BE-2

DEBRA R. COMER Crossroads

A Case Against Workplace Drug Testing

Debra R. Comer 228 WellerHall, 134 Hofstra University, Hempstead, New York 11550

Abstract Workplace drug testing, particularly urinalysis, has prolifer- ated in the last few years. Despite widespread support for biological testing, research suggests that not all drug use diminishes performance and that testing may fail to deter the most potentially harmful substance abuse. There is no solid empirical evidence that drug testing is associated with en- hanced organizational productivity and safety, and findings that persons who fail drug tests are inferior workers may be rooted in ethnic discrimination. Further, because drug testing detects exposure to a drug but cannot assess an individual's ability to perform, it is an inappropriate gauge for judging the suitability of employees or applicants. Drug tests may violate current and prospective employees' right to privacy and, according to a growing body of literature, may adversely affect their work attitudes and behaviors. Skills testing, which assesses employees' performance fitness less intrusively, is discussed as an alternative to biological testing. {Drug Testing; Privacy; Employee Attitudes)

In 1957, Chris Argyris advocated a form of organiza- tion in which managers would provide opportunities for employee need fulfillment and creative expression. He believed respecting individuals and cultivating their talents and inputs would in turn benefit organizational goal attainment. Similarly, McGregor (I960) advised managers that, if employees were trusted to find and apply their own approach to doing work, they would strive responsibly to achieve organizational objectives without close monitoring and direction. Likert (1961, 1967), too, stressed the need for organizations to en- courage, not squelch, individuals' initiatives and unique contributions.

A generation of theory and research in organiza- tional behavior and human resources management has since been informed by this notion that appreciating employees is not just an employee entitlement, but a workable organizational strategy. In the last few years, however, individual needs have been increasingly com-

promised by workplace programs of testing for drug use, particularly urine testing for the use of illicit substances. Indeed, the American Management Asso- ciation has reported that 63% of its surveyed members conduct some type of drug testing, a 200% increase since 1987 (Greenberg 1991).

This article argues against workplace drug testing. It identifies inisconceptions about drug use and drug test- ing, underscores the technological limitations of test- ing, contests the moral appropriateness of biological testing, and reviews research on individuals' negative response to workplace drug testing. Perfonnance test- ing is examined as an alternative to drug testing.

Problems with Drug Testing 1. We Do Not Have Conclusive Evidence That Drug

Testing Enhances Organizational Effectiveness Advocates of testing claim it can reduce absenteeism, theft, mistakes, accidents, and medical insurance ex- penses (see, e.g., Elliott 1989, Harwood et al. 1984). McDaniel (1988), using self-report data about the drug use of applicants for military service, observed that individuals who had never used any drugs before enlist- ing were less apt to have been discharged for unsatis- factory performance within four and a half years after their application. Moreover, early first use of a drug and frequent use of a drug during an individual's life both increased the likelihood of being discharged for performance reasons. In a different study, municipal workers who reported having used drugs in the last year or at any point in their lives were more likely than nonusers to report psychological and physical with- drawal from work and to have exhibited antagonistic job behaviors (Lehman and Simpson 1992).

However, hierarchical regression analyses reveal that, after control for the effects of personal and job factors, drug use uniquely contributes only a small portion of the variance in these behaviors. Also, self-reported use in the last year does not have a significant beta weight

ORGANtzATiON SctENCE/Vol. 5, No. 2, May 1994 259

DEBRA R. COMER Crossroads

in the regression equations for any of the criterion behaviors.* Economic analyses of data sampled from the 1984 National Longitudinal Survey of Youth, which also controlled for factors that could simultaneously affect one's decision to use drugs and one's employ- ment, yielded some equally interesting findings: First, Gill and Michaels (1992) found that drug users earned higher wages than nonusers. Second, Register and Williams (1992) observed that although long-term as well as on-the-job use of marijuana had a negative effect on wages, use of marijuana in general (including short-term and nonwork use) had a positive effect on wages. (No significant relationship was found between cocaine use and wages.) The latter team of researchers noted that their findings "should at least give pause to those whose advocacy of drug testing rests wholly or largely on the argument that drug use harms productiv- ity" (Register and Williams 1992, p. 447).

These findings alert us to the possibility that poor job performance may be related to certain social and/or individual factors in the lives of drug users rather than drug usage itself. Despite conventional wisdom, drug testing has not been definitively linked to organizational gains in safety or productivity (see Cropanzano and Konovsky 1993, Harris and Heft 1992, Hoffman and Lovler 1989, Morgan 1991, Thompson et al. 1991). Contributors to a monograph published by the National Institute for Drug Abuse associated drug testing with organizational benefits (Gust and Walsh 1989). However, Morgan's (1991) analysis of the data uncovered several methodological problems with their research, such as (1) an inflated rate of initial positives, (2) withheld information, and (3) competing factors.

Two studies of preemployment drug testing have been conducted with postal employees. Normand et al. (1990) reported that applicants at the Lf.S. Postal Ser- vice in Washington, D.C, who tested positive for illegal drug use (but were hired for the purposes of data collection) later had higher rates of absenteeism and involuntary turnover than applicants with negative drug test results. No significant relationships were detected between drug test results and employees' rate of in- juries or accidents. Interestingly, 85% of the individu- als testing positive had not been fired a year later, and hence were apparently doing their jobs at least ade- quately.

