The usefulness of criminal profiling

Craig Jackson, David Wilson and Baljit Kaur Rana review some of the evidence on the profiling process

In 2010 we published a critical review of a book by the legendary FBI profiler John Douglas, covering his role in the hunt for the ‘Bind, Torture Kill’ (‘BTK’) serial murderer, in Wichita, Kansas, USA (Douglas and Dodd, 2007). Our article was essentially critical of the role of ‘embedded profiling’ where experts, such as Douglas, use emotive aspects of cases to help develop profiles of active offenders, especially in the context of serial murder investigations.

Our central contention was that such profiling, where Douglas and Dodd (2007) claim to ‘delve inside the swamp-like minds of murderers’, was certainly not scientific psychology, and we questioned if any empirical evidence existed to support its claimed usefulness. In the 30-year case of the ‘BTK’, investigators possibly had the richest ever assembly of clues, direct correspondence, crime-scene details, physical descriptions and access to FBI expertise ever amassed in an ongoing serial murder investigation. It was therefore surely the best chance that proponents of criminal profiling (CP) would have in proving its usefulness as an applied technique. It was both surprising and disappointing then that CP failed to lead to the direct apprehension of the ‘BTK’ killer.

'Educated guesswork'

Additionally we also set a challenge at a press conference for the British Festival of Science in September 2010 at the Royal Institution, for anyone to provide us with evidence where a criminal profile has led to the direct apprehension and sound conviction of a suspect in a serial murder enquiry. Douglas is unapologetic for the embedded nature of his brand of profiling, pointing out similarities between profilers and physicians who ‘learn skills through brainstorming, intuition, and educated guesswork’. Our article (Wilson et al., 2010) caught the imaginations of journalists and was widely reported in many broadsheet and tabloid newspapers and sources – becoming known as the ‘cracker-basher’ article – referencing the TV series of the same name while providing acknowledgement to the inaccurate and limited view of the role of profilers that the public have. One small intricacy in our article is that we contained our criticism to the context of serial murder investigations, although the journalist reports of our paper broadened our criticism to all investigations using profiling. A re-cap on the usefulness of CP is therefore probably timely.

Allegations

Douglas et al. (1986) argued that a criminal's personality as well as behavioural and demographic details could be predicted from crime scene evidence, and further, despite little supporting evidence, CP became widely accepted and used (Snook et al., 2007). The allegations we lodged at profilers were by no means new, and such criticism has been aired previously by journalists as well as other academics (Snook et al., 2008) who went so far as to declare the successful acceptance of CP by the public as the ‘profiling illusion’ – the success of which was down to how the public chose to accept some fields of applied science in favour of others that they do not adopt.

Bias

An acknowledgement of a divergence in CP must be made. Some approaches have a statistical and probabilistic leaning, with other areas being based more on clinical and intuitive premises, and indeed it is this form of CP that uses ‘wet-skills’ and ‘street smarts’ (that Douglas and Dodd (2007) claim only those from a police background truly master). It appears to be the brand of CP (and related academic testing of it) that uses statistical empirical evidence that seems to hold greater longevity and influence in this field. However, regardless of the type of CP used, most surveys of police personnel show that the majority of officers polled found CP to have some level of operational usefulness (see for example, Trager and Brewster, 2001).

Concerns

The methodological concerns about bias in these studies, and the relatively small sample sizes in some research should be considered. As CP has evolved, and as the statistical approaches of ‘investigative psychology’ have become more widespread, some of the early principles adopted by proponents have been modified or outgrown. The main outdated principle here being the binary notion of offenders belonging to organised or disorganised trait sub-types, and that such traits can be predictive of offence mechanisms. Research has shown this to be an outmoded concept of little predictive usefulness (Canter et al., 2004). It is widely acknowledged that offender traits are not reliably predictive of the crimes they commit, and given that murder may often primarily be an ill-thought-through response to a highly-charged emotional situation, it is intuitive that the usefulness of trait-based approaches will be limited.

That demographic features could be predicted from an assessment of particular configurations of specific behaviours occurring in short-term, highly traumatic situations seems an ambitious and unlikely possibility. Until the process is more formally verified, the evidential usefulness of profiles should be treated cautiously, or even entirely excluded from consideration in court.

Intuitive

A similar fate awaited the ‘intuitive’ classification of serial sexual murders by Keppel and Walter (1999) as being power-assertive, power-reassuring, retaliatory-angered or excited-angered; as such classifications often did not bear fruit under statistical analysis.

