Those "Little White Lies" of Honesty Test Vendors

Article excerpt

Those "Little White Lies" Of Honesty Test Vendors

So-called "honesty" tests are selling like proverbial hot cakes--and the sales pitch is great. You want better employees. The vendors can help you. You want a "yes-no" answer. They're willing to give it, and it will not be anything like those "wishy-washy" psychologists' reports that actually put the burden of making the hiring decision back on you.

Yes, you will get a "go-no go," "honest-dishonest," "violent-nonviolent" rating that is "validated" and "accurate" and "reliable." Not only that, but if you want a short test (even if that means the test can't possibly be "reliable"), you can administer as few as 30 questions to get your answers. Maybe next week you can purchase a 10-item test!

But what exactly is behind the glossy brochures and fancy sales talk? In many cases, it's a bunch of "little white lies" and carefully worded promises from vendors who will admit privately that they're just trying to cash in on the increasing demand for a replacement of the polygraph and the decreasing availability of high-quality job applicants.

The good news for them is that there is no regulation in this industry and little likelihood that any regulations will be put into effect for a long, long time. The bad news is that, sooner or later, some intelligent people will realize that there is little substance behind the spoken and implied claims that these tests effectively predict those individuals who will later steal from their employers.

Let's keep the record straight. I, myself, am a test vendor--one with a Ph.D. in psychology from Columbia University and a diplomate from the American Board of Professional Psychology and the American Board of Forensic Psychology. After more than ten years of conducting predictive research studies to see exactly how well we test developers can foretell the future, I am truly amazed and embarrassed by what has been going on in the testing industry.

It's time someone told the truth about the misinformation being spread to the public about "honesty tests." I do not mean to criticize any particular test, since no existing tests are flawless. Neither do I mean to disparage the well-meaning efforts of some individuals to help employers reduce internal theft. Rather, I am commenting on the unfortunate manner in which many tests have been presented to the public and hope the following points will be viewed in that light.

Therefore, without naming names, I have set out to debunk those little, and often not so little, white lies you may be hearing from honesty test vendors. Following are a few that are doing quite well out there.

1. "Oh yes! Our test has been fully validated." Response: There is no such thing as "fully validated" since validation is a process that does not have a definable end. The most that should be said about a test is that it has "X" number of independently published or presented predictive validity studies, from which the findings are thus and so. The determination of whether or not a test has been validated "enough" to be used in the marketplace should be left to independent panels of test and measurement experts. To date, no such panels exist in this industry.

What you should do: Ask for the published research studies and the technical manual. Ask exactly what it has been validated for. Listen for statements implying that the test will predict those who will steal from you or who will be dishonest. Then ask for the proof in published research studies and let an unbiased university professor review the studies with you.

Watch for: Studies that have been printed by the test company and not actually published in "blind-reviewed" professional journals. There is no reason for a company to avoid this avenue for gaining professional respectability unless they do not have the studies or the research is so poorly designed that it would never be accepted by those who are "in the know" about how tests should be designed and studied. …