Zwerling et al. (1990) similarly followed Boston postal employees who had been screened for drug use. A little more than one year later, 6.4% of those testing negative for drug use had been terminated, versus 13.6% of those testing positive for marijuana and 7.3% of those testing positive for cocaine (the termination

rate for marijuana-positive employees was statistically significant). Absenteeism rates for employees with pos- itive marijuana tests (7.1%) and those with positive cocaine tests (9.8%) were significantly higher than the rate for those with negative results (4.0%). Those test- ing positive for either drug were apt to be involved in accidents, injuries, and disciplinary actions earlier in their employment than those testing negative. The differences were statistically significant for employees with positive marijuana tests, but for those with posi- tive cocaine tests the differences were significant only for injury risks. However, few of these problems oc- curred overall, and the researchers themselves noted that the differences they found were much smaller than those often claimed by advocates of applicant screen- ing.̂

2. Drug Testing Does Not Deter the Kind of Drug Abuse Most Detrimental to Organizations

Whether drug testing contributes to improved perfor- mance and effectiveness in organizations is not yet clear. Even if it could predict poor performance in the workplace, it would have to do so on a timely enough basis to safeguard individuals and organizations from the most harmful consequences of on-the-job drug impairment, namely major accidents.

Preemployment testing may help screen out addicts, who cannot suspend their drug use even temporarily, at the time they apply for work. Yet such testing cannot prevent workplace drug use by individuals who were hired before preemployment screening was adopted or those who develop serious drug habits after gaining employment. Workplace testing is likely to deter incumbents who decide to curb their infrequent, casual, off-hours drug use (which would not compro- mise their job performance) to avoid the repercussions of being detected. Thus, such testing may effectively reduce the kind of drug use that would affect work behavior minimally (if at all), but not the at-work drug abuse that could be most deleterious to organizational functioning.

It is bold optimism, if not wishful thinking, to expect current employees who imperil themselves and others by using performance-impairing drugs at work to be daunted by the risk of losing their jobs. Such employ- ees may thwart detectors by altering their own urine or even substituting drug-free urine samples (Bearman 1988, Crown and Rosse 1991, Hanson 1988). But even if they do not tamper with their urine specimens, by the time their test results have been interpreted as positive, any drug-impaired behavior could already have had its negative effect.

260 ORGANIZATION S C I E N C E / V O I . 5, No. 2, May 1994

DEBRA R. COMER Crossroads

3. The Relationship Between Positive Drug Test Results and Dismissal May Be Spurious, Actually Attributable to Each Variable's Relationship with Employee Ethnicity

The post office studies raise the question of possible ethnic discrimination in drug testing (Horgan 1991, Morgan 1991). In the Normand et al. (1990) study, a higher incidence of positive drug test results was ob- served for African-Americans than for other ethnic groups (but note that the authors did not control for demographic factors). Moreover, in the Zwerling et al. (1990) study, the positive test rate was twice as high for African-American postal workers (6% of the study) as it was for white workers. Despite having lower absen- teeism, fewer injuries, and no more accidents, the African-American workers were 143% more likely to be terminated.

One explanation for these differences in outcomes is that biased post office supervisors might have been quicker to dismiss their African-American subordi- nates, who, despite testing positive for drug use, ap- pear to have performed at least as well as their white coworkers according to objective measures (Horgan 1991, Morgan 1991). (The authors did not provide supervisors' subjective assessment of the postal em- ployees' performance.) Experiments (Howitt and Owusu-Bempah 1990, McRae 1991) and surveys (Brown et al. 1991, Greenhaus et al. 1990, Mueller et al. 1989) suggest that African-Americans still receive biased per- formance appraisals. Possibly, African-Americans do have a higher incidence of drug use (see Blank and Fenton 1989, Newcomb 1988). However, if organiza- tions are denying or terminating employment on the basis of drug tests that are not valid performance indicators, those tests are likely to discriminate against any group, such as African-Americans, in which drug use may be more prevalent than it is in other groups.

testing policies (Murphy and Thornton 1992) indicated that confirmatory testing of positive results is not uni- versal (although some states require it). Overall, about two-thirds of the organizations sampled checked posi- tive test results with a more rigorous second test, but retest rates were much lower in certain industries. Further, because false positives are possible even with gas chromatography/mass spectrometry (Abelson 1991), confirmed positives should be analyzed carefully to find the underlying cause.

Laboratory error also contributes to unreliable test results (Abbasi et al. 1988, Elliott 1989, Feit and Holosko 1990). Although the standards for laboratory certification set by the National Institute for Drug Abuse can promote appropriate handling of specimens and minimize human error, Brookler (1992) has re- ported that less than 7% of the nation's drug testing labs meet these standards. Consequently, some of the people who receive positive drug test results may be (potentially) competent performers who are mistakenly denied or dismissed from employment.

