Polygraph Numerical Scoring Explained: Scales & Thresholds

From Subjective Judgment to Objective Measurement

In the earliest days of polygraph testing, examiners relied almost entirely on their subjective interpretation of ink tracings on a scrolling paper chart — an approach known as the "Clinical" or "Global Analysis" method [2]Verified The Numerical Scoring Analysis Predecessors
Confirms Backster's numerical scoring was more evolutionary than revolutionary, adapted from Arther's check-mark system. One examiner might see deception where another saw an ambiguous reaction, creating significant inconsistency.

The development of numerical scoring systems transformed the profession. Around 1960, Cleve Backster introduced the Backster Zone Comparison Technique, which incorporated a new polygraph chart evaluation method that numerically quantified the magnitude of examinees' psychophysiological responses to relevant questions compared to adjacent comparison questions [2]Verified The Numerical Scoring Analysis Predecessors
Confirms Backster's numerical scoring was more evolutionary than revolutionary, adapted from Arther's check-mark system. This was the first comparison question test to incorporate a numerical scoring system, using a seven-point rating scale applied to each physiological measure [13]Verified Appendix A: Polygraph Questioning and Techniques — The Polygraph and Lie Detection
Confirms the zone comparison test developed by Backster was the first comparison question test to incorporate a numerical scoring system using a seven-point scale. Research by Tuvya Amsel (2023) has shown that Backster's numerical scoring was more evolutionary than revolutionary — he likely adapted Richard Arther's check-mark system by replacing check-marks with numerical figures to create the formal Numerical Analysis system [2]Verified The Numerical Scoring Analysis Predecessors
Confirms Backster's numerical scoring was more evolutionary than revolutionary, adapted from Arther's check-mark system. You can explore how Backster's broader work influenced the field — and the rivalry that shaped modern techniques — in our article on the polygraph's development during the 1960s.

Today, every polygraph examination conducted according to American Polygraph Association (APA) standards involves numerical scoring. The system transforms analog physiological data into discrete numerical values that can be summed, compared, and evaluated against predetermined thresholds. This minimizes subjectivity and provides a transparent framework for reaching a final opinion.

The Core Principle: Relative Response Strength

The fundamental idea behind polygraph numerical scoring is relative response comparison. The examiner does not measure how nervous a person is in absolute terms. Instead, they compare the physiological response to a relevant question against the response to a carefully crafted comparison question [1]Verified The Utah Numerical Scoring System
Confirms the Utah 7-position scoring system achieves inter-rater reliability exceeding.90 and decision agreement exceeding 95% across five laboratory studies.

If the response to the relevant question is stronger, the comparison receives a negative score. If the response to the comparison question is stronger, it receives a positive score. If the two responses are essentially equal, the score is zero [1]Verified The Utah Numerical Scoring System
Confirms the Utah 7-position scoring system achieves inter-rater reliability exceeding.90 and decision agreement exceeding 95% across five laboratory studies. This relative approach accounts for individual differences — some people are naturally more reactive than others. By comparing responses within the same person during the same test, the system controls for baseline physiological variability.

This foundational principle is rooted in the psychophysiological basis of the comparison question technique, which holds that truthful persons should react more strongly to comparison questions while deceptive persons react more strongly to relevant questions. You can also learn more about how the autonomic nervous system drives these responses in our guide to electrodermal activity (EDA).

Three Primary Data Streams

A polygraph instrument simultaneously records multiple physiological parameters. The three channels that form the basis of numerical scoring are:

Pneumograph (Respiration): Two pneumograph tubes wrapped around the upper chest and abdomen measure breathing patterns. Examiners evaluate changes in rate, depth, amplitude, and baseline shifts. Deceptive responses often manifest as suppressed breathing, shallower breaths, or a momentary pause in the respiratory cycle. Swinford (1999) documented 12 distinct respiration criteria used in numerical scoring at the Department of Defense Polygraph Institute [7]Verified Manually Scoring Polygraph Charts Utilizing the Seven-Position Numerical Analysis Scale at the Department of Defense Polygraph Institute
Confirms 22 distinct physiological criteria (12 respiration, 3 EDA, 7 cardiovascular) and hierarchical scoring rules.

