Effectiveness of the Air Force Modified General Question Test Format

Validation of the USAF Modified General Question Test format achieving 83.8% total accuracy.

Field study using actual criminal investigations or operational examinations with confirmed outcomes serving as ground truth.

Published in 2008 in Polygraph, 37(3), 173–192, this work by Senter and colleagues examined effectiveness of the air force modified general question test format. Key findings include: Validated the USAF-MGQT format with total accuracy of 83.8% and definitive accuracy of 84.9%. Demonstrated effectiveness of the military’s standard examination format. As a field study conducted with real cases, this research provides ecological validity that laboratory studies cannot match.

Validation of the military's standard format provides quality assurance data and benchmarks for ongoing program assessment.

Background & Context

The United States Air Force Modified General Question Test (USAF-MGQT) represents a modern evolution of comparison question testing developed to meet the specific operational needs of military and federal polygraph programs. This technique became widely adopted across federal agencies due to its efficient structure, adherence to accepted validation principles, and flexibility to adapt to both multi-facet investigative needs and multi-issue screening contexts.

Prior to the Senter, Waller, and Krapohl study, validation research on older MGQT variants existed, but only one published study had specifically addressed the criterion validity of the modernized USAF-MGQT format. This research gap was significant given the technique's widespread use throughout federal agencies. The 2008 validation filled a critical need for empirical evidence supporting the AFMGQT's diagnostic accuracy in controlled conditions.

The AFMGQT had become a polygraph technique widely used across the United States Federal Government for both criminal specific and counterintelligence screening purposes, making rigorous validation essential for maintaining quality assurance standards and establishing performance benchmarks.

Research Design & Methodology

This controlled laboratory study provided evidence for the diagnostic value afforded by the Air Force Modified General Question Test format. The research employed a mock crime scenario paradigm to establish ground truth, allowing researchers to know with certainty which participants were truthful and which were deceptive.

Key methodological features included:

• The laboratory study included N=69 participants who were assigned to either truthful or deceptive conditions
• The base rate of deception in the study was approximately 50%, providing balanced representation
• The study employed blind scoring procedures to eliminate examiner bias and ensure objective evaluation
• Charts were evaluated using the standard seven-position numerical scoring system used by federal examiners
• Research by Pollina, Dollins, Senter, Krapohl, and Ryan (2004) provided evidence for physiological similarities between laboratory and field-based research, supporting the ecological validity of the laboratory approach

The USAF-MGQT format tested in this study follows the federal standard protocol with directed-lie comparison questions, relevant questions targeting the test issue, and appropriate neutral/sacrifice questions to establish baseline physiological responses.

Results & Key Findings

The study demonstrated that the AFMGQT produced total and definitive accuracy rates that significantly exceeded chance levels for both truthful and deceptive participants. The headline finding showed the technique achieved 83.8% total accuracy and 84.9% definitive accuracy (excluding inconclusive decisions).

Detailed accuracy metrics included:

• Total Accuracy: 83.8% — Overall correct classification rate including all decisions
• Definitive Accuracy: 84.9% — Accuracy rate excluding "no opinion" decisions from calculations
• Mean blind-scoring criterion accuracy level of .849 when inconclusive results were excluded
• Both original examiners and blind scorers achieved statistically significant accuracy levels

The study employed rigorous statistical analysis including effect size calculations to assess the magnitude of accuracy differences. Multi-variate analysis from extended analyses of this data found no significant differences between total and subtotal scores of the laboratory sample and those from field investigation cases using the same technique, suggesting results generalize beyond laboratory conditions.

These accuracy rates positioned the USAF-MGQT within the performance range of other validated comparison question techniques, providing empirical support for its continued operational use.

Discussion & Significance

This validation study represented a landmark achievement for military polygraph programs by providing the first comprehensive, controlled evaluation of the modernized AFMGQT format. The accuracy rates achieved — exceeding 83% — demonstrated that the technique performs at levels consistent with other validated polygraph formats and well above chance prediction.

The significance extends beyond simple accuracy metrics. By employing blind scoring procedures, the study eliminated potential examiner bias and demonstrated that the physiological data itself contained sufficient diagnostic information for accurate classification. This methodological rigor strengthened confidence in the technique's validity and established baseline performance expectations for quality control programs.

For federal polygraph programs, these findings provided crucial benchmarking data. Additional and continuous research was recommended to expand the body of knowledge pertaining to the variety of uses afforded by the AFMGQT. The study became a foundational reference for subsequent AFMGQT research and helped establish the technique's place among validated polygraph methods recognized by professional standards.

Limitations & Considerations

The limitations of using laboratory scenarios in validation studies are well documented, and should be taken into consideration. While laboratory research offers controlled conditions and certainty about ground truth, it may not fully replicate the psychological stakes and complexities of real-world investigations.

The degree of generalizeability of the results to other scenarios and real world contexts is unknown. Ultimately it is unknown how the AFMGQT would perform with other types of scenarios and operational situations. Different crime scenarios, varying levels of consequence, and diverse examinee populations might yield different accuracy rates.

In the real world the base rate of deception might be significantly different from the 50% value used in the study. Base rate variations affect the predictive value of test outcomes, with implications for operational decision-making that extend beyond raw accuracy percentages.

The primary limitation involves the relatively small sample size, though the N=69 provides adequate statistical power for the analyses conducted. Larger samples would strengthen confidence intervals and enable more detailed subgroup analyses.

Practical Applications

For polygraph examiners and federal agencies, this research validates the continued use of the USAF-MGQT as a scientifically supported testing format. The demonstrated accuracy rates exceed minimum standards and provide quality assurance benchmarks against which individual examiners and programs can evaluate their performance. Agencies can cite these findings when defending the technique's reliability in administrative or legal contexts.

For consumers of polygraph services, these findings indicate that when conducted by trained examiners following standardized protocols, the AFMGQT achieves accuracy rates substantially better than chance. However, the laboratory setting limitations mean that real-world accuracy may vary based on case factors, examiner skill, and examinee characteristics. The technique performs best when applied to specific incident investigations where ground truth can eventually be established and quality control maintained through ongoing monitoring and blind rescoring programs.