David Canter took up the case of the 'Railway Rapist' in 1986, to help search for a serial rapist, who had also committed 2 murders. Originally, John Duffy was 1505th in a list of about 2000 suspects, but after the profile was developed he was investigated and subsequently convicted. David Canter was involved as part of the 'bottom up' approach used by the British police. It is called this because they employ private criminal psychologists to create profiles for them, as opposed to the FBI system, the 'top down' approach where everything is internal.
From the evidence left at the scene and the way the attacks were carried out he created an offender profile, which subsequently aided the police to arrest John Duffy, who was later convicted of 2 murders, 5 rapes (although he may have committed more - 33 door keys from victims were found in his house, kept as souvenirs) and given 7 life sentences. Canter constructed the profile by using 5 factors, and these are:
Applied offender profiling is created by a profiler and used by the police to create a particular set of characteristics about a suspect which may assist investigations, such as if the offender may be violent or may have previous convictions, and what area they are likely to live in. A lack of a consistent method for creating profiles in the UK can lead to discrepancies (such as whether a number of offences were committed by the same offender or different offenders), between profilers and there are disagreements between profilers as to what method is the best or how a profile should be created.
This raises an issue because profiling may be more effective if it was applied in a methodical, standardised way (Harrower, 1996), but at the moment profilers use more subjective methods, and profile construction methods differ between profilers, for example some use purely psychological principles, and others use personal experience of other cases alongside their psychological knowledge, thus allowing a large margin for discrepancies between profiles.
Two UK profilers, for example Canter and Gregory could work on the same case, and yet still come up with completely different profiles, and because of the great differences between profiling methods in the UK and the USA, a profile by an FBI profiler and a UK profiler would be even more different. Therefore, although the police are attempting to interpret a profile as giving a definite guilty or not guilty answer, or as identifying a particular suspect, the information the profiler used may not be completely accurate or may not be used in the most accurate way.
A profile created by an individual profiler may be more helpful if used as more of a guide for investigations. Perhaps the methods of FBI profiling could be why USA/FBI profiling is directly useful in 3% more cases than in the UK, as a larger numbers of cases can be compared for similar behaviours, rather than just from individual cases a particular profiler has worked on. There is also no universally accepted method of evaluating whether or not profiling is successful, therefore there is no way of telling if profiling should continue to be used, different surveys yield different results.
In 1981, the FBI claimed that profiles were of direct use in 17% of cases, and in the UK in 1995 Copson's survey of police forces showed that profiling was of direct use in just 14% of cases. However, Canter said that surveys do not show why or why not profiles work, and this could prove a problem in analysing problems with profiles and so improvements may be hard to suggest. This also leads to a problem in that the validity and usefulness of profiles cannot be determined, so it cannot be determined how effective profiling is, and if it is effective, in what context.
c). What practical improvements could be made to UK offender profiling? (8 marks) Police and criminal investigators should be aware that offender profiling is not always as helpful to an investigation as they might think. Profiling in the UK has only been directly helpful 14% of the time, and has only led to identification of the offender in 3% of cases. The police, however, tend to use it as confirmation that their investigation is on the right track (Copson, 1995).
An improvement to this could be to change people's views of profiling, and perhaps allow them to realise that a profile cannot identify a specific person, it can only narrow down the search, and even then sometimes the profile may not be correct. Then a situation could be developed such as the one found in the Dutch Profiling Unit, where the profile is seen as an investigative tool, and police and profilers work together on the case.
Also, Ainsworth (2000) suggested that because a variety of people could fit a profile, police should be cautious of presuming a particular person is guilty just because they fit the profile. The profile should be seen as a tool, not the basis of an investigation. Another way of using profiling is to incorporate both styles of approach, the FBI and the UK approach. This would include individual psychologists still working with the police, but each result from their cases go into a central database, so that crimes can be compared for patterns and estimates of behaviour.
In this way, the database would be constantly updated and the results in it would be more accurate, as they would be police results from the final case, not speculation from convicted criminals' interviews. It would also give future profilers more help and could lead to further insight into criminal behaviour patterns and future estimates of behaviour. Use of a database could also provide a unified approach to profiling and perhaps reduce inconsistencies.