An even more disturbing aspect of drug testing tech- nology is that a test can distinguish only between people who have been exposed to the drug being tested and those who have not. A positive result does not indicate patterns of drug use, abuse, or dependency (Morgan 1987). Metabolites can appear in an individual's urine long after the drug has ceased to affect his or her behavior, and differences in physi- ology can affect an individual's metabolism of a drug (Lundberger 1986). A urine test cannot ascertain the quantity of a drug consumed, the time of consumption, or its effect on the user. In short, it cannot demon- strate the very performance impairment its proponents seek to deter or detect for the sake of productivity and safety, and is therefore an inappropriate basis for as- sessing an employee's ability to perform.

4. The Technology of Urinalysis Limits Its Usefulness for Ensuring Safety and Productivity in the Workplace

The failure of drug testing to promote organizational effectiveness is due partly to its technological limita- tions. One problem is that false positives for some drugs can result from eating certain foods or taking prescription or over-the-counter cold pills or painkillers (DeCresce et al. 1989). Yet some companies determine the fate of current and prospective employees solely on the basis of an immunoassay test, without seeking confirmation through the more expensive, but more specific and rigorous, gas chromatography/mass spec- trometry method. A recent survey of employee drug

5. Drug Testing Violates Individuals' Right to Privacy Even if drug testing did enhance workplace effective- ness, it would have to be morally defensible. It is important not to confound the morality of drug testing with its legality; in fact, the courts have ruled in favor of the legitimacy of drug testing in many situations. Until 1989, the ability of government employers to conduct drug testing had been limited by the Fourth Amendment's prohibition of unreasonable search and seizure (Elliott 1989, Haas 1990). But the United States Supreme Court ruled that year that suspicionless test- ing is not unconstitutional in cases involving the Special needs of public safety (see Skinner v. Railway Labor

ORGANIZATION SCIENCE/VOI. 5, No. 2, May 1994 261

DEBRA R. COMER Crossroads

Executives' Association 1989, allowing testing of train crew members after a serious accident, and National Treasury Employees Union v. Von Raab 1989, allowing testing of Customs Service agents seeking promotions to jobs involving drug interdiction or the use of firearms).

A few states have laws constraining the conditions under which employees can be tested for drugs (see Berlin 1991). However, because private sector organi- zations (other than those in highly regulated industries, such as transportation) may act as private individuals in establishing drug testing programs,^ private sector em- ployees have usually turned to civil rights laws, union contracts, and tort suits for protection from drug test- ing (Abbasi et al. 1988, Yurow 1989). Nonetheless, "In general, unless a test violates state law, litigation chal- lenging the right to test has favored the employer" (Blum 1991, p. 1).

Is legal drug testing morally right? A philosophical treatment of drug testing must consider individual rights as well as community welfare (safety). Whereas deontologists focus on whether an act itself is right or wrong, utilitarians consider the consequences of the act and judge it morally right if it results in the greatest good (Pojman 1990). A deontologist would argue against drug testing on the grounds that it is morally inappropriate to deprive an individual of the right to privacy. A utilitarian, in contrast, would contend that an organization is warranted in sacrificing the rights of some individuals to enhance the security of their co- workers and the public. Determining just what is best for the most is an ill-defined and formidable task, and utilitarian reasoning has been criticized for endorsing the use of immoral or unjust actions to achieve the greatest good (Pojman 1990, Rosen 1993).

Business ethicists have applied a deontological perspective to the question of drug testing. Their insistence on respecting and protecting employees' in- dividual rights has led to their rejection of workplace testing. According to DesJardins and Duska (1987, p. 4):

An employee's right to privacy is violated whenever personal infonnation is requested, collected and/or used by an em- ployer in a way or for any purpose that is irrelevant to or in violation of the contractual relationship that exists between employer and employee.""

That is, employees are entitled not to disclose certain private information that is of no concern to their employers. Even if drug use does affect job perfor- mance, so long as employees can function satisfactorily such drug use is not germane to their employers.^

Caste (1992, p. 305) agrees: "The intrusion into the private lives of the employee that is occasioned by drug testing wrongly appropriates time which [sic] was not purchased." If an employee cannot perform ade- quately. Caste (1992) and DesJardins and Duska (1987) reason that an employer may rightfully exercise sanc- tions, but the cause of the problem performance is not job-relevant and therefore need not be revealed.

DesJardins and Duska (1987) acknowledge the po- tential of an employee's impaired behavior to endanger others. But they argue that employers must uphold individual rights and therefore must rely on procedures less invasive than drug tests to reduce such risks. Moore (1989) likewise emphasizes that organizations' responsibility for any harmful acts their employees commit while intoxicated in no way gives them carte blanche to control employees. She, too, asserts that organizations should take pains to identify impaired performance through means that protect employees' privacy rights.*

6. Many Employees Respond Negatively to Drug Testing Do job applicants and current employees actually view drug testing as invasive and unjust? Crant and Bate- man (1989), applying the concepts of organizational justice, assert that individuals' perceptions of the fair- ness of both the process and outcomes of drug testing programs affect their attitudinal and behavioral re- sponses to testing. These perceptions are influenced by such factors as the employees' assessment of the justi- fiability of having their privacy invaded.

