Introduction:Offender Profiling is “the collection of empirical data in order to collate a picture of the characteristics of those involved in a certain type of crime” (Howitt, 2015, p258). Offender Profiling is performed by either psychologists or professional profilers. Offender profiling was used for the first time in 1943 by Walter Langer who was asked to create a profile of Adolf Hitler in America , however this was increasingly used from the 1970s onwards. Before offender profiling was developed many investigations relied on the limited forensic evidence present at the crime scenes before a search for the offender could start and sometimes forensic evidence would not be present or there would be too little evidence to be used, therefore the offender would not be convicted or would be wrongly convicted. An example of an offender wrongly convicted was George Stinney (Eleftheriou-Smith, 2014) a 14 year old boy who was tried, convicted and executed for the murder of two girls aged 11 and 8 years old in 1944 . There was no physical evidence and later had his name cleared in 2014. Stinney was convicted from an apparent confession that there was no record of.
At this time offender profiling had only just been used for the first time when profiling Hitler, therefore Stinney’s case would have been considered minor compared to others. Profiling describes the type of person who would have committed the crime which could have eliminated Stinney from the suspect list. The use of offender profiling could have caused the officers investigating to have more suspects and to have possibly found the correct offender.Before a profile can be made the psychologists have to follow the process of data assimilation, crime classification, and crime reconstruction. This allows a large amount of information to be collected before a profile is generated so that the profile can be as accurate as possible from the evidence provided. Profilers use information collected from crime scenes to form an outline of the type of person responsible for the crime being investigated. The methods used give an insight into the behaviours of criminals to help provide possible suspects for the crime.
There are two main types of profiling used in the USA and the UK. The types are FBI crime scene analysis (top-down approach) and investigative psychology (bottom-up approach). The top-down approach, used in the USA, compares evidence from the crime scene to similar previous crimes to make links and form an idea of whether another crime may be committed and where they could take place next.
The bottom-up approach, used in the UK, uses the criminal consistency hypothesis and can predict the possible area the offender lives or works in and possibly how the offender behaves in other areas of his life. The bottom-up approach is the type of offender profiling that will be used in this essay due to it being the main method used which gives a profile of the offender rather than predicting if or when the next crime will be committed and is the method studied when researching its usefulness or reliability.This essay will be a literature review which will come to a conclusion on the research question: To What Extent is Offender Profiling Useful In Application? Therefore a balanced discussion will arise to form a judgement on the argument present. Supporting The Usefulness Of Offender Profiling:Law enforcement officers in the USA and the UK generally find profiling useful whether the profile was accurate or not (Alison, 2003). This is mainly due to the fact that it provides a starting point for the officers in high profile cases such as murders or rapes.Alison et al (2003), cited in Howitt (2015), aimed to test how accurate and useful police officers found offender profiles. The researchers provided police officers and forensic professionals with a genuine profile and the information of the offender, and asked them to evaluate the accuracy of the profile.
The study was furthered by providing a similar group of officers an accurate and inaccurate description of the offender. The researchers found that the accuracy of the profiles did not have an affect on how useful the officers found the profile, with the inaccurate profile being described as equally useful. This demonstrates that an inaccurate profile can also provide a base for law enforcement to begin their search with. In addition, this represents that profiles are not always consistent in accuracy but will always be consistent in usefulness. The second half of the study only demonstrated how useful police officers found the profiles, however, previously in the study, other forensic professionals were used along with a group of police officers.
It was found that the forensic professionals were less ‘enthusiastic’ than the police officers were to label the profile as useful. The use of law enforcement and forensic professionals provides different perceptions of profiling suggesting that it is important to investigate profiles from different perspectives to fully evaluate their usefulness. Independent measures design was used in this study meaning that different groups of officers were used in each condition. Whilst different groups were used, the usefulness of the profile did not vary significantly even when the profile was inaccurate, suggesting that the hypothesis, accuracy of a profile does not affect its usefulness, is consistent throughout most police officers.
The case provided to all participants in the first half was different to the case used in the second half of the study which can be classed as an inconsistency and suggests the study is less reliable. However, the use of different cases causes increased generalisability and reliability as the two cases illustrate a consistent finding. The results of the study are based off of the participants’ opinions which can allow the effect of social desirability to take place.
