This is the second of three chapters on "Hot Spot Analysis" from the user manual of CrimeStat IV, a spatial statistics package that can analyze crime incident location data.
This chapter, "Hot Spot Analysis of Points: II," continues the previous chapter's discussion of hot spots. Two additional routines are discussed: the STAC (the Spatial and Temporal Analysis of Crime) routine and the K-Means routine. STAC, which was developed by the Illinois Criminal Justice Information Authority and integrated into CrimeStat II, is a quick, visual, easy-to-use program for identifying hot spots. In CrimeStat, this routine searches for and identifies the densest clusters of incidents based on the scatter of points on the map. Steps are outlined in using STAC. As an example, a STAC analysis of 1999 Chicago street robberies is described. The advantages of STAC as a clustering algorithm are listed, along with its limitations. The section of the chapter on K-means clustering notes that it is a partitioning procedure in which the data are grouped into K groups defined by the user. A specified number of seed locations, K, are defined by the user (Fisher, 1958; MacQueen, 1967; Aldenderfer and Blashfield, 1984; Systat, 2008). This routine attempts to find the best positioning of the K centers and then assigns each point to the center that is nearest. Similar to the nearest neighbor hierarchical routine, the K-means assigns points to only one cluster; however, all points are assigned to clusters. Thus, there is no hierarchy to the assignment. This technique is useful when a user wants to control the grouping. Examples are provided of circumstances in which this routine might be used. As an example of the use of this routine, it is applied to Baltimore County street robberies. 27 references, extensive figures, and 2 attachments that pertain to hot spot analysis