The main purpose of the paper is to provide a greater understanding of the factors that influence heterogeneous distributions of crime. The authors focus on the comparative analysis of three quantitative approaches: Ordinary Least Squares (OLS), Geographically Weighted Regression (GWR) and Data Envelopment Analysis (DEA). The above methods are then applied to explore the structural theories of violence in Cincinnati, Ohio by block group. To predict assault rates between January and June of 2008 the authors use 2006 population density from Caliper Corporation, density of alcohol outlets from the Ohio Division of Liquor Control, and social disorganization calculated from socio-economic disadvantage, female headed-households, and residential instability, as independent variables.

The presentation can be accessed here and the more detailed review can be accessed here.

Bibliography

Grubesic, T. H., Mack, E. A., Kaylen, M. T. (2012). Comparative modeling approaches for understanding urban violence. Social Science Research, 41(1), 92-109.doi:10.1016 /j.ssresearch.2011.07.004