An element of demand characteristics may also be present which may reduce the study’s reliability as the participants may answer in the way they think they should answer. However, since the study is a field experiment, the use of genuine cases allows insight into how the officers would interpret the profile if it was presented to them in a real case. This helps to increase the study’s reliability and validity.Copson (1995) aimed to evaluate the potential of offender profilings use in police investigations. He surveyed detectives from 48 UK police forces who have worked with offender profiling. 184 cases were surveyed, of which 105 were solved and 79 unsolved.
Copson (1995) found that only 17% found offender profiling useless in their cases. Therefore, 83% found profiling useful. In 14% of these cases the profile offered direct help in solving the crime and in 3% of that 14% the profile directly identified the offender. 69% of participants said that they would use offender profiling again.
Copson (1995) concluded that the participants saw the benefit of using offender profiling mostly in an indirect way. Copson (1995) surveyed law enforcement which allows a greater insight into what detectives’ opinions are on the profiles that they are working with directly, which helps to progress research into the process’ usefulness, in its natural context. The results also demonstrate that whilst there are a range of ways offender profiling can help the case, if it does not help the case the profile is still found as useful. The researcher used a large amount of cases ranging from murders to abduction, and threatening phone calls, this provides a greater ability to generalise the findings to other cases throughout the country and therefore improves the reliability of the study. About 40% of the cases were unsolved which could account for the 31% who would not use offender profiling again or the 17% who do not find offender profiling useful. However, this could be seen as a reductionist view suggesting that the case being unsolved is the only reason for profiling to not be seen as useful. The use of both solved and unsolved cases further increases the study’s generalisability and provides a general insight into all types of profiles which may or may not be classed as successes or failures. This demonstrates how the study has reduced bias even though there are more solved cases than unsolved cases.
However, the ratio of cases may have been chosen to reflect the ratio of solved to unsolved cases in real life. Furthermore, 29 profilers overall created the profiles used in the study, however two profilers of the 29 made almost half of the profiles, this could suggest that the two profilers may be been very inaccurate and the profiles created by them could account for the unsolved cases or vice versa. However, the survey is based on opinions, suggesting that some of the participants may not have recognised the profiles usefulness or may have only seen the useful areas, this is known as confirmation bias. Previous experiences and opinions on profiles could have also affect the results of this study. Torres et al (2006) further supports Copson (1995), however the researchers did not survey law enforcement officers. Torres et al’s (2006) survey judged how useful forensic psychologists and psychiatrists thought offender profiling was in law enforcement in America. The researchers found that 95% of psychiatrists and 85% of psychologists thought that offender profiling was useful but over 97% said that for it to be reliable or scientific it needed more research to support it, a minority thought it was scientifically valid and even less thought it was a scientifically reliable method for linking the potential offender to the crime that was committed.
Torres et al (2006) supports the idea that offender profiling was found useful even if it was not believed to be a reliable source. However, this study does not investigate the opinions of people working directly with it but with people who would understand the theories behind the creation of the profile. This can aid in demonstrating its scientific usefulness, which is also researched in the study. Furthermore, The study used a self report method, gained an extremely low feedback percentage of 9.9% and uses American participants therefore it may not be able to be generalised to other UK findings even though the results support those found in a previous study using a different target population and culture. The low response rate limits the generalisability greatly as there is a much smaller range of participants than expected. This could have been due to the professionals that were contacted being inactive in profiling or feeling like they did not know enough about the subject to provide reliable answers to the questions which could have made a large difference in the results. The low feedback percentage could suggest a lack of interest from professionals which could provide more information into how they really feel about profiling.
Whilst this samples professionals who may be creating profiles, the researchers also found that only about one quarter of the participants labeled themselves as knowledgeable about offender profiling which suggest that the results could have been different if more of the participants were knowledgeable in the subject area. The sample population can therefore be interpreted as having mainly participants who do not have any contact with profiling which suggests the results are made up of many opinions that are less relevant when researching profiling. Canter’s (1994) profile of John Duffy was a very accurate profile which demonstrates the usefulness that this profile has and other profiles can potentially have. Before the offender was known he was called the Railway Rapist as a series of sexual offences, that were committed between 1975 and 1986, were committed around railway stations. From descriptions of the offender and patterns between the cases David Canter was able to produce a profile. The profile stated that the offender lived in the same area as the first three offences, was married, had no children, and had an interest in martial arts, among many other points.
