Case Studies in Crime Travel Demand Modeling II: Application of Travel Demand Behavior Model on Crime Data From Las Vegas, Nevada (CrimeStat IV: A Spatial Statistics Program for the Analysis of Crime Incident Locations, Version 4.0)

• NCJ Number: 242992
• Author(s): Dan Helms
• Date Published: June 2013
• Length: 40 pages
• Grant Number(s): 2005-IJ-CX-K037
• Agencies: NIJ-Sponsored
• Publication Type: Technical Assistance; Instructional Material (Programmed)
• Award: The Expansion of the Crimestat Spatial Statistics Program to Incorporate Spatial Regression, Space-Time Clustering, Bayesian Journey to Crime Modeling, Incident Detection Methodology and Sparse Matrix

Annotation

This is the final chapter in a series of eight chapters on "Crime Travel Demand Modeling" from the user manual for CrimeStat IV, a spatial statistics package that can analyze crime incident location data.

Abstract

This chapter presents a case study of crime travel demand in Las Vegas, NV, in order to illustrate how crime travel demand modeling can be applied to a city where travel is primarily by motor vehicle. The model is applied using data from the metropolitan Las Vegas area over a 3-year period. The model overall performed well for some crime types, but was weaker for others. One of the most troubling issues in the evaluation of the network assignment stage of the model (prediction of route used to a crime by an offender based on transportation mode) was the lack of any good final metric other than visual approximation for determining the value of the resulting prediction. Some measurement of congruence is required to make the determination of usefulness reliable and valid. Performance and usefulness are examined for each stage of CrimeStat's crime travel demand model; and suggestions are offered for how performance at each stage might be improved. The most successful predictive variables for estimating crime trip production, whether of origins or destination for crime trips, were total population, total employment, and income. Inclusion of additional variables distorted rather than improved the predictive value of the model. Regarding the mechanical aspects of the model as implemented in CrimeStat IV, more work is required in order to better calibrate and implement the model so as to make it an effective tool for law enforcement analysis and planning. 3 tables, 20 figures, and 1 reference