Article

Linking Crimes Using Behavioural Clues: Current Levels of Linking Accuracy and Strategies for Moving Forward

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

The number of published studies examining crime linkage analysis has grown rapidly over the last decade, to the point where a special issue of this journal has recently been dedicated to the topic. Many of these studies have used a particular measure (the area under the receiver operating characteristic curve, or the AUC) to quantify the degree to which it is possible to link crimes. This article reviews studies that have utilised the AUC and examines how good we are currently at linking crimes (within the context of these research studies) and what factors impact linking accuracy. The results of the review suggest that, in the majority of cases, moderate levels of linking accuracy are achieved. Of the various factors that have been examined that might impact linking accuracy, the three factors that appear to have the most significant impact are crime type, behavioural domain, and jurisdiction. We discuss how generalisable these results are to naturalistic investigative settings. We also highlight some of the important limitations of the linking studies that we reviewed and offer up some strategies for moving this area of research forward.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... In contrast, crime linkage analysis can be carried out using behavioral, temporal, and/or geographic evidence that can potentially be collected at much lower cost and analyzed more quickly (Bennell et al., 2014). i In recent years, there has been an increased focus on studying the processes underlying the crime linkage task, and in evaluating the degree to which it is possible to accurately link crimes using behavioral, temporal, and geographic information (Bennell et al., 2014). One of the most popular methods for assessing the degree to which this task can be accomplished accurately involves the use of receiver operating characteristic (ROC) analysis (e.g., Bennell & Canter, 2002;Davidson & Petherick, 2020;Tonkin et al., 2011). ...
... Like other twoalternative (yes-no) type diagnostic decisions that must be made based on ambiguous evidence, the goals when using this approach are to identify what behaviors are best suited for distinguishing between crimes committed by the same offender versus different offenders, and to determine how similar two crimes should be before a decision is made that they have been committed by the same offender (i.e., establish an appropriate decision threshold; Bennell, 2005;Swets et al., 2000). Research has shown that it is possible to achieve both these goals and to accomplish the pairwise crime linkage task with a reasonable degree of accuracy (Bennell et al., 2014). ...
... In this case, across-crime similarity scores refer to scores derived from some type of similarity index, which is used to quantify the degree of similarity that exists across a crime pair. When relying on behavioral information, numerous similarity coefficients can potentially be used for this purpose (e.g., simple matching coefficient, taxonomic similarity index; Ellingwood et al., 2013;Melnyk et al., 2011), but Jaccard's coefficient is the most commonly used measure for crime linkage analysis (Bennell et al., 2014). ii When temporal or geographic information is being relied on, across-crime similarity can be established using simple measures of time differences or inter-crime distances. ...
Article
Deciding whether two crimes have been committed by the same offender or different offenders is an important investigative task. Crime linkage researchers commonly use receiver operating characteristic (ROC) analysis to assess the accuracy of linkage decisions. Accuracy metrics derived from ROC analysis—such as the area under the curve (AUC)—offer certain advantages, but also have limitations. This paper describes the benefits that crime linkage researchers attribute to the AUC. We also discuss several limitations in crime linkage papers that rely on the AUC. We end by presenting suggestions for researchers who use ROC analysis to report on crime linkage. These suggestions aim to enhance the information presented to readers, derive more meaningful conclusions from analyses, and propose more informed recommendations for practitioners involved in crime linkage tasks. Our reflections may also benefit researchers from other areas of psychology who use ROC analysis in a wide range of prediction tasks.
... Studies over the last two decades have shown that serial offenders exhibit a reasonable degree of behavioural similarity and distinctiveness across their crimes, which makes it possible to link serial offences in a reasonably accurate fashion. Indeed, in a review of 19 studies that used receiver operating characteristic (ROC) analysis to assess the accuracy of statistical linking methods, Bennell et al. (2014) found accuracy levels (defined as the area under the ROC curve, or AUC) ranging from 0.45 (just below chance level accuracy) to 0.96 (near perfect accuracy). Using pre-established guidelines for interpreting AUCs (Swets 1988), 36% of the AUCs included in the review were in the low range (0.50-0.70), 49% were in the moderate range (0.70-0.90), and 14% were in the high range (0.90-1.00). ...
... One of the most common statistical methods examined in crime linkage studies is logistic regression (LR) analysis (Bennell et al. 2014). This method is used to determine the linkage status of crime pairs (whether they were committed by the same vs. ...
... To make binary, forced-choice linkage decisions, 1 any crime pair associated with a probability value above a pre-specified threshold is predicted to be linked. 2 While LR analysis is often associated with moderate to high AUCs when used for this purpose (Bennell et al. 2014), it typically results in nomothetic linking strategies (i.e. the same predictor variables are applied the same way in every case) that will miss potential links (Tonkin et al. 2012b). For example, a LR model might always result in a prediction that two crimes are linked when the distance between those crimes is small (e.g. ...
Article
Full-text available
Studies have shown that it is possible to link serial crimes in an accurate fashion based on the statistical analysis of crime scene information. Logistic regression (LR) is one of the most common statistical methods in use and yields relatively accurate linking decisions. However, some research suggests there may be added value in using classification tree (CT) analysis to discriminate between offences committed by the same vs. different offenders. This study explored how three variations of CT analysis can be applied to the crime linkage task. Drawing on a sample of serial sexual assaults from Quebec, Canada, we examine the predictive accuracy of standard, iterative, and multiple CTs, and we contrast the results with LR analysis. Our results revealed that all statistical approaches achieved relatively high (and similar) levels of predictive accuracy, but CTs produce idiographic linking strategies that may be more appealing to practitioners. Future research will need to examine if and how these CTs can be useful as decision aides in operational settings.
... The past 15-20 years have seen case linkage research and practice flourish, with support for linkage practice found across high-volume offences, as well as interpersonal offences (Bennell et al., 2010(Bennell et al., , 2014Burrell and Bull, 2011;Grubin et al., 2001;Labuschagne, 2006;Santtila et al., 2008;Slater et al., 2015;Tonkin and Woodhams, 2017;Tonkin et al., 2011Tonkin et al., , 2012Winter et al., 2013;Woodhams et al., 2007bWoodhams et al., , 2019Woodhams and Labuschagne, 2012;Woodhams and Toye, 2007). As research and investigative methods continue to develop, specialised databases have been created to assist in the case linkage process, such as the Canadian Violent Crime Linkage Analysis System (ViCLAS), the Violent Criminal Apprehension Program of the United States of America (USA) and Australia's Violent and Sexual Crimes Database (VSCD). ...
... Research using behavioural evidence supports linkage accuracy to varying degrees across a range of offences. Bennell et al. (2014) found that linking accuracy was higher for interpersonal crime, such as homicide and sexual assault than it was for property crime. The highest accuracy when using individual variables and behaviours occurred when inter-crime distance and temporal proximity were used compared to modus operandi (MO) behaviours (Bennell et al., 2014), although the inclusion of MO behaviours, as well as temporal and intercrime distance has resulted in greater linking accuracy (Tonkin and Woodhams, 2017). ...
... Bennell et al. (2014) found that linking accuracy was higher for interpersonal crime, such as homicide and sexual assault than it was for property crime. The highest accuracy when using individual variables and behaviours occurred when inter-crime distance and temporal proximity were used compared to modus operandi (MO) behaviours (Bennell et al., 2014), although the inclusion of MO behaviours, as well as temporal and intercrime distance has resulted in greater linking accuracy (Tonkin and Woodhams, 2017). In general, using MO behaviours results in higher linking success for interpersonal crimes such as rape and homicide, while temporal and spatial behaviours have more success in crimes such as burglary and robbery (Woodhams and Bennell, 2015;Woodhams and Toye, 2007). ...
Article
Full-text available
Purpose Case linkage theory and practice have received growing empirical support; however, they have yet to be examined fully within Australia. For sexual assault case linkage to be successful, it is assumed that a serial rapist will behave relatively consistently across offences yet distinctively compared to other offenders. The purpose of this paper is to test the underlying principles of case linkage, behavioural consistency and distinctiveness, as well as distinguishing accuracy. Design/methodology/approach In total, 250 solved stranger rapes by 171 offenders (46 serial rapists, 125 one-off rapists) were taken from Queensland Police Service (QPS) crime records. All possible crime pairings were created and cross-crime similarity was assessed using Jaccard’s coefficient. Receiver operating characteristics (ROC) analysis was used to examine the ability to distinguish between linked and unlinked offence pairs. Findings Serial linked pairs had the highest Jaccard’s coefficient (0.456), followed by non-serial unlinked (0.253) and finally, serial unlinked pairs (0.247). Within the ROC analysis, an area under the curve value was found of 0.913, indicating excellent distinguishing accuracy. Both the underlying principles of behavioural consistency and distinctiveness were supported through theoretical and practical methods. This paper provides the first analysis of serial rape case linkage in Australia, adding validity to this practice. Research limitations/implications The authors wish to acknowledge the support and assistance from the QPS in undertaking this research. The views expressed in this publication are not necessarily those of the QPS and any errors of omission or commission are the responsibility of the authors. Practical implications This paper provides validity to the practice of case linkage using a database within Australia. The results of this paper can be used to inform investigators of serial offender behaviours. The theories of offender consistency and distinctiveness are supported, highlighting the importance of behavioural evidence for practitioners. This paper provided a practical increase of the quantity and quality of offences uploaded on the Australian violent and sexual crimes database, which will assist further linkage efforts. Originality/value This paper is the first in Australia to examine consistency, distinctiveness and case linkage of serial stranger rape. Thus is contributes significantly not only to an increased understanding of serial rape and case linkage in Australia but also brings Australia closer to modern research practices in this field.
... Receiver Operating Characteristic (ROC) analysis is the preferred measure of predictive accuracy in forensic psychology (Harris & Rice, 1995) and is commonly used to quantify the accuracy with which behavioural similarity can be used to predict series membership or linkage status (linked/unlinked) (see Bennell, Mugford, Ellingwood, &Woodhams, 2014, andWinter et al., 2013). It has four possible outcomes: a hit (where a pair is predicted to have been committed by the same offender and was), a false alarm (where a pair is predicted to have been committed by the same offender but was not), a correct rejection (where the two crimes in a pair are predicted to have been committed by two different offenders and they were) and a miss (where the two crimes in a pair are predicted to have been committed by two different offenders and were committed by the same offender). ...