Electrodermal Activity (EDA/GSR): Electrodes attached to the fingers measure changes in skin conductance. When a person experiences psychological stress or arousal, sweat gland activity increases, producing measurable changes in electrical conductance. Research consistently shows that EDA data has a stronger correlation with the external criterion of deception compared to other data recorded during comparison question testing [5]Verified Numerical scoring of polygraph charts: What examiners really do
Confirms examiners achieved 82% overall accuracy in blind chart review with respiratory suppression as most frequent criterion. In field examiner surveys, over 40% of total scoring points were assigned to the EDA sensor data [5]Verified Numerical scoring of polygraph charts: What examiners really do
Confirms examiners achieved 82% overall accuracy in blind chart review with respiratory suppression as most frequent criterion. The development of this measurement has deep historical roots, explored further in our article on what electrodermal activity is and how it works.

Cardiovascular (Cardio): A blood pressure cuff placed on the upper arm captures relative changes in blood pressure and pulse rate. The examiner evaluates the tracing for baseline changes, phasic responses, sustained elevated blood pressure, and changes in pulse rate or amplitude. Swinford identified 7 distinct cardiovascular scoring criteria [7]Verified Manually Scoring Polygraph Charts Utilizing the Seven-Position Numerical Analysis Scale at the Department of Defense Polygraph Institute
Confirms 22 distinct physiological criteria (12 respiration, 3 EDA, 7 cardiovascular) and hierarchical scoring rules. These innovations in cardiovascular measurement trace back to pioneers like John Reid and his polygraph chair.

Some modern systems incorporate a fourth sensor, such as a photoplethysmograph (PPG) on a fingertip to measure peripheral vasomotor activity. Nelson (2017) demonstrated that updated scoring systems like the ESS-M can successfully incorporate vasomotor sensor data while maintaining diagnostic accuracy [6]Verified Updated numerical distributions for the Empirical Scoring System
Confirms the updated ESS-M maintained diagnostic accuracy while successfully incorporating vasomotor sensor data. ASTM International's standard E2229 establishes procedures for the systematic interpretation and analysis of polygraph data including both global and numerical methods [14]Verified ASTM E2229 — Standard Practices for Interpretation of Psychophysiological Detection of Deception (Polygraph) Data
Confirms ASTM establishes both global and numerical procedures for systematic interpretation and analysis of PDD data.

The Backbone of Modern Polygraph Scoring

Numerical scoring operates in the context of a comparison question technique. Most validated polygraph formats — including the Zone Comparison Test (ZCT), the Utah Comparison Question Test, the Federal Zone Comparison Test, and the Directed Lie Screening Test — are built around deliberate question pairing [9]Verified The Comparison Question Polygraph Test: A Contrast of Methods and Scoring
Confirms substantial main effects of guilt in both OSS2 computer and human scoring in 250-participant mock crime study.

In each test, there are typically two to four relevant questions that directly address the issue being investigated (e.g., "Did you steal the $5,000 from the safe?"). Paired with each relevant question is a comparison question that is broader and deliberately less specific (e.g., "Before age 25, did you ever take something that did not belong to you?"). For a deeper understanding of how different question types are structured, see our guide to sacrifice relevant questions in polygraph testing.

The theory is straightforward: A truthful person should react more strongly to the comparison question because the relevant question holds no emotional threat, while the comparison question taps into broader anxieties about their past. A deceptive person should react more strongly to the relevant question because it directly addresses the lie they are trying to conceal [9]Verified The Comparison Question Polygraph Test: A Contrast of Methods and Scoring
Confirms substantial main effects of guilt in both OSS2 computer and human scoring in 250-participant mock crime study.

