USING PSYCHOLOGICAL TESTS IN EMPLOYMENT II

Leonard D. Goodstein

Psichometrics, International, LLC.

There is considerable concern among HR professionals about the legitimacy of using psychological tests and other selection procedures in employment settings.  The purpose of this brief essay is to allay these concerns by providing a clear analysis of the underlying issues.

I believe that there are two central points that need to be understood.  One is that there is no legislation nor administrative ruling to indicates that there is a mandate to employ incompetent persons—individuals who will fail on the job.  Second, however, there is a mandate not to discriminate on the basis of age, gender, race, religion, national origin, or disability.  But it is incumbent on the employer to demonstrate that whatever selection processes used to choose candidates from the applicant pool truly pick those persons who are most likely to succeed on the job. 

Clearly, not all hiring decisions, however conscientious, turn out not as expected, and indeed the courts have not set an unattainable standard of perfection in the hiring process.  What the courts have demanded, however, is that employers must be able to demonstrate that they have a responsible, good faith effort to: (1) identify the core skills necessary to succeed on the job, and (2) select applicants who possess those skills.

In other words, whatever techniques are used to select people must be shown to be related to on-the-job success, that is, are valid.  In my opinion, if we select candidates who are most likely to be successful then we have met the test of “business necessity” as stipulated in the various pieces of legislation, administrative rulings, and court decisions.

Let’s assume that we set possession of a college degree as a minimum requirement for selection.  Applicants without such a degree can question why such a requirement has been set.  Failure to demonstrate that college graduates actually do better on the job than non-college applicants leaves the employers open to charges of discrimination, as minority candidates are less likely to have a college degree than majority candidates.  In one sense this is an easy problem to solve; it simply requires that employers correlate levels of education with job performance.  If such a relationship can be demonstrated, then there is a ‘business necessity” for demanding a college degree.

In another example, public safety jobs (police, firefighters, etc.) have well-established physical requirements—height, weight, strength, stamina, and so on.  When these requirements have been challenged, as they have been, those employers who have used such standards they have been required to demonstrate that these requirements are related to on-the-job performance.  Many of these challenges have led to a modification of the standards that have allowed women and minority candidates job opportunities in the public safety field.

The message here should be clear.  It has become mandatory for employers to systematically establish the validity of their selection processes.  That is, employers need to be able to show clearly that their selections from the applicant pool are based on empirical evidence that they are attempting to select the best candidates, those most likely to success—that they have adopted a process based on business necessity.

Another issue that requires constant attention in evaluating job candidates is the necessity of using appropriate norms.  The issues here are relatively straightforward.  Applicants should be compared with comparable others; in other words, candidates should be compared with candidates, not with employees or college students.

 One important reason, among others, for not using the Minnesota Multiphasic Personality Inventory (MMPI) for employment purposes is that the norms are based on the responses of psychiatric patients, not applicants for employment.

Another issue that requires attention is the recency of the norms.  Norms collected decades ago may not accurately characterize the performance of current candidates because there are changes in average performance on many tests over time, especially tests of cognitive ability, reading levels, mathematical skills, and the like. 

Also, it should be noted that using different norms for men and women in selecting employees is explicitly prohibited by Section 106 of the Civil Rights Act of 1991.  It is important to recognize that some tests, particularly ability tests such as those measuring mechanical comprehension, do have large differences in favor of male respondents.  The users of such tests should pay careful attention to the issue of gender differences in scores and avoid using those with such differences in order to avoid charges of gender discrimination.

Our position is that there are no prohibitions against using psychological tests or other selection procedures that have been properly validated and can be shown to be related to on-the-job success.  Psichometrics International, LLC is committed to working with its clients to provide such validation and to collaborating in developing and maintaining a discrimination-free workplace.