posted 12-14-2007 01:27 AM            

I think they have been doing some experiments with Dan Sosnowski (hey, that may explain some things...) - no, seriously, polygraph experiments, using monitoring test and drug tests.

Madsen presented their data, and stated to the ATSA conference participants that their study was hampered by "not enough liars," because of the comparatively few number of offenders who were caught actually using drugs, then he presented an average of decision accuracy across truthful and deceptive results, and offered an accuracy estimation of 75%.

Madsen is neither stupid nor uneducated. He knows that his math was dishonest and misleading - he should have used weighted averaging - because means are non-resistant to base rates.

He may be an OK guy, but I think its worth watching carefully what he does.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)