Test Policy and the Politics of Opportunity Allocation: The Workplace and the Law

    Test Policy and the Politics 
of Opportunity Allocation: 
The Workplace and the Law 

edited by 

Bemard R. Gifford 

Graduate School of Education 
University of California, Berkeley 

Vv 
WN 

Kluwer Academic Publishers 
Boston Dordrecht London 

 



  

Distributors for North America: 
Kluwer Acagemic Pubhshers 

'01 Philip Drive 

AssSinioDI Park 

Norwell. Massacnusetts 02061 USA 

Distributors for ail other countries: 

Kluwer Academic Pubhisners Grouo 

Distnbution Centre 

Post Office Box 322 

3300 AH Dorcrecnt THE NETHERLANDS 

Library of Congress Catsioging-in-Publication Data   

Test policy and the pontics of allocation ‘he workplace ang ‘ne law 
/ (eaitea] bv Bernard R Gittora 

0. <M — (Evaluation in egucalion ang human services) 
Proceeaings creparea unger tre aegis of the Nationa: Commission 

on Testing ana Puonc Policy —introd 

Incluges ingex 
SBN 0-7923-3515-6 (v 

! Emplovment ‘ests—iaw and 'egisiaton—yUnitea States — 
Congresses _ Zcucalional ‘ests ang measurements—._aw and 
egisiation—niteg States—Congresses. 2 Employment tesis—united 

States—Ccngresses <4 Egucational tests and measurements —ounited 

States—CC~cresses Gitforc Sernarg R i Nationa 

Commussicn on “asting ara Pubic Poncy (JS) ti Series 

ETI 2 ATeva~ +000 

344 73C12386—1:C"9 

1347 3C412596] 39-2381 

CiP 

Cooyrignt - "389 cv Kiuwer Acagemic PLphsners 

All ngnts reserved NO cart of *™s pupbicalion May De reproguced. siored in a 

retrieval svstem or ‘fansmitted in anv ‘orm or Dy any Means ~ecnamcal 

DNOLOCOOVING. “2CCraINg. Or C!Irernwise. ~Ithout ‘Ne Dror written permission of 

‘ne publisher <. wer Acagemc Pupusners ‘C1 Php Drive, ASSInoDi Park 

Norwell. Massacnuse'ts C2061 

Pnnted in "he Uniteg S:ates of Amenca 

 



Contents 

List of Contributors 

Introduction 

I 
GROUNDING TESTING POLICY: THREE PERSPECTIVES 

The Allocation of Opportunities and the Politics of 

Testing: A Policy Analytic Perspective 

Bernard R. Gifford 

The Mandarin Mentality: Civil Service and University 

Admissions Testing in Europe and Asia 

Carolyn Webber 

Testing Companies, Trends, and Policy Issues: 

A Current View from the Testing Industry 

John |. Fremer 

i 

TESTING AND THE LAW: 

TITLE VII AND THE FEDERAL GUIDELINES 

Employment Testing and Title VII of the Civil Rights Act of 1964 

Patrick O. Patterson 

Non-Discriminatory Use of Personnel Tests 

Conference Remarks 

Donald |. Schwartz 

The Uniform Guidelines and Subjective Selection Criteria and Procedures 

Conference Remarks 

Robert Gelerter  



  

hi 

TESTING AND THE LAW: THE ROLE OF THE COURTS 

Testing, Public Policy, and the Courts 

Michael A. Rebell 

Testing in Elementary and Secondary Schools: 
Can Misuse Be Avoided? 

Norman |. Chachkin 

Iv 

TESTING IN THE WORKPLACE: 
THEORETICAL AND PRACTICAL PERSPECTIVES 

Economic Models of Discrimination, Testing, and Public Policy 
Robert F. Adams 

Ability Testing for Job Selection: Are the Economic Claims Justified? 
Henry M. Levin 

Examples of Testing Programs in the Insurance Industry and a 
Discussion of Employment Testing Policy Issues 
Andrew G. Neiner and William D. Love 

Test Scores and Evaluation: The Military as Data 

John Sibley Butler 

Los Angeles Testing Policies 
Conference Remarks 

Raymond C. Fisher 

Index 

163 

191 

211 

265 

293 

 



  

Contributing Authors 

Robert F. Adams, Professor, Board of Studies in Economics, Crown College, 

University of California, Santa Cruz 

John Sibley Butler, Associate Professor, Department of Sociology, University 

of Texas, Austin 

Norman J. Chachkin, Assistant Counsel, NAACP Legal Defense and 

Educational Fund, Inc., New York 

Raymond C. Fisher, Member and Past President, Los Angeles Civil Service 

Commission; Attorney at Law, Heller Ehrman, Los Angeles, California 

John J. Fremer, Senior Development Leader, Educational Testing Service, 

Princeton, New Jersey 

Robert Gelerter, Chief, Special Studies and ADP Section, Office of Federal 

Contract Compliance Programs, U.S. Department of Labor, Washington, D.C. 

Bernard R. Gifford, Dean and Chancellor's Professor of Education, Graduate 

School of Education, University of California, Berkeley 

Henry M. Levin, Director, Center for Educational Research at Stanford; 

Professor of Education and Economics, Graduate School of Education, Stanford 

University, Stanford, California 

William D. Love, Assistant Vice-President, Agent Selection Services, 

Life Insurance Marketing and Research Association, Hartford, Connecticut 

Andrew G. Neiner, Director, Special Services, Human Resources Division, 

Life Office Management Association, Atlanta, Georgia 

Patrick O. Patterson, Co-Director, Western Regional Office, NAACP Legal 

Defense and Educational Fund, Inc., Los Angeles, California 

Michael A. Rebell, Attorney at Law, Rebell and Katzive, New York 

Donald J. Schwartz, Staff Psychologist, Research and Analytic Services, 

Equal Employment Opportunity Commission, Washington, D.C. 

Carolyn Webber, Research Associate, Institute of Urban and Regional 

Development, University of California, Berkeley 

 



    

  
 



  

Employment Testing and Title VII 
of the Civil Rights Act of 1964 
Patrick O. Patterson! 

I. INTRODUCTION 

Long before employment testing became a matter of concern to civil rights law 

and lawyers, psychologists and social scientists knew that certain kinds of tests 

disproportionately screened out minorities and women. The standard texts on 

psychological testing have recognized for many years that as groups, blacks, 

Hispanics, and other minorities generally do not perform as well as Anglos on 

standardized tests and other selection criteria that emphasize verbal skills and 

mastery of the dominant culture.2 They have also recognized that women, as a 

group, generally do not perform as well as men on tests that emphasize certain 

mechanical and physical skills.3 

While questions of causation and remediation continue to be contested, 

there is no longer any serious doubt as to the existence of these longstanding 

patterns of differential test performance. As stated in the report on testing 

issued in 1982 by the National Research Council and the National Academy of 

Sciences: 

The salient social fact today about the use of ability tests is that 
blacks, Hispanics, and native Americans do not, as groups, score 

as well as do white applicants as a group. When candidates are 
ranked according to test score and when test results are a 

  

1 The author presented this paper in December 1986, on behalf of the Lawyers’ 
Committee for Civil Rights Under Law and the NAACP Legal Defense and Educational 
Fund, Inc., to the Planning Conference for the National Commission on Testing and 
Public Policy. Subsequent developments in the law are noted at the conclusion of the 
paper. The author wishes to acknowledge the valuable advice and assistance of Richard 

T. Seymour of the Lawyers’ Committee and Barry L. Goldstein of the NAACP Legal 

Defense Fund. 

2 See A. Anastasi, Psychological Testing 343-45 (Sth ed. 1982); L. Cronbach, Essentials 

of Psychological Testing 383 (4th ed. 1984). 

3H 

4 Sce, for example, Haney, “Employment Tests and Employment Discrimination: A 

Dissenting Psychological Opinion,” 5 Indus. Rel. L.]. 1, 26-27 and n.131 (1982); unter 

and Schmidt, “Ability Tests: Economic Benefits Versus the Issue of Fairness,” 21 Indus. 

Rel. 293, 294 (1982); Lerner, “Employment Discrimination: Adverse Impact, Validity, 

and Equality,” 1979 Sup. Ct. Rev. 17, 41-42. 

 



Patrick O. Patterson 

  

determinant in the employment decision, a comparatively large 
fraction of blacks and Hispanics are screened out. . . . 

So long as the[se] groups . . . continue to have a relatively high 
proportion of less educated and more disadvantaged members 
than the general population, those social facts are likely to be 
reflected in test scores. That is, even highly valid tests will have 
adverse impact.’ 

With the enactment of Title VII of the Civil Rights Act of 1964, the law 

and lawyers began to recognize this serious social problem as a civil rights 

problem, and they began to develop legal strategies and principles to solve it.” 

This paper traces the development of the law of employment testing under 

Title VII, and it outlines some of the important legal issues that remain 
undecided. 

II. THE DEVELOPMENT OF TITLE VII TESTING LAW 

Title VII of the Civil Rights Act of 1964, as amended, prohibits discrimination in 

public and private employment on the basis of race, color, religion, sex, and 

national origin. 8 In the more than twenty years that have passed since the 

enactment of Title VII, courts and enforcement agencies have resolved 

thousands of claims asserting that the use of tests or other employee selection 

procedures has resulted in unlawful discrimination. In the process of resolving 

these claims, the courts and agencies have developed—and are continuing to 

develop—legal principles under Title VII governing the use of tests and other 

selection procedures for hiring, promotion, and other employment decisions. A 

complete discussion of the historical development of those principles is 

  

> Committee on Ability Testing, National Academy of Sciences/National Research 
Council, Ability Testing: Uses, Consequences, and Controversies 143, 146 (1982) 

(“NAS/NRC Report on Ability Testing”). Cf. Brunet v. City of Columbus, 41 E.P.D. para. 

36,498 (S.D. Ohio 1986) (physical test and mechanical reasoning test administered to 
firefighter applicants had an adverse impact on women); Burney v. City of Pawtucket, 

559 F. Supp. 1089 (D. R.I. 1983) (physical agility requirements and police academy 

physical training program had an adverse impact on female police officer applicants); 
Berkman v. City of New York, 536 F. Supp. 177 (E.D. N.Y. 1982), aff'd, 705 F.2d 584 (2d Cir. 

1983) (physical test administered to firefighter applicants had an adverse impact on 
women). 

6 42 US.C §§2000e et seq. 

7 See Cooper and Sobol, “Seniority and Testing Under Fair Employment Laws: A 

General Approach to Objective Criteria of Hiring and Promotion,” 82 Haro. L. Rev. 1598, 

1638-41 (1969); Note, “Legal Implications of the Use of Standardized Ability Tests in 

Employment and Education,” 68 Colum. L. Rev. 691, 692-95 (1968). 

8 42U.s. C.§§2000e-2(a)-(d) 

    

  
 



  
  

Employment Testing and Title VII 85 

beyond the scope of this paper.” However, some knowledge of how Title VII 

testing law has developed will be helpful in understanding what it has become. 

Statutory Language and Legislative History 

The doctrinal development of Title VII testing law has focused on two 

subsections of the statute: §703(a), which prohibits “discrimination;”10 and 

§703(h), which provides an exception for “professionally developed ability 

tests” that are not “designed, intended, or used to discriminate.” 11 

The language of §703(a) originated as an amendment by the House 

Judiciary Committee to H.R. 7152,12 the Kennedy Administration’s omnibus 

1963 civil rights bill.13 The language of this section today is essentially the 

samel? as the language reported by the House Judiciary Committee in 19631° 

  

9 For other discussions of the development of legal principles governing 
employment testing under Title VII, see Gold, “Griggs” Folly: An Essay on the Theory, 
Problems, and Origin of the Adverse Impact Definition of Employment Discrimination 
and a Recommendation for Reform,” 7 Indus. Rel. L.]. 429 (1985); Haney, “Employment 

Tests and Employment Discrimination: A Dissenting Psychological Opinion,” 5 Indus. 
Rel. L.]. 1 (1982); Booth and Mackay, “Legal Constraints on Employment Testing,” 29 
Emory L.J]. 121 (1980); Note, “The Uniform Guidelines on Employee Selection 

Procedures: Compromises and Controversies,” 28 Cath. U. L. Rev. 605 (1979). Sce also B. 

Schlei and P. Grossman, Employment Discrimination Law 80-205 (1983). 

10 section 703(a)(1) states it is unlawful for an employer “to fail or refuse to hire. . . 
any individual, or otherwise to discriminate against any individual with respect to his 
compensation, terms, conditions, or privileges of employment, because of such 

individual's race, color, religion, sex, or national origin.” 42 U.S.C. §2000¢-2(a)(1). 

Section 703(a)(2) states that it is unlawful for an employer “to limit, segregate, or classify 

his employees or applicants for employment in any way which would deprive or tend 

to deprive any individual of employment opportunities or otherwise adversely affect 

his status as an employee, because of such individual's race, color, religion, sex, or 

national origin.” 42 U.S.C. §2000¢-2(a)(2). Other provisions of Title VII use similar 

language to describe unlawful practices of labor organizations and employment 
agencies. See 42 U.S.C. §§2000e-2(b), 2(c), and 2(d). 