... Receiver Operating Characteristic (ROC) analysis is the preferred measure of predictive accuracy in forensic psychology (Harris & Rice, 1995) and is commonly used to quantify the accuracy with which behavioural similarity can be used to predict series membership or linkage status (linked/unlinked) (see Bennell, Mugford, Ellingwood, &Woodhams, 2014, andWinter et al., 2013). It has four possible outcomes: a hit (where a pair is predicted to have been committed by the same offender and was), a false alarm (where a pair is predicted to have been committed by the same offender but was not), a correct rejection (where the two crimes in a pair are predicted to have been committed by two different offenders and they were) and a miss (where the two crimes in a pair are predicted to have been committed by two different offenders and were committed by the same offender). ...
... More than a decade of research testing the crime linkage principles exists and the general conclusion from this body of research is that the principles are empirically supported to an extent (Bennell et al., 2014): some serial offenders show sufficient behavioural consistency and distinctiveness for their crimes to be linked; however, some offenders and some series are characterised by inconsistent and/or indistinct behaviour (Slater, Woodhams, & Hamilton-Giachristis, 2015). However, most of these research studies have sampled series confirmed by conviction. ...
Article
Full-text available
Purpose To conduct a test of the principles underpinning crime linkage (behavioural consistency and distinctiveness) with a sample more closely reflecting the volume and nature of sexual crimes with which practitioners work, and to assess whether solved series are characterized by greater behavioural similarity than unsolved series. Method A sample of 3,364 sexual crimes (including 668 series) was collated from five countries. For the first time, the sample included solved and unsolved but linked‐by‐DNA sexual offence series, as well as solved one‐off offences. All possible crime pairings in the data set were created, and the degree of similarity in crime scene behaviour shared by the crimes in each pair was quantified using Jaccard's coefficient. The ability to distinguish same‐offender and different‐offender pairs using similarity in crime scene behaviour was assessed using Receiver Operating Characteristic analysis. The relative amount of behavioural similarity and distinctiveness seen in solved and unsolved crime pairs was assessed. Results An Area Under the Curve of .86 was found, which represents an excellent level of discrimination accuracy. This decreased to .85 when using a data set that contained one‐off offences, and both one‐off offences and unsolved crime series. Discrimination accuracy also decreased when using a sample composed solely of unsolved but linked‐by‐DNA series (AUC = .79). Conclusions Crime linkage is practised by police forces globally, and its use in legal proceedings requires demonstration that its underlying principles are reliable. Support was found for its two underpinning principles with a more ecologically valid sample.
... delved into the underlying assumptions of computerized crime linkage systems. In a subsequent work, authors in [38] analyzed evaluation measures of crime linkage, particularly focusing on the AUC (Area Under the Curve), and identified crime type, behavior domain, and distance as key factors affecting performance. While predominantly focused on criminal profiling, Fox et al. [5] dedicated a section to discussing the 17 papers compiled by Bennell et al. (2014), with a focus on the effect size of the studied samples. ...
... In a subsequent work, authors in [38] analyzed evaluation measures of crime linkage, particularly focusing on the AUC (Area Under the Curve), and identified crime type, behavior domain, and distance as key factors affecting performance. While predominantly focused on criminal profiling, Fox et al. [5] dedicated a section to discussing the 17 papers compiled by Bennell et al. (2014), with a focus on the effect size of the studied samples. Davies et al. [39] conducted the most recent literature review on the topic, describing the challenges to obtain crime linkage in practice. ...
Preprint
Full-text available
Crime linkage is the process of analyzing criminal behavior data to determine whether a pair or group of crime cases are connected or belong to a series of offenses. This domain has been extensively studied by researchers in sociology, psychology, and statistics. More recently, it has drawn interest from computer scientists, especially with advances in artificial intelligence. Despite this, the literature indicates that work in this latter discipline is still in its early stages. This study aims to understand the challenges faced by machine learning approaches in crime linkage and to support foundational knowledge for future data-driven methods. To achieve this goal, we conducted a comprehensive survey of the main literature on the topic and developed a general framework for crime linkage processes, thoroughly describing each step. Our goal was to unify insights from diverse fields into a shared terminology to enhance the research landscape for those intrigued by this subject.
... One key difference in methodological approaches is whether researchers are using algorithms to distinguish between linked and unlinked pairs (e.g. see [3], for a review of such studies) or whether they are using them to assign a given crime to a particular series, e.g. [23]. ...
... The reason for this is that the first step of ILASP's algorithm is always to compute the hypothesis space in full. 3 In recent years the FastLAS systems [14] have been developed to solve restricted versions of ILASP's learning task without the need to compute the full hypothesis space. Until now, FastLAS has not been able to learn weak constraints. ...
Chapter
Comparative Case Analysis is an analytical process used to detect serial offending. It focuses on identifying distinctive behaviour that an offender displays consistently when committing their crimes. In practice, crime analysts consider the context in which each behaviour occurs to determine its distinctiveness, which subsequently impacts on their determination of whether crimes are committed by the same person or not. Existing algorithms do not currently consider context in this way when generating linkage predictions.This paper presents the first learning-based approach aimed at identifying contexts within which behaviour may be considered more distinctive. We show how this problem can be modelled as that of learning preferences (in answer set programming) from examples of ordered pairs of contexts in which a behaviour was observed. In this setting, a context is preferred to another context if the behaviour is rarer in the first context. We make novel use of odds ratios to determine which examples are used for learning. Our approach has been applied to a real dataset of serious sexual offences provided by the UK National Crime Agency. The approach provides (i) a systematic methodology for selecting examples from which to learn preferences; (ii) novel insights for practitioners into the contexts under which an exhibited behaviour is more rare.KeywordsCrime LinkageAnswer Set ProgrammingInductive Logic Programming
... Although forensic evidence (e.g., DNA) is obviously the preferred method for establishing crime linkages, sufficient evidence is often not available in cases involving sexual violence to carry out linkage analysis in this way (House of Commons, 2005). Given this unfortunate fact, procedures have been developed for establishing crime linkages on the basis of behavioral (and other non-forensic) evidence (Woodhams & Bennell, 2014). ...
... Since the time of Grubin et al.'s study, the number of empirical studies examining behavioral crime linkage analysis in cases of sexual violence has grown (e.g., Bennell et al., 2009;Deslauriers-Varin, & Beauregard, 2014a;Harbers, Deslauriers-Varin, Beauregard, & van der Kemp, 2012;Slater, Woodhams, & Hamilton-Giachritsis, 2015;Woodhams, Grant, & Price, 2007). Generally speaking, this research suggests that it may be possible to use behavioral evidence to link such crimes (Bennell, Mugford, Ellingwood, & Woodhams, 2014). This appears to be particularly true when using certain types of behavioral information. ...
Chapter
Sexual assault and abuse are among the most heinous crimes that can be committed. However, these crimes are largely underreported worldwide. Despite this, research on the investigation of sexual assault and abuse cases has been slow to emerge and there exists an important gap between the current knowledge on sexual victimization and the police response to it. Research is required on many topics related to the investigation of sexual assault and abuse, including research that can improve the effectiveness of investigative interviews and the accurate identification and prioritization of suspects. The goal of the current chapter is to review current research efforts in these two areas –investigative interviewing and suspect identification and prioritization procedures – and to discuss some of the challenges associated with these tasks , and the investigation of sexual assault and abuse more generally (e.g., false allegations, false confessions, cognitive biases).
... Behavioural crime linking has been a fast growing topic of interest in the field of forensic psychology research (Bennell et al., 2014). However, a vast majority of the studies conducted have looked at serial offences only. ...
... Woodhams et al., 2007). A number of studies have found evidence for the assumptions of crime linkage in serial crime (for a review of the research to date on behavioural crime linkage, see Bennell et al., 2014). While serial homicide has been studied less than, for example, rape, a few studies have found serial homicide offenders to be somewhat consistent in their behaviour over the course of their series (Bateman and Salfati, 2007;Salfati and Bateman, 2005;Salfati et al., 2015;Santtila et al., 2008;Sorochinski and Salfati, 2010). ...
Article
Full-text available
Purpose – The purpose of this paper is to explore the differences (if any) between serial and hard-to-solve one-off homicides, and to determine if it is possible to distinguish the two types of homicides based on offence behaviours and victim characteristics. Design/methodology/approach – A sample of 116 Italian serial homicides was compared to 45 hard-to-solve one-off homicides. Hard-to-solve one-off homicides were defined as having at least 72 hours pass between when the offence came to the knowledge of the police and when the offender was caught. Logistic regression was used to predict whether a killing was part of a series or a one-off offence. Findings – The serial killers targeted more strangers and prostitutes, displayed a higher level of forensic awareness both before and after the killing, and had more often an apparent sexual element in their offence. Conversely, the one-off homicides were found to include more traits indicative of impulsive and expressive behaviour. The model demonstrated a good ability (AUC=0.88) to predict whether a homicide belonged to the serial or one-off category. Research limitations/implications – The findings should be replicated using local homicide data to maximise the validity of the model in countries outside of Italy. Practical implications – Being able to distinguish between serial and one-off homicides based on information available at a new crime scene could be practically useful for homicide investigators managing finite resources. Originality/value – Studies comparing serial homicides to one-off homicides are scarce, and there are no studies explicitly trying to predict whether a homicide is an isolated case or part of a series.
... The studies conducted in this period have provided some evidence for the assumptions underpinning crime linkage (behavioral consistency and distinctiveness) with a variety of different crime types, and research has demonstrated that predictions about linkage status can be made that are significantly better than chance (e.g., Bennell & Canter, 2002;Bennell & Jones, 2005;Burrell, Bull, & Bond, 2012;Davies, Tonkin, Bull, & Bond, 2012;Ellingwood, Mugford, Bennell, Melnyk, & Fritzon, 2013;Salo et al., 2012;Santtila, Junkkila, & Sandnabba, 2005;Santtila et al., 2008;Tonkin, Santtila, & Bull, 2011;Tonkin, Woodhams, Bull, Bond, & Palmer, 2011;Woodhams & Labuschagne, 2012;Woodhams & Toye, 2007). However, a thorough qualitative review of 19 published studies by Bennell, Mugford, Ellingwood, and Woodhams (2014) revealed a moderate range of crime linkage accuracy for different types of offending. In addition, there are several limitations with regard to crime linkage studies that make this enterprise for researchers and practitioners alike less straightforward than it might seem. ...
... This raises the question of whether it is necessary to code for a large number of behaviors in police databases for the purpose of carrying out crime linkage analysis instead of relying on a smaller, but more reliable number of offense characteristics (Grubin, Kelly, & Brunsdon, 2001;Santtila et al., 2008). * * However, it should be noted, as Bennell et al. (2014) pointed out, that researchers have mostly focused on behavioral information and have tended to neglect other available nonbehavioral features (such as ethnicity, distinctive physical features, type of clothing, vehicle characteristics, and weapon description) that could potentially play a pivotal role in real-life crime linkage and that are stored in such databases. ...