Although the expansion of drug testing implies a failure of organizations to consider any negative effects of testing on employees, research undertaken in the last few years suggests that employees' perceptions of drug testing have an important impact on their atti- tudes and behaviors. Blue-collar employees at a manu- facturing firm who read a hypothetical scenario in which employees with positive test results were fired viewed testing negatively, and less favorably than those who read a scenario in which such employees were placed in rehabilitation (Stone and Kotch 1989). Simi- larly, college students had negative attitudes toward terminating an employee for a positive test result, as well as toward random testing (Murphy et al. 1990). Drug testing was also opposed when it was not limited to individuals performing dangerous or safety-related jobs (Murphy et al. 1991). In another investigation, undergraduate business students generally deemed drug testing less appropriate as an employer's right than "normative" activities such as assigning work, but not so inappropriate as "intrusive" activities such as

262 ORGANIZATION S C I E N C E / V O I . 5, No. 2, May 1994

DEBRA R. COMER Crossroads

inquiring about an employee's sexual preference or religious identification (Garland et al. 1989). Yet they viewed probing employees' off-the-job drug use as less acceptable than assessing their on-the-job use, al- though (as they were apparently unaware) urine testing cannot determine the time of drug use.

Some research has investigated employees' percep- tions of the actual testing practices used by their own organizations. Hanson (1990) reported that railroad and chemical workers generally perceived testing as justifiable only for employees who seem to be under the influence. His respondents were especially opposed to being subjected to post-accident testing when the mishap clearly resulted from nonhuman error. Their comments also revealed that they viewed random or periodic testing without reasonable suspicion as a hu- miliating intrusion on their privacy, as well as a sign that managers distrusted them and discounted their years of good service. Additionally, some railroad em- ployees feared that drug testing was being used to reduce the workforce.

Empirical evidence indicates that drug testing affects job applicants as well as current employees. Students who read a scenario about an organization that did not conduct drug testing had more positive attitudes to- ward the organization and reported greater intentions to apply for a job than did those who read about an organization that did test (Crant and Bateman 1990). In other research (Murphy et al. 1990), students viewed negatively the automatic rejection of a candidate with a positive drug test result. Stone and Bowden (1989) focused on how individuals' perceptions of hypothetical drug testing policies differ as a result of the timing of the test (before/after making a job offer) and the method of choosing the testees (testing a l l / randomly/on suspicion). It is notable that individuals had negative attitudes toward five of the combinations, and only barely above-average attitudes toward the sixth.

The studies cited suggest that drug testing programs not just for incumbents, but applicants too, may lead to negative attitudes and behaviors, especially when the process and/or outcome seems invasive and/or unfair (e.g., when testing seem(s) unnecessary or denial of or dismissal from employment is based on a positive test result). Such programs may cause some talented indi- viduals to refrain, on principle, from applying to orga- nizations that test for drug use. However, although most of the unionized public sector employees queried by Le Roy (1990) rejected random drug testing, most accepted other types of testing as long as certain proce- dures (e.g., maintaining confidentiality and carefully

handling and analyzing specimens) were followed (see also Konovsky and Cropanzano 1991, Verespej 1992).

The Performance Testing Alternative The case argued here against drug testing is based on six points: (1) there is little evidence that drug testing increases organizational effectiveness, (2) drug testing may be least effective in deterring the kind of drug abuse that is most deleterious in the workplace, (3) the scant evidence linking positive test results with im- paired performance may be due to ethnic bias, (4) the technology of urinalysis precludes it from ensuring workplace safety and productivity, (5) testing may com- promise employees' rights, and (6) testing may evoke negative employee responses. An alternative to drug testing is therefore needed.

Performance testing is one alternative to (biological) drug testing as a way to minimize the substandard work that should ultimately concern managers, or to detect it before it occurs, while protecting employees' rights. For example, the computer-based critical tracking test is a video game in which employees apply the psy- chomotor skills required in their jobs (Frieden 1990, Maltby 1990). Test-takers manually try to keep an ever-moving, randomly careening cursor centered on their monitor, and their speed and accuracy scores on each daily trial are compared with their average perfor- mance on previous trials. The test is designed so that an employee cannot dupe the computer into establish- ing a low baseline (to pass the test later while im- paired) because an unusual lack of progress on initial trials is detectable (Fine 1992).

Skills tests assess job-relevant reaction time and co- ordination, which can be affected by illness, sleep de- privation, or emotional preoccupation as well as by drug use. Hence, they are more informative about employees' job abilities than urine tests. Compared to urinalysis, the critical tracking test requires less data collection time and is less costly to administer and interpret: only $100 or $200 per employee per year rather than $40 to $60 per test per employee (Warshauer 1991). Moreover, it provides immediate results (Frieden 1990). Performance testing has even detected drug use that has eluded urinalysis (Maltby 1990). For example, cocaine will not show up in the urine of an employee who has just used it, but a performance test will demonstrate skill impairment. Organizations that use performance tests report them to be more effective and efficient than drug tests, and better received by employees (McGinley 1992).

ORGANIZATION S C I E N C E / V O I . 5, No. 2, May 1994 263

DEBRA R. COMER Crossroads

The critical tracking test is typically recommended for employees whose mistakes of hand-eye coordina- tion would be potentially life-threatening. Another computer-based test battery may have applications for a broader variety of occupations. This new means of testing performance can assay 25 skills, including deci- sion-making, spatial relations, and psychomotor func- tioning (McGinley 1992).

