Printer Friendly

Integrity tests and the law.

FOR MANY YEARS PSYchologists, sociologists, and criminologists have attempted to predict delinquent behavior on the basis of psychological tests. In the past, the tests were usually used for diagnostic and treatment purposes, but more recently they have been used for predicting theft. The potential value of this type of test to employers is readily apparent, but employers must make sure tests comply with the legal restrictions.

The form of psychological test most widely used in employment situations is a pencil-and-paper honesty questionnaire. The first such test, the Reid Report, was developed by John E. Reid and became available to the business community in 1951. It was followed by other similar tests. All of them are known as proprietary tests because they are administered by the employer.

The Reid Report, which is used here to illustrate discussion points, consists of a questionnaire with approximately eighty questions that are answered by a yes or no response, accompanied by biographical data questions, and by a list of thefts or theft-related acts that the candidate may have committed. The questionnaire consists of four parts.

In Part 1, trustworthiness is determined. The test asks a series of questions, such as, "Do you believe a person should be fired by a company if it is found that he helped another employee take a little merchandise from the company?" The answers to such questions reveal a respondent's concept of punitiveness. Part 1 also examines the candidate's projective appraisal of his or her trustworthiness with questions such as, "Did you ever think about committing a burglary?" No one answer or group of answers determines a person's honesty. A person's trustworthiness is only determined after analyzing a total test score for which the chosen answers are keyed toward one conclusion or the other.

Part 2 is a criminal admissions questionnaire that inquiries about the candidate's possible commission of or conviction for illegal activities. Incredible as it may seem, many people taking the test admit to committing not only minor thefts but also serious crimes. The presumed reason for this phenomenon is that test takers view their illegal conduct as normal.

Part 3 seeks information about the candidate's frequency of recent drug use, which is often a contributing factor in incidents of employee theft. Here again, a surprising number of those taking the test admit to recently ingesting illegal substances during or just before going to work.

Part 4 examines work history. This section is similar to many job applications in the information it provides; however, test administrators have found it useful in detecting inconsistencies in the candidate's job history.

The Reid Report can be scored by the employer in a variety of ways, the most common of which is communication with the test supplier by telephone or by computer software. Regardless of the procedure used, however, the scoring key is not released to the employer or to anyone else.

Although the Reid Report was designed to identify applicants who have unfavorable attitudes toward honesty, some research has demonstrated that it is also predictive of other employment-related behaviors, such as productivity, time theft, absenteeism, and even aspects of emotional stability.(1) Reliability. Major test publishers consider tests reliable, but all test manuals recommend that test scores not be used alone to determine a candidate's eligibility for employment. Three studies, reported in the early 1990s, have been done by objective organizations to determine the validity and reliability of such tests. One was conducted by a task force of the American Psychological Association (APA), which is comprised of 108,000 psychologists; another by the Office of Technology Assessment (OTA); and a third by researchers at the University of Iowa.

In 1990, the Office of Technology Assessment (OTA) issued its report, The Use of Integrity Tests for Employment Screening. The OTA report was particularly critical of pencil-and-paper tests, but it concluded that, "The research on integrity tests has not produced data that clearly supports or dismisses the assertion that these tests can predict dishonest behavior."

Although the report stated that sophisticated analytical methods would have to be applied to ascertain integrity test reliability, the agency relied primarily on a single experiment in which 3,790 employees were given the test and hired regardless of their performance on the test.

The OTA reports that subsequent investigations by management revealed that 91 employees had committed some type of theft. Of that group, 75 had failed the integrity test and 16 had passed it. Among the 3,699 for whom the investigation did not reveal any theft, 2,145 had failed the test and 1,554 passed. Thus, 75 of those taking the test (2 percent of the total 3,790) are known to have been characterized correctly by the test, and 16 are known to have been characterized incorrectly. The report raised the question, "What about the rest?" The report also stated that, if the 3,699 employees not detected as thieves are assumed to be honest, then 2,145 (58 percent) were misclassified. If a substantial number of them were indeed thieves, the observed correlation between the test and the outcome measure could be higher, lower, or equal to the actual correlation.

Considerable criticism has been expressed about the OTA study, both as to its methodology and its inferences from the results. For example, criticism of the OTA report appears in the 1992 Annual Review of Psychology. In its section on personnel selection, it labels the OTA report as "superficial and in part clearly erroneous."

In 1991, the APA issued its report, The Prediction of Trustworthiness in Pre-Employment Selection Decisions. The guideline for any such study, urged the report, was to evaluate the tests not against an absolute, but comparatively against the validities of other procedures.

The report stated that the error problem regarding pencil-and-paper tests is similar to that of other selection instruments, such as aptitude testing, interviews, or educational requirements. All are fallible to some degree.

After having considered the available data regarding the reliability of pencil-and-paper tests and other relevant factors, the APA task force found no basis for prohibiting their use. To do so, stated the report, would only invite alternative forms of preemployment screening that would be less reliable and less controllable.

The most recent and by far the most extensive appraisal of integrity tests was conducted by researchers at the University of Iowa (reported in the Journal of Applied Psychology in 1992). They analyzed 665 test takers chosen from a database of half a million. The results of the study disclosed that the tests were able to predict job performance and counterproductive behaviors on the job, such as theft, disciplinary problems, and absenteeism.

