Breadcrumb

  1. Home
  2. Meetings of the Commission
  3. Meeting of May 16, 2007 - Employment Testing and Screening
  4. Statement of Richard Tonowski, EEOC Office of General Counsel

Statement of Richard Tonowski, EEOC Office of General Counsel

The U.S. Equal Employment Opportunity Commission

Meeting of May 16, 2007 - Employment Testing and Screening

TESTING AND EQUAL EMPLOYMENT OPPORTUNITY:
THREATS AND PROMISES

Good morning, Madam Chair, Commissioners, and distinguished panelists. My name is Richard Tonowski. I am Chief Psychologist in the Office of General Counsel, and I will provide you with a brief overview of some employment testing issues.

An employment test is any procedure used to make an employment decision. There are the familiar multiple choice tests of job knowledge or basic cognitive abilities such as doing arithmetic or understanding written instructions. There are also physical ability tests, which can be as simple as moving a box from here to there, or as sophisticated as measuring the strength of specific muscle groups. Beyond assessing strength and endurance is the issue of pre-employment medical inquiries for both current and potential conditions, both physical and psychological. Background checks are extensively used, and sometimes include examination of credit history. Finally, for the last 20 years, multiple choice personality and integrity tests have become increasingly popular. In all, employment testing is widespread and increasing.

A mature technology of testing promises readily-available methods that serve as a check against both traditional forms of discrimination as well as the workings of unconscious bias. If that is the promise, then the threat comes from institutionalizing technical problems not yet fully addressed, the undermining of equal employment opportunity under the guise of sound selection practice, and the unintended introduction of new problems that will require resolution to safeguard test-takers and test-users.

Since the Supreme Court’s landmark decisions in the 1970s, which Carol will reference, some notable things not fully envisioned have happened to testing.

  • Understanding of cognitive ability in jobs and the tests to measure it greatly increased. Statistical approaches to summarizing results across many separate studies to reach general conclusions became widely used. The limit to these generalizations remains a matter of contention.
  • How test scores are reported became hotly debated. Congress made unlawful test score adjustments based on protected class. Some psychologists proposed banded, rather than discrete, scores as a means of promoting both diversity and test utility. Opponents have objected on both technical and legal grounds.
  • The enactment of the Americans with Disabilities Act (ADA) restricted medical exams and disability-related inquiries for applicants and employees. There remains the issue of differentiating proper tests of competencies from unlawful medical investigations.

Sometimes promise and threat arrive together. Personality testing has been hailed by some as a means for a more complete, and thus more valid, assessment of potential employees. It may reduce the adverse impact associated with cognitive ability testing used alone. But when does a legitimate inquiry into an applicant’s qualifications become an intrusive search for medical conditions? The Seventh Circuit held in Karraker v. Rent-A-Center, Inc. that the line is crossed when the personality assessment tool is an instrument such as the Minnesota Multiphasic Personality Inventory (MMPI). 411 F.3d 831(7th Cir. 2005). The MMPI is widely used for clinical diagnosis and was originally normed on hospitalized psychiatric patients. However, researchers have combined its questions in new ways to measure a variety of traits, not all of them clinical. Some of its current 567 questions explore whether the individual sees things that others do not, has laughing or crying fits, or was compelled to do things under hypnosis. The Court of Appeals held that the MMPI by its nature was a medical examination violating the ADA. An earlier case (Soroka v. Dayton Hudson Inc.) that arose in California and was ultimately settled raised similar issues, as well as views of test usage at the individual item level that alarmed psychologists. This is but one instance where science and law intersect, and where the outcome has real consequences to employers and potential employees.

The EEOC sees a limited, but increasing, number of tests in the course of its investigations. By the nature of our work, we scrutinize evidence of the test’s worth with a critical eye. Sometimes we are gratified by the care given to both technical and EEO considerations. But there are times when superficial work and unsupported conclusions come from consultants who should know better. Today you will hear about some testing practices that should not have happened. This kind of work constitutes a threat to job applicants, employers, and all concerned with good selection practice.

On the plaintiff’s side, occasionally there are arguments, backed by a highly selective reading of the research literature, that whatever the employer did was unacceptable simply because the employer might have done something else. This also presents a threat to good selection practice.

But first -- the legal bases from the Office of Legal Counsel.


This page was last modified on May 16, 2007.