Skills tests hold promise for employers who care more about their employees' fltness to perform on the job than what they do in their private time. However, because such tests have just recently been adapted for the workplace, their effectiveness must be evaluated systematically (Harris and Heft 1992). The critical tracking test may require some debugging, as evi- denced by the experience of one transportation com- pany that used but ultimately abandoned it. The company's general manager explained that test-taking anxiety impeded the performance of some of his older, less computer-literate employees. He also reported that for employees of all ages, the difficulty of "beating" the cursor varied dramatically from day to day. When the test was especially tough, many people failed it despite their apparent fitness for work. In contrast, when the test was easier, even a compromised per- former could succeed. One evening when this manager thought he had drunk too much alcohol to risk driving home from a restaurant, he decided to satisfy his curiosity about the effectiveness of the critical tracking test. He asked his dinner companion to drive him to his office, where he sat in front of the video monitor and passed the test on his first attempt.

Ethical issues in the use of performance testing must also be resolved. For example, what happens when an employee fails several consecutive tests? One could argue that the employer of this hypothetical employee would not be entitled to inquire into the cause of these failures. Yet, some performance-testing organizations require such an employee to submit to a urine test, even though stress or illness, rather than drug use, may be the reason for the performance failure.

Conclusion Workplace drug testing is proliferating. Yet we do not have compelling evidence linking it to organizational effectiveness; employees' perceptions of its invasive- ness and unfairness may adversely affect their job- related attitudes and behavior; and performance test- ing, once fine-tuned, may be a more effective, efficient, and respectful alternative. Prasad et al. (1992) argue that organizations have endorsed drug testing for pur-

poses more symbolic than practical. They believe drug testing is used to create the impression that manage- ment is applying current scientific methods to control a perceived crisis.̂ In reality, however, testing alone can hardly be expected to conquer drug abuse. Drug-de- pendent individuals are unlikely to recover without the benefit of education, counseling, and rehabilitation (Crown and Rosse 1991, Stevens et al. 1989, Wrich 1988).* But less than half of the organizations surveyed by Murphy and Thornton (1992) provide such em- ployee-centered services to persons who test positive for drug use.

Ironically, some of the very scholars who have warned of the undesirable consequences of employees' nega- tive views of drug testing have apparently accepted testing as a given. They have identified which features of drug testing individuals deem least objectionable and have recommended designing programs that will meet the least employee resistance (see, e.g., Konovsky and Cropanzano 1991, Murphy et al. 1991, Stone and Kotch 1989). Before condoning or even tacitly allowing drug testing for job incumbents and applicants, organi- zational scientists should reexamine assumptions about the soundness and impact of testing. Moreover, we must be wary of the tendency to "so closely identify with the group that has power" that we downplay employees' interests, not to mention what makes sense (Brief and Dukerich 1991, p. 345). Workplace drug testing, instituted to foster the achievement of organi- zational goals, seems to signal a disregard for the individual respect championed by Argyris (1957), McGregor (1960), and Likert (1961, 1967). Even if drug testing could guarantee improved safety and productiv- ity, employers should be obligated to reject it for depriving employees of their right to privacy. Instead of advising organizations as to which drug testing prac- tices will least offend and unsettle their employees, we should be educating them about the failings of drug testing and recommending that they refine the poten- tially preferable technique of performance testing.'

Endnotes 'interpretation of Lehman and Simpson's (1992) findings about at-work drug use is obscured because their operationalization com- bined illicit drug use with alcohol use. ^As Horgan (1991) observes, the operationalizations of these prob- lems seem odd. According to their coding scheme, the authors would have considered incurring a paper cut on one's first day on the job more incriminating than improperly sorting an entire neighborhood's mail six months later. ^United States v. Jacobsen (1984) rules that the Fourth Amendment does not apply to the actions of private individuals, and Monroe v.

264 ORGANIZATION S C I E N C E / V O 1 . 5, No. 2, May 1994

DEBRA R. COMER Crossroads

Consolidated Freightways, Inc. (1987) specifically applied Jacobsen to private sector drug testing. This explanation is consistent with Stone and Stone's (1990) concep-

tualization of privacy as information control, regulation of interac- tions with others, and freedom from control by others. Further, drug tests can reveal medical conditions that individuals

may prefer not to announce to their current or prospective employ- ers. If an employee's use of a prescription medication has no nega- tive impact on his or her work performance, the employee should not be required to discuss such use with an employer. * Moreover, the (utilitarian) rationale, that sacrificing individuals' privacy rights by drug testing is necessary to maximize societal good, is based on fallacious assumptions. As discussed previously, work- place testing may not actually deter the most deleterious drug use or detect it in time to prevent it from jeopardizing safety.

Likewise, Guthrie and Olian (1991) found that drug testing pro- grams were twice as prevalent as alcohol testing programs among the Fortune 1,000 organizations in their sample, and that employees with positive results for alcohol were granted more leeway than those with positive results for drugs, even though workplace alcohol abuse is much more common and expensive than drug abuse. They con- cluded, "It appears that substance abuse testing programs are driven more by the strong anti-drug social/political climate and perhaps the illegality of drugs rather than productivity or cost concerns" (p. 230). Ironically, testing may actually aggravate an employee's drug prob-

lem, rather than deter or eliminate it. Urine testing can heighten job stress (Crant and Bateman 1989), which in turn has been linked to substance abuse (Sauter et al. 1990). 'The author gratefully acknowledges the recommendations of Peter J. Frost and an anonymous reviewer, and the research assistance of Shuja Karim.

