In some cases, police may rely more on who they think wrote the profile than on the contents of report itself, according to new research.
A study of police officers in Australia found that police rated the profile of a murderer as more accurate when they thought it was written by a professional profiler than when they thought it was written by someone else – even though the profiles were exactly the same.
The results suggest that even professionals like police officers can be swayed by the wrong things when making some judgments, said Andrew Hayes, co-author of the study and assistant professor of communication at Ohio State University.
"Police officers who believe criminal profiling is a useful tool may be predisposed to believe that a professional profiler will produce accurate results," Hayes said. "Our results suggest that police officers' beliefs about the effectiveness of profiling may lead them to see things that aren't really there."
Hayes conducted the study with Richard Kocsis, a forensic psychologist in Australia. Their results were published in a recent issue of the International Journal of Offender Therapy and Comparative Criminology.
Hayes said that while the use of criminal profiling by police worldwide has grown in recent years, there is a lack of research examining whether it actually helps apprehend criminals. While this study can't answer whether profiling works, he said it does suggest caution when interpreting criminal profiles.
"Some police may believe that professional profilers know what they are doing, so their profiles must be accurate," Hayes said. "In these cases, they may not look closely enough at the profile itself."
The participants were 59 police officers in Australia with an average of 12 years of experience on the force. They were each given a criminal profile written about a real (but now solved) homicide case that had taken place in the country. The profile described what the author believed the suspect was like, in terms of age, physical description, and psychological makeup, among other things.
Half of the officers were told the author of the profile was a professional profiler that the investigator consulted. The other half were simply told the author of the profile was "someone the investigator consulted."
To deal with the possibility that the results of the study may be specific to the content of a specific profile, the researchers used three different profiles, all written about the same case by different authors. However, all participants were told the profile was written by either a professional profiler or someone else.
After the officers read the profile, the were asked to rate it on several different measures, such as its coherence, level of detail, and how helpful it would be in narrowing a list of suspects.
Results showed that officers rated the profiles similarly, whether they believed the profiles were written by a professional profiler or someone else.
But the key finding came in the next part of the study.
The officers were then given a written description of the actual perpetrator of the homicide (information which was available because the case had been solved). The officers were asked to rate the accuracy of the profile they read in terms of how closely it matched the perpetrator. Participants could view the description and the profile side-by-side to compare.
Officers who were told the profile was written by a professional profiler rated the profile as more accurate than those who were told the profile was written by "someone the investigator consulted."
"These officers were reading the exact same profile and the exact same description of the subject who was arrested," Hayes said.
"However, police thought the profile was more accurate when they believed it was written by a professional profiler."
Hayes noted that profilers received better marks for accuracy than others, even though the police rated the content of the profiles as similar, no matter who they thought wrote them.
One explanation is that officers had relatively objective ways of evaluating the linguistic features of the profiles, so their ratings couldn't be swayed by who they thought wrote the profile. However, the match between the profile and the perpetrator is more open to subjective interpretation.
"When evaluating the accuracy of the profiles, officers could use their own subjective criteria to determine whether the profile fit the real description," he said.
Since some officers want to believe that profiling works, and the profiles are at least somewhat ambiguous, they could resolve the ambiguities in favor of what they want to believe.
"Profiles are, in a way, like horoscopes. They are general enough that they can fit a lot of people," Hayes said. "If officers want to believe the profile is accurate, they can find the evidence they are looking for."
Hayes noted that the study itself offers no prescriptive advice about whether profiling should be used in criminal investigation.
"A profile is not intended to identify a specific person. A profile is only one of many tools in an investigator's arsenal, and it is not my business to tell investigators what tools they should use." But the results do suggest that anecdotal accounts of the accuracy of a profile are not a good basis for arguing that profiling is actually useful, he said.
Contact: Andrew Hayes, (614) 688-3027; Hayes.338@osu.edu
Written by Jeff Grabmeier, (614) 292-8457; Grabmeier.1@osu.edu
Journal
International Journal of Offender Therapy and Comparative Criminology