NCJ Number
244321
Journal
Legal and Criminological Psychology Volume: 18 Issue: 2 Dated: September 2013 Pages: 356-370
Date Published
September 2013
Length
15 pages
Annotation
The study extends research by Santtila et al. (2008) by investigating the effectiveness of linking cases of serial homicide using behavioural patterns of offenders, analysed through Bayesian reasoning. The study also investigates the informative value of individual behavioural variables in the linking process.
Abstract
Offender behavior was coded from official documents relating to 116 solved homicide cases belonging to 19 separate series. The basis of the linkage analyses was 92 behaviors coded as present or absent in the case based on investigator observations on the crime scene. The authors developed a Bayesian method for linking crime cases and judged its accuracy using cross-validation. The authors explored the information added by individual behavioral variables, first, by testing if the variable represented purely noise with respect to classification, and second, by excluding variables from the original model, one by one, by choosing the behavior that had the smallest effect on classification accuracy. The model achieved a classification accuracy of 83.6 percent whereas chance expectancy was 5.3 percent. In simulated scenarios of only one and two known cases in a series, the accuracy was 59.0 and 69.2 percent, respectively. No behavioral variable represented pure noise but the same level of accuracy was achieved by analyzing a set of 15, as analyzing all 92 variables. The study illustrates the utility of analyzing individual behavioral variables through Bayesian reasoning for crime linking. Feasible applied use of the approach is illustrated by the effectiveness of analyzing a small set of carefully chosen variables. Abstract published by arrangement with John Wiley & Sons.