... The last two decades has seen a sharp increase in BCL research. Earlier reviews, focusing exclusively on BCL (Woodhams, Hollin, & Bull, 2007;Bennell, Mugford, Ellingwood, & Woodhams, 2014), have concluded that both the assumptions of consistency and distinctiveness have been supported by a good amount of empirical evidence. BCL has been shown to be viable across a variety of crime types, in samples from all over the world: the United Kingdom, the United States, Canada, Finland, South Africa, Japan, Italy and Australia. ...
Thesis
Full-text available
Behavioural crime linking refers to the practice of trying to tie two or more offences to the same offender using behaviour observable at the crime scene. It rests on the assumptions that offenders behave consistently enough from one offence to another, and distinctively enough from other offenders allowing offences to be successfully linked together. Conceptualised in the 70s, and developed methodologically with increased scientific rigour from the 90s, the last decade has seen a sharp rise in published studies on behavioural crime linking. From empirical validation of the underlying assumptions to mapping out practice and more ecologically valid tests of linkage accuracy, the field has developed considerably. Considering that investigating homicide is resource intensive, not to mention serial homicide, reliable and valid behavioural crime linking has the potential to aid and prioritise investigative avenues and help solve serial homicide. Most studies on serial homicide have been carried out on North American samples. While some research has studied the consistency and distinctiveness of serial homicide offenders, few have empirically tested models of behavioural crime linking and linkage accuracy with serial homicide. Another shortcoming in behavioural crime linking research is the use of mostly serial cases to model crime linking, even though real crime databases include both serial and one-off offences. Some studies have tested the effect of added one-offs on the linkage accuracy of burglary and rape, but none so far the effect this would have on homicide. Additionally, while some studies have compared serial homicide offences to one-off homicides, none have tested whether it would be possible to predict whether a homicide belongs to a series or is a singular offence. Cognitive bias, especially confirmation bias or the expectancy effect, has been shown to have a considerable effect on crime investigation. No studies to date have explored the effect of such bias in behavioural crime linking. The general aim of the thesis was to increase ecological validity of behavioural crime linking research, especially with regard to sampling choices and analyses that strive to answer questions relevant for homicide investigation. The main sample consisted of 116 Italian serial homicides, committed in 23 separate series of homicide. Additionally, information about 45 cases of hard-to-solve one-off homicide was gathered, coded, and added to the sample. Study I found seven behavioural dimensions of offending (e.g., sexually motivated homicides and aspects of control-behaviour) in line with previous research. Notably, also other motives than sexual were found in the killings. A majority of offences (63%) were correctly classified to their actual series in the predictive part of the study. Study II was an experiment that investigated whether knowledge of series membership increased perceived (coded) behavioural similarity in homicides committed by the same offender. While no support was found for a strong expectancy effect, the experimental task may have lacked in sufficient complexity, and replication is thus needed. Study III found several key differences between serial and singular homicides and was able to successfully use these differences to predict with good accuracy whether an offence was part of a series. Study IV combined all the advances in the methodology thus far and showed that behavioural crime linking was still viable even with a large proportion (10:1) of one-off homicides added into the sample. As a function of added one-off homicides, the specificity of the model worsened (more false positives), as did the proportion of offences belonging to a series found near the top of a ranked listing from more behaviourally similar to less behaviourally similar. Overall model accuracy remained good, though, further validating the practice of behavioural crime linking with more ecologically valid data. The studies of the present thesis contribute to the methodology of behavioural crime linking research. Replication on local crime databases is needed to maximise the practical usefulness of the models in different jurisdictions. Going forward, a close-knit collaboration between researchers and practitioners is called for, to keep the research relevant for practice and to develop evidence-based practice. As we gain a clearer picture of the accuracy and error rate of behavioural crime linking models, their usefulness increase in both the criminal investigative phase and in the trial phase with behavioural crime linking being presented as expert evidence.
... For instance, most of the early offender profiles, such as those proposed by the Federal Bureau of Investigation (FBI)'s Behavioral Science Unit (BSU), utilize a clinical approach to OP, where the profiles are developed on a case-by-case basis by a "profiler" with considerable training, experience, and practice in the field; however, no consistent process or empirical methodology is utilized (Fox & Farrington, 2018, see also Holmes & Holmes, 1996). More recently, profiles have been developed using more objective, replicable, and scientific methods in sub-fields such as investigative psychology (Canter, 1995(Canter, , 2011, case linkage analysis (Bennell & Canter, 2002;Bennell et al., 2014), and evidence-based profiling (Farrington & Lambert, 2000;Fox & Farrington, 2012, 2015Fox et al., 2020). ...
Chapter
Full-text available
Offender profiling has been used by law enforcement agencies around the world for decades. Following homicide, offenses involving sexual assault are the second most common type of crime examined in offender profiling. Profiling sex crimes has successfully assisted investigators to find and convict sexual offenders and to deter future crimes. In this chapter, we focus on studies that developed profiles of sexual homicide and sexual assault to identify the personality characteristics, backgrounds, and mental states among offenders for these crimes, as well as analyze the similarities among sexual homicide and assault offenders, unique characteristics of these offenders, and how law enforcement can use these profiles to catch and prevent these offenders in the future.
... CLA, first developed by Bennell and Canter (2002), uses statistical analysis to link crimes to a single offender on the basis of commonalities seen in various crime scene behaviors. CLA typically uses receiver operating characteristic (ROC) analysis to evaluate behavioral similarity across offenses and determines the likelihood of case linkage (Bennell, Mugford, Ellingwood, & Woodhams, 2014). Next, an area under the ROC curve (AUC) analysis is used to determine the accuracy of the decision to link crimes to a serial offender . ...
Chapter
National security priorities, result-oriented pressures, and cost sensitivity are common features of contemporary policing. While the global shift to evidence-based policing (EBP) increased police reliance on behavioral science research on interrogation and interviews, skepticism about the effectiveness of “soft” science is pervasive and “hard” sciences have been privileged. Psychologists have consequently been challenged to fulfill their roles in compliance with the four key principles that underpin psychologists' codes of ethics, namely, respect for rights and dignity, competent caring, integrity, and social responsibility. This chapter reviews the alignment of these principles with the relational skills implicit in the four tenets of the leading theory in international police practice, procedural justice (PJ). An analysis of research constructs applied in contemporary interviewing research demonstrated the integral connection between these relational skills and effective interviewing of high-value detainees. These links are present both in a “ticking bomb” scenario as well as less exigent contexts. By mapping the links between ethical principles, PJ tenets, relational research constructs and outcomes, this chapter offers a potential framework for behavioral scientists in policing contexts to develop their ethical literacy and better articulate and evaluate potential ethical issues in their practice. Adherence to PJ tenets can reduce psychologists' role conflicts and facilitate the ethical practice of psychology and EBP.
... MDS vs. LCA typology) to predict binary diagnostic (e.g. cancer vs. non-cancer) decisions, is used here to evaluate the accuracy of the typologies (Bennell et al., 2014;Steadman et al., 2000). ...
Article
While there are a variety of statistical classification techniques available, the most prominent in the behavioral sciences are Cluster Analysis (CA), Multidimensional Scaling (MDS), and Latent Class Analysis (LCA). Researchers often rely on person-oriented statistical classification techniques to identify and understand the latent heterogeneity in the complex individuals that we study. Using data on 405 randomly selected solved burglaries committed in Florida, this study is the first to conduct a head-to-head comparison of the benefits and weakness of each analysis and evaluate the resultant typologies and predictive validity when all three analyses are applied to the same dataset. Findings suggest that the number and nature of resultant subtypes differ depending on the statistical classification technique employed. We conclude that LCA is superior to MDS and CA due to its ability to objectively evaluate model fit and handle missing data, balance of parsimony and complexity in the results, and reliability and accuracy stemming from the first test of predictive validity among the three statistical classification techniques. Implications for future research and the application and testing of statistical classification techniques are also discussed.
... One approach is a process created on a case-by-case basis by a profiler who has considerable training, experience, and practice in the field, with no clear process, methodology or systematic approach for integration across cases. 11,12 The other main approach involves profiling by using objective and replicable measures, such as Canter's investigative psychology 13,14 or case linkage analysis, 15,16 or evidence-based profiling. 17, 18 The purpose of this article is to propose the Trinity-approach for criminal profiling. ...
Article
Full-text available
Criminal profiling has been through many different periods during the time the field has existed, but none of which has resulted in a scientific validation of the field and thereby shown how it could be effectively used in the capable hands of law enforcement. Part of the problem is that there is no proper collaboration between law enforcement and the academic world from which the scientific validation could arise. This brief communication will give an overview of the main approaches used today, as well as explain the overall building blocks of a proposed Trinity-approach. The Trinity-approach consists of a geographical profile, which should lay the foundation of any profile. The reason for the geographical profile to be the foundation is that this type of profiling has shown reliable results compared to other types of profiling. The next step should be a profile of the victim and finally followed by the profile of the offender. The profile of the victim, as well as that of offender should come from a database containing extensive scientific studies within the field of criminal profiling. The strength of the Trinity-approach is a build-in scientific foundation, based on scientific studies that have undergone proper scientific review. The hope is to generate interest and debate, and call upon collaboration between law enforcement and academics to develop a scientific basis for the effectiveness of profiling as well as evaluate the reliability and validity of the Trinity-approach.
... As a limitation of the present study, it should also be noted that the present study was based on a limited sample in one rural area in Japan. Bennell et al. (2014) displayed the variation of linkage accuracy across countries and even across jurisdictions in the same country. Thus, the further comparisons of the effectiveness between the shortest route distance and the Euclidean distance in crime linkage analysis under different conditions (e.g. ...
Article
Purpose The purpose of this paper was to compare the accuracy of linking crimes using geographical proximity between three distance measures: Euclidean (distance measured by the length of a straight line between two locations), Manhattan (distance obtained by summing north-south distance and east-west distance) and the shortest route distances. Design/methodology/approach A total of 194 cases committed by 97 serial residential burglars in Aomori Prefecture in Japan between 2004 and 2015 were used in the present study. The Mann–Whitney U test was used to compare linked (two offenses committed by the same offender) and unlinked (two offenses committed by different offenders) pairs for each distance measure. Discrimination accuracy between linked and unlinked crime pairs was evaluated using area under the receiver operating characteristic curve (AUC). Findings The Mann–Whitney U test showed that the distances of the linked pairs were significantly shorter than those of the unlinked pairs for all distance measures. Comparison of the AUCs showed that the shortest route distance achieved significantly higher accuracy compared with the Euclidean distance, whereas there was no significant difference between the Euclidean and the Manhattan distance or between the Manhattan and the shortest route distance. These findings give partial support to the idea that distance measures taking the impact of environmental factors into consideration might be able to identify a crime series more accurately than Euclidean distances. Research limitations/implications Although the results suggested a difference between the Euclidean and the shortest route distance, it was small, and all distance measures resulted in outstanding AUC values, probably because of the ceiling effects. Further investigation that makes the same comparison in a narrower area is needed to avoid this potential inflation of discrimination accuracy. Practical implications The shortest route distance might contribute to improving the accuracy of crime linkage based on geographical proximity. However, further investigation is needed to recommend using the shortest route distance in practice. Given that the targeted area in the present study was relatively large, the findings may contribute especially to improve the accuracy of proactive comparative case analysis for estimating the whole picture of the distribution of serial crimes in the region by selecting more effective distance measure. Social implications Implications to improve the accuracy in linking crimes may contribute to assisting crime investigations and the earlier arrest of offenders. Originality/value The results of the present study provide an initial indication of the efficacy of using distance measures taking environmental factors into account.