John Duffy was found in the list of 2000 suspects as suspect number 1505. He was the only suspect who fit almost all of the profile. The part of the profile which did not match was the physical descriptions obtained from the victims. This profile demonstrates how accurate profiles can actually be as Canter’s predictions were almost completely correct.The study can be generalised through theoretical generalisability as it can be applied to information and theories surrounding the usefulness and accuracy of profiling due to the information that it demonstrates within itself. The study can help to emphasise the links that can be formed through profiling and support studies that are investigating this such as another of Canters studies about organised and disorganised characteristics. The findings can be applied to other similar cases and the theories or techniques used to create the profile can be extrapolated to help further research regarding profiles. As this is only one case it could perhaps be a one off extremely accurate profile as other profiles are known to have been extremely inaccurate, therefore this study could be seen as very limited when demonstrating the usefulness of offender profiling.
While the profile was extremely accurate and immediately helped find the actual offender there were still inaccuracies. However these inaccuracies were in the description of the offender as the victims descriptions had a wide range when it came to hair colour and the height of the offender was overestimated which is suggested to be due to the weapon focus effect which can cause an inaccuracy when remembering details of an event. This suggests that while victims’ descriptions can be inaccurate, profiling is still extremely useful in finding the correct offender and this information can be applied to other profiles that may have been formed around the descriptions of witnesses and have perhaps been labelled as failed profiles due to this.Against The Usefulness Of Offender Profiling:While there are many cases that were successes and provide evidence for the usefulness of offender profiling there are also cases that have been ‘failures’, that have not helped the case or have been ignored due to other evidence or the profile being inaccurate. Researchers, such as O’Keefe and Alison (2001), have suggested other reasons for profiles not being as useful as they could be and in someway becoming useless.
O’Keefe and Alison (2001) investigated the accuracy of psychic detectives’ profiles as people saw profilers working in the same way as psychics. They demonstrated that profilers produce more accurate profiles and therefore more useful profiles. However it is suggested that accuracy does not affect usefulness in previous studies. O’Keefe and Alison (2001) suggested that profiling may cause the Barnum effect and if a detective was under pressure to find the offender they may see things in the profile that may not be there. This suggests that most profiles may just be a general statement but seen just as useful as others, such as David Canter’s profile of John Duffy. This also demonstrates that the police officers may experience the Barnum effect and therefore may have found an incorrect offender due inferring the text incorrectly. Whilst this study does not directly reference the usefulness of profiling it demonstrates how profiles can be seen as more useful than they actually are, therefore emphasising that profiles are not useful and can cause the arrest of the incorrect offender when applied to real life.
The study is a laboratory experiment and does not reflect real life as profiles are used alone in cases, therefore the study has low ecological validity. Profiles will usually be considered by many officers or detectives which suggests that the Barnum effect may not be happening as all of the officers would have to have the same experience when putting the profile into application. This would suggest that the researchers reductionist view on the cause of profiles being seen as useful could be incorrect. The conclusion of the Barnum effect being the only effect limits the study as the usefulness of offender profiling is not researched in a holistic way. This study is therefore less reliable than other studies due to its view and how it was investigated which could have been improved by using real cases that used offender profiling.
A real life example of a useless or inaccurate profile is the profile made of ‘the Boston strangler’ Albert DeSalvo (Canter, 1994. and Gadson et al, 2005). The profile described the Boston Strangler as a homosexual school teacher who was living alone when in reality he was a heterosexual construction worker living with his family. This case is very inaccurate and demonstrates how wrong a profile can be.
This could have been caused by many things, such as an inexperienced profiler or a offender who is familiar with how profilers work. Other studies support this with more inaccurate cases however these are not as inaccurate as the Boston strangler’s profile. Albert DeSalvo was already known to the police as ‘The Measuring Man’ and was later identified and arrested as ‘The Green Man’, therefore DeSalvo was not actually convicted for the Boston strangling. DeSalvo later confessed in detail to the murders, however he was not identified by witnesses who had seen the Boston strangler so it is still unknown whether Albert DeSalvo was really the Boston strangler. This suggests that the profile made may not have been inaccurate and with the many other limitations that this case has, it may not be reliable in representing the uselessness of offender profiling.Another example of a failed profile is from the Rachel Nickell murder (Gadson et al, 2005), however this profile was not unsuccessful because of an inaccurate profile, rather because of the unreliability of offender profiling at the time. Before the profile was produced, the investigators were at a loss due to the lack of forensic evidence and witnesses. Once the profile was made and broadcasted the police were able to name a person who fit exactly to the profile and was found due to help from witnesses who placed him near there at the time.