11 section 703(h) states that, “[nJotwithstanding any other provision of this [title], 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 dis 

criminate because of race, color, religion, sex, or national origin.” 42 U.S.C. §2000c-2(h). 

12. see H.R. Rep. No. 570, 88th Cong,, 1st Sess. (1963). 

13 see Vaas, “Title VIL: Legislative History,” 7 B.C. Indus. and Com. L. Rev. 431, 433 
(1966). 

14 The language prohibiting “discrimination” was amended on the louse floor by the 

addition of “sex” as a protected class. 110 Cong. Rec. 2577-84, 2718, 2720-21 (1964). Scc 

Vaas, “Title VII: Legislative History,” 7 B.C. Indus. and Com. L. Rev. 431, 439-42 (1966). 

Section 703(a) was later amended by the Equal Employment Opportunity Act of 1972, 

 



  

Patrick O. Patterson 

and passed by the House in 1964.16 The House bill, however, did not contain 

§703(h) or any other provision that specifically addressed testing. That 

provision was added by the “Tower Amendment” in the Senate.1” 

In the first years after enactment of Title VII in 1964, the congressional 

purpose underlying these sections became a subject of debate in the courts 

and law journals. Everyone agreed that the statute prohibited the intentionally 

discriminatory use of employment tests, but most courts!8 and writers!? 

rejected the view that Title VII imposed any requirement that tests be related 

to the jobs for which they were used. Some courts20 and writers,2! on the other 

hand, had begun to recognize that many tests had an adverse impact on blacks 

and other minorities, and they argued that Title VII prohibited the use of tests 

that were not sufficiently job related. Bits and pieces of evidence bearing on 

the elusive question of legislative purpose were marshalled on both sides of 

this debate.’ Although there was some evidence that Congress had meant to 

  

Pub. L. No. 92-261, expressly to prohibit discrimination against “applicants for 

employment” as well as discrimination against “employees.” See H.R. Rep. No. 92-238, 

92d Cong, 1st Sess. 30 (1971); S. Rep. No. 92-415, 92d Cong, 1st Sess. 43 (1971). Congress 

regarded this amendment as “declaratory of existing law.” Id. at 43. See also 

Subcommittee on Labor, House Committee on Labor and Public Welfare, Legislative 

History of the Equal Employment Opportunity Act of 1972 1849 (1972) (section-by-section 

analysis of H.R. 1746 as reported by the Conference Committee citing, inter alia, Phillips 

v. Martin-Marietta Corp., 400 U.S. 542 [1971]). The language of §703(a) otherwise 

remains the same today as when it was first enacted in 1964. 

15 See H.R. Rep. No. 914, 88th Cong., 1st Sess. 10 (1963). 

16 See 110 Cong. Rec. 2804 (1964). 

17 110 Cong. Rec. 13492-505, 13274 (1964). See Vaas, supra note 14, at 449. 

18 gee, for example, Griggs v. Duke Power Co., 292 F. Supp. 243, 250 M.D. N.C. 1968), 

aff'd in part and rev'd in part, 420 F.2d 1225 (4th Cir. 1970), rev'd, 401 U.S. 424 (1971). 

19 See, for example, M. Sovern, Legal Restraints on Racial Discrimination in 

I'mployment 73 (1966); Rachlin, “Title VII: Limitations and Qualifications,” 7 B.C. Indus. 

and Com. L. Rev. 473, 486-90 (1966). 

20 Gee United States v. H.K. Porter Co., 296 F. Supp. 40, 78 (N.D. Ala. 1968); Dobbins v. 

Electrical Workers Local 212, 292 F. Supp. 413, 433-34, 439 (S.D. Ohio 1968). 

21 See, for example, Cooper and Sobol, “Seniority and Testing Under Fair 

Employment Laws: A General Approach to Objective Criteria of Hiring and 

Promotion,” 82 Harv. L. Rev. 1598, 1649-54 (1969); Note, “Legal Implications of the Use 

of Standardized Ability Tests in Employment and Education,” 68 Colum. L. Rev. 691, 

706-11 (1968). 

    
 



  

Employment Testing and Title VII 87 

enact a job-relatedness standard,22 many statements in the legislative record 

seemed to suggest the opposite conclusion.23 

1966 EEOC Guidelines 

In 1966, the Equal Employment Opportunity Commission (EEOC)—the agency 

created by Congress to enforce Title V1I124—reviewed the legislative history and 

decided to adopt the job-relatedness standard in its first “Guidelines on 

Employment Testing Procedures.”2> These Guidelines stated in part that 

[tlhe Commission . . . interprets “professionally developed ability 
test” to mean a test which fairly measures the knowledge or skills 
required by the particular job or class of jobs which the applicant 
sceks, or which fairly affords the employer a chance to measure 
the applicants’ ability to perform a particular job or class of jobs. 
The fact that a test was prepared by an individual or organization 
claiming expertise in test preparation does not, without more, 

justify its use within the meaning of Title VIIL.26 

The 1966 EEOC Guidelines, like the early court decisions suggesting a job- 

relatedness standard, did not distinguish between those tests that did and 

those that did not have an adverse impact on minorities. Through its 

experience in investigating and attempting to conciliate complaints of 

employment discrimination, the EEOC “began to notice a pattern of test uscage 

[sic] which was excluding minorities without serving any job related purpose or 

  

22 gee supra note 21. 

23 see Gold, “Griggs’ Folly: An Essay on the Theory, Problems, and Origin of the 
Adverse Impact Definition of Employment Discrimination and a Recommendation for 

Reform,” 7 Indus. Rel. L.]. 429, 489-578 (1985). 

24 gee 42 U.S.C. §2000c-4 and 5. 

25 EEOC Guidelines on Employment Testing Procedures, reprinted in CCH Empl. Prac. 

Guide para. 16,904 (1967). 

26 14. 

 



  

Patrick O. Patterson 

business need. . . .”27 When the EEOC responded by adopting the 1966 

Guidelines, it focused not on the question of whether a particular test had an 

adverse impact, but rather “on the employer's actual job needs and the 

reasonableness of the test in measuring those needs.”28 Instead of treating 

adverse impact as a preliminary inquiry that must be made with respect to a 

particular test before the question of job-relatedness would arise, the 1966 

Guidelines appeared to regard every test as suspect under Title VII and to 

require that all tests be job related 2? 
The 1966 EEOC Guidelines, based on the recommendations of “a panel of 

outstanding psychologists”30 whose report was issued along with the 

Guidelines, endorsed the Standards for Educational and Psychological Tests 

and Manuals (APA Standards), which had been adopted earlier in 1966 by the 

  

27 Robertson, A Staff Analysis of History of EEOC Guidelines on Employee Selection 

Procedures 15 (1976) (unpublished report submitted to the General Accounting Office by 
Peter C. Robertson, Director, Federal Liaison, Equal Employment Opportunity 

Commission, August 1976) (on file with the author). This report further states that: 

Very early in its operations EEOC discovered that employers were 

utilizing discriminatory employment testing which had proved to be a 

major barrier to minority advancement. Written findings of probable 
cause that these tests were discriminatory formed the basis of early 

conciliation attempts, but employers who disagreed with EEOC’s 

perception of discrimination refused to conciliate. An ad hoc approach 
left the impression that individual conciliators were taking a personal 
position that . . . the tests were discriminatory and had to be eliminated. 

This was replaced with a formal institutionalized position in which the 
policymakers defined agency policy through Commission Guidelines. . . . 

When EEOC began to notice a pattern of test useage [sic] which was 

excluding minorities without serving any job related purpose or business 
need[,] it consulted with experts in the design and study of mechanisms 
to identify employee capability. Specifically, EEOC asked a panel of 

psychologists to advise it with respect to issues relating to the 
development, introduction and administration of test of aptitude and/or 

ability in industrial settings as related to problems of employment 

discrimination. . . . 

Id. at 14-15 (footnotes omitted). See also Blumrosen, “Strangers in Paradise: Griggs 

v. Duke Power Co. and the Concept of Employment Discrimination,” 71 Mich. L. Rev. 

59, 59-60 (1972) (EEOC officials “knew that many companies had introduced tests in the 

1950s and early 1960s when they could no longer legally restrict opportunities of blacks 

and other minority workers and that the tests had proved to be major barriers to 

minority advancement”). 

28 Cooper and Sobol, supra note 22, at 1654. 

29 1d. But cf. Cooper and Sobol supra note 22 at 1664-65. 

30. 1 

    
 



  

Employment Testing and Title VII 89 

American Psychological Association and other professional organizations 

concerned with testing standards and practices 3! The EEOC offered not only its 

Guidelines but also the APA Standards to employers as “a scientifically sound, 

industrially-proven, and equitable basis for matching manpower requirements 

with human aptitudes and abilitics.”32 Over the next twenty years, the 

professional standards of psychologists—including subsequent versions of the 

APA Standards33 as well as the Principles for the Validation and Use of 

Personnel Selection Procedures (“Division 14 Principles”) adopted by the APA's 

Division of Industrial-Organizational Psychology (Division 14)34—have 

remained influential in shaping Title VII testing law and agency guidelines. 

1970 EEOC Guidelines 

In 1970, the EEOC issued new guidelines that were described by one of their 

drafters as “a more detailed version of the [1966] guidelines. . . 35 In fact, the 

1970 Guidelines3® departed from the 1966 Guidelines in a number of significant 

respects, including the adoption of far more specific and technically complex 

validation standards.3” The most fundamental change in the 1970 Guidelines, 

however, was the EEOC’s explicit adoption of the concept of adverse impact as 

part of the inquiry to be made in determining compliance with Title VII. 

  

31 American Psychological Association, American Educational Research Association, 
and National Council on Measurement in Education, Standards for Educational and 
Psychological Tests and Manuals (1966). 

32 1966 Guidelines. In 1968 and 1969, respectively, the Department of Labor's Office of 

Federal Contract Compliance, see 33 Fed. Reg. 11392 (1968), and the Civil Service 

Commission, see FPB Supp. 335.1 (1969), issued their own testing guidelines. 

33 American Psychological Association, American Educational Research Association, 

National Council on Measurement in Education, Standards for Educational and 

Psychological Tests (1974); American Educational Research Association, American 
Psychological Association, National Council on Measurement in Education, Standards 

for Educational and Psychological Testing (1985). (Known as APA Standards.) 

34 Division of Industrial-Organizational Psychology, American Psychological 
Association, Principles for the Validation and Use of Personnel Selection Procedures 

(1975); Division of Industrial-Organizational Psychology, American Psychological 

Association, Principles for the Validation and Use of Personnel Selection Procedures: 
Second Edition (1980) (“Division 14 Principles”). 

35 Blumrosen, supra note 27, at 60 n.5. 

36 EEOC Guidelines on Employee Selection Procedures, 35 Fed. Reg. 12333 (1970), 

codified at 29 C.F.R. §§1607.1 et seq. (1970). 

37 1d. §§1607.4-1607.8. The 1970 Guidelines continued to rely on the APA Standards. 
Id., §1607.5. 

 



Patrick O. Patterson 

  

According to the 1970 Guidelines, the Commission had detected a 
decided increase in total test usage and a marked increase in 
doubtful testing practices which, based on our experience, tend to 
have discriminatory effects. In many cases, persons have come to 
rely almost exclusively on tests as the basis for making the 
decision to hire, transfer, [or] promote . . . , with the result that 

candidates are selected or rejected on the basis of a single test 
score. Where tests are so used, minority candidates frequently 
experience disproportionately high rates of rejection by failing to 
attain score levels that have been established as minimum 
standards for qualification. 

It has also become clear that in many instances persons are using 
tests as the basis for employment decisions without evidence that 
they are valid predictors of employee job performance. . . . A test 
lacking demonstrated validity (i.e., having no known significant 
relationship to job behavior) and yielding lower scores for classes 
protected by Title VII may result in the rejection of many who 

have necessary qualifications for successful work performance. 38 

In keeping with these findings, the 1970 Guidelines adopted an 

expansive definition of the term “test,” 9 and they defined “discrimination” as 

[tIhe use of any test which adversely affects hiring, promotion, 

transfer or any other employment or membership opportunity of 

classes protected by Title VII ... unless: (a) the test has been 
validated and evidences a high degree of utility . . ., and (b) the 

person giving or acting upon the results of the particular test can 

demonstrate that alternative suitable hiring, transfer or 

promotion procedures are unavailable for his use. 0 

The 1970 EEOC Guidelines thus abandoned the apparent view of the 1966 

Guidelines that Title VII required every test to be job related; they provided 

  

38 14.,§1607.1(b) 

39 « _ [Tlhe term ‘test’ is defined as any paper-and-pencil or performance measure 

used as a basis for any employment decision. . . . This definition includes, but is not 

restricted to, measures of general intelligence, mental ability and learning ability; 

specific intellectual abilities; mechanical, clerical and other aptitudes; dexterity and 

coordination; knowledge and proficiency; occupational and other interests; and 

attitudes, personality or temperament. The term ‘test’ includes all formal, scored, 

quantified or standardized techniques of assessing job suitability including, in addition 

to the above, specific qualifying or disqualifying personal history or background 

requirements, specific educational or work history requirements, scored interviews, 

biographical information blanks, interviewers’ rating scales, scored application forms, 

etc.” Id., §1607.2. 

