Offender–Offence Profiling: Improving Burglary Solvability and Detection

Abstract

This chapter considers the feasibility of predicting the subsets of offenders likely to be responsible for different sorts of unsolved burglaries by examining the distinctive associations between types of domestic burglaries and types of offenders. It extends our understanding of offender profiling to a high-volume offence and, by drawing on high-quality police data from South Australia, helps place a profiling approach to the improvement of crime solvability and detection on stronger empirical foundations. This approach is reliant on the small proportion of solved burglary cases where the culprits have been identified and will enhance ‘known offender targeting’. The analysis uses a random sample of 349 offenders convicted for committing domestic burglary in Adelaide in 2012 for which there are key data measuring offender characteristics, offending histories, and offence and dwelling characteristics. Using latent class analysis, the findings identify five offence types, three offender types and six offending history types. It was possible to link offence types with offender types, and offender types with offending history clusters, but a link directly from offence type to offending history type was not significant. Furthermore, the effect size of the relationship between offence type and offender type is relatively weak, and it appears doubtful whether it can be used to identify with sufficient accuracy the offender subsets in Adelaide’s locales. Offender–offence profiling may, therefore, make only small improvements to the solvability of high-volume crimes like domestic burglary.