The IACP Research Advisory Committee is proud to offer the monthly “Research in Brief” column. This column features evidence-based research summaries that highlight actionable recommendations for Police Chief magazine readers to consider within their own agencies. The goal of the column is to feature research that is innovative, credible, and relevant to a diverse law enforcement audience.
any questions regarding research on police methods have traditionally been raised, tested, and evaluated by academics. In this study, the research team did not comprise academics but rather practitioners. The 90-day Sacramento Police Department (SPD) hot-spot study was completely designed, implemented, and analyzed by personnel within the SPD with the guidance of researchers from George Mason University in Fairfax, Virginia. Special thanks to David Weisburd, PhD; Cynthia Lum, PhD; Christopher Koper, PhD; and Cody Telep. Additionally, this hot-spot study was conducted without external funding.
Hot-spot policing has become an accepted practice in policing, focusing police resources in small areas such as addresses, street blocks, or clusters of addresses or street blocks.1 The first hot-spot study conducted in Minneapolis, Minnesota, during 1995 revealed that 3 percent of the addresses in Minneapolis accounted for 50 percent of the crime calls to the police.2 In Sacramento, 4.7 percent of the street segments accounted for 50 percent of the crime calls for service, leading the SPD to believe it was imperative to focus its police resources on these so-called hot areas.3 Further analysis of the Minneapolis study established the optimal amount of time to visit a hot spot was 12 to 16 minutes, or approximately 15 minutes based on the Koper curve theory.4 As such, the SPD implemented a randomized control trial designed to answer the question, “Will visiting hot spots in a random, intermittent order for 12- to 16-minute increments reduce crime and calls for service in Sacramento?”
The SPD studied data from two districts (out of six) for hot-spots data collection. The SPD examined all of the computer-aided dispatch data for Districts 3 and 6 from January 1, 2009, to December 31, 2010. Analysts retrieved only calls generated by citizens, removed all supplemental calls to a primary call for service, and excluded calls that were geocoded to an intersection, so as to create a 100-block hot spot rather than an intersection. Addresses that did not meet hot-spot criteria were removed—that is, addresses where crime occurred in public and could reasonably be deterred by police presence.5 Forty-two hot spots were identified; 21 were randomly designated as treatment areas, and they received random intermittent patrol services for 15-minute periods each day. Every day, the officers were given a computer-generated random order to treat the hot spots. The other 21 were designated as nontreatment areas and received normal patrol services. Randomized intermittent treatment creates uncertainty in the mind of the offender, thus increasing the perception of risk and potentially reducing criminal activity.6
A comparison of the calls for service in 2011 to the same three-month period in 2010 indicates a strong treatment effect. On average, each treatment hot spot had a decline of 3.57 calls for service (comparing 2011 to 2010), while each control hot spot had an average increase of 4.43 calls. Thus overall, calls for service declined by about 7.68 percent in the treatment group and increased by about 10.90 percent in the control hot spots. Part I crime incidents showed a somewhat similar pattern to calls for service. During the experimental period, treatment hot spots experienced fewer total Part I incidents (105) than the control hot spots (121). In the same period in 2010, the treatment hot spots had 140 Part I incidents, compared to 95 in the control hot spots. Thus, during the experimental period, the treatment group experienced a 25 percent decrease in Part I incidents, while the control group experienced a 27.37 percent increase in Part I incidents.7
In addition to studying treatment effects, the SPD analyzed officer activity during the study. Overall patrol response times to calls for service did not increase. Crime displacement was not an issue, and, most impressively, officers were 163.6 percent more proactive in District 3 and 72.9 percent more proactive in District 6. These statistics suggest that incorporating an intermittent, random Koper curve approach to a patrol strategy is an effective and efficient way to reduce crime and calls for service. ♦
1David Weisburd and Anthony A. Braga, “Advocate: Hot Spots Policing as a Model for Police Innovation,” in Police Innovation: Contrasting Perspectives, eds. David Weisburd and Anthony Braga (Cambridge: Cambridge University Press, 2006).
2Lawrence W. Sherman, Patrick R. Gartin, and Michael E. Buerger, “Hot Spots of Predatory Crime: Routine Activities and the Criminology of Place,” Criminology 27, no. 1 (February 1989): 27-55.
3Sacramento Police Department Crime Analysis Unit internal database, accessed by Jason Rohde, December 28, 2012.
4Christopher S. Koper, “Just Enough Police Presence: Reducing Crime and Disorderly Behavior by Optimizing Patrol Time in Crime Hot Spots,” Justice Quarterly 12, no. 4 (December 1995): 649-672.
5Lawrence W. Sherman and David Weisburd, (1995) “General Deterrent Effects of Police Patrol in Crime ‘Hot Spots’: A Randomized Controlled Trial,” Justice Quarterly 12, no. 4 (December 1995): 625-648.
6Thomas A. Loughran, Raymond Paternoster, Alex R. Piquero, and Greg Pogarsky, “On Ambiguity in Perceptions of Risk: Implications for Criminal Decision Making and Deterrence,” Criminology, 49, no. 4 (November 2011): 1029-1062.
7Cody W. Telep, Renée J. Mitchell, and David Weisburd, “How Much Time Should the Police Spend at Crime Hot Spots? An Answer from a Police Agency Directed Randomized Field Trial in Sacramento, California,” Justice Quarterly (forthcoming).
Please cite as:
Renée J. Mitchell, "Hot-Spot Randomized Control Works for Sacramento," Research In Brief, The Police Chief 80 (February 2013): 12.