LieBabyCryBaby wrote on Nov 27
th, 2006 at 11:34pm:
Well, that depends on how you look at it, George. I don't believe that all of the "false" positives are actually false, and I don't believe false positives are as common in the real world as you want the average reader to believe. The key word is IF. IF some of you are actually false positives, then you are indeed victims of an imperfect instrument or an imperfect polygraph examiner. IF that is the case, as sad as that may be, you have to move on. At least most of you aren't the victim of a botched medical procedure, mechanical failure in an automobile or airplane, an industrial accident, an accidental firearm discharge, a lightning strike, or a meteor shower. The world isn't perfect, nor or the machines made by men. I know from experience that the polygraph works almost all of the time in field conditions, and that it takes much more to be a false positive than you would like the average reader to believe. When the "cost of doing business" is viewed through my lenses rather than yours, it is much easier to swallow.
LCBC,
False positives and negatives are a function of both the
accuracy and the
base rate of what you're trying to detect...
If your beloved polygraph has only 90% accuracy, it doesn't matter what you believe, reality wills out...
If the base rate of deception is low, you will have a significant amount of false positives (e.g., employment screening)...
If the base rate of deception is high, you will have a significant amount of false negatives (e.g., sex offenders)...
The only time these balance out is if the base rate of deception is 50%. Surely you can't believe that 50% of all applicants for law enforcement are lying or that only 50% of sex offenders are lying. You're only deceiving yourself if you believe otherwise...
In addition, the rest of your comparisons of the polygraph failure to illdone medical procedures, acts of god, or mechanical failure are simply misdirection. A better comparison would be to medical screening which has significantly better methods (e.g., independent sequential testing, more accurate tests) to reduce the occurence of false positives and negative to your wishful thinking of below 1/1,000,000...
You can tap dance all you want and claim that only experience matters, but the laws of probability are pretty much immutable. Not only does the CQT polygraph have no scientific basis, its self proclaimed accuracy reduces its usefulness to nil...