Snook et al. (2007) conducted a systematic review using 130 studies in the field and were able to conclude in some respects that CP relies on being justified by common sense – or as Pierre Laplace may have called it, ‘common sense reduced to calculation’. In addition, despite the methodological and theoretical arguments, a further complication in the review of CP arises with the argument of who can best apply such techniques in the field. Snook et al., (Ibid) used a meta-analysis of the literature to show that profilers were often little better – if at all – than non-experts and laypersons in predicting offender characteristics from crime-scenes. Unsurprisingly, their conclusions about the existence and future of CP as a pseudoscience were even bleaker than the ones we expressed in response to Douglas and Dodd's (2007) account of their ‘BTK’ experiences.

In terms of the value of evidence provided by CP, the field will struggle to prove its relevance. Not many profilers would assert that CP can identify a single actual perpetrator, preferring to state that its role is to indicate the type of person responsible, being the person most likely to commit a crime with such specific and unique characteristics. This inability to deliver an identifying profile is the hurdle most guaranteed to cause CP to stumble.

Contradictory

In conclusion, much of the literature in the area is contradictory, which in its own way is a positive and healthy way for the discipline and the methods used in evaluating it to develop and evolve. Some reviews of CP have used systematic approaches, and others are more narrative and theoretical approaches, but they often arrive at the same conclusions: despite limitations in the research corpus, CP shows it has the promise to evolve into something more robust and useful. However, in the absence of that evolutionary step just yet, it must be said that there is very little evidence of the compelling kind that would actually support CP in enjoying the large level of public and police acceptance that it currently enjoys. Its appeal still remains in its potential.

Further, while some of the fundamental principles of CP that have been proved outdated by other areas of empirical psychology and behavioural science standards still remain within the applied field, the scrutiny and criticism of the whole field will remain.

Other academics in the field also suggest that the public have a role in shaping how such science is shaped, by being more discerning in how they perceive and adopt accepted wisdom. However, some would argue that such public education and discernment is not necessary in order for the discipline to become more refined; after all, how many laypersons understand the intricacies of DNA matching but yet have absolute faith in its probabilistic certainties and correct application? CP can hopefully prove itself by the virtues of large-scale ideographic research, rather than by the anecdotal nomothetic case studies that linger in the memory. But until then it still promises much more than it currently delivers, and its role as just another tool to be used by senior investigating officers in murder investigations will not develop any further.

Craig Jackson is Professor of Workplace Health Psychology and Head of the Psychology Division, David Wilson is Professor of Criminology and Head of the Centre for Applied Criminology and Baljit Kaur Rana is Senior Lecturer in Organisational Psychology, in the Psychology Division, at Birmingham City University.

References

Canter, D.V., Alison, L.J., Alison, E. and Wentink, N. (2004) The organized/ disorganized typology of serial murder: Myth or model?’. Psychology, Public Policy and Law, 10: 293–320.

Douglas , J. and Dodd , J. (2007), Inside the Mind of the BTK: The True Story behind the Thirty Year Hunt for the Notorious Wichita Serial Killer , San Francisco : Jossey-Bass . 

Douglas, J., Ressler, R., Burgess, A. and Hartman, C. (1986) 'Criminal profiling from crime scene analysis’. Behavioral Sciences and the Law, 4: 401–421. 

Keppel, R.D. and Walter, R. (1999) Profiling killers: A revised classification model for understanding sexual murder’. International Journal of Offender Therapy and Comparative Criminology, 43: 417–434. 

Snook, B., Cullen, R.M., Bennell, C., Taylor, P.J. and Gendreau, P. (2008) The criminal profiling illusion: What's behind the smoke and mirrors’. Criminal Justice and Behavior, 35: 1257–1276. 

Snook, B., Eastwood, J., Gendreau, P., Coggin, C. and Cullen, R.M. (2007) Taking stock of criminal profiling: A narrative review and metal-analysis’. Criminal Justice and Behavior, 34: 437–453. 

Trager, J. and Brewster, J. (2001) 'The effectiveness of psychological profiles’. Journal of Police and Criminal Psychology, 16: 20–28. 

Wilson, D., Jackson, C.A. and Rana, B. (2010) 'Against the medical- psychological tradition of understanding serial killing by studying the killers: The case of BTK’. Amicus Journal, 22: 8–16.

Edition reference: