Validity of polygraph techniques and decision methods

This comprehensive review chapter by Raskin and Kircher synthesizes decades of polygraph research on the Comparison Question Test (CQT), examining methods, scoring systems, decision rules, and validity evidence from laboratory and field studies. The authors argue that CQT polygraph testing achieves approximately 89% accuracy in criminal investigations when properly conducted, refuting the 2003 National Research Council critique.

Comprehensive review and synthesis of laboratory analog studies and field investigations, analyzing CQT methodology variants, numerical scoring systems including Utah method and OSS2 computerized scoring, decision rules, and validity evidence from multiple decades of research.

This 2014 chapter by Raskin and Kircher synthesizes decades of polygraph research to comprehensively address the validity and decision methods of the Comparison Question Test. The authors systematically review CQT methodology, scoring systems including computerized algorithms, and decision rules while analyzing accumulated validity evidence from laboratory experiments and field investigations. Results indicate approximately 89% accuracy for criminal investigations and 85% for screening applications, with field studies demonstrating accuracy exceeding 90% under proper conditions. The authors directly refute the National Research Council's 2003 critique and compare polygraph diagnostic reliability favorably to other psychological and medical diagnostic procedures. This work represents a major defense of polygraph validity and updates earlier handbooks with emerging methods and contemporary research evidence.

The comprehensive validity evidence supports the continued use of properly administered CQT procedures in criminal investigations and screening contexts, with implications for forensic practice standards and the need for skilled examiner training. Findings challenge scientific critiques of polygraph testing and establish benchmarks for acceptable accuracy in operational settings.

Background & Context

The 2014 Raskin and Kircher chapter represents a comprehensive synthesis of decades of polygraph research by two of the most influential figures in the field. David Raskin and John Kircher, both from the University of Utah, have been at the forefront of polygraph research since the 1970s, developing the Utah approach to comparison question testing and pioneering computerized scoring systems.

This chapter describes current methods and uses of polygraph techniques for detection of deception, provides a detailed description of the Comparison Question Test (CQT) and various analytic methods, and analyzes the scientific research and validity of the CQT while comparing it to other psychological and medical diagnostic procedures. The work explicitly refutes the National Research Council's 2003 critique of polygraph testing, representing a major defense of polygraph validity in the ongoing scientific debate.

Published in the Elsevier volume "Credibility Assessment: Scientific Research and Applications," this chapter updated the 2001 Handbook of Polygraph Testing and expanded coverage to include emerging biometric and behavioral methods alongside traditional polygraph techniques.

Research Design & Methodology

This is a comprehensive review and synthesis chapter rather than an empirical study. The authors analyzed the accumulated body of polygraph research, including both laboratory analog studies and field investigations conducted over several decades.

The chapter systematically examines:

• CQT methodology and variants — Detailed analysis of probable-lie and directed-lie comparison question formats
• Scoring systems — Review of numerical scoring methods including the Utah system and computerized algorithms like OSS2 (Objective Scoring System Version 2)
• Decision rules and cut-scores — Analysis of various methods for rendering decisions from physiological data
• Validity evidence — Synthesis of laboratory experiments and confirmed field studies with ground truth criteria

The authors provided extensive review of CQT methods, scoring techniques, decision rules and classification accuracy, with evidence that specific issue evidentiary exams correctly classify approximately 92% of computer-scored cases. The review integrates research from the University of Utah laboratory studies, government-sponsored field research, and independent investigations.

Results & Key Findings

The estimated percentage of correct decisions for polygraph tests is approximately 89% for specific-incident criminal investigations and 85% for screening applications. These accuracy estimates are based on synthesis of multiple studies with verified ground truth.

Key findings from the review include:

• In laboratory studies using the Utah numerical scoring system, the overall percentage of correct decisions was 91% for guilty subjects and 89% for innocent subjects, excluding inconclusive outcomes
• Field research demonstrates that the accuracy of the CQT can exceed 90% when properly employed by skilled examiners
• The interrater reliability of the Utah scoring system exceeded .90, with percent agreement on decisions exceeding 95% when both evaluators reached definite decisions
• The Objective Scoring System Version 2 (OSS2) was identified as the most accurate of currently available computer-based polygraph scoring algorithms

The authors compare the diagnostic reliability and validity of polygraph tests to other commonly used psychological and medical diagnostic procedures, with findings refuting the National Research Council's attack on the polygraph and including discussion of accuracy on psychopaths, victims of crimes, confidential tests, and government uses.

Discussion & Significance

This chapter represents the most comprehensive defense of CQT polygraph validity published by leading proponents in response to the 2003 National Academy of Sciences report, which had concluded that the scientific basis of the CQT was weak. The work is cited alongside other major reviews including the National Research Council (2003) and the American Polygraph Association (2011) meta-analyses, with reviews generally producing overall accuracy estimates of over 85%.

The authors argue that properly conducted CQT examinations, using validated scoring methods and skilled examiners, achieve accuracy levels comparable to many accepted diagnostic procedures in psychology and medicine. The findings indicate that laboratory experiments can shed light on processes involved in CQT use on criminal suspects, though adjustments in decision criteria should reduce false positive errors caused by overprediction, and computer methods can increase objectivity and accuracy.

The chapter has become a foundational reference in polygraph literature, frequently cited in both practitioner guidelines and academic debates about polygraph validity. It provides the theoretical and empirical foundation for the Utah approach to polygraph testing, which remains influential in both government and private sector applications.

Limitations & Considerations

As a review chapter rather than original empirical research, the work synthesizes existing studies with varying methodological quality. Critics have noted that the authors are proponents of polygraph testing with commercial interests in polygraph technology, which may influence interpretation of the evidence.

The accuracy estimates are contested by polygraph critics, who argue that field study accuracy may be inflated by confirmation bias, selective sampling, and lack of truly independent ground truth verification. The 2003 NAS report had concluded that the scientific basis of the CQT was weak, extant research was of low quality, polygraph profession's claims for high accuracy were unfounded, and although the CQT has greater than chance accuracy, its error rate is unknown — criticisms this chapter directly challenges but which remain points of ongoing scientific debate.

Practical Applications

This chapter provides the scientific foundation for modern comparison question polygraph testing in criminal investigation, security screening, and other forensic applications. The review supports the use of standardized numerical scoring systems, particularly the Utah method and computerized algorithms like OSS2, which offer improved reliability and objectivity over subjective chart interpretation.

For practitioners, the chapter emphasizes the importance of proper training, skilled administration, and adherence to validated protocols to achieve the accuracy levels reported in quality research. For consumers and policy makers, it argues that polygraph testing, when properly conducted, can be a valuable forensic tool with accuracy comparable to other accepted diagnostic procedures, though the ongoing scientific controversy means that reliance on polygraph results should be informed by awareness of both the supporting evidence and continuing criticisms.