Honts and Reavy (2015) conducted a mock crime experiment with 250 paid participants contrasting probable-lie and directed-lie CQT variants, finding substantial main effects of guilt in both OSS2 computer and human scoring, with no differences between technique types [9]Verified The Comparison Question Polygraph Test: A Contrast of Methods and Scoring
Confirms substantial main effects of guilt in both OSS2 computer and human scoring in 250-participant mock crime study. This is particularly relevant for single-issue polygraph tests, which consistently produce the highest accuracy rates. The examiner scores each comparison pair independently, for each channel, on each chart run — creating a matrix of scores that are ultimately summed.

Understanding the Two Primary Scoring Systems

The polygraph profession primarily uses two numerical scales. Both operate on the same fundamental principle, but they differ in granularity.

The 3-Position Scale uses values of -1, 0, and +1 only. If the relevant question produces a noticeably stronger response, the examiner assigns -1. If the comparison question produces a noticeably stronger response, +1. If the two responses are roughly equal, 0. The 3-position scale is simpler to apply but captures less nuance in response magnitude [1]Verified The Utah Numerical Scoring System
Confirms the Utah 7-position scoring system achieves inter-rater reliability exceeding.90 and decision agreement exceeding 95% across five laboratory studies.

The 7-Position Scale uses values from -3 to +3, allowing the examiner to express the degree of difference between responses. A score of -1 indicates a slight predominance in the relevant response, while -3 indicates a dramatic, unmistakable difference [1]Verified The Utah Numerical Scoring System
Confirms the Utah 7-position scoring system achieves inter-rater reliability exceeding.90 and decision agreement exceeding 95% across five laboratory studies. The 7-position scale provides greater measurement precision and is recommended by the APA for most validated techniques. The U.S. Defense Academy for Credibility Assessment (DACA) supports the use of the seven-position scale only for its examinations [13]Verified Appendix A: Polygraph Questioning and Techniques — The Polygraph and Lie Detection
Confirms the zone comparison test developed by Backster was the first comparison question test to incorporate a numerical scoring system using a seven-point scale.

The Utah Numerical Scoring System, formalized by Bell, Raskin, Honts, and Kircher (1999), resulted from over 30 years of scientific research and provided standardized procedures achieving the highest rates of criterion accuracy and inter-rater reliability [1]Verified The Utah Numerical Scoring System
Confirms the Utah 7-position scoring system achieves inter-rater reliability exceeding.90 and decision agreement exceeding 95% across five laboratory studies. In five laboratory studies, the interrater reliability of the Utah system exceeded.90 as measured by correlations between total numerical scores, and decision agreement exceeded 95% when both evaluators reached a definite decision [1]Verified The Utah Numerical Scoring System
Confirms the Utah 7-position scoring system achieves inter-rater reliability exceeding.90 and decision agreement exceeding 95% across five laboratory studies.

Swinford (1999) documented 22 distinct physiological reaction criteria used with the 7-position scale at the Department of Defense Polygraph Institute — 12 for respiration, 3 for electrodermal activity, and 7 for cardiovascular responses — and established hierarchical scoring rules based on the number and quality of criteria present [7]Verified Manually Scoring Polygraph Charts Utilizing the Seven-Position Numerical Analysis Scale at the Department of Defense Polygraph Institute
Confirms 22 distinct physiological criteria (12 respiration, 3 EDA, 7 cardiovascular) and hierarchical scoring rules.

Score Meanings on the 7-Position Scale

+3 — The comparison question produced a dramatically stronger response. Strong indicator of truthfulness on the relevant question. +2 — The comparison question produced a clearly stronger response. +1 — The comparison question produced a slightly stronger response. 0 — No discernible difference between the two responses, or the data is unreadable. -1 — The relevant question produced a slightly stronger response. -2 — The relevant question produced a clearly stronger response. -3 — The relevant question produced a dramatically stronger response. Strong indicator of deception.

