Criminal profiling text., Summaries of Criminology

Download Criminal profiling text. and more Summaries Criminology in PDF only on Docsity!

B. E. Turvey,

Criminal profiling

Criminal Profiling

Learning Objectives

• What is criminal profiling?
• History of profiling
• Types of profiling
• Offender profiling
• Geographical profiling

Defining Crime

• What do we mean by a crime?
• Deviance = nonconformist behaviour that produces a negative societal reaction, and in which members of society attempt to assert some control to reduce it. (e.g., disapproval)
• Crime = deviance as defined by formalized law.

Defining Profiling

• A process in which “offender characteristics can be deduced from a detailed knowledge of offence characteristics” (Ainsworth, 2000)
• BUT profiling can also include o Understanding an offender's behaviour or future level of risk (e.g., will they reoffend?) o Giving advice on how to interview a suspect o Determining whether offences are linked through analysing crime scene behaviour (Copson, 1995)
• “Profiling does not provide the specific identity of the offender. Rather, it indicates the kind of person most likely to have committed a crime by focussing on the certain behavioural and personality characteristics” - Douglas, Ressler, Burgess & Hartman (1986)
• Technique for predicting the personality, behavioural, and demographic characteristics of an individual based upon analysis of the crimes they commit
• Notion that behaviours will reflect the personality of the offenders (Douglas, 2001)
• Typically used in low volume serious crimes o Serial murder, serial rape, serial arson

Motivation for Using Profiling

• Suspect prioritization
• New lines of enquiry
• Interview strategies
• Predict dangerousness
• Flush out the offender

History of Profiling

• 1970s o Criminal profiling program developed by the FBI o Today ▪ Mixed use of profiling internationally o Snook et al (2008) ▪ FBI: Around 1000 profiles/year in USA ▪ UK: 242 profiles between 1981 - 1995

Usefulness of Profiling

• UK Police Officers (Copson, 1995; n=182) o 83% - “operationally useful” o 69% - definitely use profiling again o 14% helped solve case o 2.7% helped identify offender
• Canadian Police Officers (Snook et al., 2007; n=51) o 88% - valuable investigative tool o 84% - further understanding of case o 45% - profilers accurate o 78% - can misdirect investigation

Psychological Research Supporting Underlying Concepts

of Offender Profiling

• Offender Profiling (Gosling et al., 2002) o Environment provides lens for indirect observations of underlying constructs o 4 mechanisms ▪ Self-directed identity claims - Sentimental

• Manifests behaviours such as collecting memorabilia
• Evidence example: pebble from favourite beach ▪ Other-directed identity claims
• Socially responsible
• Manifests behaviours such as voting
• Evidence example: “I voted” sticker ▪ Interior behavioural residue
• Conscientious
• Manifests behaviour such as organising belongings
• Evidence example: organised workplace ▪ Exterior behavioural residue
• Sensation seeking
• Manifests behaviours such as going snowboarding
• Evidence example: snowboard in room o Room with a clue ▪ Is there a link between individuals and their physical environment? ▪ Can observers make predictions about people based on environments? ▪ Study 1 (Offices), Study 2 (Bedrooms)
• Consensus – do observers agree in ratings?
• Accuracy – are observers correct?
• Cues – do observers use similar valid cues? ▪ Research provides strong support for the assumption underlying this test – much can be learned about persons from the spaces in which they dwell

Types of Profiling

• Inductive Profiling – bottom up o Profile built from a database of what is known about other similar offenders (typically associated with developments in the UK) o Example: ▪ 80% of offenders who attack people in parking lots are white males ▪ Our offender has attacked three people in different parking lots, and therefore it is likely our offender is a white male o Statistical

