Offender Profiling - Week 21 PDF

Offender Profiling “Offender profiling is the stuff myths are made of” Grubin (1995) Start Recording! Equitable Learning Plans • Presentations / group work – please speak to your tutor ASAP • Mid-term test – watch for announcement Content warning! • Some of the content of this lecture and othe...

Offender Profiling “Offender profiling is the stuff myths are made of” Grubin (1995) Start Recording! Equitable Learning Plans • Presentations / group work – please speak to your tutor ASAP • Mid-term test – watch for announcement Content warning! • Some of the content of this lecture and other lectures and tutorials in tis course discuss issues that some people may find distressing • This includes mention of the nature of violent and sexual offences and the characteristics of offenders • If you feel that you do not want to be exposed to this information you may like to consider not taking this course. Offender Profiling • There is no commonly accepted definition • Douglas et al (1986) “attempting to produce a description of the perpetrator(s) of a criminal offence on the basis of analyses of characteristics of the incident” • Copson (1995) “ any predictions, recommendations and observations based on inference of offender characteristics from behaviour exhibited in a crime or series of crimes, and offered to the investigators” What do we call it • Has been called – – – – Offender Profiling Criminal Profiling Criminal Personality Profiling Behavioural Profiling Homant and Kennedy (1998) attempt to distinguish between: • Crime scene profiling – Uses info form scene to generate full picture of the unknown offender • Offender Profiling – The collection of empirical data in order to describe those involved in a particaulr type of crime • Psychological Profiling – Use of personality tests and interviews to determine if an individual matches the characteristics of a class of offender (eg sex offender) • Not all would agree! Offender Profiling • Is probably what most people imagine forensic psychologists spend most of their time doing • Very few psychologists regularly engage in profiling. • Very few people in Au or UK are employed as profilers. • In UK only a handful of people regularly produce profiles – often only get paid expenses • True situation in USA is difficult to discern And what do we use it for? • Copson (1995) surveyed UK police officers – they wanted 4 types of information 1. 2. 3. 4. description of characteristics of unknown offender enhanced understanding of offenders behavior and likely future threat guidance on interview strategies linking of a series of offences on basis of common behavioral characteristic • Some psychologists work with police to provide information related to 2 and 3 above. Some also undertake “psychological autopsy” work. • Some psychologists work with police to undertake risk analyses on violent offenders who are being monitored in the community A brief history • There are many fictional profilers from Sherlock Holmes (Doyle) to Fitz to Criminal minds • Jack the Ripper (next slide) • Most famous early case was that of Dr James Brussel’s (1968) profile of George Metesky the “Mad Bomber” First Profile? Canter (2004) suggests this profile of Jack the Ripper offered by Dr Thomas Bond in 1888 may be first profile: • ..must have been a man of physical strength and great coolness and daring. There is no evidence he had an accomplice. He must in my opinion be a man subject to periodic attacks of homicidal and erotic mania… The murder in external appearance is quite likely to be a quiet inoffensively looking man probably middle aged and neatly and respectably dressed. (…) He would be solitary and eccentric in his habits, also he is likely to be a man without regular occupation, but with a small income or pension. The Mad Bomber • • • A series of small bombs started to be placed in New York from 1940 onwards. Many of the total of over 50 bombs placed over 17 years did not explode. Notes were left with the bombs and letters sent to newspapers. He explained that would cease bombing during the war for patriotic reasons! The police asked psychiatrist Brussel for help in 1956. Metesky was arrested in 1957 Brussel’s profile • Man aged 35-50 • Mentally paranoid and with a chronic heart condition • Slavic origin, well educated but not to University level • Would wear a double breasted suit with jacket buttoned up • Would live in Westchester or Connecticut and be living with older female relative Metesky • George Metesky was – – – – – 50 years old Slavic by birth Lived with his two sisters Had TB (not heart condition) Was wearing a double breasted suit when arrested (apparently) • "One thing I can't understand is why the newspaper labeled me the Mad Bomber. That was unkind." -George Metesky But on closer examination.. • Letters well written but clear that author was not native English speaker • Most letters posted in Westchester (!) – many eastern Europeans living in this area • Letters suggested paranoia, and bombs meticulously constructed – hence suggestion re clothing. • Letters suggest paranoia – particularly targeted at “Con Edison crooks” – a reference to the Consolidated Edison Company that supplied electricity. Inspection of records of claims against the company led to arrest • Brussel thought would live with older females because he wrote “W” like two curved Us (i.e. like breasts) suggesting was underlying sexual problems that motivated crime!!! • Brussel wrote up the case 11 years after the arrest When profiling goes bad • July 1992 23-year-old Rachel Nickell was brutally murdered on Wimbledon Common in the presence of her two-year-old son. • She was dragged into the undergrowth, sexually assaulted and stabbed 49 times. • When found by a dog walker her son was clinging to her body Colin Stagg • Under pressure to solve the crime police sought assistance from psychologist and profiler, Paul Britton. • Britton described the murderer as a sexually repressed loner who lived on his own