What about the effectiveness and accuracy of the polygraph?

What about the effectiveness and accuracy of the polygraph?

What about the effectiveness and accuracy of the polygraph?


It is important to note that "Accuracy" is a combination of "Reliability" and "Validity". A polygraph test is a standardized, evidence-based test of the margin of uncertainty or level of confidence surrounding a categorical conclusion of deception or the possession of knowledge or information regarding a test target issue.  Over the past 75 years, more than 250 studies have been conducted on the validity, accuracy and reliability of polygraph testing (American Polygraph Association 1996 Polygraph Issues & Answers). Based on twelve separate studies involving 2174 real cases, evidence suggests that qualified field polygraph examiners are up to 98% accurate in their overall decisions (Norman Ansley, "The validity and reliability of polygraph decisions in real cases", Polygraph, v.19, 1990). This level of confidence is well within the guidelines for scientific measurements in other fields.  While the polygraph technique is not infallible, research clearly indicates that when administered by a competent examiner, the polygraph test is one of the most accurate means available to establish truth and deception. Several different factors that influence the outcome of the tests are, for example, what examination technique was used by the examiner? Did the examiner ask the correct questions? Did the examiner do a thorough background interview with the client before the test? Did the subject fully understand and comprehend the questions asked? Did the examiner interpret the data on the charts recorded correctly?

Some truthful subjects are concerned that their nervous tension may cause them to fail.  However, the polygraph technique allows for a person's general nervous tension. Most people will be nervous before and during a polygraph examination. This only affects the level from which recordings are made and does not affect the result of the examination. In other words, the level of nervousness will affect the individual's physiological baseline, but it will not be affected by the variations on the baseline and consequent result of the examination.  If nervousness adversely affect the overall result of a polygraph test - then no one would ever pass a polygraph test! 

Can someone be forced to take a polygraph test?

What about the effectiveness and accuracy of the polygraph?

What about the effectiveness and accuracy of the polygraph?


It is against the Constitution of South Africa to force a person to undergo a polygraph examination. The test must be taken voluntarily and a informed consent form must be signed by both the polygraph examiner and the subject. This is also confirmed verbally which is also recorded on camera.  However, a court may rule that a polygraph must be conducted. For staff of a company, it is advisable that a clause, reserving the employer’s right to polygraph employees is included in the employment contracts of all employees or employees in high risk positions. This must be done in accordance with South African Labour Law. If they fail the test, the employer can argue break of trust relationship. If they refuse, the employer can argue breach of contract (In that respect, the employee who signed a contract including a polygraph requirement is "forced" to undergo testing if required to do so).

Generally, employers are permitted to use the polygraph to investigate incidents where:

 [1] Employees had access to the property which is the subject of the investigation;

[2] There is a reasonable suspicion that the employee was involved in the incident;

[3] There has been economic loss or injury to the employer's business like theft of company property;

[4] The employer is combating dishonesty in positions of trust;

[5] The employer is combating serious alcohol, illegal drugs or narcotics abuse and fraudulent or illicit behaviour within the company;

[6] The employer is combating deliberate falsification of documents and lies regarding true identity of the people involved;

The polygraph examination results can only be released to the person that has been authorized. It can be the person who has undergone the polygraph test, or anyone designated in writing by the subject, firm, corporation or government department that requested the examination be done. 

What about the use of the polygraph in courts or the CCMA?

What about the effectiveness and accuracy of the polygraph?

What about the use of the polygraph in courts or the CCMA?


In South Africa, there is currently no legislation regulating the use of the polygraph. Therefore, there is nothing prohibiting or preventing the submission of polygraph results to corroborate other evidence in our Courts of Law. This will be considered by the magistrate or judge. It is also legal to request employees to voluntarily subject themselves to polygraph examinations. The polygraph test is generally used as supporting evidence and not as sole evidence.  The polygraph is an investigative tool and polygraph results should be used primarily to assist investigators in narrowing investigations by identifying suspects and collecting information relevant to the investigation. Polygraph results should therefore be submitted to a Court only when they can corroborate other admissible evidence and assist the Court in reaching a verdict.  