▪ What happened before, during and after the crime? ▪ Organized v disorganised; motivation for crime; crime scene dynamics based on experience ▪ Modus Operandi – what method was used to commit the crime? o Stage 4: Criminal profile stage ▪ Demographics; physical characteristics; habits; beliefs and values; pre- offense behaviour leading to the crime; post-offense behaviour o Stage 5: Investigation stage ▪ Used by investigators to match suspects ▪ Feedback to stage 2 if new evidence ▪ Used to aid in interviewing potential suspects o Stage 6: Apprehension stage ▪ Admission of guilt and validity checking of profile

Limitations of Profiling

• When offender profiling goes wrong: o Murders of 5 females in Fairbanks Alaska ▪ Profile: - Single - Aged around 40 - Not in the armed forces ▪ Offender (Thomas Bunday) - Married with children - Aged 34 - Airforce instructor o Rachel Nickell murder (London, 1992) ▪ Murdered on a weekday morning while walking her dog with her young son ▪ No forensic evidence left at the scene ▪ Colin Stagg- identified as a suspect after a photofit was released and members of the public came forward. ▪ A search of his flat found some unusual inappropriate items. - However, NO forensic evidence. ▪ Forensic advice was then sought from profiler Paul Britton ▪ Profile: lonely, socially/sexually inept, live nearby alone, interested in unusual hobbies (Britton, 1997) ▪ Colin Stagg lived alone, was a 31-year-old virgin and had a brother convicted for rape ▪ Operation Ezdell initiated – honeytrap investigation designed to get Stagg to confess to murder ▪ Case was acquitted in court ▪ 2008: Robert Napper convicted of murder ▪ Was interviewed at the time but enquiry discontinued. Carried out a double murder 2 years later in 1993
• Alison et al. (2003) – Personality Paradox

o Profiling relies on a naïve trait approach t hat people will behave consistently across different situations o According to Dougal et al. 1992, it is assumed that “the crime scene is presumed to reflect a murderer’s behaviour and personality, in the same way furnishings reveal the homeowner’s character” (pg. 21). o But there is a personality paradox ▪ Yes, people do have discernible personality traits BUT there is a need to acknowledge cross-situational inconsistencies in behaviours ▪ Your behaviour may differ depending on the scenario you are in

• Profiling assumptions o Consistency – behaviour is consistent across offences (if an offender commits several offences, they will share more similarity than offences committed by different people) ▪ E.g., there is limited evidence that different burglars can be singled out by comparing methods of entry, value of property taken (Green, Booth & Biderman, 1976) o Homology – Similar offence styles will be associated with similar background characteristics (e.g., an aggressive person will commit an aggressive offence) ▪ Some early evidence for this (Davies et al., 1998) ▪ But recent work has largely disproved this notion (Mokros & Alison, 2002) - Sample of 100 stranger rapists, offenders with similar offences did not share socio-demographic characteristics of previous offence history

Contemporary approaches

• Rather than viewing traits as table across contexts (traditional view), we instead account for the situation o If a victim resists = aggressive behaviour and violence o If a victim is passive = no violence will occur o Person x situation Interpreting The Accuracy of Offender Profiles
• Alison, Smith & Morgan (2003) o Study 1: N=46 police officers. ▪ Provided with two possible offenders
• 1. Given bogus profile (designed specifically to be ambiguous) o The offender is an inappropriately immature man for his years. He is psychologically and emotionally retarded for his age o Although the act of strangulation served the offender’s convenient removal of the victim, this act represents a powerful symbol of the extension of sexual dominance over the victim ▪ No difference between groups in accuracy rating (M=5.3) ▪ Ratings for “very accurate fit”

▪ Generally, profilers were no better than lay public at predicting offender ▪ Field is improving, but offender profiling has not yet risen to level of ‘science’ ▪ Fox & Farrington (2018) o Meta-analysis of 426 publications (1976-2016) o Methodological quality score (sample size, peer review, stats) ▪ Mean of 4.38/ ▪ 50% did not use a sample; 57% no reported sampling strategy; over 50% used no stats o Improvements still needed ▪ Needs to be scientifically credible using advanced stats, large data sets and peer review ▪ Focus on testing reliability/error rate in police investigations rather than developing new profiles ▪ Do we think profilers are better than bar tenders? o Not really o Evidence of success based on anecdotes o Biased interpretation of ambiguous predictions – Barnum Effect! o Inflated confidence in ”expertise” o Assumption of ‘nothing to lose’ by generating profile