For respiration, electrodermal activity, relative blood pressure, and peripheral vasomotor activity, a score from +3 to -3 is assigned for each presentation of a relevant question [1]Verified The Utah Numerical Scoring System
Confirms the Utah 7-position scoring system achieves inter-rater reliability exceeding.90 and decision agreement exceeding 95% across five laboratory studies. The reaction to the relevant question is compared to the reaction to a nearby comparison question. The overall score for each spot is calculated by summing the assigned values across charts for each channel.

How an Examiner Actually Scores a Polygraph Chart

The scoring process follows a systematic, structured procedure that transforms raw physiological data into a clear decision. Understanding these steps can help anyone preparing for a polygraph — whether for a pre-employment screening or a private examination.

Step 1 — Collect Multiple Charts: The examiner runs the same set of questions a minimum of three times. APA standards require at least three presentations of the question sequence. Multiple charts are essential because physiological responses naturally fluctuate, and examiners need to observe consistent patterns across repetitions.

Step 2 — Identify Comparison Pairs: For each chart, the examiner identifies the paired question sets: Relevant Question 1 vs. Comparison Question 1, and so on. In a Zone Comparison Test, the pairing follows a defined sequence; in the Federal Zone Comparison Test, the zones are structured differently.

Step 3 — Score Each Channel Independently: The examiner evaluates the pneumograph, EDA, and cardiovascular channels separately for each comparison pair [5]Verified Numerical scoring of polygraph charts: What examiners really do
Confirms examiners achieved 82% overall accuracy in blind chart review with respiratory suppression as most frequent criterion. They visually inspect the tracings, compare the magnitude and timing of the physiological response, and assign a numerical score. Capps and Ansley (1992) found that field examiners achieved 82% overall accuracy in blind chart review, with respiratory suppression being the most frequently cited criterion [5]Verified Numerical scoring of polygraph charts: What examiners really do
Confirms examiners achieved 82% overall accuracy in blind chart review with respiratory suppression as most frequent criterion.

Step 4 — Sum Across Charts: The examiner sums scores as a "spot score" (total for each relevant question across all charts) or as a "grand total" (total for all relevant questions across all charts and channels) [1]Verified The Utah Numerical Scoring System
Confirms the Utah 7-position scoring system achieves inter-rater reliability exceeding.90 and decision agreement exceeding 95% across five laboratory studies. The method depends on the specific validated technique being used. A grand total score of +6 and greater indicates nondeception, -6 and less deception, and anything in between is considered an inconclusive result.

Step 5 — Apply the Cutoff Threshold: The cumulative total is compared against predetermined cutoff thresholds. If the total meets or exceeds the positive threshold, the call is "No Deception Indicated" (NDI). If it meets or falls below the negative threshold, the call is "Deception Indicated" (DI). If it falls in between, the result is "Inconclusive" (INC).

Step 6 — Run Computer Algorithm: Most modern polygraph systems include computer scoring software that independently analyzes the chart data [3]Verified Brute-force comparison: A Monte Carlo study of the Objective Scoring System version 3 (OSS-3) and human polygraph scorers
Confirms OSS-3 exceeded the decision accuracy of 10 human scorers across six accuracy dimensions with perfect reliability. The examiner compares their hand-scored result against the algorithm output. Agreement between the two methods strengthens confidence in the final call.

Where the Numbers Meet the Decision

The cutoff threshold is the numerical boundary that separates a truthful call from a deceptive call, with an inconclusive zone in between. These thresholds are established through empirical research using confirmed-outcome studies and are calibrated to balance sensitivity (ability to detect deception) against specificity (ability to correctly identify truthful examinees).

Common threshold values for the 7-position scale with three charts are:

Grand Total of +6 or higher — No Deception Indicated (NDI / Truthful) Grand Total of -6 or lower — Deception Indicated (DI / Deceptive) Grand Total between -5 and +5 — Inconclusive (INC)

For spot scoring (evaluating each relevant question independently), the thresholds are typically: Spot total of +3 or higher — NDI on that question Spot total of -3 or lower — DI on that question Spot total between -2 and +2 — Inconclusive for that question

These thresholds reflect the optimal balance point identified by research. Moving the thresholds closer to zero would catch more deceptive individuals but increase false positives. Moving them further from zero would reduce false positives but let more deceptive individuals pass as inconclusive.

