In Part three of CrimeStat II, a spatial statistics software program supported by the National Institute of Justice and designed to provide statistical summaries and models of crime incident data, spatial modeling of CrimeStat is presented with a focus on the kernel density estimation approach, the Journey to Crime estimation technique, and space-time analysis techniques.
Supported by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice, CrimeStat II is a software program designed to provide statistical summaries and models of crime incident data. It offers crime analysts and researchers with a vast range of spatial statistical procedures that can be linked to a geographic information system (GIS). CrimeStat II is a full-featured Windows 2000 program using a graphical interface with database and expanded statistical functions. CrimeStat provides descriptions of the spatial arrangements of crime incidents. There are a variety of tools that can be used to describe these arrangements and they are useful in helping crime analysts detect patterns of crime and provide different perspectives on the arrangements, analyzing one or two variables affecting crime incidence. This report represents part three in a three part series providing information on the spatial modeling of CrimeStat. The spatial modeling is comprised of three sections beginning with a discussion of the tools aimed at interpolating incidents, using the kernel density approach. Interpolation is a technique for generalizing incident locations to an entire area. This is followed with a discussion on the Journey to Crime estimation technique which is a distance-based method that makes estimates about the likely residential location of a serial offender. Lastly, three techniques are discussed that are used to analyze the relationship between space and time. Part one and part two of CrimeStat cover and focus on the program overview and spatial description. Figures, references and Appendices A-B