Background

In 1988, the Polygraph Protection Act made it illegal to use polygraph exams to screen applicants for most private sector jobs. Congress disallowed the use of lie detectors because the preponderance of scientific evidence indicated they did not work very well – if at all.

Even before the law was passed, entrepreneurial polygraph examiners began to market "honesty tests" patterned after the questions used in lie detector interviews.

These so-called "honesty tests" take two approaches to revealing a person’s tendency to steal or engage in other dysfunctional workplace behaviors:

The first is to directly ask a subject if they have committed such infractions under the assumption that they will tell you the truth about their past behavior. This approach is often referred to as ‘overt-' or ‘admission-based’ testing.

The second is to ask about a subject’s beliefs and attitudes concerning the frequency, severity or commonality of criminal behavior, and about how perpetrators should be dealt with. This approach is based on the assumption that a subject’s attitudes about these behaviors and punishment are predictive of their actual work behavior, and is often referred to as ‘covert’ or ‘attitudinal-based’ testing.

Testing for "Honesty"

The market acceptance of honesty and integrity testing products in the 1980s prompted the field of industrial psychology to begin studying the validity and utility of these tests.

This line of research has produced a body of published and unpublished studies from a variety of organizations – psychologists and publishers with interests or expertise in the domain of employment testing.

And with all scientific research, the quality and credibility of an individual study can be evaluated using accepted indicators of scientific rigor.

In general, trustworthy scientific research results in the following:

• Is conducted by qualified scientists with expertise in the phenomena being studied.

• Employs methods of data collection and analysis that are recognized by other scientists as appropriate to the investigation.

• Is reported in a manner that clearly reveals hypotheses, data sources, analytic procedures and other features of the study in a manner that would allow other scientists to independently replicate the study and verify reported findings.

Investigator Bias

Another useful factor to consider in evaluating scientific research is the possible bias or vested interests held by the investigators.

Findings reported in peer reviewed scientific journals or textbooks and research conducted by scientists or institutions with no financial or political stake in the results of a study are often more credible than unpublished, privately-funded studies conducted by employees of test publishers or service providers.

Caveats stated, this article is intended to summarize the major criticisms observed and published in the scientific literature related to the use of pre-employment testing based on self-report behavioral admissions and attitudes related to workplace delinquency.

And in this context, CoreCentive’s assessment strategy related to the prediction of employee delinquency is contrasted and described.

An Influential Exam

One of the most influential examinations of honesty testing was conducted in 1990 by a special task force of the American Psychological Association.

This blue-ribbon panel made up of leading industrial psychologists and academicians summarized their comprehensive review of this practice with the following statement:

We found that publishers of honesty tests are highly variable in the ways that they describe their tests, in the ways that they document them, and in the evidence that they assemble to substantiate claims of their validity and utility.

For some published measures, almost no evidence at all is available beyond assurances that evidence exists. For a few measures, extensive research has been completed. However, even in the latter cases, test publishers have relied on the cloak of proprietary interests to withhold information concerning the development and scoring of the tests, along with other basic psychometric information.

The entire field of honesty testing would benefit greatly from substantially increased openness.

Many test publishers have not come close to meeting standards for the publication, sale and use of psychological measures and these deficiencies should be promptly corrected.

Many of these patterns of behavior may or may not be consistent with measured attitudes, values and beliefs.

It's this emphasis on behavior and the predictors of behavior that differentiates credible scientific research – research built on the prediction of employee delinquency from the simplistic assumptions of admission-based and attitudinal honesty testing.

Look for the next in the series from CoreCentive, www.corecentive.com : Behavioral Assessment & Dependability.