close to the scene of the crime, would be in his twenties or thirties, with an interest in the occult and in knives. • Detectives knocked on doors around Wimbledon Common and questioned 32 men. They produced a photofit which four callers identified as Mr Stagg, who often walked his beloved dog Brandy on the Common. • Colin Stagg was 29, unemployed, a virgin, and a pagan. A woman with whom he had corresponded via a lonely heart ad contacted police to say he had shared graphic sexual fantasies. A witness picked him out of an identity parade. • He had no previous convictions but told police he sometimes sunbathed naked on the heathland. The police became convinced they had found their man. • “I think the police had convinced themselves that Stagg murdered Rachel. They became so desperate to solve the murder that they couldn’t imagine it being anyone else.” (Stagg’s defense barrister) Paul Britton • Police enlisted profiler Britton to design a covert operation to prove or disprove their suspect’s guilt. • A covert operation was designed: An undercover female police officer befriended Stagg in the hope of getting him to admit to the murder. (Britton denies this was his idea) • Over five months of letters, phone calls and four in-person meetings, she feigned sexual interest in him and he admitted to sexual fantasies involving violence. According to reports he told the officer he had fantasized about Miss Nickell’s killing – but he never admitted responsibility. • • • • • In fact, when she wrote: ‘If only you had done the Wimbledon Common murder. If only you had killed her, it would be all right’, he replied: ‘I’m terribly sorry, but I haven’t.’ But at trial the judge ruled the police had shown "excessive zeal” and had tried to incriminate Stagg by “deceptive conduct of the grossest kind". He excluded the entrapment evidence and the prosecution withdrew its case. Stagg was eventually released and awarded over 700,000 pounds (about $1.4 million) damages. 2008 another man with a history of violent sexual assaults in the area was convicted of Rachel’s murder. This individual had continued to offend while Stagg was being investigated Paul Britton Different Approaches to Profiling Several different approaches can be identified, including: • Clinical approach • FBI approach • Geographical approach • Statistical approach The Clinical approach • A general term used to describe the work of profilers who rely on clinical experience rather than statistical analysis or theory • Advocates describe it as “more of an art than a science” The FBI approach • First attempt to develop systematic approach was by Behavioral Science Unit • Interviewed 36 convicted sexually motivated murderers. • This led to the distinction between Organised and Disorganised murders FBI categorization of sexual murders Organised • • • • • • • Planned in advance, specific victim targeted – is “personalized” controlled conversation crime scene shows signs of control Restraint used sexual acts after death Body moved and hidden weapon removed Disorganised • • • • • • • • • • Spontaneous offence location and or victim known depersonalizes victim minimal conversation chaotic scene sudden violence to victim little use of restraints aggressive acts before death body left in view at scene weapon present FBI • Once we can categories crime as organized or disorganized can then move to describe perpetrator – Organised – intelligent, socially competent, skilled job, sexually competent, lives with partner, follows details of crime in news may move following crime etc – Disorganised – average IQ, immature, poor work history, sexually incompetent, lives alone, lives near crime scene, little interest in media, little change in life style after murder etc Alternative methods • Geographical approach (Rossmo, 2000) Seeks to predict base from location of series of crimes. Attempts to reverse the process of environmental criminology Geographical profiling Geographical profiling • • • "Geographic profiling a serial criminal makes a lot more sense than the more traditional psychological profiling. (…) I fail to easily recall an instance where psychological profiling has assisted an officer in solving a series of serious crimes. Whereas geographical profiling is a much more exact science that can help target resources and target our enquiries into a specific place.” Colin Sutton, former senior investigating officer Met Police Murder Squad. Wired.co.uk, 10 November 2014 Statistical approach • Main advocate is David Canter, Liverpool UK • Collates data from solved cases. Data is analysed using a form of multivariate analysis called multi dimensional scaling and smallest space analysis to reveal patterns and relationship in the data. The idea is that groups of actions will tend to occur together allowing one to reveal a typology. • This is more evidence-based than some other approaches, but the analysis is only as good as the data fed into it. It is not always obvious what characteristics of a crime to record. • Also, this type of analysis can be statistically unstable unless we have a very large data set (should have at least 10 cases for every variable included in the analysis) Profiling : An evaluation • Should it work? Given what we know about the determinants of behaviour, should we be able to work backwards to identify characteristics of an offender from evidence left at scene? • Does it work? How can we evaluate the work of profilers, and what are the problems we are likely to face in doing this research? Should it work? Canter (2011) Resolving the Offender ‘Profiling Equations’ Canter argues that there is some reason for optimism and that there are some established scientific frameworks for drawing inferences about offenders characteristics from criminal actions The action–characteristics (A > C) equation, a model of what is involved in ‘‘profiling’’ an unknown offender. Inference must be made about the offender from information about how the crime was committed, including its location, time, and