The APA's 2011 meta-analytic survey, involving 38 qualifying studies and 3,723 examinations, found that the combination of all validated techniques produced an aggregated decision accuracy of 87% (confidence interval 80%–94%) with an inconclusive rate of 13% [11]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms aggregated 89% accuracy for single-issue tests and 87% overall with 11-13% inconclusive rates across 38 studies and 3,723 examinations. For single-issue diagnostic testing specifically, the aggregated decision accuracy was 89% (confidence interval 83%–95%) with an inconclusive rate of 11% [11]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms aggregated 89% accuracy for single-issue tests and 87% overall with 11-13% inconclusive rates across 38 studies and 3,723 examinations. These findings were consistent with the National Research Council's (2003) conclusions regarding polygraph accuracy [11]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms aggregated 89% accuracy for single-issue tests and 87% overall with 11-13% inconclusive rates across 38 studies and 3,723 examinations. Understanding the differences between private and court-ordered polygraph examinations can help you anticipate how threshold standards are applied in different contexts.

How Software Independently Analyzes Polygraph Data

Computer scoring algorithms represent one of the most significant advances in polygraph science. Rather than relying solely on a human examiner's visual evaluation, these algorithms apply mathematical and statistical models to raw physiological data. Three primary algorithms dominate the professional landscape:

PolyScore: Developed at the Johns Hopkins University Applied Physics Laboratory (JHU-APL), PolyScore uses a logistic regression model to analyze features extracted from EDA, cardiovascular, and respiratory channels and produces a probability estimate of deception versus truthfulness [12]Verified Appendix F: Computerized Scoring of Polygraph Data — The Polygraph and Lie Detection
Confirms PolyScore uses logistic regression, CPS uses discriminant function analysis developed at University of Utah, and CPS relies on Stoelting instruments. PolyScore 3.2 uses a logistic regression model incorporating ten features: three each that describe galvanic skin response and blood volume and two each that describe pulse and respiration [12]Verified Appendix F: Computerized Scoring of Polygraph Data — The Polygraph and Lie Detection
Confirms PolyScore uses logistic regression, CPS uses discriminant function analysis developed at University of Utah, and CPS relies on Stoelting instruments. The 1994 patent by Harris and Olsen confirmed the logistic regression architecture [12]Verified Appendix F: Computerized Scoring of Polygraph Data — The Polygraph and Lie Detection
Confirms PolyScore uses logistic regression, CPS uses discriminant function analysis developed at University of Utah, and CPS relies on Stoelting instruments. PolyScore 3.3 achieved 90.9% accuracy on ZCT examinations in a study by Blackwell (1999), while human PDD examiners achieved 82.3% accuracy on MGQT examinations using the 7-position scale [4]Verified PolyScore 3.3 and psychophysiological detection of deception examiner rates of accuracy
Confirms PolyScore 3.3 achieved 90.9% accuracy on ZCT examinations and human examiners achieved 82.3% on MGQT.

CPS (Computerized Polygraph System): Developed by Scientific Assessment Technologies based on research conducted at the University of Utah psychology laboratory by John Kircher and David Raskin, CPS is built into Stoelting Company's polygraph software suite [12]Verified Appendix F: Computerized Scoring of Polygraph Data — The Polygraph and Lie Detection
Confirms PolyScore uses logistic regression, CPS uses discriminant function analysis developed at University of Utah, and CPS relies on Stoelting instruments. The CPS scoring algorithm is based on standard multivariate linear discriminant function analysis followed by a calculation that produces an estimate of the probability of truthfulness [12]Verified Appendix F: Computerized Scoring of Polygraph Data — The Polygraph and Lie Detection
Confirms PolyScore uses logistic regression, CPS uses discriminant function analysis developed at University of Utah, and CPS relies on Stoelting instruments. Today, Stoelting's CPS Fusion and CPS Elite software continue this lineage with updated scoring algorithms [15]Verified Stoelting CPS Elite Polygraph Systems
Confirms Stoelting's CPS Elite/Fusion software includes scientifically-validated scoring algorithms and innovative numerical scoring features.