References Abbasi, S. M., K. W. Hollman, and J. H. Murrey, Jr. (1988), "Drug

Testing: The Moral, Constitutional, and Accuracy Issues," Jour- nal of Collective Negotiations, 17, 3, 221-235.

Abelson, J. L. (1991), "Urine Drug Testing—Watch What You Eat," X4A^, 266, 22, 3130-3131.

Argyris, C. (1957), Personality and Organization, New York: Harper & Row Publishers, Inc.

Bearman, D. (1988), "The Medical Case Against Drug Testing," Harvard Business Review, 66, 1, 126.

Berlin, P. E. (1991, July 8), "More States Now Regulate Drug Testing in the Workplace," The National Law Journal, 13, 44, 19-21.

Blank, D. L. and J. W. Fenton (1989), "Early Employment Testing for Marijuana: Demographic and Employee Retention Patterns," Research and Evaluation Data, Rockville, MD: National Insti- tute on Drug Abuse, 139-150.

Blum, A. (1991, December 11), "State Drug Test Rules Are Varied," The National Law Journal, 14, 15, 1, 43.

Brief, A. P. and J. M. Dukerich (1991), "Theory in Organizational Behavior: Can It Be Useful?" in B. M. Staw (Ed.), Research in Organizational Behavior, 13, Greenwich, CT: JAI Press, Inc., 327-352.

Brookler, R. (1992), "Industry Standards in Workplace Drug Test- ing," Personnel Journal, 71, 4, 128-132.

Brown, D., C. W. Minor, and D. A. Jepsen (1991), "The Opinions of Minorities About Preparing for Work: Report of the Second NCDA National Survey," Career Development Quarterly, 40, 1, 5-19.

Caste, N. J. (1992), "Drug Testing and Productivity," Joumai of Business Ethics, 11, 301-306.

Crant, J. M. and T. S. Bateman (1989), "A Model of Employee Responses to Drug-testing Programs," Employee Responsibilities and Rights Joumai, 2, 3, 173-190.

and (1990), "An Experimental Test of the Impact of Drug-Testing Programs on Potential Job Applicants' Attitudes and Intentions," Joumai of Applied Psychology, 75, 2, 127-131.

Cropanzano, R. and M. Konovsky (1993), "Drug Use and Its Impli- cations for Employee Drug Testing," in G. Ferris and M. Rowland (Eds.), Research in Personnel and Human Resources Management, 11, Greenwich, CT: JAI Press, Inc., 207-257.

Crown, D. F. and J. G. Rosse (1991), "Critical Issues in Drug Testing," in J. W. Jones, B. D. Steffy, and D. W. Bray (Eds.), Applying Psychology in Business, Lexington, MA: D. C. Heath & Company, 260-274.

DeCresce, R., M. S. Lifshitz, A. C. Mazura, and J. E. Tilson (1989), Drug Testing in the Workplace, Chicago, IL: American Society of Clinical Pathologists Press.

DesJardins, J. and R. Duska (1987), "Drug Testing in Employment," Business & Professional Ethics Journal, 6, 3, 3-21.

Elliott, R. H. (1989), "Drug Testing and Public Personnel Adminis- tration," Review of Public Personnel Administration, 9, 3, 15-31.

Feit, M. D. and M. J. Holosko (1990), "Drug Testing: A Research Strategy," Compensation & Benefits Management, 6, 3, 219-224.

Fine, C. R. (1992), "Video Tests Are the New Frontier in Drug Detection," Personnel Joumai, 71, 6, 149-161.

Frieden, J. (1990), "Corporate America's Response to Substance Abuse," Business & Health, 8, 7, 32-42.

Garland, H., J. Giacobbe, and J. L. French (1989), "Attitudes To- ward Employee and Employer Rights in the Workplace," Em- ployee Responsibilities and Rights Joumai, 2, 1, 49-59.

Gill, A. M. and R. J. Michaels (1992), "Does Drug Use Lower Wages?" Industrial and Labor Relations Review, 45, 3, 419-434.

Greenberg, E. R. (1991), 1991 AMA Survey on Workplace Drug Testing and Drug Abuse Policies, New York: American Manage- ment Association.

Greenhaus, J. H., S. Parasuraman, and W. M. Wormley (1990), "Effects of Race in Organizational Experiences, Job Perfor- mance Evaluations, and Career Outcomes," Academy of Man- agement Joumai, 33, 1, 64-86.

Gust, S. W. and J. M. Walsh (1989), Drugs in the Workplace, NIDA Research Monograph Series # 1 , Washington, DC: U.S. Govern- ment Printing Office.

Guthrie, J. P. and J. D. Olian (1991), "Drug and Alcohol Testing Programs: Do Firms Consider Their Operating Environment?" Human Resource Planning, 14, 3, 221-232.