Although the Lower Court has been slow to admit polygraph evidence and the High Court has yet to rule on the admissibility of polygraph results and establish what weight should be placed upon it, the CCMA is acknowledging it as a forensic tool. The CCMA have therefore accepted Forensic Psychophysiologist (Polygraph Examiners) as expert witnesses. The duty of the commissioner is to determine the admissibility and reliability of the evidence. 

Polygraph test may not be interpreted as implying guilt but may be regarded as an aggravating factor especially where there is other evidence of misconduct. In other words, polygraph test results, on their own, are not a basis for a finding of guilt. It can be used in support of other evidence. 

Please note that unlike the Polygraph, Voice Stress Analysis (VSA) is not accepted by the CCMA. Polygraph testing should not be confused with VSA (also called CVSA). These tests are not polygraphs and have as yet not obtained any scientific validation in lie detection based on any independent scientific studies.   

The issue of trust between employer and employee?

The issue of trust between employer and employee?

What about the use of the polygraph in courts or the CCMA?


Most IR practitioners are familiar with the case of Harmse & Rainbow Farms (Pty) Ltd. In this case the employee was one of 15 who had access or potential access to the missing equipment. When offered an exculpatory polygraph examination, Mr Harmse at first declined, then changed his mind and did complete the test which he was the only person to fail. When offered another examination which could have led to his total pardon, he declined the opportunity and was then dismissed by the employer on the basis of not theft, but breach of trust relationship. This position and decision of the employer were upheld by the CCMA. 

Furthermore it was noted that:

[1] The employer has the right to dismiss an employee whom it can no longer trust provided there are reasonable grounds for such loss of trust based on the actions of the employee.

[2] The employer had reason to request polygraph examination as a result of suffering a substantial financial loss.

[3] An employer has a right in circumstances of such loss to exert some pressure on employees to cooperate in the investigation of the misconduct or crime.

[4] Granting of the option of a re-examination may be seen as a show of good faith by the employer.

[5] The polygraph examination phase was seen as an integral and crucial part of the investigation.

[6] The advance notification in writing of the examinees of the intention of the employer to institute polygraph examination as part of the investigation was seen to be fair.

[7] The procedure was not viewed as an unfair labour practice nor as an infringement of the fundamental rights of the examinee.

[8] The procedure underscored the fact that the examination was considered as voluntary.

It is strongly advised that employers have an agreement to undergo polygraph testing as part of their employment contracts.  If the employee then refuses, a charge of breach of contract can also be used against them. 

What exactly is a polygraph and how does it work?

The issue of trust between employer and employee?

What exactly is a polygraph and how does it work?

The term "Polygraph" literally means "many writings". The name refers to the manner in which selected physiological activities are simultaneously recorded. The data is then used to verify a person's truthfulness or deceptiveness, and is often called a "lie detector" test. For example, if you remember an experience where something frightened you, you may recall feeling a cold chill, a cold sweat, or a general quavering of the muscles. These symptoms were caused by the danger your mind perceived and if someone asks you a question about a wrongful conduct which you have in fact committed, you cannot help but remember what you have done. If you admit it, you are inviting punishment. If you deny it, knowing that you are lying, you risk getting caught. During a polygraph test, the risk is that your body will give you away.

Polygraph instruments will collect the physiological data from at least three systems in the human body. Sensors that are placed over the subject's chest and abdominal area (Pneumograph) will record muscle activity. Two small sensors (Electro Dermal Activity Sensors) attached to the fingers, will record electrical activity (Galvanic Skin Response) in the skin, and a blood pressure sensor (Cardio Sensor) will record changes in cardiovascular activity. These are accompanied by a Activity Sensor and other sensors may also be used, such as a Plethysmograph which measures changes in pulse blood volume.


A typical polygraph examination will include a pre-test interview phase, an examination phase and a data analysis phase. In the pre-test phase, the polygraph examiner will complete the required paperwork and talk with the subject about the examination in detail. This is when confessions or admissions can be made by the subject. During this period, the examiner will also discuss the exact questions to be asked during the test and familiarize the subject with the examination phase procedure and conduct a practice test. Involuntary physiological reactions are then recorded by the instruments, and interpreted by the examiner during the data analysis phase to reach an opinion as to the truthfulness or deceptiveness of the subject. 