0 pg (emphasis added). 

  

  
   



  

  

Employment Testing and Title VII 91 

instead that, at least in the absence of intentional discrimination,41 an inquiry 

into job-relatedness or validation would be appropriate only after it was 

determined that the use of a test had an adverse impact on a protected 

group.#2 

Griggs v. Duke Power Company 

The 1970 EEOC Guidelines did not put an end to the debate over the proper 

application of Title VII to employment tests. In 1971, the issue found its way to 

the Supreme Court in the case of Griggs v. Duke Power Co. 43 which became 

not only the most important testing case ever decided under Title VII, but “the 

most important court decision in employment discrimination law.”44 

The employer in Griggs had required applicants for all jobs in its higher 

paying departments (including the Coal Handling, Operations, and 

Maintenance Departments) to have a high school diploma and to receive 

satisfactory scores on two standardized aptitude tests—the Wonderlic 

Personnel Test and the Bennett Mechanical Comprehension Test. Although 

the employer had not adopted these requirements for the purpose of 

discriminating against black applicants, both requirements had the effect of 

“renderling] incligible a markedly disproportionate number of Negroes . . . rhs 

and neither requirement was “shown to bear a demonstrable relationship to 

successful performance of the jobs for which it was used.”46 Indeed, the record 

showed that the requirements had been adopted “without meaningful study of 

their relationship to job-performance ability. Rather, a vice president of the 

company testified, the requircments were instituted on the company’s 

judgment that they generally would improve the overall quality of the work 

force.”47 

  

41 gee Id., §1607.11 (“Disparate Treatment”). 

42 In 1971, the Office of Federal Contract Compliance adopted additional testing 

guidelines which endorsed an adverse impact interpretation of Executive Order 11246 

See 36 Fed. Reg. 19307 (1971). In 1972, the Civil Service Commission adopted new 

testing guidelines which, while stating a policy of nondiscrimination, did not mention 

adverse impact analysis. See 37 Fed. Reg. 12984 (1972). 

43 401 US. 424 (1971). 

44 B schlei and P. Grossman, supra note 9, at 5. 

45 401 US. at 429. 

46 14 at 431. 

47 1a. 

 



  

Patrick O. Patterson 

In a sweeping opinion authored by Chief Justice Burger, the Supreme 

Court unanimously endorsed the adverse impact interpretation of Title VII 

that had been adopted by the EEOC: 

Congress has now provided that tests or criteria for employment 
or promotion may not provide equality of opportunity only in the 
sense of the fabled offer of milk to the stork and the fox. On the 
contrary, Congress has now required that the posture and 
condition of the job secker be taken into account. It has—to resort 
again to the fable—provided that the vessel in which the milk is 
proffered be one all seekers can use. The Act proscribes not only 
overt discrimination but also practices that are fair in form, but 

discriminatory in operation. The touchstone is business necessity. 
If an employment practice which operates to exclude Negroes 
cannot be shown to be related to job performance, the practice is 

prohibited. 4? 

* * * 

The facts of this case demonstrate the inadequacy of broad and 
general testing devices as well as the infirmity of using diplomas 
or degrees as fixed measures of capability. History is filled with 
examples of men and women who rendered highly effective 
performance without the conventional badges of accomplish- 
ment in terms of certificates, diplomas, or degrees. Diplomas and 

tests are useful servants, but Congress has mandated the 

common-sense proposition that they are not to become the 

masters of reality.>0 

The Court in Griggs found that the 1966 EEOC Guidelines—as 

“elaborated” in the 1970 Guidelines®l—were “entitled to great deference” by 

the courts: “Since the Act and its legislative history support the Commission's 

construction, this affords good reason to treat the Guidelines as expressing the 

will of Congress.”>2 Thus, the Court joined the EEOC in holding that, under Title 

VII, an employer may not use a test that has an adverse impact on a protected 

group unless the employer can prove that the test “bear{s] a demonstrable 

relationship to successful performance of the jobs for which it [is] used.”>3 

  

48 

49 

50 

51 

52 

53 

See B. Woodward and S. Armstrong, The Brethren 122-23 (1979). 

401 USS. at 431. 

Id. at 433. 

Id. at 434 n.9. 

Id. at 434. 

Id. at 431. 

    

 



  

Employment Testing and Title VII 93 

Echoing the principles underlying the Guidelines, the Griggs opinion 

repeatedly emphasized that Title VII, in secking “to achieve equality of 

employment opportunities and remove barriers that have operated in the past 

to favor an identifiable group of white employees over other employees,”>¥ was 

not limited to a prohibition of intentional discrimination:>> 

[Glood intent or absence of discriminatory intent does not 
redeem employment procedures or testing mechanisms that 
operate as “built-in headwinds” for minority groups and are 
unrelated to measuring job capability. . . . 

Congress directed the thrust of the Act to the consequences of 
employment practices, not simply the motivation. More than that, 

Congress placed on the employer the burden of showing that any 
given requirement must have a manifest relationship to the 

employment in question.%6 

The Equal Employment Opportunity Act of 1972 

Some have argued that the Griggs decision misread the language and purpose 

of Title VII; the legislative history available to the Court at the time of the 

Griggs decision, according to this argument, showed that Congress meant to 

prohibit only intentional discrimination, not to regulate tests or other practices 

that had an adverse but unintended impact on minorities or women.” As 

noted above, the legislative evidence as to what Congress intended in 1964 was 

at best inconclusive.”® Soon after the Court decided Griggs, however, Congress 

disposed of any lingering doubts by enacting the Equal Employment 

Opportunity Act of 1972.5 
The 1972 Act amended Title VII in a number of respects and extended 

its coverage to federal, state, and local government employment. The House 

and Senate committee reports on this legislation expressly recognized and 

approved the interpretation of Title VII that had been developed by the EEOC 

  

54 14. at 429-30. 

55 The Supreme Court has suggested, but has never decided, that the adverse impact 

standard of Griggs might be limited to claims under §703(a)(2), and that §703(a)(1) might 
prohibit only intentional discrimination. See General Electric Co. v. Gilbert, 429 US. 

125, 137 and n.13 (1979). 

56 1d. at432 (emphasis in original). 

57 See Gold, supra note 23, at 489-578; Lyons, “An Agency with a Mind of Its Own: 
The EEOC’s Guidelines on Employment Testing,” 17 New Perspectives 20 (1985). 

58 see supra notes 18-23. 

59 pub. L. No. 92-261, 86 Stat. 103 (1972). 

 



  

Patrick O. Patterson 

in its guidelines and had been endorsed by the Supreme Court in Griggs. The 

repert of the House Committee on Education and Labor stated in part: 

Employment discrimination, as we know today, is a far more 
complex and pervasive phenomenon [than previously believed]. 
Experts familiar with the subject generally describe the problem 
in terms of “systems” and “effects” rather than simply intentional 
wrongs. The literature on the subject is replete with discussions of 
the mechanics of seniority and lines of progression, perpetuation 

of the present effects of earlier discriminatory practices through 
various institutional devices, and testing and validation 
requirements. . . . A recent striking example was provided by the 
U.S. Supreme Court in its decision in Griggs v. Duke Power Co. . . . 
where the Court held that the use of employment tests as 
determinants of an applicant's job qualification, even when 

nondiscriminatory and applied in good faith by the employer, was 
in violation of Title VII if such tests work a discriminatory effect in 
hiring patterns and there is no showing of an overriding business 

necessity for the use of such criteria 60 

The report of the Senate Committee on Labor and Public Welfare 

included similar language,®! and additionally directed the federal Civil Service 

Commission 

to develop more expertise in recognizing and isolating the various 
forms of discrimination which exist in the system it administers. . . 
The Commission should not assume that employment 
discrimination in the Federal Government is solely a matter of 
malicious intent on the part of individuals. It apparently has not 
fully recognized that the general rules and procedures that it has 
promulgated may in themselves constitute systemic barriers to 
minorities and women. Civil Service selection and promotion 
techniques and requirements are replete with artificial 
requirements that place a premium on “paper” credentials. 
Similar requirements in the private sectors of business have often 
proven of questionable value in predicting job performance and 

have often resulted in perpetuating existing patterns of 
discrimination (see e.g. Griggs v. Duke Power Co.). The inevitable 
consequence of this kind of a technique in Federal employment, 

as it has been in the private sector, is that classes of persons who 

are socio-economically or educationally disadvantaged suffer a 

heavy burden in trying to meet such artificial qualifications. 

  

60 

61 

H.R. Rep. No. 92-238, 92d Cong,, 1st Sess. 8 (1971) (footnote and citations omitted). 

See S. Rep. No. 92-415, 92d Cong,, 1st Sess. 5 (1971). 

    

 



  

Employment Testing and Title VII 95 

It is in these and other areas where discrimination is institutional, 

rather than merely a matter of bad faith, that corrective measures 

appear to be urgently required. For example, the Committee 
expects the Civil Service Commission to undertake a thorough re- 

examination of its entire testing and qualification program to 
ensure that the standards enunciated in the Griggs case are fully 

met.62 

Thus, in enacting the Equal Employment Opportunity Act of 1972, 

Congress unequivocally ratified the interpretation of Title VII that had been 

adopted by the EEOC in its guidelines and by the Supreme Court in Griggs. 

Albemarle Paper Company v. Moody 

The carly guidelines, the Griggs opinion, and the 1972 statutory amendment 

sketched in the broad outlines of Title VII testing law. The Supreme Court's 

1975 decision in Albemarle Paper Co. v. Moody®3 provided some of the finer 

details. 

The Court in Albemarle began by reaffirming its holding in Griggs: 

“Title VII forbids the usc of employment tests that are discriminatory in cffect 

unless the employer meets ‘the burden of showing that any given requirement 

[has] . . . a manifest relation to the employment in question.’”64 The Court 

noted that the employer's burden of justification arises only after the 

complaining party “has shown that the tests in question select applicants for 

hire or promotion in a racial pattern significantly different from that of the pool 

of applicants.”6° The opinion went on to state that, even if the employer meets 

its burden of showing that its tests are job related, the complaining party may 

still prove a violation of Title VII by showing “that other tests or selection 

devices, without a similarly undesirable racial effect, would also serve the 

employer's legitimate interest in ‘efficient and trustworthy workmanship. ”60 

Like the employer in Griggs, the employer in Albemarle required 

applicants for jobs in higher paying lines of progression in an industrial plant to 

obtain minimum scores on two standardized “general ability tests”67 the 

Beta Examination and the Wonderlic Test. There was little or no dispute that 

  

62 14 at 14-15. 

63 422 US. 405 (1975). 

64 Id. at 425, quoting Griggs v. Duke Power Co., 401 U.S. 424, 432 (1971). 

65 Ja. 

66 14, quoting McDonnell Douglas Corp. v. Green, 411 U.S. 792, 801 (1973). 

67 1d.ar427. 

 



Patrick O. Patterson 

  

the use of these tests had an adverse impact on blacks.®3 The Court's opinion 
therefore focused primarily on the issue of job-relatedness. Restating the view 
expressed in Griggs that the EEOC Guidelines were entitled to great deference, 
the Court in Albemarle concluded that 

[tlhe message of these [1970] Guidelines is the same as that of the 
Griggs case—that discriminatory tests are impermissible unless 
shown, by professionally acceptable methods, to be “predictive of 
or significantly correlated with important elements of work 
behavior which comprise or are relevant to the job or jobs for 

which candidates are being evaluated.”69 

Drawing upon those Guidelines and the 1966 and 1974 APA Standards, 

and expressing its skepticism about a validation study conducted by a party to 
litigation on the eve of trial,”? the Court in Albemarle set forth a relatively 
detailed and technical critique of several aspects of the employer's validation 
evidence.” 

First, the Court criticized the “odd patchwork of results” demonstrated 

by the employer's concurrent criterion-related validity study. The study found 
significant correlations between test scores and job performance for some jobs 
but not for others; since there was no analysis of the jobs involved, the Court 

found “no basis for concluding that ‘no significant differences’ exist among the 

lines of progression, or among distinct job groupings within the studied lines of 

progression. Indeed, the study's checkered results appear to compel the 

  

68 Until 1964, the company had a formal policy of racially segregating the lines of 
progression. By 1971, a study of 105 incumbent employees in relatively high ranking 
jobs included 101 whites and 4 blacks. 422 U.S. at 429 n.25. 

69 422 U.S. at 431, quoting 1970 EEOC Guidelines, 29 C.F.R. §1607.4(c). 

70 “It cannot escape notice that Albemarle’s study was conducted by plant officials, 
without neutral, on-the-scene oversight, at a time when this litigation was about to 

come to trial. Studies so closely controlled by an interested party in litigation must be 
examined with great care.” 422 US. at 433 n.32. 