... This is a universally used method in the analysis of perpetrators' behaviours (e.g. Bennell et al., 2014), and is also gaining popularity among Polish researchers (e.g. Gierowski, Jaśkiewicz-Obydzińska, 2002). ...
Article
The aim of the article was to study the distribution of distances travelled for a sample of the Polish population of perpetrators. The analysis covered burglaries and robberies in Warsaw. The results point to the key role of the analysis of the behaviour at the crime scene and its characteristics in the estimation of the journey to crime distance, which is partly consistent with the hypotheses put forward. The results show that perpetrators are ready to travel longer distances if the event is associated with a greater profit or increased risk. However, differences in the distribution of the distances travelled are not as significant as suggested by previous research. The reasons for this and potential consequences are discussed in this paper.
... The difficulty of translating research into practice has been highlighted by , who explained that research findings may not be useful in investigative settings because (a) the samples used in research may not be representative of samples found in real-life settings and (b) the nature of the linking tasks used in theoretical research may be different to those within a real-world investigative setting. Indeed, although some theoretical research does note the differences in crime linkage approaches (e.g., Bennell, Mugford, Ellingwood, & Woodhams, 2014), these practical differences are not always considered, meaning it may not be a straightforward process to infer the utility of theoretical research to crime linkage practice. ...
Article
Crime linkage has been the subject of increasing attention in academic research. Research has found support for the principles of behavioural consistency and distinctiveness, which underpin crime linkage, but this does not provide direct evidence as to whether crime linkage is useful in practice. This literature review draws together documentation that refers to the practice of crime linkage, from assessing analysts' efficacy, to discussing the usage of computerised tools to assist with the linkage process, to providing a comprehensive outline of the process itself. The implications of the amount and type of information currently available are discussed, including the variations in practice and terminology that were explored. Avenues for future investigation and the manner in which future research could be conducted are set out in a research agenda.
... Conversely, others conduct and have begun to evaluate OP using objective and replicable measures and 1 It is important to note that (1) there is not a single agreed upon definition for offender profiling used by all in the field, and (2) there are other related (and arguably synonymous) terms/concepts, including criminal profiling, criminal investigative analysis, and personality profiling, which are also used in the field based solely upon author preference and background, not substantive differences in terminology or meaning. scientific methods in subfields such as investigative psychology (Canter, 1995a;Canter, 2011), case linkage analysis (Bennell & Canter, 2002;Bennell, Mugford, Ellingwood, & Woodhams, 2014), and evidence-based profiling (Farrington & Lambert, 2000;Fox & Farrington, 2012. ...
Article
Full-text available
In the 4 decades since offender profiling (OP) was established, hundreds of journal articles, books, book chapters, reports, and magazine articles have been published on the topic, and the technique has been used by countless law enforcement agencies around the globe. However, despite the popularity and extensive literature published on OP, very little is known about its evolution, current state, or findings of the field to date. Therefore, this study presents a systematic review and meta-analysis of 426 publications on OP from 1976 through 2016. Results of this systematic review suggest that there have been considerable improvements in the scientific rigor and self-assessment being conducted in the discipline, although in total, few studies have used a strong empirical approach to develop new profiles. Even fewer evaluations of the effectiveness of OP have been conducted. The first summary of offender profiles proposed for major crimes in OP literature is also presented, with results indicating some recurrent themes in profiles, but wide variations in the number, name, and description of the profiles often found. A meta-analysis of case linkage analysis research indicates that this area is statistically sophisticated, and has yielded moderate to strong accuracy rates for linking crimes to a single offender. Finally, the first analysis of the most prolific authors, researchers, departments, and outlets for OP research, and the methods, approach, and most cited publications in OP are identified. Suggestions for future research on OP and the potential impact that this may have on policy and practice are also discussed. (PsycINFO Database Record (c) 2018 APA, all rights reserved).
... The task of determining whether multiple crimes were committed by the same offender is challenging [2,4]. The behaviors exhibited by offenders across the crimes are often used to establish crime linkages [31,44]. In [3,32] researchers use statistical modeling to detect changes in criminal activities. ...
Article
Full-text available
The paper introduces a solution to the criminal prediction problem using Naïve Bayes theory. The criminal prediction problem is stated as finding the most likely criminal of a particular crime incident when the history of crime incidents is given with the incident-level crime data. The incident-level crime data are assumed to be given as a crime dataset where the incident date and location, crime type, criminal ID and the acquaintances are the attributes or crime parameters considered in the paper. The acquaintances are the suspects whose names are either directly involved in the incident or indirectly the acquaintances of the criminal. Acquiring the crime dataset is a difficult process in practice due to confidentiality principle. So the crime dataset is generated synthetically using the state-of-the-art methods. The proposed system is tested for the criminal prediction problem using the cross-validation, and the experimental results show that the proposed system provides high scores in finding of suspected criminals.
... A DT has been defined as 'a way to represent rules underlying data with hierarchical, sequential structures that recursively partition the data' (Murthy, 1998, p. 345). DT is a technique that learns to recognise patterns in data and has performed well in various areas of application (Quinlan, 1993;Rokach, 2007;Wu & Kumar, 2009), including criminal and forensic investigations (Abbasi & Chen, 2005;Bennell, Mugford, Ellingwood, & Woodhams, 2014;Berk & Bleich, 2013;Gutierrez & Leroy, 2008;Ngai, Hu, Wong, Chen, & Sun, 2011;Yang & Olafsson, 2011). Furthermore, DTs are used as pedagogical support tools to produce easy-to-interpret models generally (Anaya, Luque, & García-Saiz, 2013;Breiman, 2001b;Jormanainen & Sutinen, 2012). ...
Article
Full-text available
The aim of this paper was to create a decision tree (DT) to identify personality profiles of offenders against public safety. A technique meeting this requirement was proposed that uses the C4.5 algorithm to derive decision rules for personality profiling of public safety offenders. The Mini-Mult test was used to measure the personality profiles of 238 individuals. With the test results as our database, a C4.5 DT was applied to construct rules that classify each profile into one of two groups, those without and those with records of offences against public safety. The model correctly classified 80% of the personality profiles and delivered a set of decision rules for distinguishing the profiles by group, and the principal personality profiles were interpreted. We conclude that DTs are a promising technique for analysing personality profiles by their offender or non-offender status. Finally, we believe that the development of a classifying model using DT may have practical applications in the Colombian prison system.
... Consequently, large AUC values that statistically exceed chance (AUC = 0.50) indicate that offender behavioural consistency and distinctiveness exist at a level that is sufficient to allow linked crimes to be accurately distinguished from unlinked crimes. 7 A recent review of the BCL literature using ROC analysis (Bennell, Mugford, Ellingwood, & Woodhams, 2014 ) suggested that 2% of the AUCs reported fall in the noninformative range (AUC < .50), 29% fall in the low range (AUC = 0.50 – 0.70), 54% fall in the moderate range (AUC = 0.70 – 0.90), and 15% fall in the high range (AUC > 0.90) (Swets, 1988). ...
Article
Purpose: To test whether geographical, temporal and Modus Operandi (MO) crime scene behaviours can be used to support behavioural case linkage (BCL) with crime series that contain several different types of offence. Methods: Crime scene data relating to 749 solved commercial burglaries and robberies were extracted from the databases of the Metropolitan Police Service, London, England. From these data, 2,231 linked crime pairs (containing two crimes committed by the same offender) and 273,422 unlinked crime pairs were created (two crimes committed by different offenders). Three measures of similarity were calculated for each crime pair: 1) the kilometre-distance between crimes (inter-crime distance); 2) the number of days between crimes (temporal proximity); and 3) a statistical measure of similarity in MO behaviour (Jaccard’s coefficient). Statistical tests of difference, binary leave-one-out logistic regression, and Receiver Operating Characteristic analysis were used to determine whether the three measures of similarity could be used to distinguish between linked and unlinked crime pairs, some containing only burglaries (burglary pairs), some containing only robberies (robbery pairs) and some containing both burglaries and robberies (cross-crime pairs). Results: Linked and unlinked crime pairs could be distinguished with a high level of accuracy (AUCs > .90), with the highest accuracy when combining inter-crime distance, temporal proximity and Jaccard’s coefficient. These findings were replicated with the burglary pairs, robbery pairs and cross-crime pairs. Conclusions: Offender behaviour is sufficiently consistent and distinctive to support the use of BCL with versatile crime series, as well as with burglary crime series and robbery crime series.
... Another literature that informs the question of whether offenders return to previously targeted objects or areas is that on behavioral consistency in offending behavior (Bennell, Mugford, Ellingwood, & Woodhams, 2014;Woodhams, Hollin, & Bull, 2007). The purpose of studies of behavioral consistency in offending is to inform crime linkage procedures, i.e. to determine which unsolved crimes are likely to have been committed by the same offenders. ...
Article
Informed by a growing literature on space-time patterns of repeat and near repeat burglary victimization, a crime location choice model was used to test whether burglars are attracted to areas they previously targeted. Using data in 3337 detected burglaries from one UK police force, and accounting for the distance to the offender's residence, and for other factors that make target areas attractive to burglars, it was demonstrated that burglars were more likely to commit a burglary in an area they had targeted before. This was particularly the case if the prior burglary was (very) recent. Areas near to those in which burglaries had been committed were also more likely to be selected.
... In fact, and acknowledged that the relative scarcity of empirical research regarding the veracity of linkage analysis, and the variety of techniques used in prior research, posed a barrier to its more widespread acceptance. However, a recent increase in published research on this issue (Bennell, Mugford, Ellingwood, & Woodhams, 2013) could somewhat alleviate this concern. Nonetheless, further research into the accuracy of linkage analysis may allow a body of supportive literature for the various techniques involved to be established and thus assist in allaying concerns over its scientific rigour and accuracy. ...