OSS-3 (Objective Scoring System version 3): Developed by Nelson, Krapohl, and Handler (2008) at the Department of Defense, OSS-3 exceeded the decision accuracy of 10 human scorers across six accuracy dimensions and demonstrated perfect reliability [3]Verified Brute-force comparison: A Monte Carlo study of the Objective Scoring System version 3 (OSS-3) and human polygraph scorers
Confirms OSS-3 exceeded the decision accuracy of 10 human scorers across six accuracy dimensions with perfect reliability. A 2024 study by Donald Krapohl and Dutton compared OSS-3 and PolyScore on field data, finding that OSS-3 correctly identified 71% of truthful and 96% of deceptive subjects with a 14% inconclusive rate [10]Verified Comparison of OSS-3 and PolyScore in Multi-Issue and Single-Issue Field Exams
Confirms OSS-3 correctly identified 71% truthful and 96% deceptive with 14% inconclusive rate in field data comparison.

Ansley and Weir (1976) were among the earliest researchers to propose a numerical scoring approach for polygraph tests, introducing a system for Relevant-Irrelevant tests to address standardization concerns [8]Verified A numerical scoring system for Relevant-Irrelevant polygraph tests
Confirms early proposal for numerical scoring standardization of R-I polygraph tests. The ESS (Empirical Scoring System) offers an additional manual scoring alternative that does not require a computer, designed as a more evidence-based option compared to traditional seven-point scoring [6]Verified Updated numerical distributions for the Empirical Scoring System
Confirms the updated ESS-M maintained diagnostic accuracy while successfully incorporating vasomotor sensor data.

Agreement Rates and Complementary Strengths

Published research demonstrates strong concordance between human examiners and computer algorithms. A landmark U.S. Secret Service study found that the accuracy of human and computer interpretations was very high, ranging from 91% to 96% correct on confirmed truthful answers, and 85% to 95% correct on confirmed deceptive answers [12]Verified Appendix F: Computerized Scoring of Polygraph Data — The Polygraph and Lie Detection
Confirms PolyScore uses logistic regression, CPS uses discriminant function analysis developed at University of Utah, and CPS relies on Stoelting instruments. More recently, one study found that the OSS-3 algorithm was in concordance with human scorers approximately 89.5% of the time when excluding inconclusive results [10]Verified Comparison of OSS-3 and PolyScore in Multi-Issue and Single-Issue Field Exams
Confirms OSS-3 correctly identified 71% truthful and 96% deceptive with 14% inconclusive rate in field data comparison.

Nelson, Krapohl, and Handler (2008) demonstrated in a Monte Carlo study that OSS-3 exceeded the decision accuracy of 10 human scorers and 9 out of 10 individual scorers across six accuracy dimensions, while achieving perfect reliability compared to human interrater variability [3]Verified Brute-force comparison: A Monte Carlo study of the Objective Scoring System version 3 (OSS-3) and human polygraph scorers
Confirms OSS-3 exceeded the decision accuracy of 10 human scorers across six accuracy dimensions with perfect reliability.

Human examiners bring contextual awareness — they can identify artifacts like coughs, movement, or irregular breathing that might distort the data. Computer algorithms bring consistency — they apply the same mathematical model every time without fatigue or bias. When a confirmed-outcome study by Dollins, Krapohl, and Dutton (2000) tested five computer scoring algorithms on 97 cases, the proportion of correct decisions ranged from.88 to.91 when inconclusive results were excluded [10]Verified Comparison of OSS-3 and PolyScore in Multi-Issue and Single-Issue Field Exams
Confirms OSS-3 correctly identified 71% truthful and 96% deceptive with 14% inconclusive rate in field data comparison.

