This is an archive page that is no longer being updated. It may contain outdated information and links may no longer function as originally intended.
As innovations in technology make it easier for even small law enforcement agencies to track data related to repeat crime patterns, agencies increasingly try to mine this data for information that will help them allocate limited staffing resources and potentially keep crime from happening.
To predict crime risk potential, a team of researchers from the Temple University Center for Security and Crime Science have begun to explore various combinations of U.S. Census Bureau information for the city of Philadelphia and crime data from the city’s Police Department. The research team examined statistics for crimes to include homicide, rape, aggravated assault and robbery, and property crimes such as burglary and motor vehicle theft. They also considered long-term and short-term risk and looked at near-repeat crimes on a two-week timeframe. (See pages 4-6 of the full report for additional discussion.) The researchers found that a model combining community structural characteristics, crime counts from the previous year, and an estimate of near repeat activity generated the best results overall. It is important to note that the tool is predicting the potential risk of the type and place of a crime, not a specific person or persons likely to commit the crime.
The research team partnered with Azavea, a Philadelphia-based company that creates civic geospatial software and data analytics for the web, to develop a software tool known as PROVE (Prediction of Repeat Offending and Victimization in the Environment). The program, which will be offered to law enforcement agencies free of charge, has the capability to predict future crime risk potential based on the research detailed in this report. It produces both a visual and a statistical report that agencies can use to help inform their crime prediction efforts.
The authors do caution that PROVE, which was still in the beta test phase at the time of publication, is based on calculations generated by only Philadelphia data, and care should be taken in applying results in other areas. Philadelphia, with a population of approximately 1.5 million, is the fifth-largest city in the country and approximately 25 percent of the population lives below the poverty line. Also, although the report offers a link to the beta test version for informational purposes, it is possible that the final version to be released later in 2017 may differ significantly from that version due to changes implemented as a result of the beta test.
About this Article
The work described in this article was supported by NIJ cooperative agreement number 2010-DE-BX-K004, awarded to Temple University.
This article is based on the grantee report Predictive Modeling Combining Short- and Long-Term Crime Risk Potential, Final Report (pdf, 131 pages).