Program Goals/Target Sites
Operation Safe Streets was a law enforcement and crime prevention initiative implemented in
The sites identified for intervention were the highest drug-activity locations in the city. Using crime data, arrest data, firearm seizure data and informant data, the Philadelphia Police Department identified 214 locations throughout the city to receive the intervention.
Program Components/Key Personnel
Starting on May 1, 2002, the Philadelphia Police Department stationed pairs of officers at the locations identified. The locations, mainly drug corners, were policed around the clock in order to avoid their use by drug dealers and buyers and dissuade violent crime related to the drug trade (turf wars, gang conflict, etc). The operation required the involvement of more than 600 officers and close to 24-hour-a-day, 7-day-a-week policing of the target locations. The resources, both in terms of funds and required dedicated staff, were substantial.
Geographically targeted crackdowns are an example of problem-solving policing, a strategy which uses a preventive approach to address the underlying issues of crime. The strategies utilized in the Operation Safe Streets program are based on the idea that the underlying issues of the drug trade tend to be localized and dependent on a number of environmentally specific factors. Specifically, the positioning of police officers at specific corners is used to suppress the establishment and operation of local drug markets, as well as to prevent the proliferation of drug-related crimes.
The treatment group consisted of 214 high-crime areas identified over a period of 139 weeks, with 121 weeks of crime data before the intervention and 18 weeks of crime data during Operation Safe Streets. (While the newspaper requested additional data from the PPD to extend the observation of the treatment period for analysis, this request was denied.) The authors additionally selected 73 comparison areas, which also reported high-rates of drug crime but which were not targeted by the intervention. This comparison group was used only descriptively and not in any statistical tests. This was because direct comparison was not possible due to the intervention areas being identified specifically for being the highest drug-crime areas. The data used contained all reported drug and violent crimes from Jan. 1, 2000, through Aug. 31, 2002, with Operation Safe Streets being implemented on May 2, 2002.
The study used a time series design using Auto-Regressive Integrated Moving Average (or ARIMA) models to establish the impact of the intervention on violent crime, homicide, and drug crime citywide. The homicide category included all homicides and manslaughter; the violent crime category included all rapes, robberies, and aggravated assaults in addition to homicides; and the drug crime category included all types of selling, possession, and manufacturing of illegal drugs. Weekly counts of each type of crime were aggregated and coded using
The study additionally used Geographic Information Software to geocode the recorded crimes throughout the observation period. This allowed a further time series analysis of the local impact of the treatment, but also to look at potential benefits and displacement effects in adjoining neighborhoods. For the geographically focused localized analyses, weekly crime counts were transformed into weekly counts/kilometer² (km²) to account for the differences in sizes of areas. An analysis of homicide counts in localized areas was not possible due to a small number of homicide cases.
This study was limited to the first 4 months of treatment data, which did not permit for analysis of the potential for waning program effects or lasting effects posttreatment.
The results of
Citywide Drug Crime
There was a decrease of approximately 10 drug crimes a week, but these declines were not found to be statistically significant.
Citywide Violent Crime
Violent crime declined citywide after the intervention, with about 16 fewer violent crimes each week. These declines were not found to be statistically significant, however.
The analysis showed approximately one less homicide every 2 weeks after the intervention was implemented, but these declines were not statistically significant.
Localized Drug Crime
Localized Violent Crime
The results of the analysis indicated that the intervention prevented on average one violent crime a week per kilometer² (km²) within 0.1 mile of the intervention site. At 0.1–0.2 miles from the intervention site, the results reveal a smaller, but still significant, reduction of approximately 0.3 violent crimes a week per km².