71 Chief Justice Burger dissented from this part of the Court’s opinion on the ground 

that the majority’s analysis was “based upon a wooden application of the EEOC 

Guidelines.” 422 U.S. at 451. The Chief Justice complained that, unlike the 1966 EEOC 

Guidelines approved in Griggs, the 1970 Guidelines “relate[d] to methods for proving job 

relatedness; they interpret no section of Title VII and are nowhere referred to in its 

legislative history. Moreover, they are not federal regulations which have been 
submitted to public comment and scrutiny as required by the Admininstrative 

Procedure Act. Thus, slavish adherence to the EEOC Guidelines regarding test 

validation should not be required.” Id. at 452 (emphasis in original; footnote omitted). 

    

  

 



  

  

Employment Testing and Title VII 97 

opposite conclusion.””2 Second, noting the possibility of bias,”3 the Court 

disapproved the study’s use of vague and subjective supervisory ratings as 

performance measures.’ Third, the Court found fault with the study's focus 

on job groups near the top of the lines of progression, rather than on the entry 

level jobs. The Court endorsed the view of the 1970 EEOC Guidelines that 

performance measures should be based on higher level jobs only where the 

employer can show that “new employees will probably, within a reasonable 

period of time and in a great majority of cases, progress to a higher level 70 

Finally, the Court noted that the study “dealt only with job-experienced, white 

workers; but the tests themselves are given to new job applicants, who are 

younger, largely inexperienced, and in many instances nonwhite.””6 Relying 

on both the APA Standards and the EEOC Guidelines, the Court indicated that 

validation studies should focus on persons similar to those with whom the tests 

are used, and that differential validity studies should be done on minority and 

nonminority groups wherever technically feasible.” 

Although the Supreme Court has reaffirmed and applied the basic 

principles of Griggs and Albemarle in a number of subsequent cases,’ the 

opinion in Albemarle remains the Court’s most detailed discussion to date of 

Title VII's job-relatedness requirement. 

Uniform Guidelines on Employee Selection Procedures 

By the time Congress amended Title VII in 1972, the EEOC, the Civil Scrvice 

Commission, and the Department of Labor each had its own separate set of 

employment testing guidelines.”? The existence of potentially conflicting 

guidelines had by then become a matter of concern to enforcement agencies, 

employers, unions, civil rights groups, and others. In response to such 

  

72 422 USS. at 432, quoting 1970 EEOC Guidelines, 29 C.F.R. §1607.4(c)(2). 

73 Id. at 432-33 and n.30, quoting 1970 EEOC Guidelines, 29 C.F.R. §§1607.5(b)(3) and (4). 

74 Supervisors were told to “determine which ones [employees] they felt irrespective 

of the job that they were actually doing, but in their respective jobs, did a better job than 

the person they were rating against.” 422 U.S. at 433. 

75 Id. at 434, quoting 1970 EEOC Guidelines, 29 C.F.R. §1607.4(c)(1). 

76 14. at 435. 

77 Id. at 435, quoting 1966 APA Standards, para. C5.4, and 1970 EEOC Guidelines, 29 

C.F.R. §1607.5(b). 

78 See International Brotherhood of Teamsters v. United States, 431 U.S. 324 (1977); 

Dothard v. Rawlinson, 433 U.S. 321 (1977); New York City Transit Authority v. Beazer, 

440 U.S. 568 (1979); Connecticut v. Teal, 457 U.S. 440 (1982). 

79 See supra notes 32 and 42. 

 



Patrick O. Patterson 

  

concerns, in 1972 Congress established the Equal Employment Opportunity 

Coordinating Council and charged it with responsibility for developing and 

implementing uniform enforcement policies.30 The agencies thereafter 

intensified efforts they had previously undertaken to develop a single set of 

uniform guidelines on employee selection procedures, but they were unable to 

accomplish this goal. In 1976, after several years of unsuccessful efforts, three 

of these agencies—the Department of Justice, the Department of Labor, and 

the Civil Service Commission—reached agreement and issued the Federal 

Executive Agency Guidelines on Employee Selection Procedures (“FEA 

Guidelines”).81 The EEOC, disagreeing with some provisions of the “FEA 

Guidelines,” then reissued its own 1970 Guidelines.82 

Despite substantial discord among the agencies on a number of issues 

during this period, there was complete agreement as to the correctness and 

applicability of the basic principles enunciated by all the agencies in their 

previous guidelines, approved by the Supreme Court in Griggs and Albemarle, 

and reaffirmed by Congress in the Equal Employment Opportunity Act of 1972. 

The agencies agreed that Title VII prohibits selection procedures that have an 

adverse impact unless those procedures are shown to be job related; they 

agreed that statistical evidence should be maintained by employers and 

should be used to determine adverse impact; and they agreed that job- 

relatedness should be shown by professionally acceptable methods. The 

disagreements among the agencies did not concern these fundamental 

principles, but centered instead on the details of their implementation—that 

is, the types of statistical analyses that should be used in assessing adverse 

impact, and the technical details of the validity studies neccessary to 

demonstrate job-relatedness.83 

More than a year later, after further debate and negotiation, all the 

agencies finally reached agreement and jointly published a proposed draft of 

the Uniform Guidelines on Employee Selection Procedures in December 

1977.84 The agencies also published a notice of proposed rulemaking, solicited 

  

80 See 42 U.S.C. §2000-14. 

81 41 Fed. Reg. 51737 (1976). 

82 41 Fed. Reg. 51984 (1976). 

83 see Booth and Mackay, supra note 9, at 124-40; Note, The Uniform Guidelines on 

Employee Selection Procedures: Compromises and Controversies, supra note 9, 

at 607-10. 

84 42 Fed. Reg. 65542 (1977). Before publishing this proposed draft, the enforcement 

agencies circulated an earlier draft and obtained comments from representatives of state 
and local governments, psychologists, private employers, and civil rights groups. See 

“Uniform Guidelines, Supplementary Information: Analysis of Comments,” 43 Fed. 

Reg. 38292 (1978). 

  

   



  

Employment Testing and Title VII 99 

written comments, and held a public hearing and meeting at which testimony 

was given by representatives of private industry, state and local governments, 

labor organizations, civil rights groups, and psychologists.8° After considering 

the written comments submitted by more than two hundred organizations and 

individuals, the testimony elicited at the public hearing and meeting, and the 

views expressed in informal consultations,836 the EEOC, the Civil Service 

Commission (now the Office of Personnel Management), and the 

Departments of Justice, Labor, and the Treasury revised the proposed draft 

and adopted the Uniform Guidelines in September 1978.87 

The Uniform Guidelines state that they “are designed to provide a 

framework for determining the proper use of tests and other selection 

procedures.” 88 They are “built upon court decisions, the previously issued 

guidelines of the agencies, and the practical experience of the agencies, as well 

as the standards of the psychological profession,” and they are “intended to be 

consistent with existing law.”89 

Section 3 of the Uniform Guidelines restates the Griggs-Albemarle 

interpretation of Title VII, which underlies all the remaining provisions: “The 

use of any sclection procedure which has an adverse impact . . . will be 

considered to be discriminatory and inconsistent with these guidelines, unless 

the procedure has been validated.”90 Section 4 contains record-keeping 

provisions? and methods for determining whether a selection procedure has 

an adverse impact.92 Under section 6, employers “may choose to utilize 

  

85 See “Notice of Proposed Rulemaking,” 42 Fed. Reg. 65542 (1977); “Notice of Issues 

of Particular Interest for Public Hearing and Meeting,” 43 Fed. Reg. 11812 (1978); 

“Uniform Guidelines, Supplementary Information: Analysis of Comments,” 43 Fed. 

Reg. 38292-93 (1978). 

86 sce “Uniform Guidelines, Supplementary Information: Analysis of Comments,” 

43 Fed. Reg. 38292-93 (1978). 

87 sce 43 Fed. Reg. 38312 (1978) (EEOC); 43 Fed. Reg. 38310 (1978) (Civil Service 
Commission [Office of Personnel Management]); 43 Fed. Reg. 38311 (1978) (Department 

of Justice); 43 Fed. Reg. 38314 (1978) (Department of Labor); 43 Fed. Reg. 38309 (1978) 

(Department of the Treasury). The Uniform Guidelines are codified in 29 C.F.R. §1607 
(EEOC); 5 C.F.R. §300.103(c) (Civil Service Commission [Office of Personnel 

Management]); 28 C.F.R. §50.14 (Department of Justice); 41 C.F.R. §60-3 (Department of 

Labor). This paper will use the EEOC codification in 29 C.F.R. §1607 for citations to the 

Uniform Guidelines. 

88 29 C.F.R. §1607.1B. 

89 29 CFR.§1607.1C. 

90 29 CFR. §1607.3A. 

91 29 C.F.R. §§1607.4A-B. 

92 29 CFR. §§1607.4C-E. 

 



  

Patrick O. Patterson 

alternative selection procedures in order to eliminate adverse impact.”3 Thus, 
when an employer finds that its selection procedures have an adverse impact, 
it has two options under the Uniform Guidelines: Either validate the 
procedures, or eliminate the impact.’ An employer choosing to validate its 
procedures will find general standards for criterion-related, content, and 
construct validity studies in section 5.9° More detailed technical standards for 
validity studies are set forth in section 15.96 Additionally, under section 3B a 
validity study “should include . . . an investigation of suitable alternative 
selection procedures and suitable alternative methods of using the selection 
procedure which have as little adverse impact as possible.”%7 

The Uniform Guidelines have remained in effect since their adoption in 
1978.98 Several provisions of the Guidelines, however, have been challenged 
on grounds that they are inconsistent with Title VII case law or incompatible 
with accepted professional standards and practices. Some of those provisions 
are discussed in the remainder of this paper. 

III. CURRENT ISSUES IN TITLE VII TESTING LAW 

The purpose of this part of the paper is to outline some of the important legal 
issues that are currently before courts and administrative agencies. The 
discussion will be suggestive, not exhaustive. 

The legal principles that have developed and are developing under Title 
VII continue to generate a good deal of controversy, much of which in recent 
years has focused on the provisions of the Uniform Guidelines. Section 3A of 
those Guidelines states the fundamental principle, drawn from the Supreme 
Court's decisions in Griggs and Albemarle, that “[t]he use of any selection 
procedure which has an adverse impact . . . will be considered to be 
discriminatory . . . unless the procedure has been validated .”%? Although this 

  

93 29 CFR. §1607.6A. 

94 See Bushey v. New York State Civil Service Commission, 733 F.2d 220, 224-28 (2d 
Cir. 1984), cert. denied, 105 S. Ct. 803 (1985). 

9 29 CER. §1607.5. 

9% 29 C.F.R. §1607.15. 

97 29 C.F.R. §1607.3B. 

9% The enforcement agencies have subsequently issued “Questions and Answers to 

Clarify and Provide a Common Interpretation of the Uniform Guidelines on Employee 

Selection Procedures.” See 44 Fed. Reg. 11996 (March 2, 1979); 45 Fed. Reg. 29530 (May 2, 

1980). 

9 MCFR §1607.3A.This principle, in one form or another, has been incorporated in 
testing guidelines that have been issued at least ten times, by five different federal 
agencies, through two Democratic and two Republican administrations. The 1966 EEOC 

    

  

 



  

  

Employment Testing and Title VII 101 

principle was called into question for a time by EEOC Chairman Clarence 

Thomas, 100 it does not at present seem to be seriously disputed.101 However, 

many other questions remain concerning the content and meaning of Title VII 

testing law. As discussed below, those questions often focus on claims that 

provisions of the Uniform Guidelines are inconsistent with Title VII or 

incompatible with accepted professional standards and practices.102 

The Griggs Principle and Subjective Procedures 

For some years after the Supreme Court approved the disparate impact 

interpretation of Title VII in Griggs, many courts took an expansive view of its 

application, holding that adverse impact could be shown in a variety of ways 

  

Guidelines and the 1968 OFCC Guidelines—which appeared to require validation even 

in the absence of adverse impact—were issued during the Johnson administration. Sec 

supra notes 25-32. The 1970 EEOC Guidelines and the 1971 OFCC Guidelines were issued 

during the Nixon administration. See supra notes 35-42. The 1976 FEA Guidelines were 

issued by the Department of Justice, the Department of Labor, and the Civil Service 

Commission during the Ford administration. Sece supra note 81. The 1970 EEOC 

Guidelines were also reissued in 1976 during the Ford administration. See supra note 

82. And the 1978 Uniform Guidelines were issued by the EEOC, the Civil Service 

Commission, and the Departments of Justice, Labor, and the Treasury during the Carter 

administration. See supra notes 84-98. 

100 Mr. Thomas argued in 1985 that the Uniform Guidelines reflected a 

“fundamentally flawed approach to enforcement of the anti-discrimination statutes,” 

and announced that he therefore intended to propose major substantive revisions of 

the Guidelines. Office of Management and Budget, Regula:ory Program of the United 

States Government at 523 (Aug. 8, 1985) (Statement of Clarence Thomas). At that time, 

Mr. Thomas expressed the view that the Guidelines were “founded on the premise that 

but for unlawful discrimination, there would not be variations in the rates of hire or 

promotion of people of different races, sexes, and national origins.” Id. at 526. lle also 

stated that the Guidelines “seem[ed] to assume some inherent inferiority of blacks, 

Hispanics, other minorities, and women by suggesting that they should not be held to 

the same standards as other people, even if those standards are race- and sex- 

neutral.” Id. 

