![]() |
home | contact us | |
| w w w . m a t t e p o l y g r a p h . c o m |
![]() An Assessment of the Backster "Either-Or" Rule in Polygraph Scoring
by Eldad Meiron, Donald J. Krapohl and Tzachi Ashkenazi We investigated the Backster “Either-Or” Rule (EOR) using 100 Backster field cases which were blind scored by field examiners who used either the Backster scoring system or the US federal scoring system. The Backster scorers indicated those scores which were the product of the EOR. Without the EOR the Backster scorers made an average of 70% correct decisions, 3% errors, and 27% inconclusive results. With the EOR they rendered an average of 80% correct decisions, 3% errors, and 17% inconclusives. All of the increase in decision accuracy and reduction in inconclusive rate was attributable to improvement in classifying deceptive cases. There was no effect on truthful cases. The scores from the US federal system produced an average of 86% correct decisions, 5% errors, and 9% inconclusives. On the second research question of the relative accuracy of the Backster and Federal Scoring systems, without inconclusives the Backster scorers were correct in 96% of their decisions (16% inconclusive) and the Federal scorers were correct in 94% of their decisions (9% inconclusive). Decision accuracy, inconclusives and errors between the US federal and the Backster (with EOR) were not significantly different, though both performed better than the Backster method without EOR. The implications and limitations of these findings are discussed. The above abstract is from a research report presented at the 43rd Annual Seminar/Workshop of the American Polygraph Association (APA) in Indianapolis, Indiana in August 2008. The complete study can be found in Volume 37, Number 4, 2008, of Polygraph, Journal of the American Polygraph Association. www.polygraph.org |
Copyright © 1996 by James Allan Matte. All Rights Reserved. 43 Brookside Drive | Williamsville, NY 14221-6915 | (716) 634-6645 | FAX: (716) 634-7204E-mail: JamesAllanMatte@mattepolygraph.com |