http://www.washingtonpost.com/wp-dyn/content/article/2006/04/30/AR2006043001 006.html
You may need to register for NYTimes to access the article. According to a quick read, this is the federal govt take on the validity: In a sample of 10,000 which contains 10 'spies' 2 spies would be missed and 1600 innocents would be incorrectly nabbed. I'm off to class, will figure the false + and - there ... BUT, one would have to be very foolish to base any major decision on one piece of evidence - and this includes p values (a lengthy thread in itself). On the other hand, would one be equally foolish to discard bits of evidence that predict better than chance? ----------------------------- John W. Kulig Professor of Psychology Director, Psychology Honors Plymouth State University Plymouth NH 03264 ----------------------------- --- To make changes to your subscription go to: http://acsun.frostburg.edu/cgi-bin/lyris.pl?enter=tips&text_mode=0&lang=english