Article
Linkage analysis has, albeit occasionally, been presented in courts across the world as evidence that a series of offences possess behavioural similarities and distinctiveness from other offences, meaning they have probably been committed by the same individual. It is therefore imperative to ascertain how linkage analysis is regarded by juries within the context of deliberations. Three groups of participants (N = 22) eligible for jury duty in England and Wales viewed a simulated rape and murder trial derived from an actual South African case. Linkage analysis formed the sole evidence against the defendant in the two later offences, although DNA matches and eyewitness identifications of the defendant were present in the two earlier offences. Participant deliberations were recorded and subjected to thematic analysis. Five themes were discovered: behavioural consistencies and inconsistencies, physical versus case linkage evidence, barriers to admissibility, potential uses of linkage analysis, and dependence of lay knowledge. Jurors' over-reliance on erroneous lay knowledge contributed to their conclusion that linkage analysis is, at present, unrepresentative evidence that cannot independently indicate a defendant's culpability. However, participants believed that linkage analysis could be a useful tool within investigations and, with further research evidence, in court in England and Wales. Copyright © 2014 John Wiley & Sons, Ltd.
Chapter
Offender profiling is an investigative tool used by police, consultants, and academics to identify the most likely personality, behavioral, and demographic characteristics that an offender will possess, based on an analysis of their crime scene behaviors. While offender profiling has been popularized by TV shows and movies such as Criminal Minds, Silence of the Lambs, Mindhunter, and more, the accuracy and real-world impact of OP is largely unknown. The goal of this chapter is to provide an overview of the history and current status of offender profiling, and to summarize the research on predominant offender profiling methods. This chapter also reviews the accuracy and applied impact of each approach to offender profiling, and highlights the need for an evidence-based offender profiling technique, based upon statistical and scientific methods. A discussion of the future of offender profiling, particularly in terms of scientific research and evaluations, and the implications for law enforcement investigations, are also provided, with recommendations made to improve the field of offender profiling in the future.
Article
Purpose The purpose of this paper is to understand (i) how crime linkage is currently performed with residential burglaries in New Zealand, (ii) the factors that promote/hinder accurate crime linkage and (iii)whether computerised decision-support tools might assist crime linkage practice. Design/methodology/approach A total of 39 New Zealand Police staff completed a questionnaire/interview/focus group relating to the process, challenges, products and uses of crime linkage with residential burglary in New Zealand. These data (alongside four redacted crime linkage reports) were subjected to thematic analysis. Findings The data clearly indicated wide variation in crime linkage process, methods and products (Theme 1). Furthermore, a number of factors were identified that impacted on crime linkage practice (Theme 2). Research limitations/implications Future research should develop computerised crime linkage decision-support tools and evaluate their ability to enhance crime linkage practice. Also, researchers should explore the use of crime linkage in court proceedings. Practical implications To overcome the barriers identified in the current study, greater training in and understanding of crime linkage is needed. Moreover, efforts to enhance the quality of crime data recorded by the police will only serve to enhance crime linkage practice. Social implications By enhancing crime linkage practice, opportunities to reduce crime, protect the public and deliver justice for victims will be maximised. Originality/value The practice of crime linkage is under-researched, which makes it difficult to determine if/how existing empirical research can be used to support ongoing police investigations. The current project fills that gap by providing a national overview of crime linkage practice in New Zealand, a country where crime linkage is regularly conducted by the police, but no published linkage research exists.
Article
Full-text available
Purpose – Crime linkage analysis (CLA) can be applied in the police investigation-phase to sift through a database to find behaviorally similar cases to the one under investigation and in the trial-phase to try to prove that the perpetrator of two or more offences is the same, by showing similarity and distinctiveness in the offences. Lately, research has moved toward more naturalistic settings, analyzing data sets that are as similar to actual crime databases as possible. One such step has been to include one-off offences in the data sets, but this has not yet been done with homicide. The purpose of this paper is to investigate how linking accuracy of serial homicide is affected as a function of added hard-to-solve one-off offences. Design/methodology/approach – A sample (N = 117–1160) of Italian serial homicides (n = 116) and hard-to-solve one-off homicides (n = 1–1044, simulated from 45 cases) was analyzed using a Bayesian approach to identify series membership, and a case by case comparison of similarity using Jaccard’s coefficient. Linking accuracy was evaluated using receiver operating characteristics and by examining the sensitivity and specificity of the model. Findings – After an initial dip in linking accuracy (as measured by the AUC), the accuracy increased as more one-offs were added to the data. While adding one-offs made it easier to identify correct series (increased sensitivity), there was an increase in false positives (decreased specificity) in the linkage decisions. When rank ordering cases according to similarity, linkage accuracy was affected negatively as a function of added non-serial cases. Practical implications – While using a more natural data set, in terms of adding a significant portion of non-serial homicides into the mix, does introduce error into the linkage decision, the authors conclude that taken overall, the findings still support the validity of CLA in practice. Originality/value – This is the first crime linkage study on homicide to investigate how linking accuracy is affected as a function of non-serial cases being introduced into the data.
Article
Full-text available
This study compared the ability of seven statistical models to distinguish between linked and unlinked crimes. The seven models utilised geographical, temporal, and modus operandi information relating to residential burglaries (n = 180), commercial robberies, (n = 118), and car thefts (n = 376). Model performance was assessed using receiver operating characteristic analysis and by examining the success with which the seven models could successfully prioritise linked over unlinked crimes. The regression‐based and probabilistic models achieved comparable accuracy and were generally more accurate than the tree‐based models tested in this study. The Logistic algorithm achieved the highest area under the curve (AUC) for residential burglary (AUC = 0.903) and commercial robbery (AUC = 0.830) and the SimpleLogistic algorithm achieving the highest for car theft (AUC = 0.820). The findings also indicated that discrimination accuracy is maximised (in some situations) if behavioural domains are utilised rather than individual crime scene behaviours and that the AUC should not be used as the sole measure of accuracy in behavioural crime linkage research.
Chapter
Evidence suggests that the majority of crime is committed by a minority of prolific serial offenders who pose a substantial risk to society. Behavioural Crime Linkage (BCL) has been proposed as one method to more effectively bring serial offenders to justice. BCL is a form of behavioural analysis that seeks to identify similarities in offender crime scene behaviour across two or more crimes so that evidence collected across multiple investigations can be combined, and thus helps the police to work in a more efficient way (thereby saving time and money). This chapter will introduce the concept of BCL, describe the different scenarios in which it is used during live police investigations, and briefly discuss empirical research in the area. The chapter will also explore how BCL relates to the concept of primary, secondary, and tertiary prevention, how BCL fits within the wider criminal justice process, and how BCL can support the work of other criminal justice agencies.
Article
Full-text available
Human behavior is characterized by inherent regularity, and previous studies have confirmed that criminal behavior is no exception. In particular, three distinguishable aspects of geographic offending behavior may be characterized by consistency: destination, distance, and direction. In this study, which focused on the aspect of direction, we carried out Van Daele and Bernasco's (2012) additional tests concerning residential burglary cases in Japan. Similar to Van Daele and Bernasco, we proposed the directional consistency could be verified by mean angulation calculated with the average angle between all possible pairs of offenses in a series and compared these to the thresholds that were simulated using the angles chosen at random by the previous authors. Our results confirmed directional consistency in the cases of residential burglary in Japan. We expect this information will improve the accuracy of estimating the location of the offender's residence in geographic profiling. However, the offenders did not show as much consistency as those observed in the previous study. Furthermore, these findings do not take into account preferences in crime patterns or the environmental characteristics. Therefore, it is necessary to verify the impact of these factors on directional consistency.
Article
Full-text available
Purpose: This study compared the utility of different statistical methods in differentiating sexual crimes committed by the same person from sexual crimes committed by different persons. Methods: Logistic regression, iterative classification tree (ICT), and Bayesian analysis were applied to a dataset of 3,364 solved, unsolved, serial, and apparent one-off sexual assaults committed in five countries. Receiver Operating Characteristic analysis was used to compare the statistical approaches. Results: All approaches achieved statistically significant levels of discrimination accuracy. Two out of three Bayesian methods achieved a statistically higher level of accuracy (Areas Under the Curve [AUC] = 0.89 [Bayesian coding method 1]; AUC = 0.91 [Bayesian coding method 3]) than ICT analysis (AUC = 0.88), logistic regression (AUC = 0.87), and Bayesian coding method 2 (AUC = 0.86). Conclusions: The ability to capture/utilize between-offender differences in behavioral consistency appear to be of benefit when linking sexual offenses. Statistical approaches that utilize individual offender behaviors when generating crime linkage predictions may be preferable to approaches that rely on a single summary score of behavioral similarity. Crime linkage decision-support tools should incorporate a range of statistical methods and future research must compare these methods in terms of accuracy, usability, and suitability for practice.
Article
This study examined the effectiveness of multiple correspondence analysis (MCA) for linking crimes. The data consisted of 720 incidents committed by 360 individual offenders of serial criminal sexual contact cases (i.e., rape and forced indecency). According to the measured similarity obtained from the Euclidean distance in the MCA configuration, a rank-score was assigned to each linked pair and unlinked pair. The validity of MCA for linking crimes was evaluated with a 10-fold cross-validation procedure and a receiver operating characteristic analysis. The t-test of rank scores showed a statistically significant difference between linked pairs and unlinked pairs (t(641.43) = −18.53, p < .001, d = 1.38). The value of the area under the receiver operating characteristic curve was 0.83 (95% CI = 0.80–0.86), indicating a high level of discriminative accuracy in distinguishing offences committed by the same person from those committed by different persons.
Article
Full-text available
A number of studies have investigated behavioral case linkage between crimes of a specified type, such as linking one residential burglary with another residential burglary. However, only a few studies have investigated the effectiveness of case linkage across crime types, which have been limited to the UK. This study examined whether linking across crime types using spatio-temporal proximity was possible in samples that were different from the UK in terms of the structure of crime classification. This was accomplished by calculating the discrimination accuracy between linked crime pairs (two offenses committed by the same offender) and unlinked crime pairs (two offenses committed by different offenders) using geographical (inter-crime distance) and temporal proximity (number of days between offenses) across various crimes committed in Japan. Both the geographical proximity and temporal proximity had statistically significant levels of discrimination accuracy across crime types as assessed by Receiver Operating Characteristic (ROC) analysis. This suggests the possibility of identifying a crime series by geographical and temporal proximity across multiple crime types in Japan.
