STUDIES PROVE POLYGRAPH RESULTS ARE DECEPTIVE & UNRELIABLE, YET TESTS ARE STILL IN WIDESPREAD USE

A few months ago while wired up to a polygraph machine your favorite Clearwater Criminal Defense Attorney examined reasons why you should never take a polygraph test if asked by law enforcement officers. Having fully recovered from the experience of telling the truth albeit for only an hour, today we'll look at the scientific studies that establish the unreliability of polygraph tests. 

Cezanne, Truth or Polygraph, 1900

In Federal Courts polygraph evidence is never admissible. Why? Because the accuracy of polygraph results has never been established with supporting scientific evidence. In Florida Courts polygraph evidence may be admissible but only if the prosecution and the defense agree. Yet it's interesting to note Florida law requires that convicted sex offenders take at least one annual polygraph every year for "risk management and treatment and to reduce the sex offender's denial mechanisms..."; clearly with the hope that the convicted sex offenders will give accurate information because of the threat that inaccurate information will be ferreted out during polygraph exams. 
Some of the factors that affect polygraph examination validity include the setting of the enquiry,  the quality of the examiner, the quality of the examiner's questions technique, the subject's countermeasures as well as the subject's mental and physical state. All of these variables result in inaccurate tests, which an evaluation study of all of the scientific studies gave as an accuracy range as low as 64% to as high as 98% without any perfect scores. The National Research Counsel of the National Academy of Sciences also looked at the evidence finding that polygraph gave inaccurate, questionable results.
Even the folks who license polygraph examiners at the American Polygraph Association, in putting forth their best arguments as to the validity of the exams state the following:

The data showed that techniques intended for event-specific (single issue) diagnostic testing produced an aggregated decision accuracy of 89% (confidence interval of 83% - 95%), with an estimated inconclusive rate of 11%.  Polygraph techniques in which multiple issues were encompassed by the relevant questions produced an aggregated decision accuracy of 85% (confidence interval 77% - 93%) with an inconclusive rate of 13%.

If a hundred innocent folks are asked questions concerning a murder under a polygraph exam, twenty-three of the innocents will be falsely accused and if someone with guilty knowledge is questioned she (because the best murderers are always ladies) has nearly a one if four chance of not being found out. 
So how can law enforcement get away with using such a flawed polygraph test system? Because it's what they've got. 
Which brings your Clearwater Criminal Defense Lawyer to the paradox of lie detector testing: clearly it can be a very effective tool for law enforcement when someone suspects the lie detector will be accurate, because the subject will parse words carefully in hiding deception or in admitting incriminating facts.