Monte Carlo Study of Criterion Validity of the Directed Lie Screening Test Using the Empirical Scoring System and the Objective Scoring System Version 3

• NCJ Number: 240330
• Journal: Polygraph Volume: 41 Issue: 3 Dated: 2012 Pages: 145-155
• Author(s): Raymond Nelson; Mark Handler
• Date Published: 2012
• Length: 11 pages
• Publication Format: Article
• Publication Type: Report (Study/Research)
• Language: English
• Country: United States of America

Annotation

This study is intended to extend the validation data on the DLST/TES by investigating criterion accuracy with the ESS and the OSS-3.

Abstract

Monte Carlo methods were used to calculate normative data and criterion accuracy rates for PDD screening tests conducted using the DLST format for multi-issue examinations. Decision accuracy was significantly greater than chance for all DLST cases and for the deceptive and truthful groups. There were no significant differences in unweighted decision accuracy for the two models. The OSS-3 model produced significantly fewer inconclusive results with truthful cases compared to the ESS. Dimensional profiles of criterion accuracy, including means, standard deviations, and statistical confidence intervals are shown for test sensitivity and specificity, false negative and false-positive errors, inconclusive rates for deceptive and truthful cases, positive predictive value, negative predictive value, the proportions of correct decisions for deceptive and truthful cases, and unweighted decisions accuracy. Normative lookup tables for the DLST scores with the ESS are shown in an appendix continued interest in the DLST, the ESS, and the OSS-3 is recommended. (Published Abstract)