Article
The purpose of this study was twofold: first, to create an index for a behavioral linkage analysis of serial sex crimes, and second, to construct a predictive model for the analysis. Data on 720 sex crimes (rape, indecent assault) committed by 360 offenders arrested between 1993 and 2005 throughout Japan were collected. The following seven behaviors were examined during a series of analyses aimed at illustrating the effectiveness of crime linkage in serial sex crimes: victim age group, area type, publicness of offense site, weapon, time, contact method, and day of the week. The results indicated that six of the seven behaviors (excluding “day of the week”) significantly distinguished between linked and unlinked crime pairs. Under a logistic regression of these six variables, which were dichotomously coded in terms of the concordance or discordance between each pair of incidents, the area under the receiver operating characteristic (ROC) curve was 0.85 (95% CI = 0.82-0.87), indicating a high level of discriminative accuracy in identifying disparate sex crimes committed by the same person.
Article
Full-text available
An increasing amount of research has been conducted on crime linkage, a practice that has already been presented as expert evidence in some countries; however it is questionable whether standards of admissibility, applied in some jurisdictions, have been achieved (e.g., the Daubert criteria). Much research has assessed the two basic assumptions underpinning this practice: that offenders are consistent in the way they commit their crimes and that offenders commit their crimes in a relatively distinctive manner. While studies of these assumptions with stranger sex offenses exist, they are problematic for two reasons: (1) small samples (usually N=50 series, 194 offenses; and N= 50 one-off offenses) and by sampling the offenses of both serial and one-off sex offenders, thereby representing a more ecologically valid test of the assumptions. The two assumptions were tested simultaneously by assessing how accurately 365 linked crime pairs could be differentiated from 29,281 unlinked crime pairs through the use of Leave-One-Out Cross-Validation logistic regression followed by Receiver Operating Characteristic analysis. An excellent level of predictive accuracy was achieved providing support for the assumptions underpinning crime linkage.
Article
Full-text available
There are investigative advantages to being able to determine early in a police investigation whether a rape has been committed by a serial or a one-off rapist. Previous research has found some differences in the crime-scene behaviors of serial and one-off rapists; however, this research suffers from the limitation of utilizing a sample of rapes within which there was a mixture of victim–offender relationships. To address this limitation, this study sampled 38 serial (two or more convictions) and 50 one-off (one conviction) stranger rapists and compared their crime scene behavior across four domains (control, sex, escape and style behaviors). Serial and one-off rapists differed in some control and sexual behaviors, in particular, in the type of victim targeted, the offense locations, methods of control and the sexual acts forced upon the victim. However, the results did not indicate a striking difference in the offending behavior of the two groups. The implications of these findings for criminal investigations are discussed. Copyright © 2013 John Wiley & Sons, Ltd.
Chapter
Full-text available
This chapter begins by explaining the purposes of linking crimes committed by the same offender and what case linkage can add to a police investigation and prosecution. The various steps involved in the process of case linkage are explained. The assumptions of behavioral consistency and inter-individual behavioral variation, which case linkage rests on, are outlined, and the research that has begun to test these assumptions is reported. The effect of poor-quality data on the case linkage process and on empirical research is examined. Current methods and future developments for overcoming this difficulty are described. The obstacles to identifying linked crimes across police boundaries are discussed. Case linkage research and practice are compared with various criteria for expert evidence with promising results. The chapter closes by considering future avenues for research and practice in case linkage.
Article
Full-text available
Case linkage uses crime scene behaviours to identify series of crimes committed by the same offender. This paper tests the underlying assumptions of case linkage (behavioural consistency and behavioural distinctiveness) by comparing the behavioural similarity of linked pairs of offences (i.e. two offences committed by the same offender) with the behavioural similarity of unlinked pairs of offences (i.e. two offences committed by different offenders). It is hypothesised that linked pairs will be more behaviourally similar than unlinked pairs thereby providing evidence for the two assumptions. The current research uses logistic regression and receiver operating characteristic analyses to explore which behaviours can be used to reliably link personal robbery offences using a sample of 166 solved offences committed by 83 offenders. The method of generating unlinked pairs is then refined to reflect how the police work at a local level, and the success of predictive factors re‐tested. Both phases of the research provide evidence of behavioural consistency and behavioural distinctiveness with linked pairs displaying more similarity than unlinked pairs across a range of behavioural domains. Inter‐crime distance and target selection emerge as the most useful linkage factors with promising results also found for temporal proximity and control. No evidence was found to indicate that the property stolen is useful for linkage. Copyright © 2012 John Wiley & Sons, Ltd.
Article
Full-text available
This study aimed to investigate the case linkage principles, behavioural consistency and distinctiveness, with a sample of serial car thieves. Target selection, acquisition, and disposal behaviours, as well as geographical and temporal behaviours, were examined. The effects of temporal proximity and offender expertise were also investigated as moderating factors of behavioural consistency. As in previous case linkage research, geographical and some target selection behaviours were able to predict whether crime pairs are linked or unlinked at a statistically significant level. Crucially, it was also found that temporal behaviours demonstrate a significant capability to predict linkage status, a variable which has never before been applied to the prediction of linkage in serial car theft. Furthermore, it was demonstrated that changing the operationalisation of the behavioural domains can affect the results obtained. No support was found for the moderation of behavioural consistency on the basis of temporal proximity or expertise. Overall, the results support previous case linkage studies, furthering their practical applicability within the criminal justice system. Copyright © 2012 John Wiley & Sons, Ltd.
Article
Full-text available
Much previous research on behavioural case linkage has used binary logistic regression to build predictive models that can discriminate between linked and unlinked offences. However, classification tree analysis has recently been proposed as a potential alternative owing to its ability to build user‐friendly and transparent predictive models. Building on previous research, the current study compares the relative ability of logistic regression analysis and classification tree analysis to construct predictive models for the purposes of case linkage. Two samples are utilised in this study: a sample of 376 serial car thefts committed in the UK and a sample of 160 serial residential burglaries committed in Finland. In both datasets, logistic regression and classification tree models achieve comparable levels of discrimination accuracy, but the classification tree models demonstrate problems in terms of reliability or usability that the logistic regression models do not. These findings suggest that future research is needed before classification tree analysis can be considered a viable alternative to logistic regression in behavioural case linkage. Copyright © 2012 John Wiley & Sons, Ltd.
Article
Full-text available
Computerized crime linkage systems are meant to assist the police in determining whether crimes have been committed by the same offender. In this article, the authors assess these systems critically and identify four assumptions that affect the effectiveness of these systems. These assumptions are that (a) data in the systems can be coded reliably, (b) data in the systems are accurate, (c) violent serial offenders exhibit consistent but distinctive patterns of behavior, and (d) analysts have the ability to use the data in the systems to link crimes accurately. The authors argue that there is no compelling empirical support for any of the four assumptions, and they outline a research agenda for testing each assumption. Until evidence supporting these assumptions becomes available, the value of linkage systems will remain open to debate.
Article
Full-text available
In the absence of forensic evidence (such as DNA or fingerprints), offender behavior can be used to identify crimes that have been committed by the same person (referred to as behavioral case linkage). The current study presents the first empirical test of whether it is possible to link different types of crime using simple aspects of offender behavior. The discrimination accuracy of the kilometer distance between offense locations (the intercrime distance) and the number of days between offenses (temporal proximity) was examined across a range of crimes, including violent, sexual, and property-related offenses. Both the intercrime distance and temporal proximity were able to achieve statistically significant levels of discrimination accuracy that were comparable across and within crime types and categories. The theoretical and practical implications of these findings are discussed and recommendations made for future research.
Article
Full-text available
Criminologists long have recognized the importance of field studies of active offenders. Nevertheless, the vast majority of them have shied away from researching criminals “in the wild” in the belief that doing so is impractical. This article, based on the authors' fieldwork with 105 currently active residential burglars, challenges that assumption. Specifically, it describes how the authors went about finding these offenders and obtaining their cooperation. Further, it considers the difficulties involved in maintaining an on-going field relationship with those who lead chaotic lives. And lastly, the article outlines the characteristics of the sample, noting important ways in which it differs from one collected through criminal justice channels.
Article
Full-text available
The extrapolation of characteristics of criminals from information about their crimes, as an aid to police investigation, is the essence of ‘profiling'. This paper proposes that for such extrapolations to be more than educated guesses they must be based upon knowledge of (1) coherent consistencies in criminal behaviour and (2) the relationship those behavioural consistencies have to aspects of an offender available to the police in an investigation. Hypotheses concerning behavioural consistencies are drawn from the diverse literature on sexual offences and a study is described of 66 sexual assaults committed by 27 offenders against strangers. Multivariate statistical analyses of these assaults support a five-component system of rapist behaviour, reflecting modes of interaction with the victim as a sexual object. The potential this provides for an eclectic theoretical basis to offender profiling is discussed.
Article
Full-text available
This article explores how unobtrusive research methods popularized by Webb et al. (1966) can be utilized in forensic research. In particular, the value of the approach is considered with special reference to examining investigative processes and criminal behaviour. Webb et al.’s three non-reactive types of unobtrusive measures (physical traces, archival material and simple observation) are discussed in relation to a variety of types of material collected across the course of police enquiries. The breadth and variation of this material is emphasized with special consideration of its utility for research. We illustrate how the limitations and benefits identified by Webb et al. are echoed within the forensic domain and how close attention to the methods of collecting such material developed by researchers may improve the effectiveness of investigations.
Article
Full-text available
The empirical support for linkage analysis is steadily increasing, but the question remains as to what method of linking is the most effective. We compared a more theory‐based, dimensional behavioural approach with a rather pragmatic, multivariate behavioural approach with regard to their accuracy in linking serial sexual assaults in a UK sample of serial sexual assaults (n = 90) and one‐off sexual assaults (n = 129). Their respective linkage accuracy was assessed by (1) using seven dimensions derived by non‐parametric Mokken scale analysis (MSA) as predictors in discriminant function analysis (DFA) and (2) 46 crime scene characteristics simultaneously in a naive Bayesian classifier (NBC). The dimensional scales predicted 28.9% of the series correctly, whereas the NBC correctly identified 34.5% of the series. However, a subsequent inclusion of non‐serial offences in the target group decreased the amount of correct links in the dimensional approach (MSA–DFA: 8.9%; NBC: 32.2%). Receiver operating characteristic analysis was used as a more objective comparison of the two methods under both conditions, confirming that each achieved good accuracies (AUCs = .74–.89), but the NBC performed significantly better than the dimensional approach. The consequences for the practical implementation in behavioural case linkage are discussed. Copyright © 2012 John Wiley & Sons, Ltd.