101 After the House Committee on Education and Labor held hearings on Mr. 

Thomas's proposals, and after the Senate Committee on Labor and Human Resources 

rejected the nomination of the chief architect of those proposals to the office of FEOC 

General Counsel, Mr. Thomas announced—during Senate consideration of his own 

renomination as Chairman of the EEOC—that modification of the Uniform Guidelines 

was no longer a major goal for him. 

102 see GAO Report, Uniform Guidelines on Employee Selection Procedures Should Be 

Reviewed and Revised (FPCD-82-26) (1982) (“1982 GAO Report”); Ad Hoc Group on 

Uniform Selection Guidelines, A Professional and Legal Analysis of the Uniform 

Guidelines on Employee Selection Procedures (1981) (“Ad Hoc Group Analysis”); Equal 

Employment Advisory Council, Employee Selection: Legal and Practical Alternatives to 

Compliance and Litigation (2d ed. 1986) (“EFAC Analysis”). 

 



Patrick O. Patterson 

  

and that disparate impact theory could be applied to a wide range of 

employment practices and systems. In determining adverse impact, courts 

variously considered—sometimes in the same case—differences in passing 

rates, failure rates, and selection rates, as well as statistics showing 

underutilization of minorities or women in comparison to the applicant pool, 
the general population, and differing conceptions of the relevant labor 

market.193 And courts routinely applied the Griggs principle not only to 

paper-and-pencil tests and educational requirements but also to 

multicomponent selection systems and to subjective employment criteria.104 

More recently, however, the Supreme Court has suggested a more 

restrictive approach, both to the kinds of evidence that might be required to 

prove adverse impact,10° and to the kinds of employment practices that might 

be subject to challenge under a disparate impact theory.1% Some lower courts 

have responded to these suggestions by more closely scrutinizing evidence 

offered to prove adverse impact, and by holding that disparate impact theory 

does not even apply when a multistep selection process or an individual 

subjective criterion has an adverse impact. According to these cases, the most 

prominent of which is the Fifth Circuit's decision in Pouncy v. Prudential Life 

Insurance Co., 107 the Griggs disparate impact interpretation of Title VII applies 

only “when an employer has instituted a specific procedure, usually a selection 

criterion for employment, that can be shown to have a causal connection to a 

class-based imbalance in the work force.”108 Other kinds of selection 

procedures and other employment practices—even if they have substantial 

  

18 gee B. Schlei and P. Grossman, supra note 9, at 98-100 and cases cited therein; 

Lerner, supra note 4, at 21-39; Booth and Mackay, supra note 8, at 142-51. 

104 gee B. Schlei and P. Grossman, supra note 9, at 162-205 and cases cited therein. 

105 gee New York City Transit Authority v. Beazer, 440 U.S. 568, 583-87 (1979) (statistics 
showing disproportionate percentage of minority employees referred for suspected use 
of narcotics and disproportionate percentage of minorities receiving methadone 

maintenance in public programs held insufficient to show that employer’s policy 

against employment of methadone users had an adverse impact). 

106 See Furnco Construction Corp. v. Waters, 438 U.S. 567, 575-76 and nn.7-8 (1978) 

(McDonnell Douglas standard, rather than Griggs- Albemarle standard, applied where 

case “did not involve employment tests . . . or particularized requirements such as . . . 

height and weight specifications . . . , and it was not a ‘pattern or practice’ case”). 

107 668 F.2d 795 (5th Cir. 1982). 

108 14 at 800. 

   



Employment Testing and Title VII 103 

adverse impact and even if they are not job related—will not violate Title VII 

unless they are shown to be intentionally discriminatory.10? 

The Uniform Guidelines, on the other hand, include subjective criteria 

within their broad definition of “selection procedures,”110 and they apply 

disparate impact analysis to all “tests and other selection procedures which are 
used as a basis for any employment decision.”111 A number of courts, agreeing 

with the position of the Guidelines on this question, have rejected the 

restrictive approach of Pouncy and similar cases on the ground that it would 

encourage employers to use subjective criteria and multicomponent selection 

processes having an unjustified adverse impact. These courts have held that 

disparate impact analysis applies not only to specific, objective, facially neutral 

practices, but also to more diffuse selection systems having a cumulative 

adverse impact, as well as to subjective selection criteria.112 The Supreme 

Court probably will resolve this question within the next few years. 

The Duty To Investigate Alternatives 

Another unresolved issue concerns the nature and extent of an employer's 

duty under Title VII to investigate and adopt alternative selection procedures 

or uses having less adverse impact. This is a matter of great practical 

importance for the enforcement of Title VII. Persons opposed to the imposition 

of such an obligation argue that it perpetuates: 

  

109 14. See also Antonio v. Wards Cove Packing Co., 768 F.2d 1120, 1131-33 and n.8 (9th 

Cir. 1985); Spaulding v. University of Washington, 740 F.2d 686 (9th Cir.), cert. denied, 

105 S. Ct. 511 (1984); Vuyanich v. Republic National Bank, 723 F.2d 1195 (5th Cir. 1984); 

Talley v. United States Postal Service, 720 F.2d 505 (8th Cir. 1983); EEOC v. Federal 

Reserve Bank, 698 F.2d 633 (4th Cir. 1983); Mortensen v. Callaway, 672 F.2d 822 (10th Cir. 

1982). See generally B. Schlei and P. Grossman, supra note 9, at 1287-90. Cf. Bartholet, 
“Application of Title VII to Jobs in High Places,” 95 Haro. L. Rev. 947, 959-78 (1982). 

0 The Guidelines define “selection procedures” to include “the full range of 

assessment techniques from traditional paper and pencil tests, performance tests, 

training programs, or probationary periods and physical, educational, and work 

experience requirements through informal or casual interviews and unscored 

application forms.” 29 C.F.R. §1607.16Q. 

111 29 CFR. §1607.2B. 

112 gee Griffin v. Carlin, 755 F.2d 1516, 1522-25 (11th Cir. 1985); Lasso 0. Woodmen of 
the World Life Insurance Co., 741 F.2d 1241 (10th Cir. 1984); Segar v. Smith, 738 F.2d 1249 

(D.C. Cir. 1984); Page v. U.S. Industries, Inc., 726 F.2d 1038 (Sth Cir. 1984); Wang v. 

Hoffman, 694 F.2d 1146 (9th Cir. 1982); Rowe v. Cleveland Pneumatic Co., 690 F.2d 88 

(6th Cir. 1982); Rowe v. General Motors Corp., 457 F.2d 348 (5th Cir. 1972); Allen v. Issac, 

39 E.P.D. para. 35,989 (N.D. Ill. 1986); Shidaker v. Bolger, 593 F. Supp. 823 (N.D. Ill. 1984). 

Cf. Soria v. Ozinga Brothers, Inc., 704 F.2d 990 (7th Cir. 1983) (noting but not resolving 

the issue). See generally Bartholet, supra note 109, at 978-98. 

 



Patrick O. Patterson 

  

the myth of the alternative screening-method solution to the 
problem of adverse impact, forcing employers who use valid 
selection devices to spend time and money searching for other, 
equally valid devices with lesser adverse impact. In most cases, 
this is a chimera because there is no real alternative to the need 
for literacy, numeracy, and other job- and life-relevant skills that 
valid tests measure.113 

Proponents, on the other hand, contend that the legal duty to investigate and 

adopt such reasonable alternatives is “vital” to achieving Title VII's goal of 

integrating minorities and women into the American economic 
mainstream.114 

Some recent cases illustrate the importance of this issue. In a pending 

case against the police and fire departments of Akron, Ohio, for example, the 

Lawyers” Committee for Civil Rights Under Law contends that separate item 
analyses for blacks, Hispanics, and Anglos can identify particular test items 
that are not good predictors of job performance for persons of each racial or 

ethnic group.11> When those items are rewritten to cover the same content in 

different terms, much of the tests adverse impact can be eliminated.116 

  

13 Lerner, supra note 4, at 43. See also NAS/NRC Report on Ability Testing, supra note 
5, at 144 (finding “no evidence of alternatives to testing that are equally informative, 

equally adequate technically, and also economically and politically viable”). Cf. Hunter 

and Schmidt, “Ability Tests: Economic Benefits Versus the Issue of Fairness,” 21 Indus. 

Rel. 293, 299-301 (1982) (arguing that alternative uses of cognitive ability tests result in 
reduced productivity). 

114 Bartholet, supra note 109, at 1024. See United Steelworkers v. Weber, 443 U.S. 193, 

202 (1979). 

115 An item analysis separately examines each item on a test, and compares the 

responses to that item with the responses to all other items on the test. If test takers 
who generally have made correct responses to other items are found to have made 

random or near-random responses to the item being examined, the test researcher 
concludes that the item is not useful because it does not help to distinguish between 

high and low scorers. This may be because the item was badly written, because the test 

takers were unfamiliar with the words used in the item or with culturally specific 

knowledge required by the item, or due to other reasons. When separate item analyses 
are done for different racial and ethnic groups, particular items that have an adverse 

impact on minorities can be identified and eliminated. 

116 This is essentially a variation of the “Golden Rule Procedure,” named for a 1984 
out-of-court settlement between the Educational Testing Service, the State of Illinois, 

and the Golden Rule Insurance Company concerning the Illinois Insurance Agent 

Licensing Examination. Under the settlement, ETS is required to conduct separate race- 
based item analyses and, within groups of equally difficult items in the same content 

areas, to select those items that display the least difference in correct answer rates 

between majority and minority test takers. 

    

 



  

Employment Testing and Title VII 105 

Another case, involving promotional testing in the St. Louis Fire 

Department, 117 was scttled by climinating the paper-and-pencil tests that had 

disproportionately excluded blacks from promotion and developing a new 

selection procedure based on “real world” problems. The city had previously 

followed the usual civil service practice of making promotions based on ranked 

scores on a multiple-choice test asking about such matters as the meaning of 

abstruse parts of the local fire code and the proper deployment of firefighters 

and equipment in described situations. Under the settlement, several experts 

developed real-world simulations that involved, for example, taking thc 

candidates to a building, asking them to imagine a fire of a particular type and 

intensity, telling them the available equipment, and then asking them to give 

the kinds of orders they would actually give if they were in charge of the 

firefighting effort at the building. This resulted in both the virtual elimination of 

adverse impact on minorities and the development of a much more useful 

selection procedure.118 

Another alternative to paper-and-pencil testing is the biodata approach. 

This approach is based on the theory that future performance on a job can be 

predicted more successfully by an inquiry into the applicant’s past experiences 

than by a traditional paper-and-pencil test. Studies have shown that biodata 

are related to job performance, and that their use may significantly reduce the 

degree of adverse impact on minorities. At the request of the Justice 

Department, the Lawyers’ Committee has agreed to participate in the 

development of a biodata alternative to traditional testing in a case involving 

police and fire department hiring in Birmingham and Jefferson County, 

Alabama. 
What remains unclear is whether Title VII requires employers to 

investigate and implement such alternatives, and if so, to what extent. Section 

3B of the Uniform Guidelines states in part that a validity study “should 

include . . . an investigation of suitable alternative selection procedures and 

suitable alternative methods of using the selection procedure which have as 

little adverse impact as possible.” 119 Questions have been raised as to 

whether this duty to investigate alternatives is consistent with Title VII case law. 

  

17 Firefighters Institute for Racial Equality v. City of St. Louis, Civil Action Nos. 74- 

30(c)(3) and 74-200(c)(3) (E.D. Mo.). Filed in 1974, the case went to the Eighth Circuit 

three times. The last decision on appeal is reported at 616 F.2d 350 (8th Cir. 1980), cert. 

denied, 452 U.S. 938 (1981). 

118 The parties in Hammon vo. Barry, Civil Action Nos. 84-0903, 85-0782, and 85-0797 

(D. D.C.), in accordance with recommendations of the Lawyers’ Committee and the 

Legal Defense Fund as amici curiae, have agreed to develop a similar procedure to 

resolve disputes over fire department promotions in the District of Columbia. 

119 29 CER. §1607.3B. 

 



  

Patrick O. Patterson 

For example, in its 1982 report recommending revision of the Uniform 
Guidelines, the General Accounting Office cited a “perceived inconsistency” 

between this section of the Guidelines and the following language from the 

Supreme Court's opinion in Albemarle: 

If an employer does then meet the burden of proving that its tests 
are “job related,” it remains open to the complaining party to 
show that other tests or selection devices, without a similarly 
undesirable racial effect, would also serve the employer's 

legitimate interest in “efficient and trustworthy workmanship.”120 

The Equal Employment Advisory Council (EEAC) also has taken the position 

that §3B of the Guidelines is “inconsistent with Albemarle” and should be 

“revised to conform to federal court precedent which places the burden of 

showing suitable alternative selection procedures on the plaintiff or the 

enforcement agencies.” 121 

The obligation to investigate alternatives with less adverse impact is also 

said to be inconsistent with the Supreme Court's statement in Furnco that 

“Title VII . . . does not impose a duty to adopt a hiring procedure that 

maximizes hiring of minority employees.” 122 According to this view, in 

litigating a case “an employer need do no more than show that a procedure is 

job-related, and it is improper to require a further showing that the use of the 

procedure is the use that maximizes the hiring of minorities. Griggs . . . requires 

no more than validation to dispel the implication of discrimination arising from 

a statistical showing of adverse impact.”123 

However, neither Albemarle nor Furnco directly addresses the question 

of whether Title VII imposes an affirmative duty to investigate or adopt 

rcasonable alternatives. Instead, these cases concern the order and allocation 

  

120 1982 GAO Report, supra note 102, at 8-9, quoting Albemarle, 422 U.S. at 425 (citations 

omitted). 