Legal issues. An important legal issue regarding integrity testing, or any other test of a similar nature, is compliance with the requirements of the Civil Rights Act of 1964, as amended in 1991, and comparable fair employment laws on the state level.(2) Specifically, the issue is whether such testing involves discrimination or other illegal employment practices. The Civil Rights Act provides victims of employment discrimination with legal recourse.

The federal statute is only applicable to employers of fifteen or more employees who are engaged in an industry affecting interstate commerce, but some states, such as Iowa, have laws that apply to employers with fewer employees. A Presidential Executive Order applies the essence of the Civil Rights Act to federal employers and companies that do contract work for the government. The most relevant portion of the Civil Rights Act makes it an unlawful employment practice for an employer to discriminate against anyone because of race, color, religion, sex, or national origin. The act provides, however, that "it shall not be an unlawful employment practice for an employer...to give and to act upon the results of any professionally developed ability test provided that such test, its administration or action upon the results is not designed, intended or used to discriminate because of race, color, religion, sex or national origin."

The implementation of the federal civil rights legislation is the responsibility of the Equal Employment Opportunity Commission (EEOC), which was established by Congress in the Civil Rights Act. On the state level there are agencies comparable to EEOC, known by various names, such as commission for human rights, civil rights commission, fair employment practices commission, and department of fair employment.

Discriminatory practices. Court decisions regarding the legal tests for discriminatory practices pertain primarily to ordinary employment situations rather than to psychological testing for dishonesty, but the same basic principles are applicable.

To prove discrimination in violation of the Civil Rights Act (or a comparable state statute), the plaintiff has two separate but related paths to pursue: a showing of either disparate treatment or disparate impact. Disparate treatment occurs when some employees or applicants are treated less favorably than others, based on race, color, religion, sex, or national origin. Disparate impact occurs when an employer's practices fall more harshly on one group or another, without a justification of business necessity.

Disparate treatment. In a 1973 case, McDonnell Douglas Co. v. Green, the Supreme Court outlined the procedures to be followed in cases with respect to burden of proof and the presentation of evidence. First, the plaintiff has the burden of proving a prima facie case of discrimination by showing that he or she belongs to a racial minority, for instance; applied and was qualified for an available job; was rejected despite those qualifications; and that the position remained unfilled while the employer sought applicants with similar qualifications. If the prima facie case is established, the burden shifts to the defendant to show that, as stated in the 1981 case of Texas Department of Community Affairs v. Burdine, "the plaintiff was rejected, or someone else was preferred for a legitimate, nondiscriminatory reason." However, the defendant need not persuade the court that it was actually motivated by the proffered reasons. It is sufficient if the evidence raises a genuine issue of fact as to whether there was discrimination against the plaintiff. That was the position of the Eighth Circuit Court in the 1981 case of Johnson v. Bunny Bread Co.

If the defendant's burden is met, the plaintiff has the opportunity to prove that the legitimate reasons offered by the defendant were not its true reasons but were a pretext for discrimination. Basically, in disparate treatment cases, proof of a discriminatory motive is critical. Motive can be established either by an outward intent to discriminate or by inference based on the mere fact of differences in treatment.

Disparate impact. The focus of inquiry in disparate impact cases begins with the burden on the plaintiff to show that the employment practice selects applicants for employment or promotion in a pattern significantly different from that within the pool of applicants or employees, even though the effect may be unintentional.

Three general approaches have been used to establish a prima facie case of disparate impact with regard to applicants for employment. These are explained in an article by attorneys Dean Booth and James L. Mackay of the Georgia Bar.(3) Only two of them, however, need be mentioned as relevant to psychological testing. The first approach is the applicant statistics approach, which compares the percentage of minority applicants passing the test with the percentage of majority applicants who pass. A number of courts have expressed a preference for this approach.

The second approach involves, according to Booth and Mackay, "a comparison of the percentage of minority persons employed in the specific job or jobs for which the test is required with the percentage of minority persons in the relevant geographic labor market." Most courts recognize, however, that differences in employment levels relative to the population can result from many reasons other than discrimination.

Some decisions, though approving the use of population statistics, have narrowed its application. The relevant labor pool must be based both on the geographical area from which the employer draws its labor force and on those persons in the pool who are qualified for the job.

In the 1989 case of Wards Cove Packing Co. v. Atonio, the Supreme Court made meeting the standard of proof more difficult for plaintiffs. In the 1991 amendments to the Civil Rights Act, however, Congress effectively negated the Wards Cove decision. Although litigation will probably be required to determine the ultimate impact of the Civil Rights Act of 1991, the statute generally states that the following process shall be adhered to in disparate impact cases: * The complaining party must demonstrate a prima facie case of discrimination by showing that the employer uses a selection process significantly adverse to a protected minority group in the relevant labor pool

* Once a prima facie case is established, the burden shifts to the employer to either demonstrate that the specific practice does not cause the alleged impact or that the practice is job related for the available position and consistent with business necessity

* If the test is job related, the complainant may still prevail if proof is offered that other available tests would not produce a similar disproportionate effect while at the same time serving the employer's legitimate interests. EEOC guidelines. In its 1966 guidelines, the EEOC sought, among other objectives, to establish a job-related requirement for preemployment testing. The guidelines allowed for the usage of a professionally developed ability test, "which fairly measures...the applicant's ability to perform a particular job or class of jobs."

