Citat:
Ursäkta att jag säger det, men Wall är en lightweight i mina ögon. Jag fick iallafall nog av honom när han attackerade själva fenomenet gärningsmansprofilering. Som om att det bara var nåt mumbojumbo. Just det har inte han tyngd nog att komma med.
Jordås har i mina ögon öppnat dörren til AI i MOP, det tycker jag att han ska ha beröm för. Han tog det här ett steg vidare. Kudos för det.
Jordås har i mina ögon öppnat dörren til AI i MOP, det tycker jag att han ska ha beröm för. Han tog det här ett steg vidare. Kudos för det.
Nu vet jag inte exakt hur kategoriskt han formulerar sig, men rent generellt är det väl bra att problematisera GMP-fenomenet lite och inte ta det som gospel. Det är ju inte helt okontroversiellt, speciellt inte när det gäller så få datapunkter som vi har att göra med när det gäller MOP. Det rör ju sig inte direkt om någon seriemördare som utfört tio mord på samma sätt.
Här är till exempel en forskningsrapport framställd i samverkan med fyra universitet som klassar det som pseudovetenskap:
https://web.archive.org/web/20190219192451/http://pdfs.semanticscholar.org/19b0/4f185903eea905cfd8d6d8cb75055de7527f.pdf
En snarlik rapport:
https://eprints.lancs.ac.uk/id/eprint/49930/1/Snook_et_al._CJB.pdf
Jag klistrar in lite från Wikipedia också:
In a review of the literature by Eastwood et al. (2006),[36] one of the studies noted, Pinizzotto and Finkel (1990),[55] showed that trained criminal profilers did not do any better than non-profilers in producing an accurate profile. A 2000 study also showed that profilers were not significantly better at creating a profile than any other participating groups.[56]
A survey of statements made in offender profiles done for major cases from 1992 to 2001 found that "72% included repetition of the details of what occurred in the offence (factual statements already known by the police), references to the profiler’s competence [...] or caveats about using the material in the investigation." Over 80% of the remaining statements, which made claims about the offender's characteristics, gave no justification for their conclusion.[57][22]
A 2003 study which asked two different groups of police to rate how accurately a profile matched a description of the apprehended offender, with one group given a description of a completely fabricated offender instead of the real one, found that the profile was rated equally accurate in both cases.[57][22]
There is a lack of clear, quantifiable evidence of a link between crime scene actions (A) and offender characteristics (C), a necessary supposition of the A to C paradigm proposed by Canter (1995).[58][59] A 2002 review by Alison et al. concluded, "The notion that particular configurations of demographic features can be predicted from an assessment of particular configurations of specific behaviors occurring in short-term, highly traumatic situations seems an overly ambitious and unlikely possibility. Thus, until such inferential processes can be reliably verified, such claims should be treated with great caution in investigations and should be entirely excluded from consideration in court."
The profession of criminal profiling is highly unregulated.[41] There is no governing body which determines who is and who is not qualified to be a criminal profiler, and therefore those who identify themselves as criminal profilers may range from someone with minimal to someone with extensive experience in the realm of criminal investigation.[41] In addition to the lack of criteria as to what makes an expert in the field of criminal profiling, there is little empirical evidence supporting the accuracy of criminal profiling.[42] There is an abundance of anecdotal support for criminal profiling, much of which originates from reports made by police officers and investigators regarding the performance of criminal profilers.[42] However, law enforcement agents have been found to greatly support the use of criminal profiling, but studies have shown that detectives are poor profilers themselves.[41][42]
One study presented police officers with two different profiles for the same perpetrator, each of which varied greatly from the officers’ own description.[43] It was found that the officers were unable to determine whether one profile was more accurate than the other, and felt that all profiles accurately described the perpetrator. Officers were able to find truth in whichever profile they viewed, believing it accurately described the perpetrator, demonstrating the presence of the Barnum effect.[43][44] In addition, an investigator's judgement of the accuracy of a profile is impacted by the perceived source of the information; if the officer believes that the profile was written by an “expert” or “professional”, they are likely to perceive it as more accurate than a profile written by someone who is identified as a consultant.[45] This poses a genuine problem when considering that there are no true criteria which determine who may be considered a “professional” criminal profiler, and when considering that support for criminal profiling is largely based on the opinion of police officers.[41][42]
As of 2021, although the practice of offender profiling is widely used, publicized and researched globally, there is a significant lack of empirical research or evidence to support the validity of psychological profiling in criminal investigations.[38][39] Critics question the reliability, validity, and utility of criminal profiles generally provided in police investigations. Even over the years common criminal profiling methods have changed and been looked down upon due to weak definitions that differentiate the criminal's behaviors, assumptions and their psychodynamic process of the offender actions and characteristics that occur. In other words, this leads to poor and misleading profiles on offenders because they are based on opinions and decisions made up from one profiler conducting research on the offender. Research in 2007-2008 into profiling's effectiveness have prompted researchers to label the practice as pseudoscientific.[37][40]
A 2007 meta-analysis of existing research into offender profiling noted that there was "a notable incongruity between [profiling's] lack of empirical foundation and the degree of support for the field."[37] Profiling's continued popularity has been speculatively attributed to broad use of anecdotes and testimonials, a focus on correct predictions over the number of incorrect ones, ambiguous profiles benefiting from the Barnum effect, and the popular appeal of the fantasy of a sleuth with deductive powers like Hercule Poirot and Sherlock Holmes.[35]
__________________
Senast redigerad av Knattarna 2024-10-04 kl. 12:04.
Senast redigerad av Knattarna 2024-10-04 kl. 12:04.