121 ggac Analysis, supra note 102, at 33. See also Ad Hoc Group Analysis, supra note 

102, at 41; Booth and Mackay, supra note 9, at 189-93; Thompson and Christiansen, 

“Court Acceptance of Uniform Guidelines Provisions: The Bottom Line and the Search 

for Alternatives,” 8 Empl. Rel. L.]. 587, 598-602 (1983); Rubenfeld and Crino, “The 

Uniform Guidelines: A Personnel Decision-Making Perspective,” 7 Empl. Rel. L.]. 105, 

117 (1981). 

122 Furnco Construction Corp. v. Waters, 438 U.S. 567, 578 (1978). 

123 Ad Hoc Group Analysis, supra note 102, at 41. See also EEAC Analysis, supra note 

102, at 34 (Furnco “holds that an employer is not required to use selection procedures 

that maximize the employment opportunities of minorities and women”); Booth and 

Mackay, supra note 9, at 191-92; Thompson and Christiansen, supra note 121, at 600. 

  

   



  

Employment Testing and Title VII 107 

of proof during litigation.124 As a principal critique of the Guidelines 

acknowledges, these decisions announce “only . . . rule[s] of evidence, not . . . 

rules] of law.”125 

A few courts appear to have accepted the argument that imposing a 

duty to investigate alternatives may be inconsistent with Albemarle or 

Furnco.126 Other courts, however, have held that the law does impose such a 

duty.127 Even where there is some evidence of a sclection procedure’s validity, 

some courts have ruled that failure to consider the degree of adverse impact 

resulting from cutoff scores or ranking, and failure to implement reasonable 

alternative uses having less adverse impact, may lead to a finding that the 

employer's use of the procedure is not job related.'28 Conversely, employers 

who have investigated and adopted reasonable alternative procedures or uses 

that reduce or eliminate adverse impact on minorities have been protected 

from liability in “reverse discrimination” suits.]29 Some courts have even 

based findings of intentional discrimination against minorities on an 

  

124 Sce, for example, Gillespie v. State of Wisconsin, 771 F.2d 1035, 1045-46 (7th Cir. 

1985), cert. denied, 106 S. Ct. 854 (1986) (plaintiff in litigation did not satisfy burden of 
demonstrating equally valid alternative selection procedures with less adverse impact); 
Clady v. County of Los Angeles, 770 F.2d 1421, 1432-33 (9th Cir. 1985) (same). 

125 Ad Hoc Group Analysis, supra note 102, at 41. 

126 Cormier v. PPG Industries, Inc., 519 F. Supp. 211, 281 (W.D. La. 1981), aff'd per 

curiam, 702 F.2d 567, 568 (5th Cir. 1983) (federal agencies “cannot, through the guise of 

interpreting their own regulations or by issuing ‘guidelines,’ overrule decisions of the 

Supreme Court”); M.A.G.E. v. Bailar, 518 F. Supp. 800, 811-12 (N.D. Cal. 1981) 
(“Whatever weight might be given the failure to follow that instruction [to investigate 
alternatives] in the consideration of the value of the validity studies presented in 

evidence or in consideration of suitable relief after a finding of liability, this court does 
not find that this statement in the EEOC guidelines should be enforced by [preliminary] 

injunction, particularly in the face of the case law distributing the burden of proof 

differently”). 

127 See, for example, Allen v. City of Mobile, 464 F. Supp. 433, 439-40 (S.D. Ala. 1978) 
(citing Uniform Guidelines §3B with approval and holding that defendants had not 
made an adequate showing of job-relatedness due in large part to failure to conduct a 

reasonable search for alternatives). 

128 See infra Part III(E). See, for example, Guardians Association v. Civil Service 

Commission of the City of New York, 630 F.2d 79, 100-06 (2d Cir. 1980), cert. denied, 452 

U.S. 940 (1981); Ensley Branch, NAACP v. Seibels, 616 F.2d 812, 822 (5th Cir.), cert. denied, 

449 U.S. 1061 (1980); Louisville Black Police Officers Organization v. City of louisville, 

511 F. Supp. 825, 838-39 (W.D. Ky. 1979). 

129 Soe Bushey v. New York State Civil Service Commission, 733 F.2d 220, 224-28 (2d 
Cir. 1984), cert. denied, 105 S. Ct. 803 (1985); Kirkland v. New York State Department of 

Correctional Services, 628 F.2d 796, 798-801 (2d Cir. 1980), cert. denied, 450 U.S. 980 (1981). 

 



Patrick O. Patterson 

  

employer's failure to consider and adopt reasonable alternatives with less 
adverse impact.130 The Supreme Court has yet to decide the issue. 

Validation Against Training Performance 

Prior to the Supreme Court's 1976 opinion in Washington v. Davis, 131 it 

scemed well settled that evidence, standing alone, of a relationship between 
performance on a test and performance in a training program, was not 

sufficient to show that a test was “job related” under the Title VII standards of 

Griggs and Albemarle; it was necessary for the employer to establish in 

addition that performance in the training program was itself manifestly related 

to performance on the job.132 In Washington v . Davis, however, the Court— 

suggesting in a non-Title VII case that it might be applying “standards similar 

to those obtaining under Title VII”133_announced a new and “much more 

sensible construction” of the job-relatedness requirement: A “positive 

relationship between the test and training-course performance was sufficient 

to validate the [test] wholly aside from its possible relationship to actual 

performance as a police officer.”134 Two years later in NEA v. South Carolina, 135 

a Title VII case, the Court summarily affirmed a three-judge district court 

ruling that “the decision to validate [the National Teacher Examinations] 

against the academic training program rather than job performance is 

specifically endorsed in principle in Davis.”136 In a dissenting opinion in NEA, 

however, Justice White appeared to disavow the implications of his own prior 

opinion for the majority in Davis: 

Washington v. Davis . . . was thought by the District Court [in NEA] 
to have warranted validating the test in terms of the applicant's 
training rather than against job requirements; but Washington v. 

  

130 Sce Easley v. Anheuser-Busch, Inc., 572 F. Supp. 402, 410 (E.D. Mo. 1983); Dickerson 

v. United States Steel Corp., 472 F. Supp. 1304, 1352 (E.D. Pa. 1978). 

131 426 U.S. 229 (1976). 

132 See, for example, Vulcan Society v. Civil Service Commission of the City of New 

York, 490 F.2d 387, 396 and n.11 (2d Cir. 1973); Pennsylvania v. O'Neill, 348 F. Supp. 1084 

(E.D. Pa. 1972), aff d in pertinent part, 473 F.2d 1029 (3d Cir. 1973); United States v. City of 

Chicago, 385 F. Supp. 543, 556 (N.D. Ill. 1974), aff'd in pertinent part, 549 F.2d 415 (7th 

Cir.), cert. denied, 434 U.S. 875 (1977); Officers for Justice v. Civil Service Commission of 
the City of San Francisco, 371 F. Supp. 1328, 1337 (N.D. Cal. 1973). 

133 426 US. at 249. 

134 14 at 250-51. 

135 434 US. 1026 (1978). 

136 United States v. South Carolina, 445 F. Supp. 1094, 1113 (D. S.C. 1977), aff'd sub 

nom. National Education Association v. South Carolina, 434 U.S. 1026 (1978). 

    

 



  

Employment Testing and Title VII 109 

Davis, in this respect, held only that the test there involved, which 

sought to ascertain whether the applicant had the minimum 
communication skills necessary to understand the offerings in a 
police training course, could be used to measure eligibility to 
enter that program. The case did not hold that a training course, 
the completion of which is required for employment, need not 
itself be validated in terms of job relatedness. Nor did it hold that 
a test that a job applicant must pass and that is designed to 
indicate his mastery of the materials or skills taught in the 
training course can be validated without reference to the job. 
Tests supposedly measuring an applicant’s qualifications for 
employment, if they have differential racial impact, must bear 
some “manifest relationship to the employment in question,” 
Griggs . . ., and it is insufficient for the employer “to demonstrate 
some rational basis for the challenged practices.” Washington v. 

Davis.137 

Not surprisingly, subsequent Title VII cases are divided on the question 

of whether a selection procedure may be validated against training 

performance alone. Some cases simply cite Davis and hold that “[i]t is 

sufficient that the test validly predicts performance at the police academy. It 

need not also bear a positive correlation to satisfactory performance as a police 

officer.” 138 

More of the cases, however, wrestle with the confusingly contradictory 

Supreme Court opinions, distinguish Davis as a non-Title VII case, and 

conclude that employers must show that selection procedures are related not 

just to training performance but also to job performance. For example, the 

Ninth Circuit has reasoned that, “[i]f employers were permitted to validate 

selection procedures without reference to job performance, then non-job- 

related selection devices could always be validated through the simple 

expedient of employing them at both the pre-training and training stage.” 13° 

In the Ninth Circuits view, the Supreme Court did not consider this danger in 

Davis and NEA because the danger was minimal in those cases; however, where 

(unlike NEA) both the selection procedure and the training program are under 

the employer’s control, and where (unlike Davis) the selection procedure is 

used to measure something more than minimum communication skills, the 

  

137 434 U.S. at 1027-28 (White, ]., dissenting). For differing perspectives on the 
significance of these opinions, compare Haney, supra note 4, at 16-26, with Booth and 

Mackay, supra note 9, at 128-41. 

138 Rivera v. City of Wichita Falls, 665 F.2d 531, 538 n.9 (5th Cir. 1982). See also Corley 

v. City of Jacksonville, 506 F. Supp. 528, 532-36 (M.D. Fla. 1981). 

139 Blake v. City of Los Angeles, 595 F.2d 1367, 1382 n.17 (9th Cir. 1979), cert. denied, 446 
U.S. 928 (1980). 

 



Patrick O. Patterson 

  

cmployer must show “that the test correlates significantly with important 

clements of [training] academy performance and that those elements are 

important to actual job performance.” 140 The Second Circuit has similarly 

noted that allowing employers to validate tests against training performance 

alone “would undermine Title VII's goal of eliminating arbitrary barriers to the 

employment of minorities,” and therefore has continued to require “a showing 

that examinations testing performance at training school are themselves job- 

related.” 141 

Other post-Davis court decisions have also held that selection 

procedures having an adverse impact must be shown to be related not merely 

to training performance but also to job performance.42 These decisions 

appear to be consistent with the Uniform Guidelines,143 the APA Standards, 144 

and the “Division 14 Principles.”145 Nevertheless, until the Supreme Court 

clarifies its position on this issue, the law will remain uncertain. 

Validity Generalization 

In the last few years, “validity generalization” —the theory that a finding of test 

validity for one job in one setting may be freely generalized or transported to 

other jobs in different settings—has become a focus of debate in the 

psychological community, and it is beginning to become a legal issue as well 

(sce also Henry Levin, this volume). Some historical background will be helpful 

in understanding this subject. 

  

140 Craig v. County of Los Angeles, 626 F.2d 659, 663 (9th Cir. 1980), cert. denied, 450 

U.S. 919 (1981) (emphasis in original). But see Clady v. County of Los Angeles, 770 F.2d 

1421, 1426 (9th Cir. 1985) (accepting validation against training performance alone where 

“plaintiffs did not contest the appropriateness of the performance criteria against 

which the exam was correlated”). 

141 Guardians Association v. Civil Service Commission of the City of New York, 633 F. 
2d 232, 244-45 (2d Cir. 1980), cert. denied, 103 S. Ct. 3568 (1983). 

142 oe Harless v. Duck, 619 F.2d 611, 616-17 (6th Cir.), cert. denied, 449 U.S. 872 (1980); 
Ensley Branch, NAACP v. Seibels, 616 F.2d 812, 822 n.25 (5th Cir.), cert. denied, 449 U.S. 

1061 (1980); Brunet v. City of Columbus, 41 E.P.D. para. 36,498 (S.D. Ohio 1986); Dickerson 

v. United States Steel Corp., 472 F. Supp. 1304, 1347-49 (E.D. Pa. 1978). 

143 500 29 C.F.R. §§1607.5B and 1607.14B(3). 

144 Sce 1985 APA Standards, supra note 33, at 60-62. 

145 gee 1980 “Division 14 Principles,” supra note 34, at 7, 12. 

