suspects, but increase subsequent domestic assault by unemployed suspects. We also will test the effect of marriage as a moderator of arrest effects on repeat domestic violence. Thus, we intend to address three research questions: a) main effects; b) effects conditioned on employment; c) effects conditioned on marriage.
3.
Define the population
Misdemeanor domestic assault suspects in the presence of the police, in sites where experiments on domestic assault have been conducted and reported in the English language.
2

4.
Define the intervention
4.1
Arrest of any type and length;
4.2
Any alternative to an arrest, such as mediation, separation, consultation or any other processing carried out by the police in misdemeanor domestic violence cases that does not result in the suspect being taken into police custody at a different location. Collectively, we call this comparison category any “alternative to an arrest”.
5.
Outcome(s) (what is aimed to accomplish – Primary and secondary outcomes should all be mentioned)
The primary outcome measure will be the frequency of official records of repeat arrests or reports of domestic violence with the same offender against the same victim, calculated in frequencies of re-arrest per 1000 days at risk in each group. The source of data comes from official police records
The secondary outcome measure will be the same as the primary, except that it will include domestic violence against any additional victims as well as the instant victim in the presenting case.
The tertiary measure will be the same-victim victimization reports, using prevalence measures.
6.
Methodology (What types of studies are to be included or excluded and what will be your method of synthesis? Will you use meta-analysis?)
Only randomized field trials will be included in the synthesis and the meta-analysis, under the premise that anything else lacks sufficient internal validity necessary for reaching sound conclusions based on unbiased estimates. Laboratory experiments, natural experiments and quasiexperiments will be excluded from the systematic search for eligible studies.
3

The statistical analysis that best fits this methodology is forest plots, with a calculation of the overall effect size on the basis of arrest frequencies reported in the studies. The effect sizes will be measured by a magnitude of the difference between the groups, using Hedges’ g.
7.
Do you need support in any of these areas (methodology, statistics, systematic searches, field expertise, review manager etc?)
No, thank you.
8.
Lead reviewer(s) with contact information
Barak Ariel, PhD
Email: ba285@cam.ac.uk
Jerry Lee Centre of Experimental Criminology,
Institute of Criminology, University of Cambridge, UK
4

9.
Reference List
Berk, R.A., Campbell, A., Klap, R. and Western, B. (1992a). Bayesian analysis of the
Colorado Springs spouse abuse experiment, Journal of Criminal Law and
Criminology 83: 170.
Berk, R.A., Campbell, A., Klap, R. and Western, B. (1992b). The deterrent effect of arrest in incidents of domestic violence: A Bayesian analysis of four field experiments.
American Sociological Review : 698.
Dunford, F.W. (1990a). System-initiated warrants for suspects of misdemeanor domestic assault: A pilot study. Justice Quarterly 7(4): 631.
Dunford, F.W., Huizinga, D. and Elliott, D.S. (1990b). The role of arrest in domestic assault: The Omaha police experiment. Criminology 28(2): 183.
Garner, J., Fagan, J. and Maxwell, C. (1995). Published findings from the spouse assault replication program: A critical review. Journal of Quantitative Criminology 11(1): 3.
Hirschel, J.D., Hutchison Iii, I. and Dean, C.W. (1992). The failure of arrest to deter spouse abuse. Journal of Research in Crime and Delinquency 29(1): 7.
Maxwell, C.D., Garner, J.H. and Fagan, J.A. (2002). The preventive effects of arrest on intimate partner violence: Research, policy and theory. Criminology and Public
Policy 2(1): 51.
Pate, A.M. and Hamilton, E.E. (1992). Formal and informal deterrents to domestic violence: The Dade County spouse assault experiment. American Sociological
Review : 691.
Sherman, L.W. (1992b). Policing Domestic Violence . NY: Free Press.
Sherman, L. W. and Berk, R. A. (1984). The specific deterrent effects of arrest for domestic assault. American Sociological Review: 261.
Sherman, L.W. (1992a). The influence of criminology on criminal law: Evaluating arrests for misdemeanor domestic violence. Journal of Criminal Law and Criminology 83: 1.
Sherman, L.W., Schmidt, J.D., Rogan, D.P. and Smith, D.A. (1992). Variable effects of arrest on criminal careers: The Milwaukee domestic violence experiment, the.
Journal of Criminal Law and Criminology 83: 137.
Toby, J. (1957). "Social disorganization and stake in conformity: Complementary factors in the predatory behavior of hoodlums." Journal of Criminal Law and Criminology
48: 12.
5