 The examiner, when appropriate, will offer the subject an opportunity to explain physiological responses in relation to one or more questions asked during the test.

This process can generally take between one and a half hours to three hours per subject depending on the specifics of the scenario, types of tests being conducted and how it all unfolds.

Who can benefit from the use of the polygraph?

The issue of trust between employer and employee?

What exactly is a polygraph and how does it work?


It must be noted that proper polygraph procedures comply with the Constitution and the Labour Relations Act, and actively promote fair labour practices and respect for the human rights of the examinee. It therefore affords the innocent person an opportunity to be heard and to prove their innocence. In that regard, it is highly beneficial for the innocent person to undergo the exam.

For the employer, the list of beneficial possibilities are endless. Here are some examples - 

 [1] Testing staff considered for appointment to more senior positions or new prospective staff being considered for hiring.

[2] Testing people being suspected of committing specific crimes/misconducts.

[3] Insurance companies can test people or the prevention of insurance fraud.

[4] Owners of shops and factories can test staff or loss of stock, material, equipment, money and so forth.

[5] Business related concerns – various internal departmental investigations.

[6] Transport related companies can test staff for damage to and / or loss of cargo / vehicles / cash and so forth.

[7] Banks can conduct tests for a variety of reasons including staff working with large amounts of money and highly confidential information.

[8] Security companies and other private and governmental enforcement bodies can test staff or investigating loss of firearms, evidence, dockets, possible smuggling or inside jobs pertaining to robberies, burglaries, assault, bribery, sexual harassment and other related crimes.

[9] Every normal person in dealing with their domestic affairs can consider testing for when partners are being accused of adultery, when relatives or friends are accused of child neglect, abuse, maintenance claims, animal cruelty and so forth.

Anyone seeking the truth about a specific issue or anyone wishing to prove his/her innocence regarding a specific matter of concern or investigation can benefit from to do a polygraph.


R. Nelson - American Polygraph Association - Aug 2019


Psychophysiological detection of deception (PDD) testing, also known as the polygraph test – sometimes referred to as “lie-detection” as a term of convenience – is a standardized, evidence-based test of the margin of uncertainty or level of confidence surrounding a categorical conclusion of truth-telling or deception. The analytic theory of the polygraph test is that greater changes in physiological activity are loaded at different types of test stimuli as a function of deception or truth-telling in response to relevant target stimuli. Adequately selected polygraph target issues will involve a behavioural act for which an examinee can know unequivocally whether one has engaged in that action. 

A polygraph test consists of a pre-test interview, test data acquisition, and test data analysis phases. Post-test procedures can include other ancillary activities, including additional discussion, testing and investigation. Completion of a polygraph test can take upwards of 90 minutes, and should take place in a controlled environment, so that an examinee can adequately attend to and concentrate on the test topic and test stimuli without distraction. In this way, changes in physiological activity, if they are timely and independent of other observable influence, can be reliably attributed to the test stimuli.

The psychological basis of responses to polygraph test stimuli can be thought of as involving a combination of factors including: attention, cognition, emotion, behavioural conditioning and other resources for self-control. Differences in deception and truth-telling can manifest in patterns of response that can become observable upon repeated iterations of relevant and comparison stimuli. Statistical classifiers can be calculated for observed changes in physiological activity.

Polygraph test data are a combination of physiological proxies that have been shown to correlate significantly with different types of test stimuli as a function of deception or truth-telling in response to deception or truth-telling when answering relevant investigation target stimuli. Polygraph recording sensors include respiration activity, electrodermal activity, and cardiovascular sensors in addition to an optional vasomotor activity sensor. Polygraph recording sensors are autonomic – because activity in the autonomic nervous system has been shown to be correlated with differences between deception and truth-telling. Although each of the individual sensors is itself insufficient as a strong probabilistic indicator of deception or truth-telling, each of the different sensors contributes additional information to an optimized structural or statistical model. Standards of practice today require the use of activity recording sensors to help identify faking or countermeasures.

