This is the first of three chapters on "Hot Spot Analysis" in the user manual for CrimeStat IV, a spatial statistics package that can analyze crime incident location data.
The chapter, "Hot Spot Analysis of Points: I" discusses the concept of a "hot spot" and four hot-spot techniques: the mode, fuzzy mode, nearest neighbor hierarchical clustering, and risk-adjusted nearest neighbor hierarchical clustering. "Hot spots" are defined as "concentrations of incidents within a limited geographical area that appear over time" (Braga & Weisburd, 2010). Police have learned from experience that there are particular environments that attract crimes in above-average concentrations. Hot spots may vary by types of crime. Such information is useful for law enforcement agencies, because it enables them to allocate their resources more efficiently and cost effectively. One section of this chapter describes various statistical approaches to the measurement of hot spots. Issues addressed are the types of cluster analysis (hot-spot) methods and optimization criteria. CrimeStat IV includes 10 techniques that cover the range of techniques that have been used. Each of these techniques is described in this chapter, and examples of their use are presented. The limitations of some are discussed. Eight attachments address issues in statistical techniques for hot spot analysis applied to various crime types and other issues, including homicides; crime areas along commercial corridors; arrest locations as a means for directing resources; the use of CrimeStat in crime mapping in India; identifying duplications in genomic data; the clustering of tuberculosis cases in Harris County, TX; and the seizures of tiger parts and derivatives in India during 2000-2012. Extensive figures showing computer screens, and 52 references