Craig Bennell

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.

The criminal profiling illusion: what's behind the smoke and mirrors?

There is a belief that criminal profilers can predict a criminals characteristics from crime scene evidence. In this article, the authors argue that this belief may be an illusion and explain how people may have been misled into believing that criminal profiling (CP) works despite no sound theoretical grounding and no strong empirical support for this possibility. Potentially responsible for this illusory belief is the information that people acquire about CP, which is heavily influenced by anecdotes, repetition of the message that profiling works, the expert profiler label, and a disproportionate emphasis on correct predictions. Also potentially responsible are aspects of information processing such as reasoning errors, creating meaning out of ambiguous information, imitating good ideas, and inferring fact from fiction. The authors conclude that CP should not be used as an investigative tool because it lacks scientific support.

Addressing problems with traditional crime linking methods using receiver operating characteristic analysis

Purpose. Through an examination of serial rape data, the current article presents arguments supporting the use of receiver operating characteristic (ROC) analysis over traditional methods in addressing challenges that arise when attempting to link serial crimes. Primarily, these arguments centre on the fact that traditional linking methods do not take into account how linking accuracy will vary as a function of the threshold used for determining when two crimes are similar enough to be considered linked. Methods. Considered for analysis were 27 crime scene behaviours exhibited in 126 rapes, which were committed by 42 perpetrators. Similarity scores were derived for every possible crime pair in the sample. These measures of similarity were then subjected to ROC analysis in order to (1) determine threshold-independent measures of linking accuracy and (2) set appropriate decision thresholds for linking purposes. Results. By providing a measure of linking accuracy that is not biased by threshold placement, the analysis confirmed that it is possible to link crimes at a level that significantly exceeds chance (AUC = .75). The use of ROC analysis also allowed for the identification of decision thresholds that resulted in the desired balance between various linking outcomes (e.g. hits and false alarms). Conclusions. ROC analysis is exclusive in its ability to circumvent the limitations of threshold-specific results yielded from traditional approaches to linkage analysis. Moreover, results of the current analysis provide a basis for challenging common assumptions underlying the linking task.

Sexual Offenses Against Children as the Abusive Exploitation of Conventional Adult-Child Relationships:

A study was conducted to determine whether the relationship processes underlying offender-child interactions in child sexual abuse could be considered as abusive and manipulative variants of more conventional relationship processes that exist between adults and children. To explore this possibility, it was proposed that a circular order of behaviors (i.e., a circumplex) would exist in offender-child interactions that relates to circular models found in more conventional adult-child interactions. Furthermore, just as biases are found in conventional adult-child interactions towards particular relationship patterns, it was proposed that a bias would exist in offender-child interactions reflecting the predominant way in which child sex offenses are carried out. Using a multivariate statistical analysis that geometrically represents the co-occurrence of individual actions, the behaviors from 97 British child sex offenses were analyzed. The results provide preliminary support for the idea that the coercive nature of offender-child interactions in child sexual abuse relies on the offender gaining and abusing the trust of the victim by exploiting a range of conventional adult-child relationship patterns.

The impact of data degradation and sample size on the performance of two similarity coefficients used in behavioural linkage analysis

In order to determine whether a series of unsolved crimes has been committed by the same offender, the police often must rely on an analysis of behavioural evidence. When carrying out this task, some type of similarity coefficient is typically relied on to assess the degree of behavioural stability and distinctiveness that exists across a set of crimes and questions inevitably arise as to which coefficient to use. In cases of juvenile sex offences, research has suggested that a taxonomic similarity index outperforms the most commonly used metric at the moment, Jaccards coefficient, especially under conditions of data degradation (missing data). However, recent research has failed to replicate this result in cases of serial homicide and burglary, especially when relatively large sample sizes are used. The current study provides further support for these recent findings using adult serial sexual assault data. Across a range of conditions, the current study demonstrates that Jaccards coefficient slightly outperforms the taxonomic similarity index on a measure of linking accuracy. Potential explanations for the results are provided, implications are discussed, and future research directions are presented.

It’s no Riddle, Choose the Middle The Effect of Number of Crimes and Topographical Detail on Police Officer Predictions of Serial Burglars’ Home Locations

This study examines the effect of the number of crimes and topographical detail on police officer predictions of serial burglars’ home locations. Officers are given 36 maps depicting three, five, or seven crime sites and topographical or no topographical details. They are asked to predict, by marking an X on the map, where they thought each burglar lived. After making their predictions on half of the maps, officers randomly receive either no training or training in one of two simple decision-making strategies. The accuracy of predictions at baseline and retest is measured as the distance between the predicted and actual home locations, and these accuracy scores are compared to a commonly used geographic profiling system. Results show that training significantly improved predictive accuracy, regardless of the number of crime locations or topographical detail presented. In addition, trained participants are as accurate as the geographic profiling system.

Validities and Abilities in Criminal Profiling: A Critique of the Studies Conducted by Richard Kocsis and His Colleagues

In a recent issue of this journal, Kocsis reviewed the criminal profiling research that he and his colleagues have conducted during the past 4 years. Their research examines the correlates of profile accuracy with respect to the skills of the individual constructing the profile, and it has led Kocsis to draw conclusions that are important to the profiling field. In this article, the authors review the contributions of the Kocsis studies and critique their methodological and conceptual foundations. The authors raise a number of concerns and argue that data from the Kocsis studies fail to support many of the conclusions presented in his recent review. The authors present evidence in support of their assertions and provide recommendations that will allow future research in the area to generate data that are more meaningful and generalizable.

Another look at across-crime similarity coefficients for use in behavioural linkage analysis: an attempt to replicate Woodhams, Grant, and Price (2007)

Abstract In the absence of physical evidence, investigators must often rely on offence behaviours when determining whether several crimes are linked to a common offender. A variety of factors can potentially influence the degree to which accurate linking is possible, including the similarity coefficient used to assess across-crime similarity. The current study examines the performance of two similarity coefficients that have recently been compared to one another, Jaccards coefficient (J) and the taxonomic similarity index (Δs), using samples of two crime types, serial homicide (N=237) and serial burglary (N=210). In contrast to previous research, the results indicate that Δs does not significantly outperform J with respect to linking accuracy. In addition, both coefficients lead to higher levels of linking accuracy in cases of serial homicide compared to serial burglary. Potential explanations for these findings are presented and their implications are discussed.

Problems of classification in investigative psychology

Problems of classification in the field of Investigative Psychology are defined and examples of each problem class are introduced. The problems addressed are behavioural differentiation, discrimination among alternatives, and prioritisation of investigative options. Contemporary solutions to these problems are presented that use smallest space analysis, receiver operating characteristic analysis, and probability functions.

The Violent Crime Linkage Analysis System A Test of Interrater Reliability

The interrater reliability of an internationally renowned crime linkage system—the Violent Crime Linkage Analysis System (ViCLAS)—was tested. Police officers (N = 10) were presented with a case file and asked to complete a ViCLAS booklet. The level of occurrence agreement between each officer was calculated. Results showed a 30.77% level of agreement across the 106 variables examined. Agreement ranged from 2.36% for weapon variables to 62.87% for administration variables. Only 11 (10.38%) of the variables reached an acceptable level of agreement. Concerns pertaining to the validity of inferences produced using ViCLAS data are discussed, along with potential explanations for the findings, limitations of the study, and future research directions.