Suitability for polygraph testing requires that an individual is functioning reasonably within normal limits in terms of basic physiology, psychology, cognitive functioning or level of development, and language/communication skills. Probabilistic calculations may not apply in normal ways for individuals whose functional characteristics are outside normal limits, or those for whom the basic theory does not reasonably apply. Most persons who can work, drive, attend school, live in the community and function independently are suitable for polygraph testing.

Analysis of polygraph test data, like other test data, can be accomplished manually or via automated computer algorithm. Test data analysis involves several processes including: feature extraction, numerical transformation and data reduction, use of a likelihood function to calculate a statistical value, and the use of structured rules to parse a categorical test result from the numerical and statistical test data. Whereas manual polygraph test data analysis methods may continue to rely on visual processes, computer algorithms offer the advantage of automated reliability. Reliability of manually scored test results may be strengthened when they concur with the results of automated computer scoring algorithms.

Published information indicates that event-specific single-issue polygraphs can provide point estimate accuracy rates that exceed .90, depending on the test format and other factors. Multiple-issue polygraphs are both statistically and psychologically more complex than single-issue tests, with mean accuracy rates having point estimates that can exceed .80 for some polygraph techniques. Probabilistic estimates for polygraph test results can be calculated for both groups of cases (i.e. the observed frequency of correct or incorrect outcomes with groups of exams), and for individual cases (i.e., what is the likelihood that the data for a particular examination is correct or incorrect). Like other scientific tests, polygraph test results can be thought of as either positive or negative. Understanding and making use of polygraph test results can be thought of as a process of understanding the likelihood or potential for true-positive and true-negative in addition to the potential for false-positive and false-negative outcomes.

Polygraph testing does not detect, or measure, lies or deception per se. Actual detection of lies or deception – with deterministic perfection – would be immune to both human behaviour and random variation. Because deception and truth are not manifested as unique physical phenomena there is no physical or linear measurement for deception or truth. Scientific tests are used to quantify phenomena of interest for which neither perfect deterministic observation nor direct physical measurement are possible. For this reason, all scientific test results are fundamentally probabilistic, and are not expected to be infallible. Scientific tests are expected to provide a reproducible quantification of the margin of uncertainty or level of confidence that can be attributed to a test result or conclusion.  

This description is by no means the only description or final word on the matter of polygraph testing. It is neither the most complete and comprehensive description nor the most abbreviated description. It is an evolution of numerous other attempts to describe the polygraph test in report writing, courtrooms and other areas of discussion with professionals that may wish to understand or make use of polygraph test data or test results. This description starts with an assumption of a blank-slate level understanding, and attempts the lofty goal of quickly and efficiently integrating basic concepts from psychology, physiology, decision theory and test theory. Other descriptions of the polygraph are possible, and may at times be more useful than this one. This description is offered with the hope that it may be useful in prompting discussion and stimulating additional advances in written and verbal communication among professionals who use polygraph tests. 



Choosing a Polygraph Examiner

Choosing a polygraph examiner isn't as simple as just relying on a Google search and then making a choice based on the cheapest quote. The truth is that “you get what you pay for”!

To help you ensure that you are getting a good service for your money, here are some guidelines for choosing a suitable examiner. Ask yourself the following questions:

[1] Is the examiner registered with the American Polygraph Association? 

This cannot be substituted by membership with only a South African polygraph association, although the additional memberships to such organisations are considered an advantage. The APA members are required to maintain the highest standards of ethical and professional conduct and they are further required to discharge their duties with complete impartiality, dignity and respect. Members recognize that their primary responsibility is to the person being examined, and are forbidden to allow considerations of race, religion, politics, union activity, or economic status to play any part in their examinations, and they are pledged to issue an objective and unbiased opinion and to protect the confidentiality of the examination.

