Profiling Fails

As anyone who watches television detective programmes will know, criminal profiling claims to predict the characteristics of an offender from an analysis of a crime’s circumstances. Most police forces in the developed world use profiling, … Several studies … comparing the predictions of profilers and non-profilers in mock crime situations where the characteristics of the “real” perpetrator are known. Drawing together the results of four of these studies in a meta-analysis published in 2007, Snook’s team found that the profilers did only slightly better than students without any experience of profiling, and that the predictive abilities of both were very low (Criminal Justice and Behavior, vol 34, p 437). (more)

GD Star Rating
loading...
Tagged as:
Trackback URL:
  • Jonas

    It seems there is a relationship between CP and Recruiting. Do the authors` findings have implications for the practical utility of Recruiting as well?

    Recruitment is the process of matching the characteristics of a potential employee with the circumstances of the specific employment or job.

    My estimation is that CP predictions include more uncertainty while R Predictions include more risk.

    they are part of different risk and uncertainty categories.

    of different of risk and uncertainty.

  • Nanonymous

    profilers did only slightly better than students without any experience of profiling, and that the predictive abilities of both were very low

    BS. 99.99% of NYC muggers are not elderly white women, that’s not very hard to see.

  • rapscallion

    I wonder how psychics would do against profilers.

  • http://thecoldequations.blogspot.com coldequation

    This is referring to Hannibal Lecter-style psychological profiling, not common sense stuff like “muggers tend not to be little old ladies” or “Jains are unlikely to hijack airplanes.”

  • Nanonymous

    This is referring to Hannibal Lecter-style psychological profiling, not common sense stuff

    Does it? The paper defines “criminal profiling” as “the practice of inferring personality, behavioral, and demographic characteristics of criminals based on crime scene evidence”. That’s a wide open definition. Maybe they actually restrict the analysis to psychological profiling only – I don’t know. It is not obvious from the text or the list of references included in the study.

    In any case, the findings do not in any way show that profiling does not work. All that the paper shows is that there is only a modest difference between those who call themselves profilers and those that don’t. And the modest difference doesn’t seem compeletely negligible either. “Profilers” do better in 15-30% of the cases. If you have to make a guess, slating a chance of the correct guess some 20% can in many cases be a very, very good deal. And considering that a lot of the profiling (most, I’d guess) is common sense that just about everyone is bound to get right, the “poor” performance by experts may not be as poor as it seems.

    • Taxman

      I think that the point is, if profilers only outperform students by a small margin, why bother having profilers?

      • Jordan

        Let’s say you can choose between a 10% chance of death and an 11% chance of death. That’s only a small margin, so you don’t mind if I select one at random, right?

        Of course not; a difference is still a difference.

  • Pingback: The Desert Lamp » Campus » Tuesday Morning Links (Holiday Edition)- September 7, 2010