    
 



  

Employment Testing and Title VII 111 

The origins of modern psychological testing 146 lic in nineteenth century 

theories of dispositionalism, genetic determinism, and racial inferiority.147 

Although genetic determinism and racial inferiority have been largely rejected 

by present-day psychologists and social scientists, 148 the dispositionalist view 

of human personality and behavior has to some extent remained a central 

assumption underlying many aspects of test theory.149 Early theoreticians and 

practitioners of psychological testing believed that people possessed relatively 

stable and consistent traits and abilities which varied little from one situation to 

another, and that those traits and abilities could always be measured by 

appropriate tests.1>0 “It is the presumed stability of these attributes that 

provides the rationale for their measurement; if attributes were inconsistent or 

unstable, there would be little value or utility in measuring them.” 131 

As time went on, however, evidence and experience contradicted these 

basic assumptions; over and over, industrial psychologists found that, even 

where their studies showed that a test was valid for a particular job in a 

particular setting, they could not show the same test to be valid for different 

jobs, or even for quite similar jobs in different settings.152 Psychologists 

therefore gradually moved away from the “dispositional” model of human 

behavior and toward a “situational” model, which views behavior as highly 

dependent on the immediate situation or environment in which it occurs.193 

As stated in the current edition of a leading text on psychological testing, 

  

146 Systematic employment testing is thought to have originated in China around 2200 
B.C., when the emperor began examining his officials every third year to determine 
their fitness to continue in office. The elaborate civil service testing system that 

evolved in imperial China was adapted by France in the late eighteenth century and by 
Britain in the nineteenth century (see also Carolyn Webber, this volume). The United 

States picked up the movement from Britain in the late nineteenth century, and began 
to develop and use civil service tests in the 1880s. P. DuBois, A History of Psychological 
Testing 3-6 (1970). Professional psychologists did not become involved in employment 
testing until early in the twentieth century. Id. at 82. 

147 gee Haney, supra note 4, at 3-13. Sce also S. Gould, The Mismeasure of Man 20-29 

(1981). 

148 But sce A. Jensen, Bias in Mental Testing (1979). 

149 Haney, supra note 4, at 50. 

150 gee Haney, supra note 4, at 50. The principal trait the early testers were trying to 
measure was “intelligence.” They turned to psychological tests for this purpose after 
they gave up on craniometry—the science of measuring heads. See S. Gould, supra note 

147, at 30-107. 

151 Haney, supra note 4, at 50. 

152 sce id. at 49 and nn.228-29. 

133 gee id. at 51-55. 

 



  

Patrick O. Patterson 

“[clautious professional opinion has urged for decades that any selection plan 

be validated afresh in each firm.”154 Cognitive psychologists now refer (rather 

pejoratively) to the dispositional model as “fundamental attribution error.”155 

By the time the EEOC began formulating legal standards for employment 

testing in its 1966 and 1970 Guidelines, 156 the agency and its psychologist- 

advisers were well aware of the professional consensus that validity results 

could be safely generalized only in limited circumstances. The 1966 Guidelines 

thus defined the statutory phrase “professionally developed ability test” to 

mean “a test which fairly measures the knowledge or skills required by the 

particular job or class of jobs which the applicant seeks, or which fairly affords 

the employer a chance to measure the applicant’s ability to perform a 

particular job or class of jobs.”157 The focus on the need for a match between 

a specific test and a specific job is also apparent in the 1970 EEOC Guidelines 
and in the language of the Supreme Court’s opinions in Griggs, Albemarle, and 

subsequent cases.]® Indeed, in Albemarle the Court relied on the 1970 EEOC 

Guidelines in holding that “[a] test [having an adverse impact] may be used in 

jobs other than those for which it has been professionally validated only if there 

are ‘no significant differences’ between the studied and unstudied jobs.”159 

This basically situationalist view was never seriously questioned even 

during the period of interagency conflict over different versions of the 

guidelines, and it was reaffirmed when the agencies adopted the Uniform 

Guidelines in 1978.160 Although this view finds expression in many provisions 

of the current Uniform Guidelines, it is perhaps most evident in §§7 and 8, 

which govern the “transportability” of validity evidence. These sections permit 

the transportation only of criterion-related validity evidence, and they impose 

rclatively stringent requirements on employers attempting to rely on such 

  

154 L. Cronbach, supra note 2, at 400. 

155 R. Nisbett and L. Ross, Human Inference: Strategies and Shortcomings of Social 
Judgment 31 (1980). Nisbett and Ross define “fundamental attribution error” as the 

“assumption that behavior is caused primarily by the enduring and consistent 
dispositions of the actor, as opposed to the particular characteristics of the situation to 
which the actor responds.” Id. They maintain that this inferential error is so 

“thoroughly woven into the fabric of our culture” that “children growing up in our 
culture come to hold an increasingly dispositional view of the causes of behavior.” Id. 

156 Supra notes 25 and 36. 

157 1966 EEOC Guidelines, supra note 25 (emphasis added). 

158 See supra notes 43-78. 

159 Albemarle Paper Co. v. Moody, 422 U.S. 405, 432 (1975). 

160 gee supra notes 84-98. 

    
 



  

Employment Testing and Title VII 113 

cvidence.16l The courts gencrally have adopted the cautious approach to 

transportability and validity generalization reflected in the Guidelines.162 

The dispositional model, however, now appears to be making a 

comeback among some prominent industrial and personnel psychologists. 

Frank Schmidt, John Hunter, and other psychologists have published a series 

of recent articles in which they reexamine old validity studies and conclude 

that the variance in the outcomes of those studies was not caused by actual 

differences in validity depending on the particular situations in question, but 

instcad was “largely an illusion created by statistical artifacts” resulting from 

differences between studies in test reliability, criterion reliability, range 

restriction, and sampling error.183 They use this conclusion to support a broad 

view of “validity generalization,” claiming that tests of “general cognitive 

ability” are valid for all jobs, that “massive research findings” have shown that 

such tests are not unfair to minority applicants, and that increased use of these 

tests would raise productivity in the United States by billions of dollars 

annually 164 They assert that their research not only proves that “validity 

generalization is possible on a widespread basis,” but also shows that it is 

finally possible to “develop the general principles and theories that are 

  

161 29 CFR. §§1607.7 and 1607.8. The requirements for transportability include: 

(1) “clear” evidence of validity; (2) job analyses showing that incumbents in the 
borrowing user’s jobs and incumbents in the jobs on which the study was conducted 

perform “substantially the same major work behaviors”; and (3) an investigation of test 
fairness for each race, sex, and ethnic group constituting a significant factor in the 

borrowing employer's relevant labor market. 29 C.F.R. §1607.7B. The Guidelines further 
restrict transportability to circumstances in which there are no “variables in the user's 
situation which are likely to affect validity significantly.” 29 C.F.R. §1607.8B. See also 29 
C.F.R. §§1607.7C and 1607.7D. 

162 Sce Albemarle Paper Co. v. Moody, 422 U.S. at 432; Brunet v. City of Columbus, 41 

E.P.D. para. 36,498 (S.D. Ohio 1986); Dickerson v. United States Steel Corp., 472 F. Supp 

1304, 1338-40 (E.D. Pa. 1978); Vanguard Justice Society, Inc. v. Iughes, 471 F. Supp. 670, 

732-35 (D. Md. 1979); Berkman v. City of New York, 536 F. Supp. 177, 212-16 (E.D. N.Y. 
1982), aff'd, 705 F.2d 584 (2d Cir. 1983). Cf. Rivera v. City of Wichita Falls, 665 F.2d 531, 

538 n.10 (5th Cir. 1982). 

163 Pearlman, Schmidt, and Hunter, “Validity Generalization Results for Tests Used 

To Predict Job Proficiency and Training Success in Clerical Occupations,” 65 |. App 
Psych. 373, 374-76, 399-400 (1980). See also Schmidt and Hunter, “Employment Testing: 

Old Theories and New Research Findings,” 36 Am. Psych. 1128 (1981) (bricf review of 

recent findings); Schmidt, Hunter, and Pearlman, “Task Differences and Validity of 

Aptitude Tests in Selection: A Red Herring,” 61 |. App. Psych. 166 (1981); Schmidt, 
Hunter, Pearlman, aiid Shane, “Further Tests of the Schmidt-H unter Bayesian Validity 

Generalization Procedure,” 32 Pers. Psych. 257 (1979); Schmidt and Hunter, 

“Development of a General Solution to the Problem of Validity Generalization,” 

62 J. App. Psych. 529 (1977). 

164 | lunter and Schmidt, supra note 4, at 295-98. 

 



Patrick O. Patterson 

  

necessary to take the field [of personnel psychology] beyond a mere 

technology to the status of a science.”16 

While Schmidt and Hunter take their validity generalization theory to 

an extreme, it has been accepted to some extent by a few courts!66 and by 

many psychologists. As one standard text on psychological testing states, 

Schmidt, Hunter, and their co-workers were able to show that the 

validity of tests of verbal, numerical, and reasoning aptitudes can 

be generalized far more widely across occupations than had 
heretofore been recognized. . . . Evidently, the successful 
performance of a wide variety of occupational tasks depends to a 
significant degree on a common core of cognitive skills. . . . It 
would seem that this cluster of cognitive skills and knowledge is 
broadly predictive of performance in both academic and 
occupational activities demanded in advanced technological 

societies.167 

Another standard text, however, concludes that “validities are much less 

generalizable than Schmidt's group suggests.”168 Both the 1980 “Division 14 

Principles” 169 and the 1985 APA Standards170 acknowledge the Schmidt-Hunter 
research, but they maintain a cautious approach to the subjects of 

transportability and validity generalization. The “Division 14 Principles” note 

that “ImJany questions regarding generality are still open to debate,” and they 

hopefully predict that “[cJontinued evidence . . . should enable further 

extensions of validity generalization.” 171 The APA Standards restrict the use of 

transported validity evidence to circumstances in which “it can be 

demonstrated that the test has generated a significant record of validity in 

  

165 Pearlman, Schmidt, and Hunter, supra note 163, at 400, 374. Stephen Jay Gould 
might characterize these assertions as an example of “physics envy”—the yearning of 
some social scientists to find simplifying laws and basic particles that will reduce the 
outward complexity of the social world to a simpler, underlying reality. See S. Gould, 

supra note 147, at 262. 

166 sae Friend ov. Leidinger, 588 F.2d 61, 65 (4th Cir. 1978); Pegues v. Mississippi State 

Employment Service, 488 F. Supp. 239, 254 and n.1 (M.D. Miss. 1980), aff'd on other 

grounds, 699 F.2d 760 (5th Cir.), cert. denied, 104 S. Ct. 482 (1983). 

167 A. Anastasi, supra note 2, at 143-44. See also id. at 437-38. 

168 |. Cronbach, supra note 2, at 401. See also James, Demaree, and Mulaik, “A Note 

on Validity Generalization Procedures,” 71 J. App. Psych. 440 (1986) (a critical review of 

the Schmidt-Hunter studies). 

169 1980 “Division 14 Principles,” supra note 34, at 15-16. 

170 1985 APA Standards, supra note 33, at 12 and 59-62. 

171 1980 “Division 14 Principles,” supra note 32, at 15-16. 

  
  
 



Cal
 

gy
 

Ju
i 

  
  

Employment Testing and Title VII 115 

similar job settings for highly similar pcople or that it is otherwise appropriate 

to generalize from other applications.”172 

The cautious approach of the profession to validity generalization is also 

reflected in the 1982 recommendations of the National Academy of 

Sciences /National Research Council report on ability testing: 

Government agencies concerned with fair employment practices 
should accept the principle of cooperative validation research so 

that tests validated for a job category such as fire fighter in a 
number of localities can be accepted for use in other localities on 

the basis of the cumulated evidence. It would remain incumbent 

on the user of such a test to develop a persuasive showing—based 
on close examination of the test, the work, and the applicant 

pool—that it is appropriate for use in the conditions that obtain in 

the local situation.173 

As noted above, the validity generalization debate within the 

psychological community has already begun to spill over into the courts and 

administrative agencies, which ultimately will have to decide how this 

developing concept might affect their interpretations of Title VII. 

Cutoff Scores and Ranking 

A final developing legal issue concerns the extent to which Title VII requires 

employers to justify particular methods of using otherwise valid sclection 

procedures. Such requirements have been applied primarily to methods of use 

that have greater adverse impact than other methods—especially the use of 

cutoff scores and the practice of ranking on the basis of test scores. Many of 

the issues raised by such requirements have already been discussed in 

connection with the more general obligation to investigate and implement 

reasonable alternative procedures or uses with less adverse impact.!74 These 

issues are scparately considered here for three reasons: (1) industrial and 

personnel psychologists generally regard cutoff scores and ranking as 

presenting unique problems; (2) a number of courts that have been unwilling 

or unable to master the complexities of many other testing and validation 

issues have shown that they are capable of understanding and applying the 

more basic, commonsense principles concerning the use of test scores; and (3) 

the use of cutoff scores, ranking, and related practices are ubiquitous features 

of civil service testing at all levels—and therefore affect the job opportunities of 

millions of people, including large numbers of minoritics and women. 

  

172 1985 APA Standards, supra note 31, at 59. 

173 NAS/NRC Report on Ability Testing, supra note 5, at 148. 

174 ‘gue supra Part 111(B).