[2] Does the examiner have an appropriate investigative background?;

[3] Is the examiner using the best technology such as the Lafayette LX4000, LX5000 or LX6 accompanied by at least the minimum required four validated sensors?;

[4] Is the polygraph examiner using the appropriate time scale to conduct the examination? A proper polygraph examination which has followed all scientifically validated procedures cannot be conducted in less time than dictated to by international standards of approximately 1:30 hours minimum per test and even as long as 3:00 hours per test in some cases;

[5] Is the polygraph examiner charging a market related price for his work? (Rates between R900,00 and R1350,00 are the current industry standard);

[6] Is the examiner looking into the facts of the case well enough before conducting polygraph testing to ensure that the minimum number of persons are tested, that the right persons are tested and that the rights tests are used with the right questions?;

[7] Is the polygraph examiner willing to submit the test results for quality-review if there is reason to request it?;

[8] Is the examiner audio and visually recording the entire process for record purposes?;

[9] Is the examiner asking the test subjects to sign a consent form before doing any testing on them?

[10] Is the examiner able to offer sound advice on what to do with the test results and/or how to proceed with your investigation, in follow-up to such an intervention? Is the examiner opposed to communicate with your HR practitioner?;

[11] Is the examiner using Voice Stress Analysis and calling it a polygraph? VSA is not a polygraph and there is NO properly conducted scientific research to validate the use or results of VSA. This device is also NOT accepted by the CCMA.



Noticeable Case Outcomes

[1] In Govender and Chetty and Cargo & Container Services KN4881 the Commissioner said that the failure of the two employees in two polygraph tests lends some support to the finding of the facts that they were involved in theft. 

[2] In Hotellica Trade Union and San Angelo Spur WE3799 the Commissioner found that although refusal to take a lie detector test may not be interpreted as implying guilt, it can be regarded as an aggravating factor, especially where there is other evidence of misconduct. 

[3] In CWIU and Druggist Distributors WE10734 the Commissioner said that it is not necessary to deal with the weight that should be attached to the polygraph test if the company did not rely on the test to establish the employee's guilt, but regarded it merely as additional evidence that would demonstrate whether the employee's version was truthful and provided an opportunity for the employee to disprove the company's evidence. 

[4] The following is yet another instance in which polygraph testing added value to the investigative process to the point where a favorable outcome was achieved: Lephuthing / MNT (Pty) Ltd [2012] JOL 29064 (CCMA) 

A thief who had been caught red handed with eight Apple iPhones from the warehouse, in which electronic stock worth billions of rand was stored. Both the applicant and the thief were charged criminally, arrested, but released on bail. The thief subsequently disappeared and the criminal charges against the applicant were withdrawn. The applicant failed the polygraph test, was charged with theft, and dismissed. The applicant denied involvement in the theft, and sought reinstatement. The commissioner noted that the respondent’s predicament was that, after the thief’s disappearance, it was left with only hearsay and circumstantial evidence and the results of the polygraph test. Whether the hearsay evidence concerning the thief’s allegation that the applicant was his accomplice should be admitted depended on whether it was in the interest of justice to do so. A party seeking to deviate from the rule against the admission of hearsay evidence must lay a proper basis for so doing. The commissioner decided, however, that the matter could be decided on the applicant’s version. The applicant claimed that the thief had told him that he had implicated him because they were “friends”. The applicant had put up the R1 000 bail for the thief’s release. The respondent had to prove its case on a balance of probabilities, and the result of the polygraph test tipped the scales against the applicant. 

[5] Mnube & Cash Paymaster Services (Pty) KN1583, showed that the Commissioner accepted the polygraphist as an expert witness whose evidence needed to be tested for reliability. 

[6] SACCAWU obo Chauke and Mass Discounters ( 2004) 13 CCMA 2.13.1 Expert Evidence – polygraph tests – procedure followed and technique used was satisfactorily explained so as to make the results admissible as evidence.

Where a polygraph test had been performed by a properly trained examiner and the employee had voluntarily taken the test, provided of course that the polygraph was considered admissible on the facts of the case, then the outcome is a relevant evidentiary fact and should be taken into account together with the evidence in its totality. 

[7] Harmse vs. Rainbow Farms is a very important precedent in the South African labor law environment. It is a finding that opens the door for business to utilize the full abilities of polygraph technology within the workplace in a responsible and sensible manner. 

In the aforementioned case, there has been a break-in to the offices of the employer and computer equipment to the value of some R200 000 had been stolen. There were no signs of forcible entry and the alarm did not go off. The above mentioned fact lead the police and the employer to conclude that there had been some inside involvement by an employee who had a front door key and knew the alarm code. 

The company informed all staff members who were in possession of keys and who knew the alarm code, of the company's intention to let the employees undergo a voluntary polygraph examination. These examinations were to be administered by Mr. Paul van Niekerk, an internationally trained examiner who is currently a member in good standing of the American Polygraph Association. 

All the employees to be subjected to polygraph examinations were informed of the employer’s decision to do so in writing 48 hours before the examinations were due. Of the 15 key holders Mr. Harmse was the only one to initially refuse to undertake the polygraph examination. He put his initial refusal in writing, stating religious grounds and human rights ground for his refusal. The employer wrote him a letter urging him to submit to the examination. He then decided to take the test as he felt that if he continued to refuse he would be prejudiced in the eyes of the company. All 15 of the key holders were then subjected to polygraph examinations. All the employees passed the examination except for Mr. Harmse. 

Mr. Harmse was employed in the position of buyer in the Stores Department, a position in which trustworthiness is of the utmost importance, as well as being critical to the employer's conducting of its business. Mr. Harmse indicated that he found the test an acceptable procedure and he also signed a statement that he understood all the questions asked during the test and that those questions were explained by the polygraph examiner before the examination. He also signed a statement that the polygraph examiner treated him in a fair and respectable manner. 

The employer held a disciplinary hearing and consulted with the employee, urging him to give a reasonable explanation for his deceptive responses during the examination. After a disciplinary hearing the employee was dismissed. The employee appealed to the CCMA stating that his dismissal was procedurally correct by substantially unfair.

The following aspects of the finding of CCMA Commissioner Wilson (Case no We1728, 09 July 1997) needs to be mentioned: 

· An employer is entitled to dismiss an employee whom it can no longer trust, providing that the employer has reasonable ground for losing trust in the employee. 

· “The actions of the employee must have been as such as to have reasonably caused the employer, on a balance of probabilities to feel that the employee is no longer trustworthy” 

· The company's action in conducting voluntary polygraph tests were reasonable in the light of the circumstances of the substantial financial loss. 

· The fact that the employer wrote the employee a letter urging him to undergo the polygraph examination was not construed as putting undue pressure on the employee to undergo the examination against his will. To quote directly. “I believe that in the circumstances the company was justified in exerting some pressure, which they clearly did.” 

· The dismissal of Mr. Harmse was judged substantially fair and was confirmed.



Application of the Polygraph

The general rule is that a polygraph examination can only be conducted with the consent of the examinee. A refusal to undergo a polygraph test would not enable one to draw any inference of guilt. Would this be the case if the reason for refusal - or absence of a valid reason for refusal - to assist the investigation by undergoing a polygraph test is deemed to be unreasonable/invalid? There are exceptions to this general rule which compel a person to undergo a polygraph test. 

Admissibility of the Polygraph

Law of evidence states that the prerequisite for admissibility is relevance. It follows that if the polygraph examination relates to the issue under review, then it is admissible. This would necessarily hold true for all judicial forums such as the CCMA, civil courts and criminal courts. Polygraphists have been accepted as expert witnesses whose evidence needs to be tested for reliability. Therefore the Polygraphists should be called to testify as an expert witness. He/she must testify how the test was performed, his/her qualifications, the type of test used and the questions asked. 

Weight of Polygraph Evidence

Burden of proof: 

  • Balance of probabilities - Applicable in labour matters and civil courts (The plaintiff has the burden of proving his case by a preponderance of the evidence) 
  • Beyond reasonable doubt - Criminal matters ( The burden of proving the defendant's guilt is on the prosecution, and they must establish that fact beyond a reasonable doubt) 


The methodology considered to be best practice is to ensure that there is another form of evidence which corroborates the finding made/result of the polygraph examination.  Such evidence, amongst possible others, could be: 

  • Viva voce evidence (Oral rather than written)
  • Documentary evidence (Photographs, tape recordings, films, printed emails, etc)
  • Circumstantial evidence (Relies on an inference to connect it to a conclusion of fact—such as a fingerprint at the scene of a crime, etc)


However, exceptions may be admissible and carry weight such as when a paired testing protocol is followed which, mathematically, enhances the overall reliability of the results.