FIGURE H-9 Boxplots of accuracy index (A) values for all 52 laboratory datasets, the subgroup of 14 datasets with internal validity scores better than 3, and the subgroup of 4 datasets with both internal validity and salience scores better than 3.

value, often blur in the context of polygraph research. Only three of these seven studies were funded by agencies with polygraph programs. However, one study classified as locally funded academic research was conducted in the university laboratory of a polygraph examiner and former employee of the U.S. Department of Defense Polygraph Institute. Another, with A essentially equal to 1.0 in the above figures, was the doctoral dissertation of the president of a polygraph company, based on data from polygraph tests that used a specific variant of the control question test, with an associated scoring mechanism, that the author had developed some years earlier. Of 122 polygraph tests from criminal investigations that were examined in this particular study, there were seven inconclusive tests but no false positive or false negative errors.

Due to the small number of field studies available, we use a slightly more lenient criterion in restricting to higher quality studies and consider



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement