This article reports on an evaluation of the Philadelphia Police Department's Operation Safe Streets, which involved stationing officers at 214 of the city's highest drug-activity locations 24 hours a day, 7 days a week.
Interrupted time-series analysis ARIMA was used to analyze weekly counts of homicides, violent crimes, and drug crimes. For the geographically focused analyses, weekly crime counts were transformed into weekly counts/km squared, given the varying sizes of target areas and adjoining buffers. The dataset included all reported crimes in the aforementioned categories from January 1, 2000, through August 31, 2002. The intervention began the week of April 29, 2002. This provided 139 weeks of data, with 121 of the weeks encompassing reported crime before the intervention and 18 weeks after. Efforts to obtain data over a longer follow-up period were unsuccessful. The analyses found that Operation Safe Streets failed to have a significant citywide impact on homicides, violent crime, or drug crimes; however, there were significant localized crime reductions for violent crimes and drug crimes; and areas within 0.1 of a mile of the target sites experienced significantly lower weekly crime rates. According to the ARIMA model, there was an apparent partial displacement of drug crimes. Although the localized analyses showed spatially limited program benefits in target areas for both violent-crime rates and drug-crime rates, spatial diffusion of the intervention's impacts predominated for violent-crime rates; and spatial displacement predominated for drug-crime rates. Geographically targeted interventions have a significant impact on crime in these areas, but they cannot be sustained over the long-term because of their high cost. The goal should be to devise more cost-effective crackdowns that can be sustained over time. 4 tables, 4 figures, and 55 references