Linking commercial burglaries by modus operandi: tests using regression and ROC analysis.

This paper uses statistical models to test directly the police practice of utilising modus operandi to link crimes to a common offender. Data from 86 solved commercial burglaries committed by 43 offenders are analysed using logistic regression analysis to identify behavioural features that reliably distinguish between linked and unlinked crime pairs. Receiver operating characteristic analysis is then used to assign each behavioural feature an overall level of predictive accuracy. The results indicate that certain features, in particular the distances between burglary locations, lead to high levels of predictive accuracy. This study therefore reveals some of the important consistencies in commercial burglary behaviour. These have theoretical value in helping to explain criminal activity. They also have practical value by providing the basis for a diagnostic tool that could be used in comparative case analysis.

Differentiating stranger murders: profiling offender characteristics from behavioral styles.

The literature on "offender profiling" suggests that different "styles" of homicide will reflect differences in the background characteristics of the offender. To test this suggestion, hypotheses were drawn from studies of aggression to propose that murder crime scenes would reveal stylistic distinctions in the role of the aggression in the offense. It was further hypothesized that these distinctions would be clearest for those crime scene indicators that reflect the instrument (cognitive) actions that shape the offense rather than the more expressive (impulsive and emotional) components. Hypotheses about associated offender characteristics were also deduced on the assumption that the murder scene theme revealed the killers' typical styles of interpersonal transaction. To test these hypotheses MDS analyses were carried out on the crime scene and offender characteristics derived from 82 single offender, single victim stranger homicides. A multivariate structure resulted, including all three hypothesized styles, allowing 65% of the cases to be assigned to unique styles and a further 36% to be assigned to appropriate hybrids. Offender characteristics related in the anticipated way to the different crime scene styles, providing a basis for law enforcement inferences about offenders in stranger murders.

Identifying the residential location of rapists.

The spatial offence behaviour of 45 British serial rapists was examined in order to generate a predictive model, of use to criminal investigators. Two alternative hypotheses were explored. One predicted that rapists would commute into an area to carry out their offence. The second predicted that they would "maraud" out from a fixed location. Of the 45 offenders, 39 fitted the "maurader" hypothesis. However, the area covered by this model was an average of nearly 180 square miles. A second complementary theory, developed from facets of offenders' backgrounds, was therefore used to refine the predictions of distance travelled to and between offences. This enabled the size of the residential zone predicted from the marauder model to be reduced to a mean area of just over ten square miles. Tests of these models, combined into a small scale expert system, predicted the correct area for 82% of the cases. Suggestions for the further development of this expert system are discussed.