John M. wrote on Aug 20
th, 2017 at 11:28pm:
A better question is why not?
Precisely.
Here is a yet a new wrinkle to the pseudoscience of polygraph "testing"...
The latest issue of the APA's house organ
Polygraph magazine features an article enthusiastically supporting the use of Bayes' Theorem in the analysis of polygraph "test" results.
Coincidentally, a newly published article --
Bayesian inference for interpretation of polygraph results in the courtroom -- appearing in
Law, Probability & Risk, which publisher Oxford University Press describes as "a fully refereed journal which publishes papers dealing with topics on the interface of law and probabilistic reasoning," by polygraph critic Alan Zelicoff, co-authored by Steven E. Rigdon, apparently
shoots holes in the argument espousing the use of Bayes' Theorem in connection with polygraph "testing." Here's an excerpt from the publisher's abstract of the Zelicoff/Rigdon article:
Despite more than 80 years of field use, there is very limited data that may be applied to assess the diagnostic utility of the polygraph for purposes of establishing the innocence of a defendant. In this article, we present a fully Bayesian analysis of what is probably the largest and most realistic existing data set and we obtain the posterior distribution for the PPV (the probability of guilt conditioned on ‘failing’ the polygraph) and the NPV (the probability of innocence conditioned on ‘passing’ the polygraph). We show that these quantities have a high degree of uncertainty that is often unexpressed when just point estimates are given. On its web site, the APA continues to make this claim:
"...APA examiners are able to attain accuracy rates exceeding 90 percent." Such boasting is misleading at best. Polygraph consumers worldwide should beware -- and look cautiously to any APA response to the Oxford article authored by Zelicoff and Rigdon.
Why do polygraph professionals continue to ignore the NAS report and characterize polygraph "test" accuracy in terms of a percentage? In my professional opinion -- and I'm a full member of the APA with 13 years of experience --polygraph "testing" is mainly about one thing: MONEY. Expressing polygraph accuracy in percentages is key to winning (and keeping) business.
When consumers grasp the reality of polygraph accuracy, they quickly lose interest in the "test."
As I see it, here's the bottom line: Even with the statistical contortions afforded by Bayes' Theorem, polygraph "testing" is still a SWAG -- a Scientific Wild-Ass Guess.