Current Approach To Profiling

In The UK

▪ Behavioural investigative advice (BIA) o New approach that moves away from “profiling” o Developed in response to the challenges of offender profiling o Involves those with an academic background o NOT intending to “read a criminals mind” o Behavioural Investigative Advice should be understood to consist of a range of methods used to develop advice for investigators, based on the study of behaviour exhibited in the commission of crime(s) and the drawing of inferences about the offender(s) (ACPO Crime Committee, 1999) o Intelligence is not evidence o Decision support is not decision making o “Academic research (Safarik et al., 2000) found that one of the best predictors of the offender being a stranger in such cases of female elderly murder was the presence of Postmortem activity.” o Alison et al., 2011 – pragmatic solutions of offender profiling and behavioural investigative advice ▪ Behavioural investigative advice holds some promise

▪ But the precise nature by which it is implement is not yet clear ▪ Further research is needed

Geographic Profiling

▪ An investigative process to predict the most likely “home” location of an unknown offender based on the geographical location of their crime ▪ Basic assumptions o Crimes all belong to one offender (linkage analysis) o Offenders live close and central to their crimes (safety zone) ▪ Aims to prioritize investigation ▪ Example: London rapist and murderer o Killer lived in C o Railway rapists o Some committed with an accomplice – David Mulcahy, some on own. o Around railways in London o Like the offending profiling argument – that is, the actions of the offender, in this case the locations within which they offended, may tell us something about the offender’s characteristics, in this case, their home address. o Psychological processes and arguments may help us predict where an offender lives – the investigative process of geographical profiling. o If you draw a circle around all the points on the map, the home base is within that centre – Duffy is known as a marauder. o Known to him, not far from his house – common aspects of any offenders’ movements. They usually don’t travel far from home, and they usually offend somewhere that is known to them. ▪ Geographical profiling uses: o Locations of offences o Local topography o Distance between offences o Availability of suitable targets at offence locations o Movements of offender and victim BEFORE and AFTER the offence o Temporal factors o Body disposal sites o Whether a vehicle was used o Local transport networks

Origins of Geographical Profiling

▪ Offenders do not usually travel far from home to commit crimes ▪ Shaw & McKay (1942) – Chicago studies o There are offending ‘zones’ in cities based on urban environment

▪ Evidence of ‘safe area’ – crimes occur at least 0.61 miles from home

Automated Systems for

Geographical Profiling

▪ Rise in computerized systems to automate geographical profiling based on location of crimes o Dragnet (Canter) o CrimeStat (Levine)

Human Performance vs Automated Systems for

Geographical Profiling

▪ Snook, Taylor & Bennell (2004) o Can humans perform as well as actuarial techniques to predict residence of offenders based on crime patterns? o Instructions to participants: ▪ Control: no instructions ▪ Circle group: ‘most offenders homes can be located within a circle, with its diameter defined by the distance between the offenders two furthermost crimes’ ▪ Decay group – ‘most offenders commit offences close to home’ o Training led to significant improvements in performance o No difference to actuarial tool!

Conclusion

▪ Profiling assumes you can predict offender characteristics based on behaviour at a crime scene ▪ Profilers offer “comfort” but do not help solve crimes ▪ Profiles contain few predictions about offenders, and those are often ambiguous, unjustified, or unverifiable ▪ Profiling lacks theoretical support ▪ “Expert” profilers can barely out-predict university students ▪ Some evidence that Behaviour Investigative Advice is a better approach ▪ Geographic profiling/case linkage analyses have good evidence (but they focus on logistics, not personal characteristics!)