Forensic Science Seminar
Peer-reviewed forensic science journal · Established 2011
ISSN 2157-118XFOREN SCI SEM · Volume 8

Original Research Papers · Research Paper P1

Using Offender Crime Scene Behavior to Link Stranger Sexual Assaults: A Comparison of Three Statistical Approaches

Matthew Tonkin, Tom Pakkanen, Sirén, J., Craig Bennell, Jessica Woodhams, Amy Burrell, Imre, H., Jan Martin Winter, Lam, E., ten Brinke, G., Webb, M., Labuschagne, G. N., Leah Ashmore-Hills, Jasper Van der Kemp, Lipponen, S., Rainbow, L., Salfati, C. G., Pekka Santtila

Forensic Science Seminar · 2018 · Volume 8 · Issue 2 · pp. 30–42

Peer ReviewedISSN 2157-118XIssue date 30 June 2018
Affiliations
  1. Department of Criminology, University of Leicester, UK.
  2. Department of Psychology, Åbo Akademi University.
  3. Department of Biosciences, University of Helsinki, Finland.
  4. Department of Psychology, Carleton University.
  5. School of Psychology, University of Birmingham, UK.
  6. Department of Psychology, Coventry University, UK.
  7. Belgian Federal Police, Brussels, Belgium.
  8. Dutch National Police, Zoetermeer, The Netherlands.
  9. Department of Clinical and Life Span Psychology (KLEP), Vrije Universiteit Brussel, Belgium.
  10. National Crime Agency, UK.
  11. L&S Threat Management, South Africa.
  12. Department of Psychology, Birmingham City University, UK.
  13. VU University Amsterdam, Faculty of Law, The Netherlands.
  14. Finnish National Police, Helsinki, Finland.
  15. Department of Psychology, John Jay College of Criminal Justice, City University of New York, USA.

Corresponding author: Matthew Tonkin · mjt46@le.ac.uk

Download Article PDFBack to Volume 8, Number 2

Article Summary

Summary

This study compares logistic regression, iterative classification trees and Bayesian methods for distinguishing linked from unlinked stranger sexual assaults using crime-scene behavior. The methods are evaluated on a large multinational dataset, with Bayesian approaches showing particularly strong discrimination in some comparisons. The work informs statistical decision support for behavioral crime linkage.

Keywords: Crime linkage; Comparative case analysis; Bayesian analysis; Logistic regression; Classification tree analysis; Stranger sexual assault; Forensic science

The scholarly title and author names above are normalized from the source PDF and issue contents. Affiliation superscripts are displayed only through the affiliation list and are not part of any author name.

Publication History

Received5 January 2018
Revised9 March 2018
Accepted27 March 2018
Available Online1 June 2018

Citation Information

Document codeP1 · Research Paper
JournalForensic Science Seminar
ISSN2157-118X
Volume / Issue8 / 2
Pages30–42
Official issue date30 June 2018
Article HTMLhttps://fss.xxyy.info/journal/2018/2157118X.8.2.P1.html
Full text PDFhttps://fss.xxyy.info/journal/2018/2157118X.8.2.P1.pdf