Effects of Problem-Oriented Policing on Crime and Disorder

• NCJ Number: 224990
• Author(s): David Weisburd; Cody W. Telep; Joshua C. Hinkle; John E. Eck
• Date Published: November 2008
• Length: 89 pages
• Grant Number(s): 2007-IJ-CX-0045
• Agencies: NIJ-Sponsored
• Publication Type: Report (Study/Research); Literature Review
• Award: The Campbell Crime and Justice GroupWhat Works in Crime Prevention: Answers from Systematic Reviews

Annotation

This study synthesized the existing problem-oriented policing evaluation literature and assessed the effects of problem-oriented policing on crime and disorder.

Abstract

Based on high-quality evaluations (both experimental and quasi-experimental studies) conducted thus far, problem-oriented policing has had a statistically significant but modest impact on reductions in crime and disorder. The authors urge caution in interpreting these results, however, because of the small number of methodologically rigorous studies on problem-oriented policing (n=10) and the diversity of problems and responses used in the eligible studies. As used in this study, problem-oriented policing refers to the paradigm developed by Herman Goldstein (1979), in which he replaced the reactive, incident-driven model of policing with a model that requires the police to be proactive in identifying underlying problems that could be targeted to alleviate crime and disorder at their root causes. John Eck and William Spelman (1987) drew upon Goldstein's model in developing the SARA model for problem-oriented policing. This is an acronym for "scanning" (identifying and prioritizing potential problems causing crime and disorder); "analysis" (analysis of the identified problem); "response" (design and implementation of interventions intended to solve the problem); and "assessment" (evaluating the impact of the response on the targeted problem). Eligible studies for this meta-analysis had to meet three criteria: the SARA model was used; a comparison group was included; and at least one crime or disorder outcome was reported. After an exhaustive search strategy that identified over 5,500 articles and reports, only 10 studies met the inclusion criteria. Forty-five other studies were also examined. They met all the selection criteria except the use of a comparison group. 6 tables, 10 figures, 112 references, and appended coding sheets, list of police experts contacted, and a list of effect sizes for all outcomes for eligible studies