The modern standard in forensic psychophysiology is a hybrid approach: examiner-driven analysis supplemented by computer-assisted scoring. This combination leverages the strengths of both methods and provides the strongest basis for a defensible final decision.

When the Data Falls in the Middle

An inconclusive result does not mean the test failed or that the examinee was evasive. It simply means the numerical scores fell in the zone between the positive and negative cutoff thresholds, and the data did not produce a sufficiently clear pattern for a definitive call.

The APA meta-analysis found inconclusive rates of approximately 11% for single-issue diagnostic tests and 13% for the combination of all validated techniques [11]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms aggregated 89% accuracy for single-issue tests and 87% overall with 11-13% inconclusive rates across 38 studies and 3,723 examinations. These rates represent a deliberate design feature of the scoring system — the inconclusive zone acts as a safeguard that prevents premature calls when data quality is marginal.

Common reasons for inconclusive results include general physiological reactivity that produces similar responses to both relevant and comparison questions, inconsistent reaction patterns across chart runs, physical artifacts like coughing or movement during critical question periods, and medical conditions or medications that dampen physiological responses. If you are concerned about how anxiety might affect your results, our guide on anxiety disorders and polygraph testing addresses this directly.

When an inconclusive result occurs, standard practice is to offer the examinee an additional test at a later date. Many agencies and private examiners allow a retest after a waiting period, which often produces a definitive result on the second attempt.

What the Numbers on Your Report Mean

A professional polygraph report typically includes: the examiner's final opinion (NDI, DI, or INC), numerical scores for each channel on each chart, spot totals for each relevant question, the grand total score, and the computer algorithm's probability output.

The grand total is the single most important number. It represents the sum of all individual scores across all channels, all comparison pairs, and all chart runs. If your report shows a grand total of +9 on the 7-position scale, that means the comparison questions consistently produced stronger responses than the relevant questions — a strong indicator of truthfulness.

Computer algorithm outputs are typically expressed as probabilities. For example, a PolyScore result might read "Probability of Deception: 0.02" — meaning a 2% estimated probability that the subject was deceptive, which corresponds to a truthful call. OSS-3 similarly outputs probability values calibrated against empirical thresholds [3]Verified Brute-force comparison: A Monte Carlo study of the Objective Scoring System version 3 (OSS-3) and human polygraph scorers
Confirms OSS-3 exceeded the decision accuracy of 10 human scorers across six accuracy dimensions with perfect reliability.

Attorneys and therapists should look for agreement between the hand-scored result and the computer result. When both methods point in the same direction, confidence in the final call is substantially increased. If you are reviewing a report for legal purposes, understanding what a polygraph waiver means before signing is also essential.

The Link Between Scoring Method and Test Validity

The scoring method used is one of the most important determinants of polygraph accuracy. The APA meta-analysis demonstrated that validated techniques scored with proper numerical methods produce an aggregated decision accuracy of 89% for single-issue tests (confidence interval 83%–95%) [11]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms aggregated 89% accuracy for single-issue tests and 87% overall with 11-13% inconclusive rates across 38 studies and 3,723 examinations. Reviews of the CQT literature generally produce overall accuracy estimates of over 85% [11]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms aggregated 89% accuracy for single-issue tests and 87% overall with 11-13% inconclusive rates across 38 studies and 3,723 examinations.

The Utah Numerical Scoring System has been particularly well-validated. In five laboratory studies, interrater reliability exceeded.90, and decision agreement exceeded 95% when both evaluators reached a definite decision [1]Verified The Utah Numerical Scoring System
Confirms the Utah 7-position scoring system achieves inter-rater reliability exceeding.90 and decision agreement exceeding 95% across five laboratory studies. Capps and Ansley (1992) found that field examiners achieved 82% overall accuracy — 72.2% for truthful cases and 89.3% for deceptive cases — in blind chart review using numerical scoring [5]Verified Numerical scoring of polygraph charts: What examiners really do
Confirms examiners achieved 82% overall accuracy in blind chart review with respiratory suppression as most frequent criterion.