Haas, K. C. (1990), "The Supreme Court Enters the 'Jar Wars': Drug Testing, Public Employees, and the Fourth Amendment," Dick- inson Law Review, 94, Winter, 305-371.

ORGANIZATION SCIENCE/VO1. 5, No. 2, May 1994 265

DEBRA R. COMER Crossroads

Hanson, A. (1988), "Some Social Implications of Drug Testing," The

University of Kansas Law Review, 36, 4, 899-917.

(1990), "What Employees Say About Drug Testing," Personnel,

67, 7, 32-36.

Harris, M. M. and L. L. Heft, (1992), "Alcohol and Drug Use in the Workplace: Issues, Controversies, and Directions for Future

Research," Joumai of Management, 18, 2, 239-266.

Harwood, H. S., D. M. Napolitano, P. L. Kristiansen, and J. J.

Collins (1984), Economic Costs to Society of Alcohol and Drug

Abuse and Mental Illness: 1980, Research Triangle Park, NC:

Research Triangle Institute.

Hoffman, R. G. and R. Lovler (1989), "Drug Testing: A Multi-faceted

Issue," AITH, Fall, 6-18.

Horgan, J. (1991), "Postal Mortem: Drug Test Results May Point to

Racial Discrimination," Scientific American, 264, 2, 22-23.

Howitt, D. and J. Owusu-Bempah (1990), "The Pragmatics of Insti-

tutional Racism: Beyond Words," Human Relations, 43, 9,

885-899.

Konovsky, M. A. and R. Cropanzano (1991), "Perceived Fairness of

Employee Drug Testing as a Predictor of Employee Attitudes

and Job Performance," Joumai of Applied Psychology, 76, 5,

698-707.

Lehman, W. E. K. and D. D. Simpson (1992), "Employee Substance

Use and On-the-Job Behaviors," Joumai of Applied Psychology,

11, 3, 309-321.

Le Roy, M. H. (1990), "Drug Testing in the Public Sector: Union

Member Attitudes," Joumai of Collective Negotiations in the

Public Sector, 19, 3, 165-173.

Likert, R. (1961), New Patterns of Management, New York:

McGraw-Hill Book Company.

(1967), The Human Organization, New York: McGraw-Hill

Book Company.

Lundberger, G. (1986), "Mandatory Unindicated Urine Drug

Screening: Still Chemical McCarthyism," JAMA, 256, 21,

3003-3004.

Maltby, L. L. (1990), "Put Performance to the Test," Personnel, 67,

7, 3 0 - 3 1 .

McDaniel, M. A. (1988), "Does Pre-Employment Drug Use Predict

On-the-Job Suitability?" Personnel Psychology, 41, 4, 717-729.

McGinley, L. (1992), " ' F i t n e s s ' Exams Help to Measure Worker

Acuity," Wall Street Journal, April 21, BI, BIO.

McGregor, D. (1960), The Human Side of Enterprise, T<lew York:

McGraw-Hill Book Company.

McRae, M. B. (1991), "Sex and Race Bias in Employment Decisions:

Black Women Considered," Joumai of Employment Counseling,

28, 3, 91-98. Monroe u. Consolidated Freightways, Inc. (1987) 654 Federal Supple-

ment 661.

Moore, J. (1989), "Drug Testing and Corporate Responsibility: The

'Ought Implies Can' Argument," Joumai of Business Ethics, 8, 4,

279-287.

Morgan, J. P. (1987), "Urine Testing for Abused Drugs: Technology and Problems," in S. G. Kramer (Ed.), AIDS and Drug Abuse in

the Work Place, New York: Prentice-Hall, Inc., 63-67.

(1991), Under What Influence?: A Critical Review of Reputed

Benefits of Workplace Drug Testing, unpublished paper. City

University of New York Medical School.

Mueller, C. W., T. L. Parcel, and K. Tanaka (1989), "Particularism in

Authority Outcomes of Black and White Supervisors," Social

Science Research, 18, 1, 1-20.

Murphy, K. R. and G. C. Thornton (1992), "Characteristics of

Employee Drug Testing Policies," Journal of Business and Psy-

chology, 6, 33, 295-309.

, and K. Prue (1991), "Influence of Job Characteristics on

the Acceptability of Employee Drug Testing," Joumai of Applied

Psychology, 76, 3, 447-453.

, and D. H. Reynolds (1990), "College Students' Attitudes

Toward Employee Drug Testing Programs," Personnel Psychol-

ogy, 43, 3, 615-631.

National Treasury Employees Union v. Von Raab (1989), 109 Supreme

Court 1384.

Newcomb, M. D. (1988), Drug Use in the Workplace: Risk Factors for

Disruptive Substance Use Among Young Adults, Dover, MA:

Auburn House.

Normand, J., S. D. Salyards, and J. J. Mahoney (1990), "An Evalua-

tion of Preemployment Drug Testing," Journal of Applied Psy-

chology, 75, 6, 629-639.

Pojman, L. P. (1990), Ethics: Discovering Right and Wrong, Belmont,

CA: Wadsworth Publishing Company.

Prasad, P., J. M. Cavanaugh, and J. M. Oakes (1992), "Drug Testing

in the Workplace: The Symbolic Management of an Organiza-

tional Crisis," in C. Harris and C. C. Lundberg (Eds.), Emerging

Realities in an Around Organizations: Proceedings of the Twenty-

ninth Annual Meeting of the Eastern Academy of Management,

Baltimore, MD: 46-49.