Article
Full-text available
Purpose. The current study tests whether existing behavioural case linkage findings from the United Kingdom (UK) will replicate abroad with a sample of residential burglaries committed in Finland. In addition, a previously discussed methodological issue is empirically explored. Methods. Seven measures of behavioural similarity, geographical proximity, and temporal proximity are calculated for pairs of burglary crimes committed by 117 serial burglars in Finland. The ability of these seven measures to distinguish between pairs of crimes committed by the same offender (linked pairs) and different offenders (unlinked pairs) is tested using logistic regression and receiver operating characteristic (ROC) analysis. Two methodologies for forming the unlinked pairs are compared; one representing the ‘traditional’ approach used by research and, the other, a new approach that represents a potentially more realistic and statistically sound approach to testing case linkage. Results. A wider range of offender behaviours were able to distinguish between linked and unlinked crime pairs in the current Finnish sample than in previous UK-based research. The most successful features were the kilometre-distance between crimes (the intercrime distance), the number of days separating offences (temporal proximity), and a combination of target, entry, internal, and property behaviours (the combined domain). There were no statistically significant differences between the two methodological approaches. Conclusions. The current findings demonstrate that a wider range of offender behaviours can be used to discriminate between linked and unlinked residential burglary crimes committed in Finland than in the UK. The use of a more realistic and statistically sound methodology does not lead to substantial changes in case linkage findings.
Article
Full-text available
Purpose . This paper is concerned with case linkage, a form of behavioural analysis used to identify crimes committed by the same offender, through their behavioural similarity. Whilst widely practised, relatively little has been published on the process of linking crimes. This review aims to draw together diverse published studies by outlining what the process involves, critically examining its underlying psychological assumptions and reviewing the empirical research conducted on its viability. Methods . Literature searches were completed on the electronic databases, PsychInfo and Criminal Justice Abstracts, to identify theoretical and empirical papers relating to the practice of linking crimes and to behavioural consistency. Results . The available research gives some support to the assumption of consistency in criminals' behaviour. It also suggests that in comparison with intra‐individual variation in behaviour, inter‐individual variation is sufficient for the offences of one offender to be distinguished from those of other offenders. Thus, the two fundamental assumptions underlying the practice of linking crimes, behavioural consistency and inter‐individual variation, are supported. However, not all behaviours show the same degree of consistency, with behaviours that are less situation‐dependent, and hence more offender‐initiated, showing greater consistency. Conclusions . The limited research regarding linking offenders' crimes appears promising at both a theoretical and an empirical level. There is a clear need, however, for replication studies and for research with various types of crime.
Article
Full-text available
The purpose of this study is to determine if readily available information about commercial and residential serial burglaries, in the form of the offender's modus operandi, provides a statistically significant basis for accurately linking crimes committed by the same offender. Logistic regression analysis is applied to examine the degree to which various linking features can be used to discriminate between linked and unlinked burglaries. Receiver operating characteristic (ROC) analysis is then performed to calibrate the validity of these features and to identify optimal decision thresholds for linking purposes. Contrary to crime scene behaviours traditionally examined to link serial burglaries, the distance between crime site locations demonstrated significantly greater effectiveness as a linking feature for both commercial and residential burglaries. Specifically, shorter distances between crimes signalled an increased likelihood that burglaries were linked. Thus, these results indicate that, if one examines suitable behavioural domains, high levels of stability and distinctiveness exist in the actions of serial burglars, and these actions can be used to accurately link crimes committed by the same offender. Copyright © 2005 John Wiley & Sons, Ltd.
Article
Full-text available
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.
Article
Full-text available
Geographic profilers have access to a repertoire of strategies for predicting a serial offenders home location. These strategies range in complexity—some involve more calculations to implement than others—and the assumption often made is that more complex strategies will outperform simpler strategies. In the present study, we tested the relationship between the complexity and accuracy of 11 strategies. Data were crime site and home locations of 16 UK residential burglars who had committed 10 or more crimes each. The results indicated that strategy complexity was not positively related to accuracy. This was also found to be the case across tasks that ranged in complexity (where complexity was determined by the number of crimes used to make a prediction). Implications for police policies and procedures, as well as our understanding of human decision-making, are discussed.
Article
Full-text available
Since the 1970s, a wide body of research has suggested that the accuracy of clinical risk assessments of violence might be increased if clinicians used actuarial tools. Despite considerable progress in recent years in the development of such tools for violence risk assessment, they remain primarily research instruments, largely ignored in daily clinical practice. We argue that because most existing actuarial tools are based on a main effects regression approach, they do not adequately reflect the contingent nature of the clinical assessment processes. To enhance the use of actuarial violence risk assessment tools, we propose a classification tree rather than a main effects regression approach. In addition, we suggest that by employing two decision thresholds for identifying high- and low-risk cases--instead of the standard single threshold--the use of actuarial tools to make dichotomous risk classification decisions may be further enhanced. These claims are supported with empirical data from the MacArthur Violence Risk Assessment Study.
Article
Full-text available
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.
Chapter
Full-text available
This chapter begins by explaining the purposes of linking crimes committed by the same offender and what case linkage can add to a police investigation and prosecution. The various steps involved in the process of case linkage are explained. The assumptions of behavioral consistency and inter-individual behavioral variation, which case linkage rests on, are outlined, and the research that has begun to test these assumptions is reported. The effect of poor-quality data on the case linkage process and on empirical research is examined. Current methods and future developments for overcoming this difficulty are described. The obstacles to identifying linked crimes across police boundaries are discussed. Case linkage research and practice are compared with various criteria for expert evidence with promising results. The chapter closes by considering future avenues for research and practice in case linkage.
Article
Full-text available
Case linkage involves identifying crime series on the basis of behavioral similarity and distinctiveness. Research regarding the behavioral consistency of serial rapists has accumulated; however, it has its limitations. One of these limitations is that convicted or solved crime series are exclusively sampled whereas, in practice, case linkage is applied to unsolved crimes. Further, concerns have been raised that previous studies might have reported inflated estimates of case linkage effectiveness due to sampling series that were first identified based on similar modus operandi (MO), thereby overestimating the degree of consistency and distinctiveness that would exist in naturalistic settings. We present the first study to overcome these limitations; we tested the assumptions of case linkage with a sample containing 1) offenses that remain unsolved, and 2) crime series that were first identified as possible series through DNA matches, rather than similar MO. Twenty-two series consisting of 119 rapes from South Africa were used to create a dataset of 7021 crime pairs. Comparisons of crime pairs that were linked using MO vs. DNA revealed significant, but small differences in behavioral similarity with MO-linked crimes being characterized by greater similarity. When combining these two types of crimes together, linked pairs (those committed by the same serial offender) were significantly more similar in MO behavior than unlinked pairs (those committed by two different offenders) and could be differentiated from them. These findings support the underlying assumptions of case linkage. Additional factors thought to impact on linkage accuracy were also investigated. KeywordsComparative case analysis–Linkage analysis–Behavioral linking–Sexual assault–Sexual offense
Article
Full-text available
Serial rapists have the propensity to cause harm to a significant number of victims, meaning that they are of concern to the police as well as to treatment providers. Despite the serious nature of their offending, there are surprisingly few studies that provide information regarding their characteristics, the types of victim they target, or the nature of the sexual offenses they commit, and those studies that do exist are varied in their findings. This study provides a descriptive analysis of serial rape in South Africa. One hundred and nineteen sexual offenses committed by 22 serial rapists were sampled. Information regarding the victims, the offenders, and the crimes they had committed were extracted from police files. The characteristics of victims and offenders are reported as well as the frequencies for 114 different crime scene behaviors. When compared with samples of serial sex offenders from other countries, differences emerged in victim characteristics and crime scene behaviors, including how the victims were targeted, the sexual behaviors engaged in, and the incidence of physical violence. The implications of these observed differences for practice are discussed.
Article
Full-text available
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, Jaccard's 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 Jaccard's 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.
Article
Full-text available
Case linkage, the identification of crimes suspected of being committed by the same perpetrator on the basis of behavioral similarity, and offender profiling, the inference of offender characteristics from offense behaviors, are used to advise police investigations and, in relation to case linkage, have been admitted in legal proceedings. Criteria for expert evidence, such as the Daubert criteria (Daubert v. Merrell Dow Pharmaceuticals, 1993), place stringent conditions on the admissibility of expert evidence. The future contribution of these practices to legal proceedings depends, in part, on whether they are underpinned by hypotheses that are testable and supported. The 3 hypotheses of offender behavioral consistency, of offender behavioral distinctiveness, and of a homology (direct relationship) between offender characteristics and behavior were empirically examined using a sample of serial commercial robberies. Support was found for the former 2 hypotheses but not for the last. The findings of the 2 studies have implications for the future development of these practices, for legal practitioners evaluating expert evidence, and for the implementation of public policy.
Article
Full-text available
In a study exploring the cross-situational consistency of behavior, 140 undergraduate Ss were video-taped in each of 3 laboratory settings, and personality descriptions of these Ss were obtained from friends and acquaintances. Analyses focused on the degree to which Ss maintained consistent patterns of behavior across laboratory settings and between these settings and daily life. The following conclusions were reached: (a) Behavior can exhibit impressive consistency at the level of psychological meaning, (b) psychological properties of situations can be detected from their behavioral effects, (c) cross-situational consistency and discriminativeness are independent, and (d) some behaviors are more consistent than others.
Article
Full-text available
Until very recently, there has been little evidence of the ability of either clinicians or actuarial instruments to predict violent behavior. Moreover, a confusing variety of measures have been proposed for the evaluation of the accuracy of predictions. This report demonstrates that receiver operating characteristics (ROCs) have advantages over other measures inasmuch as they are simultaneously independent of the base rate for violence in the populations studied and of the particular cutoff score chosen to classify cases as likely to be violent. In an illustration of the value of this approach, the base rates of violence were altered with the use of data from 3.5-, 6-, and 10-year follow-ups of 799 previously violent men. Base rates for the 10-year follow-up were also altered by changing the definition of violent recidivism and by examining a high-risk subgroup. The report also shows how ROC methods can be used to compare the performance of different instruments for the prediction of violence. The report illustrates how ROCs facilitate decisions about whether, at a particular base rate, the use of a prediction instrument is warranted. Finally, some of the limitations of ROCs are outlined, and some cautionary remarks are made with regard to their use.
