Behavioural case linkage with solved and unsolved crimes

• Published in: Forensic science international Vol. 222; no. 1; pp. 146 - 153
• Main Authors: Tonkin, Matthew, Woodhams, Jessica, Bull, Ray, Bond, John W
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ireland Ltd 10.10.2012; Elsevier; Elsevier Limited
• Subjects: Accuracy; Behavior; Behavioural; Behavioural case linkage; Biological and medical sciences; Categories; Cluster analysis; Comparative case analysis; Crime; Crime - statistics & numerical data; Criminal investigations; Criminals; Databases, Factual; Deoxyribonucleic acid; Discrimination; DNA; Evidence; Forensic medicine; Forensic Psychiatry; Forensic sciences; General aspects; Genetic testing; Humans; Investigative techniques, diagnostic techniques (general aspects); Linkage analysis; Linkages; Medical sciences; Murders & murder attempts; Pathology; Prosecutions; Public health. Hygiene; Public health. Hygiene-occupational medicine; Rape; Regression Analysis; Residential; Robbery; ROC Curve; Sample size; Samples; Serial crime; Studies; Temporal logic; Time Factors; United Kingdom; Unsolved crime; Unsolved crimes; United Kingdom > UK; England; Wales; Behavioural case linkage; Serial crime; Comparative case analysis; Unsolved crime; Linkage analysis; Forensic science; Legal medicine; Pathogenesis; Forensic aspect; Murder; Genetic determinism; Case study; Analysis; Crime; Genetic linkage; Genetics; Behavior; Comparative study
• ISSN: 0379-0738; 1872-6283
• DOI: 10.1016/j.forsciint.2012.05.017

Abstract Offender behaviour is used to distinguish between crimes committed by the same person (linked crimes) and crimes committed by different people (unlinked crimes) through behavioural case linkage. There is growing evidence to support the use of behavioural case linkage by investigative organisations such as the police, but this research is typically limited to samples of solved crime that do not reflect how this procedure is used in real life. The current paper extends previous research by testing the potential for behavioural case linkage in a sample containing both solved and unsolved crimes. Discrimination accuracy is examined across crime categories (e.g. a crime pair containing a car theft and a residential burglary), across crime types (e.g. a crime pair containing a residential burglary and a commercial burglary), and within crime types (e.g. a crime pair containing two residential burglaries) using the number of kilometres (intercrime distance) and the number of days (temporal proximity) between offences to distinguish between linked and unlinked crimes. The intercrime distance and/or the temporal proximity were able to achieve statistically significant levels of discrimination accuracy across crime categories, across crime types, and within crime types as measured by Receiver Operating Characteristic (ROC) analysis. This suggests that behavioural case linkage can be used to assist the investigation, detection and prosecution of prolific and versatile serial offenders.