the victim details. To make such inferences, appropriate psychological theories and models must be drawn upon. Canter (2011) But does it work? • Design a study to evaluate the value of offender profiling Does it work? • Archival analysis of past cases – survey of users – Compare profile to perpetrator in solved cases • Experimental studies – Compare profiles produced by different groups • Other – Analyse profiles – Analyse methods used by profilers Archival Studies: Coals to Newcastle • DCI Garry Copson form Met Police (1995) conducted a major survey of effectiveness of profiling • 184 profiles requested. • 54% felt had received some extra information from profile • 14% had assisted in solving case • 3% led to identification of offender (according to officers in charge of case) • Better news is that majority of officers were happy to have reassurance from expert that their own hunches were correct! Coals to Newcastle • Identified 50 solved cases and looked at profile • Profilers got 68% of info in profile correct – clinical did slightly better than statistical • However, very difficult to assess – many items in profiles not analysed as could not be classified as correct or incorrect. Experimental evaluations • Pinizzotto et al (1990) gave solved cases (one murder, one sex offense) to profilers, detectives, psychologists, and students. Profilers did better with the sex offense, but not the murder. Profilers did not use the information any differently to other groups. • Kocsis et al (2000). Used solved murder case. gave to profilers, police,psychologists, students and psychics. Psychologists slightly better than police and psychics. Profilers- some evidence that were better but very few agreed to participate (5). Psychics no good at all – relied on stereotypical views about murderers Alison et al (2003) • Analysed profiles produced by profilers working in UK and USA. • Identified over 3,000 individual statements in the profiles • The vast majority were reiterations of facts already known to police, or described the profiler’s competence (!) or the limits of the profile • Only 28% provided descriptions of the offender. • Of these most (82%) were unsubstantiated claims. Only 16% were justified. 1% were illogical. Emily and Laurence Alison, University of Liverpool, UK Organised / Disorganised? • Canter, Alison, Alison and Wentink (2004) examined validity of Organised/ disorganised dichotomy • Coded 39 aspects of 100 killings from 100 US serial killers. • Found no distinct dichotomy • Most serial killings showed some of the organised characteristics • Disorganised characteristics were much rarer Profiling – Science? • Dowden, Benell & Bloomield (2007) • Reviewed 132 published articles: – – – – Authors rarely publish multiple articles Publish in many different journals Most are discussion pieces – no data Some increase in peer-reviewed articles, but statistical sophistication “sorely lacking, with most including no statistical or formal analyses of data” The criminal profiling illusion: What’s behind the smoke and mirrors (Snook et al, 2008) • A critical analysis of the evidence for profiling. Makes the following points: – Common typologies are invalid (eg organized / disorganized) – Based on outdated personality theories (doesn’t account for substantial role of environment) – Profilers don’t often outperform nonprofilers The criminal profiling illusion: What’s behind the smoke and mirrors (Snook et al, 2008) • So why do people believe in profiling? – Anecdotal evidence can be persuasive – but not scientific – Repetition – repeating a message increases belief – Selective reporting – eg reporting only hits, ignoring errors – Cognitive biases – such as Confirmation Bias (we look for conformity instances and ignore contrary evidence) – Barnum effect – we have a tendency to interpret vague statements as if were specific (eg astrology, palm reading etc) – Media attention, TV, film etc Alison et al (2003) • UK Police officers were given details of a solved crime and either a genuine or a fabricated profile. • Police officers and “forensic professionals” rated the two versions as “somewhat accurate” – no significant differences in ratings. The criminal profiling illusion: What’s behind the smoke and mirrors (Snook et al, 2008) • Conclusions – Criminal profiling lacks scientific support and has the ability to interfere with criminal investigations – should not be used – False belief that CP works is widespread. This is problematic – Profilers should be required to prove the effectiveness of their techniques. Fallon and Snook (2019) • • • • • • • • Applied Carl Sagan's “Baloney detection kit” – a set of rules to detect fraudulent claims and pseudoscience. They noted the following characteristics which matched Sagan's guidelines: A tendency to invoke ad hoc hypotheses (evade testing or findings by saying not relevant to the methods they use) Intellectual stagnation Excessive reliance on anecdotes Evasion of scrutiny Lack of controlled experimentation Burden of proof Absence of connectivity (with other disciplines) Fox and Farrington (2018) • • • A systematic review and meta-analysis of 426 publications on Offender Profiling from 1976 through 2016. Suggested that was some evidence that case linkage analysis studies showed moderate to strong accuracy in its ability to link crimes to a single offender. However, concluded that “very few evaluations of the profiles’ accuracy or effectiveness have taken place. This lack of scientific evidence on the validity and impact of profiling, despite the technique’s widespread use, seriously questions its efficacy in real world investigations, as it currently stands.” Summary • A variety of approaches • Limited evidence that profilers actually contribute anything to investigations • Perhaps profiling is simply the application of logic by a “fresh pair of eyes” • Statistical approach has potential in “common” crime types where are more data available • Is profiling “harmless” or could it distract an investigation?