Testing the accuracy of geographical profiling considering the effect of the number of crimes.

The present research compared geographic profiling (GP) methods, including spatial distribution strategies (SDS) and probability distance strategies (PDS), to account for the effect of the number of crimes on the accuracy of GP. Comparing the accuracy of GP methods among three burglar groups (a total of 333 offenders) with different sizes of crimes (1–5, 6–15 and 16 or more) committed before arrest showed that the accuracy decreased in a group with a larger size of crime. Comparing the accuracy of GP methods among the number of crimes (3–10) using 150 offenders who committed 10 or more crimes showed a slight improvement in the accuracy with the increasing number of crimes. In addition, the PDS (vs. SDS) showed a higher accuracy with the increasing number of crimes. The findings encourage practitioners to use the PDS and research implications are discussed. [ABSTRACT FROM AUTHOR]