In 1970, a new set of EEOC regulations specified certain technical requirements for test validation. According to Booth and Mackay, the revised regulations were viewed as effectively precluding almost all usage of employment testing. Nevertheless, the Supreme Court in two subsequent decisions accorded considerable deference to the 1970 guidelines.

In the 1971 case of Griggs v. Duke Power Co., the court faulted an intelligence test on the grounds that its use had a disproportionate impact on job transfer applicants and that the employer had failed to demonstrate business necessity sufficient to justify the test usage. The decision accorded great deference to the EEOC job-relatedness requirement. According to the court, the law forbids not only overt discrimination but also practices that, while fair in form, are discriminatory in operation. If an employment selection technique has a disproportionate impact on one race or one suspect class and the employer cannot show that the requirement has a manifest relationship to the employment in question, the selection technique is prohibited. In 1976, however, the Washington v. Davis decision led the way toward a de-emphasis of the 1970 guidelines. First, the court held that disproportionate impact was not the only factor in determining racial discrimination. The decision suggested that intent to discriminate was the important factor. Second, in contrast to what was previously presumed, a training-course performance, and not just job performance, could be used in validating promotion testing procedures. This placed a less burdensome task on employers than did the 1970 guidelines.

In 1976, in realization of the Supreme Court's reluctance to fully accept the 1970 guidelines, the Departments of Justice and Labor and the Civil Service Commission proposed the "Federal Executive Agency Guidelines on Employment Selection procedures." Declining to adopt them, EEOC proceeded to republish its 1970 guidelines in substantially unaltered form. Then, in 1978, EEOC agreed to uniform guidelines that, when drafted and adopted, were similar to the standards that had been proposed by the Justice and Labor Departments and the Civil Service Commission.

Claims procedures. The procedural posture of an employment discrimination claim is as follows: Before a charging party can file a civil rights suit in a federal court, a charge must first be filed with the EEOC or a comparable state agency. If the state has such an agency, which is considered a deferral agency of EEOC, the charges must be filed with it before they can be filed with EEOC itself. After the agency process has been pursued, a complainant who wishes further consideration may file a complaint with the EEOC. Only after a disposition by EEOC may suit be filed in the federal courts.

Although the Civil Rights Act itself prescribes procedures, EEOC guidelines deal with such substantive matters as the validity of tests used for employment and promotion. The act governs exclusively the procedures for bringing suit, but the EEOC guidelines may be applied to testing procedures. Any testing technique that is used as a basis for an employment decision should first be validated, to avoid the consequences of a selection procedure being ultimately considered discriminatory because of an adverse impact. It is important to note, however, that an employer has no legal obligation, at the outset, to validate a selection procedure if the test itself does not appear to have a significant impact on a protected group.

Prohibited inquiries. Some jurisdictions place restrictions on employer access to the criminal history records of applicants and employees. Where access is curtailed, it is reasonable to assume that test questions about the protected information would also be restricted. Reid Psychological Systems, for example, limits inquiries about convictions to a period of five years prior to the testing in deference to certain state laws.

Employers must find ways to avoid hiring future problems. They must do so within the confines of restrictions placed by Congress and the courts. The right approach can help a company keep out bad apples, while keeping the company out of trouble with the law.

1 S. Kochkin, "Personality Correlates of a Measure of Honesty," Journal of Business and Psychology (1987): 236-247; R. C. Hollinger and J. P. Clark, Theft by Employees (New York: Free Press, 1983), 240.

2 States without such statues are Alabama and Arkansas. Louisiana requires "intentional discrimination." The laws in Georgia and Mississippi do not apply to private employers.

3 Dean Booth and James L. Mackay, "Legal Constraints on Employment Testing and Evolving Trends in the Law," Emory Law Journal 29 (1980): 121-194.

Fred E. Inbau is the John Henry Wigmore Professor of Law Emeritus at Northwestern University in Chicago and a consultant to Reid Psychological Systems.

Editor's Note: This article is an edited excerpt of a chapter in the forthcoming second edition of Protective Security Law. The chapter itself contains significantly greater detail, which space does not permit Security Management to include.
COPYRIGHT 1994 American Society for Industrial Security
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1994 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Preemployment Testing; psychological testing of employee's delinquent behavior potentials
Author:Inbau, Fred E.
Publication:Security Management
Date:Jan 1, 1994
Words:3031
Previous Article:The special needs of special event security.
Next Article:Psychological testing in ADA's wake.
Topics:


Related Articles
Shoring up eroding options.
More than a gut feeling.
Screening for the top.
Getting help from the outside.
Psychological testing in ADA's wake.
Pre-employment screening to weed out "bad apples".
A few good men.
Granting a stamp of approval.
What they are.
How to choose the best ones.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters