What are we critics to make of these Computer Polygraph Scoring Algorithms. I'm told there exists now the Johns Hopkins University PolyScore Ver. 5.5, Identifi 6.0, Axciton Algorithm, OSS / Objective Scoring System, Lafayette's Quest Algorithm, & others !
My question is if these Phd's writing these are using only confirmed material (validated Truthfulls and validated Deceptives) then don't they have a basis for claiming objectivity, rather than subjectivity, as they compare known solution date to the new tests / variables ?
Therefore, isn't the future of polygraph type technology headed for computer scored tests, rather than the subjectiveness which has been criticized here. Second, if and when these independant algorithms concur in their findings, then doesn't that lead to the findings being statisticly significant ?
TheNoLieGuy
Quote from: TheNoLieGuy4U on Nov 08, 2004, 04:01 PMMy question is if these Phd's writing these are using only confirmed material (validated Truthfulls and validated Deceptives) then don't they have a basis for claiming objectivity, rather than subjectivity, as they compare known solution date to the new tests / variables ?
I don't doubt that computerized scoring algorithms can remove an element of subjectivity from polygraph chart scoring. But the problem remains that the underlying procedure (CQT polygraphy) lacks scientific underpinnings.
But here is what the National Academy of Sciences concluded regarding computerized scoring algorithms (http://books.nap.edu/books/0309084369/html/320.html):
"There has yet to be a proper independent evaluation of computer scoring algorithms on a suitably selected set of cases, for either specific incidents or security screening, which would allow one to accurately assess the validity and accuracy of these algorithms."
You also ask:
QuoteTherefore, isn't the future of polygraph type technology headed for computer scored tests, rather than the subjectiveness which has been criticized here.
In the near term, I think we may well see more use of computerized scoring algorithms as the polygraph community struggles to maintain its waning ability to hoodwink the public. But the ultimate destination of polygraphy is atop the trash heap of bad ideas (alongside other pseudosciences such as phrenology and graphology).
QuoteSecond, if and when these independant algorithms concur in their findings, then doesn't that lead to the findings being statisticly significant ?
No, again because of the underlying procedure's lack of scientific underpinnings. If computerized astrological charts concur in their findings, does that mean they are statistically significant?