Article
[Correction Notice: An erratum for this article was reported in Vol 73(4) of Journal of Consulting and Clinical Psychology (see record 2007-16787-001). In this article, several errors are present on pp. 738 and 746. The corrections are listed in the erratum.] Until very recently, there has been little evidence of the ability of either clinicians or actuarial instruments to predict violent behavior. Moreover, a confusing variety of measures have been proposed for the evaluation of the accuracy of predictions. This report demonstrates that receiver operating characteristics (ROCs) have advantages over other measures inasmuch as they are simultaneously independent of the base rate for violence in the populations studied and of the particular cutoff score chosen to classify cases as likely to be violent. In an illustration of the value of this approach, the base rates of violence were altered with the use of data from 3.5-, 6-, and 10-year follow-ups of 799 previously violent men. Base rates for the 10-year follow-up were also altered by changing the definition of violent recidivism and by examining a high-risk subgroup. The report also shows how ROC methods can be used to compare the performance of different instruments for the prediction of violence. The report illustrates how ROCs facilitate decisions about whether, at a particular base rate, the use of a prediction instrument is warranted. Finally, some of the limitations of ROCs are outlined, and some cautionary remarks are made with regard to their use.
Book
From the reviews of the First Edition."An interesting, useful, and well-written book on logistic regression models . . . Hosmer and Lemeshow have used very little mathematics, have presented difficult concepts heuristically and through illustrative examples, and have included references."—Choice"Well written, clearly organized, and comprehensive . . . the authors carefully walk the reader through the estimation of interpretation of coefficients from a wide variety of logistic regression models . . . their careful explication of the quantitative re-expression of coefficients from these various models is excellent."—Contemporary Sociology"An extremely well-written book that will certainly prove an invaluable acquisition to the practicing statistician who finds other literature on analysis of discrete data hard to follow or heavily theoretical."—The StatisticianIn this revised and updated edition of their popular book, David Hosmer and Stanley Lemeshow continue to provide an amazingly accessible introduction to the logistic regression model while incorporating advances of the last decade, including a variety of software packages for the analysis of data sets. Hosmer and Lemeshow extend the discussion from biostatistics and epidemiology to cutting-edge applications in data mining and machine learning, guiding readers step-by-step through the use of modeling techniques for dichotomous data in diverse fields. Ample new topics and expanded discussions of existing material are accompanied by a wealth of real-world examples-with extensive data sets available over the Internet.
Article
When relying on crime scene behaviours to link serial crimes, linking accuracy may be influenced by the measure used to assess across-crime similarity and the types of behaviours included in the analysis. To examine these issues, the present study compared the level of linking accuracy achieved by using the simple matching index (S) to that of the commonly used Jaccard's coefficient (J) across themes of arson behaviour. The data consisted of 42 crime scene behaviours, separated into three behavioural themes, which were exhibited by 37 offenders across 114 solved arsons. The results of logistic regression and receiver op-erating characteristic analysis indicate that, with the exception of one theme where S was more effective than J at discriminating between linked and unlinked crimes, no significant differences emerged between the two similarity measures. In addition, our results suggest that thematically unrelated behaviours can be used to link crimes with the same degree of accuracy as thematic-ally related behaviours, potentially calling into the question the importance of theme-based approaches to behavioural linkage analysis.
Article
The last two decades have witnessed major changes in the way that mental health professionals assess, describe, and think about persons' risk for future violence. Psychiatrists and psychologists have gone from believing that they could not predict violence to feeling certain they can assess violence risk with well‐above‐chance accuracy. Receiver operating characteristic (ROC) analysis has played a central role in changing this view. This article reviews the key concepts underlying ROC methods, the meaning of the area under the ROC curve (AUC), the relationship between AUC and effect size d , and what these two indices tell us about evaluations of violence risk. The area under the ROC curve and d provide succinct but incomplete descriptions of discrimination capacity. These indices do not provide details about sensitivity–specificity trade‐offs; they do not tell us how to balance false‐positive and false‐negative errors; and they do not determine whether a diagnostic system is accurate enough to make practically useful distinctions between violent and non‐violent subject groups. Justifying choices or clinical practices requires a contextual investigation of outcomes, a process that takes us beyond simply knowing global indices of accuracy. Copyright © 2013 John Wiley & Sons, Ltd.
Article
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, Jaccard's 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.
Article
This study tested different Bayesian Journey-to-Crime (JTC) models on a sample group of 850 serial offenders apprehended in Baltimore County, MD from 1993 to 1997. In this research, Bayesian JTC models were being used to predict the home locations of the offenders. The sample group data included 133 assaults, 90 burglaries, 497 larcenies, 81 robberies, and 49 vehicle thefts. The main question this research aimed to answer was whether the addition of crimes of a different type to an existing crime series of a single type would result in more accurate and/or precise Bayesian JTC models. The standard practice by law enforcement has been to consider the same-type crime series only when modelling the anchor point of the offender. Similarly, in research, geographic profiles have been constructed exclusively with the same-type crime series. The results of this study clearly indicated that the inclusion of crimes of a different type into a single crime-type series will result in significantly more accurate and more precise Bayesian JTC models. In contrast, crime series with predominantly assault and burglary showed results that were inconclusive or indicated no significant differences. These results should encourage law enforcement agencies to re-evaluate their standard practice of constructing geographic profiles with only the same-type crime series. Copyright © 2009 John Wiley & Sons, Ltd.
Article
Contents:Introduction and Scope, p.1Components of Diagnostic Decision Making, p.4Statistical Machinery, p.5Examples of Enhanced Descision Making, p.10Conclusion and Discussion, p.20Acknowledgements, p.23References, p.23Appendix: Some Concepts of Possibility, p.25
Article
This study tested the Bayesian journey-to-crime (JTC) model by gender and age for serial offenders from Manchester, England. The data were 4056 crimes committed by 171 serial offenders between 2003 and 2006. The data were subdivided by gender and age group to examine whether accuracy and precision varied by these subgroups. In general, the centre of minimum distance was found to be the most accurate measure, but the Bayesian risk and product measures were found to be the most precise measurements. The traditional ‘distance decay’ type of JTC function did not produce estimates that are as accurate nor as precise as the Bayesian approach. Tests were conducted on whether specific gender and age group JTC functions and origin–destination matrices improved predictability. With the exception of juvenile male offenders, the general functions were more accurate and precise. In terms of building an accurate and precise geographical profiling methodology, the results point to the need to include information on the likely predisposition of neighbourhoods to produce offenders as well as information on the crime opportunities available to offenders. Copyright © 2009 John Wiley & Sons, Ltd.
Article
Whilst case linkage is used with serious forms of serial crime (e.g. rape and murder), the potential exists for it to be used with volume crime. This study replicates and extends previous research on the behavioural linking of burglaries. One hundred and sixty solved residential burglaries were sampled from a British police force. From these, 80 linked crime pairs (committed by the same serial offender) and 80 unlinked crime pairs (committed by two different serial offenders) were created. Following the methodology used by previous researchers, the behavioural similarity, geographical proximity, and temporal proximity of linked crime pairs were compared with those of unlinked crime pairs. Geographical and temporal proximity possessed a high degree of predictive accuracy in distinguishing linked from unlinked pairs as assessed by logistic regression and receiver operating characteristic analyses. Comparatively, other traditional modus operandi behaviours showed less potential for linkage. Whilst personality psychology literature has suggested we might expect to find a relationship between temporal proximity and behavioural consistency, such a relationship was not observed. Copyright © 2010 John Wiley & Sons, Ltd.
Article
Receiver operating characteristic (ROC) analysis is a widely used and accepted method for improving decision making performance across a range of diagnostic settings. The goal of this paper is to demonstrate how ROC analysis can be used to improve the quality of decisions made routinely in a policing context. To begin, I discuss the general principles underlying the ROC approach and demonstrate how one can conduct the analysis. Several practical applications of ROC analysis are then presented by drawing on a number of policing tasks where the procedure has been used already (bite mark identification and linking serial crimes) or where it could be used in the future (statement validity assessment and determining the veracity of suicide notes). I conclude by considering briefly some of the potential difficulties that may be encountered when using ROC analysis in the policing context and offer some possible solutions to these problems. Copyright © 2005 John Wiley & Sons, Ltd.
Article
After two years of research and development, the Royal Canadian Mounted Police (RCMP), in conjunction with the Ontario Provincial Police (OPP), introduced the Violent Crime Linkage Analysis System (ViCLAS). This computerized system allows trained specialists to link crimes such as serial homicide and serial sexual assaults. ViCLAS represents the application of the latest technology in crime linkage analysis to assist in criminal investigation. Now considered to be the “gold standard” of linkage analysis systems, it has been adopted by Australia, The Netherlands, Belgium, and Austria with 10 other nations examining the possibility of using ViCLAS. In the United States, Indiana and Tennessee use the system. This article examines the evolution and development of ViCLAS and its contribution to criminal investigation.
Article
The purpose of the present study is to test the case linkage principles of behavioural consistency and behavioural distinctiveness using serial vehicle theft data. Data from 386 solved vehicle thefts committed by 193 offenders were analysed using Jaccard's, regression and Receiver Operating Characteristic analyses to determine whether objectively observable aspects of crime scene behaviour could be used to distinguish crimes committed by the same offender from those committed by different offenders. The findings indicate that spatial behaviour, specifically the distance between theft locations and between dump locations, is a highly consistent and distinctive aspect of vehicle theft behaviour; thus, intercrime and interdump distance represent the most useful aspects of vehicle theft for the purpose of case linkage analysis. The findings have theoretical and practical implications for understanding of criminal behaviour and for the development of decision-support tools to assist police investigation and apprehension of serial vehicle theft offenders.
Article
In this paper it is proposed that the behaviour of an offender during a sexual assault can be considered as consisting of three main groupings or 'aspects'. These have been termed 'Modus Operandi', 'Sexual and Personal Gratification' and 'Attitude and Intimacy'. Although an oversimplification, such a model facilitates a preliminary description of the offender which can be of direct assistance to an investigating officer. Various factors affect the consistency and variability of behaviour and the model has formed a basis for the development of a concept as to which behaviours might be the more useful for identifying linked offences. This more systematic approach to the investigative and intelligence processes is reliant on a victim's statement containing very detailed information about the offender's actions, together with a record of as much of his speech as can be recollected. The material required is described using the model and the text is illustrated with case examples. Instances are quoted of the possible inferences and their relevance to the investigation of sexual offences.
Article
Diagnostic systems of several kinds are used to distinguish between two classes of events, essentially "signals" and "noise". For them, analysis in terms of the "relative operating characteristic" of signal detection theory provides a precise and valid measure of diagnostic accuracy. It is the only measure available that is uninfluenced by decision biases and prior probabilities, and it places the performances of diverse systems on a common, easily interpreted scale. Representative values of this measure are reported here for systems in medical imaging, materials testing, weather forecasting, information retrieval, polygraph lie detection, and aptitude testing. Though the measure itself is sound, the values obtained from tests of diagnostic systems often require qualification because the test data on which they are based are of unsure quality. A common set of problems in testing is faced in all fields. How well these problems are handled, or can be handled in a given field, determines the degree of confidence that can be placed in a measured value of accuracy. Some fields fare much better than others.