Unfortunately, the judge of the case described the profile as being pulled out of thin air and made from speculation and this unreliability as a source allowed the suspect to be released and the killer not to be found. This demonstrates how an investigation entirely based on offender profiling can easily fail as it causes the law enforcement officers to use only this ‘evidence’ to catch the offender even while the profile may be completely inaccurate like the one of the Boston Strangler. Intuition is a major factor in creating profile which is why they are seen as such unreliable sources of evidence as they are based on the individuals subjectiveness which is easily affected by their personal prejudices rather than existing scientific knowledge that is much largely accepted. This profile was extremely accurate but its scientific reliability caused it to be unsuccessful.
The scientific reliability has been questioned by forensic professionals previously in Torres et al (2006). The outcome of the case can help to improve the use of profiles in its application as it demonstrates how offender profiling alone cannot be the solution for the case however useful it is, which suggests that profiles will never be completely useful. This could suggest that scientific reliability limits the usefulness of offender profiling.
The opinions that people have on offender profiling can greatly affect how useful it can been seen by others which is demonstrated through this study as the judge’s opinion caused the profile to become useless. If this was a different judge the profile may have been accepted as reliable evidence and could have convicted the suspect. This demonstrates how the outcome of a trial can be greatly affected and emphasises the subjectivity of a profile.
Another example is of a profile made by Brussel (1968) of the Mad Bomber of New York that was highly accurate. Many bombs were placed around New York over a long period of time, some of which did not detonate and the notes attached them provided evidence to the identity of the Mad Bomber as they accused a company of being the reason for the bombings. However, the profile was based on previous knowledge of ‘bomber profiles’ and the notes that were found, rather than the style and pattern of the offence which is how profiles are more commonly produced. This profile can be seen as both a success and a failure as whilst the profile was accurate, it was not the reason that the offender, George Metesky, was identified, it was because of the letters he left and the records the company, that he previously worked for, had. The case demonstrates how profiles can be unnecessary in identifying an offender even when the profile is extremely accurate. The case study is a real life situation depicting its high ecological validity.
This allows the results from the study to be applied to similar cases and can help to decide whether profiles are necessary in certain types of crimes or crimes with certain amounts of evidence available to law enforcement. However, it is quite limited in reliability due to the large amount of confounding variables. It is impossible to know if the profile did influence Metesky’s conviction, suggesting this study is unable to illustrate the usefulness of offender profiling.
Conclusion:Offender profiling is generally described as a useful tool in its application in law enforcement. Profiling combines previous knowledge, patterns, evidence, and witness descriptions to draw up a description of an offender. The profiler is able to draw connections between behaviours in their crime and their life outside of the criminal offences to produce a description of the offender in their normal lives.Successful and unsuccessful profiles are classified mainly on their accuracy rather than helpfulness or whether the correct offender was convicted or not.
The research discussed demonstrates how the accuracy of the profile used in the case does not particularly affect how useful law enforcement finds it but how useful professionals producing the profile find it. In a large amount of cases in Copson (1995) profiling was found useful and only a very small amount of those profiles had directly solved the case which demonstrates how while the profile does not produce a solution to the case it can be useful in providing help in other ways such as narrowing down a suspect list or providing a base to start with in an investigation. As stated by Jackson and Bekarian (1997), cited in Harrower (2001), “Offender profiles do not solve crimes. Instead … profiles should be viewed as simply one more tool that can be extremely useful”.Others may see profiles as being unnecessary or useless as some profiles are completely inaccurate or can become useless in application. As this is demonstrated through case studies, it may make the conclusions less reliable and less generalisable to other studies emphasising the usefulness of offender profiling.
One study suggested a single reason for the uselessness of studies which does not look at the information in a holistic way. In general, there is a lack of studies illustrating the uselessness of profiling with studies mainly focusing on its accuracy.Many studies have been performed over the years since profiling came into use, most of which helped to develop profiling into becoming as accurate as possible. However, during this literature review, the studies researched emphasise that however accurate a profile may be this does not affect its usefulness in application.
Many of the studies can be seen as somewhat unreliable due to their reliance on opinions, whether professional or not, however, the use of real cases help to confirm the usefulness that can be achieved through an accurate profile. Throughout the research it has been found that more studies have provided evidence to support the idea of profilings usefulness and that while offender profiling may not be counted as scientifically reliable by many people, or perhaps even not accurate enough to solve a case, profiling is undoubtedly useful in application.