Register, C. A. and D. R. Williams (1992), "Labor Market Effects of

Marijuana and Cocaine Use Among Young Men," Industrial

and Labor Relations Review, 45, 3, 435-448.

Rosen, B. (1993), Ethical Theory: Strategies and Concepts, Mountain

View, CA: Mayfield Publishing Company.

Sauter, S. L., L. R. Murphy, and J. J. Hurrell, Jr. (1990), "Prevention

of Work-related Psychological Disorders," American Psycholo-

gist, 45, 10, 1146-1158.

Skinner V. Railway Labor Executives' Association (1989), 109 Supreme

Court 1402.

Stevens, G. E., C. D. Surles, and F. W. Stevens (1989), " A Better

Approach by Management in Drug Testing," Employee Respon-

sibilities and Rights Journal, 2, 1, 6 1 - 7 1 .

Stone, D. L. and C. Bowden (1989), "Effects of Job Applicant Drug

Testing Practices on Reactions to Drug Testing," in F. Hoy

(Ed.), Academy of Management Conference Best Paper Proceed-

ings Washington, DC, 290-294.

and D. A. Kotch (1989), "Individuals' Attitudes toward Organi-

zational Drug Testing Policies and Practices," Journal of Applied

Psychology, 74, 3, 518-521.

Stone, E. F. and D. L. Stone (1990), "Privacy in Organizations:

Theoretical Issues, Research Findings, and Protection Mecha-

nisms," in G. Ferris and M. Rowland (Eds.), Research in Person-

nel and Human Resources Management, 8, Greenwich, CT: JAI

Press, Inc., 349-411.

266 ORGANIZATION SCIENCE/VOI. 5, No. 2, May 1994

DEBRA R. COMER / J . MICHAEL CAVANAUGH AND PUSHKALA PRASAD Crossroads

Thompson, F. J., N. M. Riccucci, and C. Ban (1991), "Drug Testing

in the Federal Workplace: An Instrumental and Symbolic As-

sessment," Public Administration Review, 51, 6, 515-525.

United States v. Jacobsen (1984), 466 U.S. 109.

Verespej, M. A. (1992, February 17), "Drug Users—Not T e s t i n g -

Angers Workers," Industry Week, 33-34.

Warshauer, W. (1991, September 17), "Performance Testing," The

New York Times, A-20.

Wrich, J. T. (1988), "Beyond Testing: Coping with Drugs at Work,"

Harvard Business Review, 66, 1, 120-130.

Yurow, L. (1989), "Alternative Challenges to Drug Testing ot Gov-

ernment Employees: Options After Von Raab and Skinner,"

The George Washington Law Review, 58, 1, 148-179.

Zwerling, C , J. Ryan, and E. J. Orav (1990), "The Efficacy of

Preemployment Drug Screening for Marijuana and Cocaine in

Predicting Employment Outcome," //lAt4, 264, 20, 2639-2643.

Drug Testing as Symbolic Managerial Action: In Response to "A Case

Against Workplace Drug Testing"

J. Michael Cavanaugh • Pushkala Prasad School of Business, Fairfield University, Fairfield, Connecticut 06430

Faculty of Management, University of Calgary, Calgary, Canada TIN \NA

{Drugs, Workplace Drug Testing; Symbolism; Institu- tional Theory; Phenomenology)

In contrast to much of the management and organiza- tional literature supporting drug testing (Coombs and Coombs 1991, Cowan 1987, Harris and Heft 1992), Debra Comer's (1994) "A Case Against Workplace Drug Testing" presents a refreshing series of argu- ments against this practice. Comer (1994) carefully scrutinizes and summarizes a substantial body of em- pirical and conceptual literature on workplace drug testing, on the basis of which, she makes a compelling argument against it. Comer even suggests that fre- quently, drug testing can have adverse consequences for organizations in terms of hurting employee morale, productivity and performance.

For the most part, we are in agreement with Comer's findings. However, we suggest that her paper falls short of offering a convincing explanation for the continued use of drug testing in the workplace. Her case against drug testing is marshalled from two distinct vantage points: (1) normative, and (2) instrumental. Both are incomplete when it comes to explaining the prevalence

of workplace drug testing. In the remainder of this paper, we explain why we think these two perspectives are inadequate, and suggest an alternative way of look- ing at the phenomenon.

The Limitations of Normative and Instrumental Positions First of all. Comer's case against drug testing is explic- itly normative. That is, she questions the morality of drug testing by underscoring its violation of employee privacy rights. On account of these violations, she suggests that drug testing is "morally inappropriate" (Comer 1994, p. ). We have no quarrel with this posi- tion, and in fact share her values concerning this issue. Nevertheless, we also suggest that this does not contain sufficient grounds to make a convincing case against drug testing.

For one thing. Comer's view represents just one moral or normative position. Other normative positions could conceivably view drug testing quite differently. For instance, one ethical stand might view drug use itself as morally wrong or sinful, and consequently see

ORGANIZATION SCIENCE/VOI. 5, No. 2, May 1994 267