Canadian Policing Research Catalogue

The effectiveness of calibrated vs. default distance decay functions for geographic profiling / by Karla Emeno.

This page has been archived on the Web

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

Location

Canadian Policing Research

Resource

e-Books

Authors

Publishers

Bibliography

Includes bibliographical references (pages 68-74).

Description

1 online resource (77 pages)

Note

"September 2008"
Thesis (M.A.)--Carleton University, 2008.

Summary

Geographic profiling systems use distance decay functions to predict where serial offenders live. These functions assume that the likelihood of an offender residing at a particular location decreases with increasing distance from the offender's crime sites. Currently, each system relies on a default function, which was validated on data that are unrelated to the crimes being submitted to these systems during investigations. This occurs despite the possibility that a single decay function cannot be used with equal effectiveness across data collected under varying conditions. This study determined whether a decay function calibrated for a particular crime type or geographic region resulted in greater predictive power than an uncalibrated, default function. Decay functions were calibrated for three different types of serial crime (residential burglary, theft, and auto theft) collected from the same geographic region (Glendale, AZ) and for serial burglary collected from three different geographic regions (Glendale, AZ, Baltimore, MD, and Dorset, UK). The two default functions (truncated negative exponential and negative exponential) used for comparison purposes came from CrimeStat (v 3.1), a computerized geographic profiling system. The hypothesis that calibrated functions would possess more predictive power than the two default functions, as measured by error distance and hit percentage, was not supported to the extent that was expected, with the majority of analyses finding non-significant differences across the various functions within each data file. Potential explanations for these findings are provided, implications are discussed, and directions for future research are presented.

Subject

Online Access

Date modified: