The lie detector has fallen on hard times. The CBS news program "60 Minutes" recently attempted to test the polygraph and polygraphers. The producers asked three polygraphers to identify a thief among four CBS employees, none of whom had stolen anything and none of whom was lying. Furthermore, unknown to the employees, each polygrapher was told that a different one of the four employees was the probable culprit. In each case the polygrapher identified with near certainty the designated employee as the thief.
While there is some research and some basis, however slim, for defending real-life "specific issue" testing when properly done--for example, in the course of a police investigation--there is virtually no research and no scientific support to justify the most common use of the polygraph: pre-employment screening.
Pressure is mounting for the Senate to pass the Hatch-Kennedy Polygraph Protection Act, which would bar these tests in almost all private-sector industries. Since the livelihood of the typical polygrapher has depended virtually entirely on this type of screening for private-sector clients, the real question has become: What service can be substituted for the polygraph? It would have to satisfy the insatiable desire among employers to know in advance which employees can be trusted, and it would have to fulfill polygraphers' need for work.
In states that have banned pre-employment polygraph screening, it has been replaced by the marketing of a new type of "psychological test" of employee veracity. The obvious economic impact of the proposed federal anti-polygraph law is not lost on the manufacturers of the new tests. Polygraphers around the nation have been inundated with advertisements for products and seminars all promising essentially the same thing: paper-and-pencil "honesty" tests that will prove to be just as good as polygraphs, but would not violate the proposed federal law.
The answers to these tests are fed into a small computer, which looks remarkably like the polygraph equipment that it replaces, and the risk assessment of the job applicant spews forth in front of the employer-client's eyes.
To reassure polygraphers that the new tests will not be an investment vulnerable to a successful legal challenge, typical honesty-test advertisers include endorsements from lawyers. These promise that the tests conform to the Equal Employment Opportunity Commission job-relatedness test and the Civil Rights Act protections against employment discrimination.
All the manufacturers of the new tests claim that the answers to about 60 to 80 questions will enable the employer (with the help and expertise of the former polygrapher, of course) to discover vital information, such as whether the job applicant is theft-prone, has good ethical standards, will be a "troublemaker"--and even whether the applicant has "lied" in answering the questions on the test! The manufacturers promise that "honesty" may be measured just as well by "assessing attitudes concerning theft," and that this is "more predictive of actual future theft-proneness" than the polygraph. One company even boasts that its test "identified the same individuals as being high-risk, low-risk or caution-risk as did polygraph evaluations."
Aside from the obvious problem of validating one measure on the basis of another measure about to be outlawed largely because of questions of validity, one wonders why anyone ever bothered with the more intrusive and complex polygraph if a simple questionnaire would have served just as well.
But the same fundamental flaws of the pre-employment polygraph are found in the pre-employment honesty tests. There simply is no compelling evidence that such tests can predict who will be a good employee. Reliance on these new "truth" tests assumes, without any evidence, that it is possible to find out what is in the respondents' minds when they encounter the issues raised in the tests, and that their answers correspond to what they might do in an infinite range of on-the-job situations.
If Sens. Orrin G. Hatch (R-Utah) and Edward M. Kennedy (D-Mass.) are successful, we will see an even less-valid type of psychological testing, but one that is much harder to oppose. The likely result is that prospective employees may soon face more extensive testing, more rejections because they are considered "high risks" and more humiliation--all of which will be perfectly legal.