Specifically, the chapter examines block aggregation, local indicators of spatial autocorrelation (LISA), and spatial regression to show how they can be used to identify crime clusters and to test theoretical explanations for the causes of crime. In addition to crime data provided by the New York City Police Department (NYPD), the study also incorporates data sets created by the U.S. Census Bureau. The chapter concludes that as a package, the techniques described permit the gleaning of a tremendous amount of information from raw crime data. The authors started by examining temporal clustering and then used aggregation techniques to identify individual polygons with high crime rates. This spatial clustering was augmented by LISA statistics to show areas of relative clustering, i.e., where high-crime polygons were located next to other high-crime polygons. Finally, the authors used spatial regression techniques to identify variables linked to the root causes of criminal behavior. Challenges for future research are discussed. 3 figures, 1 table, 6 notes, and 10 references