Geography of Crime and Justice
Description: Neighborhood parks in urban areas are public spaces that provide a variety of recreational opportunities for residents in a natural environment. At the same time, they can become staging areas for illegal and disorderly activities. Systematic social observations describing park characteristics, US Census data quantifying neighborhood social composition, community surveys capturing social cohesion, and official crime data are analyzed using multilevel models to examine: 1) whether activity generating features of parks explain differences in crime levels across parks; 2) whether differences in parks’ neighborhood context explain differences in crime levels across parks; and 3) the extent to which park characteristics as opposed to neighborhood context explain differences in crime levels across parks. Our results are discussed in terms of their implications for theories of crime and place.
Description: We used cutting-edge predictive policing software to plot locations of increased risk of violent and property crime. The Philadelphia Police Department then applied three different types of policing activity in these areas. Across five Philadelphia Police districts, officers were made aware of the predictive policing areas on roll-call. In five others, a dedicated police car patrolled the predictive grids. And in five other districts, an unmarked car patrolled the grids. The six remaining districts were used as control locations. The results of this NIJ-funded program are currently being analyzed.
Project Title: Linking Theory to Practice: Testing Geospatial Predictive Policing
Funding Agency: National Institute of Justice</br> Affiliated Faculty: Alese Wooditch
Description: This NIJ-funded study seeks to advance the knowledge and utility of predictive policing by examining the strategy’s key components and processes. In collaboration with the Denver Police Department (DPD), this project evaluates available data sources for predictive analytics, and assesses a variety of predictive policing software programs to compare accuracy, reliability, and ease of use. The effectiveness of the best performing predictive policing software program will be conducted using a randomized control trial and the implementation of the predictive policing process by the DPD will be examined.
Project Title: Neighborhood Context and Spatiotemporal Patterns Of Crime
Affiliated Faculty: Jeffrey T. Ward
Description: With evidence for spatiotemporal patterning of crime mounting, there is a growing need to understand how neighborhood context impacts repeat/near repeat crime as well as crime that is not connected in space and time. This project has provided one of the first assessments of how social-structural and neighborhood design features similarly and differentially contribute to spatiotemporally related burglaries and spatiotemporally unrelated burglaries (Nobles, Ward, & Tillyer, in press, Journal of Research in Crime and Delinquency). Other crime types and additional sites beyond Jacksonville are now under study.
Description: This NIJ-funded study examined connections between jurisdiction-level demographic structure, land use, street networks and changing violent and property crime rates in the first decade of the 21st Century in the nine county Philadelphia Metropolitan region. Results suggest increasing spatial inequality in violent crime across the region, and impacts of large scale commercial sites on property crime rates. Implications emerge for cross-jurisdictional police cooperation.
Project Title: Community Criminology Crossroads
Affiliated Faculty: Ralph B. Taylor
Description: This review of almost 100 years of community patterns of crime and delinquency highlights four critical issues blocking further development in our understanding of the causes and consequences of crime and delinquency ecological patterns: temporal scaling, spatial scaling, ecological construct validation, and selection (NYU Press, 2015).
Project Title: Bail Prediction: Exploring the Role of Neighborhood Context in Philadelphia
Affiliated Faculty: E. Rely Vîlcică
Description: This study (2015, Criminal Justice & Behavior) examined neighborhood effects on defendant pretrial performance. Specifically, the study looked at whether there was neighborhood-level variation in defendants’ failure-to-appear for trial and pretrial criminal conduct and explored the impact on these outcomes of neighborhood structural conditions such as socioeconomic status, stability, and racial composition. Prior bail prediction research has relied only on individual-level attributes. This paper, therefore, for the first time provided a novel perspective in recognizing the role of community context in empirical prediction of bail outcomes. The findings have important implications for bail decisionmaking policy and are relevant for the validity of the contemporary theories that emphasize the community context of crime in shaping deviant behavior.