NCJ Number
223858
Journal
Journal of Quantitative Criminology Volume: 24 Issue: 3 Dated: September 2008 Pages: 285-306
Date Published
September 2008
Length
22 pages
Annotation
This study explored the usefulness of statistical measures for identifying and comparing the space-time patterns of robbery, burglary, and assault, using a comprehensive database of crime events for Cincinnati, OH.
Abstract
This study shows that burglary, robbery, and assault have relatively distinctive space-time patterns in Cincinnati. Despite some of the limitations associated with the proposed techniques for analyzing the space-time interaction of these crimes, the results indicate that the linking of space and time in crime analysis improves on traditional “hot-sport” analysis methods, which have focused on either the time or the space (when or where) of crime patterns without attempting to link the two variables. Space-time interaction occurs when events located relatively close in geographic space occur at about the same time. Studies that have examined repeat victimization suggest that there is a higher risk of burglary following an initial incident, and there are regularities in the timing and spacing of these repeat victimizations; however, questions remain regarding the degree to which these regularities exist across space and time in alternative environments. Specific to this study are questions regarding how these regularities compare among crime types (e.g., burglary, assault, and robbery) within a single environment. The study applied two data tests to the Cincinnati crime data, the Knox test (1964) and the Jacquez k-nearest neighbor test (1996). Although originally developed to detect space-time interaction for disease events, the Knox test is also applicable to crime-event data. For a set of geocoded crime events with a corresponding time variable, a test for space-time interaction can be conducted. As the name suggests, the Jacquez k-nearest neighbor test is based on nearest-neighbor distances, a relative popular approach for examining crime-point patterns. Application data included 3 sets of crime events from 17 neighborhoods in Cincinnati, OH, for 2003. 10 figures and 77 references