Computer algorithms have pushed accuracy even further. PolyScore 3.3 achieved 90.9% accuracy on ZCT examinations [4]Verified PolyScore 3.3 and psychophysiological detection of deception examiner rates of accuracy
Confirms PolyScore 3.3 achieved 90.9% accuracy on ZCT examinations and human examiners achieved 82.3% on MGQT, and OSS-3 has demonstrated the ability to exceed the accuracy of experienced human scorers across multiple accuracy dimensions [3]Verified Brute-force comparison: A Monte Carlo study of the Objective Scoring System version 3 (OSS-3) and human polygraph scorers
Confirms OSS-3 exceeded the decision accuracy of 10 human scorers across six accuracy dimensions with perfect reliability. The combination of human expertise with algorithmic precision represents the gold standard for modern polygraph practice.

Professional standards organizations like the APA require the use of validated techniques with empirically supported scoring rules. For law enforcement applicants — whether in Georgia, Utah, or Massachusetts — this means that the scoring methods used on your examination have been tested and validated through peer-reviewed research.

Regulatory Framework and Quality Assurance

Numerical scoring in polygraph is governed by multiple layers of professional standards. ASTM International Committee E52 on Forensic Psychophysiology has developed a comprehensive series of standards covering all aspects of polygraphy [14]Verified ASTM E2229 — Standard Practices for Interpretation of Psychophysiological Detection of Deception (Polygraph) Data
Confirms ASTM establishes both global and numerical procedures for systematic interpretation and analysis of PDD data. Key standards include:

ASTM E2229 — Standard Practices for Interpretation of Psychophysiological Detection of Deception (Polygraph) Data: This standard establishes both global and numerical procedures for the systematic interpretation and analysis of PDD data, requiring that examiners use methods for which they have been formally trained [14]Verified ASTM E2229 — Standard Practices for Interpretation of Psychophysiological Detection of Deception (Polygraph) Data
Confirms ASTM establishes both global and numerical procedures for systematic interpretation and analysis of PDD data.

ASTM E2031 — Standard Practice for Quality Control of Psychophysiological Detection of Deception (Polygraph) Examinations: This standard establishes procedures for quality control review of polygraph examinations by experienced examiners [14]Verified ASTM E2229 — Standard Practices for Interpretation of Psychophysiological Detection of Deception (Polygraph) Data
Confirms ASTM establishes both global and numerical procedures for systematic interpretation and analysis of PDD data.

ASTM E1954 — Standard Practice for Conduct of Research in Psychophysiological Detection of Deception: This standard governs the qualifications required of polygraphic researchers and outlines how research should be conducted [14]Verified ASTM E2229 — Standard Practices for Interpretation of Psychophysiological Detection of Deception (Polygraph) Data
Confirms ASTM establishes both global and numerical procedures for systematic interpretation and analysis of PDD data.

The APA Standards of Practice further require that members use validated techniques with published research supporting their validity [11]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms aggregated 89% accuracy for single-issue tests and 87% overall with 11-13% inconclusive rates across 38 studies and 3,723 examinations. The textbook Fundamentals of Polygraph Practice by Donald Krapohl and Pamela Shaw (2015) provides a comprehensive reference for evidence-based polygraph practice, covering data analysis, testing techniques, and scoring methodology [16]Verified Fundamentals of Polygraph Practice
Confirms comprehensive reference for evidence-based polygraph practice by Donald Krapohl and Pamela Shaw, published by Academic Press/Elsevier in 2015. Together, these standards ensure that numerical scoring is applied consistently, transparently, and in accordance with the best available scientific evidence — whether the examination is conducted for PCSOT purposes, employment screening, or private investigation.

For a complete overview of how polygraph practice has evolved from its earliest origins to the present day, visit our Polygraph History Timeline: 1875–2025.