U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Comparative Analysis of Polygraph With Other Screening and Diagnostic Tools

NCJ Number
202090
Journal
Polygraph Volume: 32 Issue: 2 Dated: 2003 Pages: 57-85
Author(s)
Philip E. Crewson
Date Published
2003
Length
29 pages
Annotation
This limited literature review, published between January 1986 and May 2001, examined the accuracy and reliability of screening and diagnostic tests in the fields of polygraphy, medicine, and psychology.
Abstract
Out of 1,158 articles and abstracts reviewed, 145 fit the objectives of the literature review, yielding data on 198 studies. For field screening assessments, the sensitivity of polygraph, medical, and psychological tools measured .59, .79, and .74, respectively. The specificity of polygraph, medical, and psychological screening measured .90, .94, and .78. For field diagnostic assessments, the sensitivity of polygraph, medical, and psychological tools was .92, .83, and .72. The specificity of polygraph, medical, and psychological diagnostic testing was .83, .88, and .67, respectively. Agreement was determined with kappa. Among readers in polygraph, medicine, and psychology, kappa was .77, .56, and .79, respectively. The polygraph's accuracy and reliability (agreement) on specific issues was apparently consistent with published studies on medical and psychological assessment tools; however, the literature manifested an enormous range of accuracy and agreement in the performance of tools in all three fields. Although there were few polygraph screening studies, accuracy reports were lower than those in medicine and psychology. Based on these findings, the author advises that it is unlikely that polygraph research will be able to reach a level of accuracy and reliability sufficient to quell opposition arguments. Polygraph testing suffers from the same flaws as the other diagnostic tools examined in this literature review. These tools do not produce 100-percent accuracy, nor will their applications from one subject to another or by one examiner to another be invariant. 12 tables, 2 figures, and a list of publications reviewed

Downloads

No download available

Availability