Patrick O. Patterson 

  

The Uniform Guidelines contain several provisions which, in essence, 
require employers to consider the degree of adverse impact and to justify the 
setting of particular cutoff scores!” and the use of ranking176 even when the 
employer's selection procedure is otherwise valid.17”7 This approach finds 
some support in the 1982 study of the National Academy of Sciences /National 
Research Council, which urged courts and enforcement agencies to 

“distinguish between the technical psychometric standards that can 
reasonably be imposed on ability tests and the legal and social policy 
requirements that more properly apply to the rules for using test scores and 
other information in selecting employees.”1”8 The study also recommended 

that federal authorities “concentrate on providing employers with guidelines 

that set out the range of legally defensible decision rules to guide their use of 

test scores.” 179 

The “Division 14 Principles”, on the other hand, adopt a dramatically 

different approach. According to the Principles, cutoff scores may be set “as 

high or as low as the purposes of the organization require, if they are based on 

valid predictors,” and “the relationship between a predictor and a criterion 

  

175 see 29 CFR. §1607.5H (cutoff scores “should normally be set so as to be reasonable 
and consistent with normal expectations of acceptable proficiency within the work 
force”; where ranking results in effectively higher cutoff scores, “the degree of adverse 

impact should be considered”). 

176 see 29 CFR. §1607.5G (“if a user decides to use a selection procedure on a ranking 
basis, and that method of use has a greater adverse impact than use on an appropriate 

pass/fail basis . . . , the user should have sufficient evidence of validity and utility to 

support the use on a ranking basis”). 

177 See also 29 C.F.R. §§1607.14B(6) and 1607.14C(9). 

178 NAs/NRC Report on Ability Testing, supra note 5, at 148. 

179 14. at 148. The report also states that “[t]he goals of efficiency and representativeness 
are more likely to be brought into a workable balance by altering the decision rule ([e.g.,] 

ranking and the rule of three) that determines how test scores are used. This might be 

in the form of a weighting formula that recognizes high ability, ethnic diversity, and 

other socially valued considerations in selecting from the portion of the applicant 

population that has demonstrated the threshold level of ability or skill necessary to 

satisfactory job performance.” Id. at 147. 

    

 



  

Employment Testing and Title VII 117 

may [usually] be assumed to be lincar.”180 [n other words, if a test is otherwise 

valid, an employer may sct any cutoff score it likes and may use the test as a 

ranking device without any further justification. This approach has been 

incorporated in the EEAC’s proposed revisions of the Uniform Guidelines.181 

In a small number of cases, the courts—usually without meaningful 

discussion or explanation—have upheld the use of cutoff scores and ranking 

solely because they found tests to be otherwise valid. 182 The great majority of 

courts, however, have endorsed the imposition of additional requirements. 

Both beforel83 and since!®4 the adoption of the Uniform Guidelines, “[clourts 

  

180 1980 “Division 14 Principles”, supra note 34, at 18. The relevant portion of the 

Principles states in full: 

Selection standards may be sect as high or as low as the purposes of the 
organization require, if they are based on valid predictors. This implies 

that (a) the purposes of selection are clear and (b) they are acceptable in 

the social and legal context in which the employing organization 
functions. In usual circumstances, the relationship between a predictor 
and a criterion may be assumed to be linear. Consequently, selecting 
from the top scorers on down is almost always the most beneficial 
procedure from the standpoint of an organization if there is an 
appropriate amount of variance in the predictor. Selection techniques 

developed by content-oriented procedures and discriminating adequately 

within the range of interest can be assumed to have a linear relationship 

to job behavior. Consequently, ranking on the basis of scores on these 
procedures is appropriate. 

181 See EEAC Analysis, supra note 102, at 47-48. 

182 gee Gillespie v. State of Wisconsin, 771 F.2d 1035, 1045 (7th Cir. 1985), cert. denied, 
106 S. Ct. 854 (1986) (upholding cutoff scores based on “an estimate of the ability levels 
needed,” an effort to “select . . . as many qualified minority applicants as possible,” and a 
policy of insuring “that the interviewers would not be overwhelmed by the sheer 
number of candidates”); Contreras v. City of Los Angeles, 656 F.2d 1267, 1284 (5th Cir. 

1981), cert. denied 455, U. S. 1021 (1982) (upholding cutoff score and ranking on the 

ground that “the purpose of a civil service examination” is “to rank applicants so that 

only the top few may be hired”); Cuesta v. State of New York Office of Court 

Administration, 571 F. Supp. 392, 395 n.3 (S.D. N.Y. 1983) (rejecting without explanation 

challenges to cutoff score and ranking); Cormier v. PPG Industries, Inc., 519 Supp. 211264 

(W.D. La. 1981), aff'd per curiam, 702 F.2d 567, 568 (5th Cir. 1983) (upholding cutoff scores 

on the basis of Dr. Schmidt's testimony that higher cutoff scores result in increased 

dollar savings in productivity). Cf. Hunter and Schmidt, supra note 4, at 300-01 (“[s]ince 

the relationship between ability and performance is linear, any departure from top 
down hiring will result in a loss in the mean productivity of the work force”). 

183 See Rogers v. International Paper Co. 510 F.2d 1340, 1351 (8th Cir. 1975); Boston 

Chapter, NAACP v. Beecher, 504 F.2d 1017, 1023 (1st Cir. 1974), cert. denied, 421 U.S. 910 

(1975); Walston v. County School Board, 492 F.2d 919, 927 (4th Cir. 1974); Kirkland v. 

New York State Department of Correctional Services, 374 F. Supp. 1361, 1377 (S.D. N.Y. 

 



Patrick O. Patterson 

  

have regularly rejected rank ordering as inadequately validated . . . ,” and 

“[clutoff scores have similarly met with strict scrutiny by the courts.” 185 

The leading case in this area is the Second Circuit's decision in 

Guardians Association v. Civil Service Commission (Guardians 1V).186 

Although the court in Guardians IV generally criticized the Uniform Guidelines 

and otherwise upheld the validity of an entry-level police officer examination, it 

expressly endorsed the Guidelines with respect to cutoff scores and ranking, 

and on that basis held that the employer's use of the exam was unlawful. The 

court found that §14C(9) of the Guidelines, providing that rank ordering should 

be used only if it can be shown that “a higher score . . . is likely to result in better 

job performance,” 187 is 

rcasonable and consistent with Title VII's provision that the 

“results” of a test may not be “used to discriminate.” 42 U.S.C. 
§2000e-2(h). If test scores do not vary directly with job 
performance, ranking the candidates on the basis of their scores 

will not select better employees. . . . 

[Cllose scrutiny is required because rank-ordering makes such a 
refined use of the test's basic power to distinguish between those 
who are qualified to perform the job and those who are not... . A 
test may have enough validity for making gross distinctions 
between those qualified and unqualified for a job, yet may be 
totally inadequate to yicld passing grades that show positive 
correlation with job performance. . .. 

  

1974); Western Addition Community Organization v. Alioto, 360 F. Supp. 733, 738 (N.D. 

Cal. 1973), appeal dismissed, 514 F.2d 542 (9th Cir.), cert. denied, 423 U.S. 1014 (1975). 

184 See Guardians Association v. Civil Service Commission of the City of New York, 
630 F.2d 79, 100-06 (2d Cir. 1980), cert. denied, 452 U.S. 940 (1981); United States v. City of 

Chicago, 631 F.2d 469, 476 (7th Cir. 1980); Firefighters Institute for Racial Equality v. City 

of St. Louis, 616 F.2d 350, 357-60 (8th Cir. 1980), cert. denied, 452 U.S. 938 (1981); Ensely 

Branch, NAACP v. Seibels, 616 F.2d 812, 822 (5th Cir.), cert. denied, 449 U.S. 1061 (1980); 

Brunet v. City of Columbus, E.P.D. para. 36, 498 (S.D. Ohio 1986); Walls v. Mississippi 

Department of Public Welfare, 542 F. Supp. 281, 313 (N.D. Miss. 1982), aff'd, 730 F.2d 306, 

320 (5th Cir. 1984); Vanguard Justice Society, Inc. v. Hughes, 592 F. Supp. 245, 266-69 (D. 

Md. 1984); Easley v. Anheuser-Busch, Inc., 572 F. Supp. 402, 409 (E.D. Mo., 1983); Burney 

v. City of Pawtucket, 559 F. Supp. 1089, 1103 (D. R. E. 1985) Berkman v. City of New York, 

536 F. Supp. 177, 210-12 (E.D. N.Y. 1982), aff'd, 795 F.2d 584 (2d Cir. 1983); Louisville Black 

Police Officers Organization, Inc. v. City of Louisville, 511 F. Supp. 825-39 (W.D. Ky. 

1979); Vanguard Justice Society, Inc., v. Hughes, 471 F. Supp. 670, 735, 741 (D. Md. 1979); 

Allen v. City of Mobile, 464 F. Supp. 433, 440-41 (S.D. Ala. 1978). 

185 B. Schlei and P. Grossman, supra note 9, at 155-56 (footnotes omitted). 

186 630 F.2d 79 (2d Cir. 1980), cert. denied, 452 U.S. 940 (1981). 

187 29 C.F.R. §1607.14C(9). 

  
  
 



  

Employment Testing and Title VII 119 

The frequency with which . . . one-point differentials are used for 
important decisions in our society, both in academic assessment 
and civil service employment, should not obscure their equally 
frequent lack of demonstrated significance. Rank-ordering 
satisfies a felt need for objectivity, but it does not necessarily 
select better job performers. In some circumstances the virtues of 
objectivity may justify the inherent artificiality of the substantively 
deficient distinctions being made. But when test scores have a 
disparate racial impact, an employer violates Title VII if he uses 
them in ways that lack significant relationship to job 
performance. 188 

Thus, although the court had concluded that there was sufficient evidence of 
validity to permit some use of the test, it held that the evidence was not 
adequate to satisfy the employer's “substantial task in demonstrating that 
rank-ordering is sufficiently justified to be used.” 189 

The court in Guardians IV also reviewed §5H of the Uniform Guidelines, 

which provides that cutoff scores “should normally be set so as to be 
reasonable and consistent with normal expectations of acceptable proficiency 
within the work force.”1%0 The court found that this provision 

also makes sense. No matter how valid the exam, it is the cutoff 

score that ultimately determines whether a person passes or fails. 

A cutoff score unrelated to job performance may well lead to the 
rejection of applicants who were fully capable of performing the 
job. When a cutoff score unrelated to job performance produces 

disparate racial results, Title VII is violated.191 

The employer in Guardians IV did not select a cutoff score on the basis 

of any relationship with job performance. Instead, like many civil service 

employers, it “merely chose as many candidates as it needed, and then set the 

cutoff score so that the remaining candidates would fail.”192 The court held 

that, since the cutoff score had an adverse impact on minorities and was not 

adequately justified, this practice was unlawful.193 

  

188 £30 F.2d at 100. 

189 630 F.2d at 103. Sce also Berkman v. City of New York, 536 F. Supp. 177, 211 (ED. 
N.Y. 1982), aff'd, 705 F.2d 584 (2d Cir. 1983) (evidence held insufficient to support the 
“extraordinary pretense at precision” reflected in rank ordering). 

190 29 CFR. §1607.511. 

191 630 F.2d at 105. 

192 14 

193 14 at 105-06. 

 



  

120 Patrick O. Patterson 

At present, the courts and enforcement agencies are nearly unanimous 

in adopting the view expressed in Guardians IV: Where cutoff scores or 

ranking cause or increase adverse impact, Title VII requires a substantial 

showing that their use is justified. However, the contrary position of many 

industrial and personnel psychologists, expressed in the Division 14 Principles, 

is likely to make this a disputed legal issue in the future. 

IV. CONCLUSION 

Psychologists and social scientists have long known that certain kinds of 

employment tests disproportionately screen out minoritics and women. With 

the enactment of Title VII of the Civil Rights Act of 1964, the law and lawyers 

began to recognize this serious social problem as a legal problem, and they 

began to develop legal strategies and principles to solve it. 

Through more than twenty years of interpretation and evolution, the 

courts and enforcement agencies have firmly established the basic principles 

of Title VII testing law. The most fundamental of those principles—repeatedly 

reaffirmed by the Supreme Court and incorporated in guidelines issued over 

the last two decades by all the federal enforcement agencies—is that, 

regardless of the presence or absence of discriminatory intent, it is unlawful to 

use a test or other selection procedure that has an adverse impact on 

minorities or women unless the selection procedure is shown to be job related. 

Many legal issues, however, remain unresolved. These issues include 

whether the Griggs “disparate impact” interpretation of Title VII applies to 

multicomponent selection systems and to subjective employment criteria; 

whether Title VII imposes a duty to investigate and adopt alternative selection 

procedures or uses having less adverse impact; whether Title VII's job- 

relatedness requirement may be satisfied by showing that test performance is 

related to training performance alone; whether the validity generalization 

concept will result in easing requirements for the transportability of validity 

evidence, and perhaps in reducing or eliminating other validation 

requirements; and whether Title VII requires employers to justify the use of 

cutoff scores or ranking where such methods of use cause or increase the 

adverse impact of an otherwise valid selection procedure. 194 As the courts and 

enforcement agencies continue to grapple with these and other issues, Title 

VII testing law will continue to evolve. 

  

194 since the writing of this chapter, the U.S. Supreme Court handed down a decision 

in Watson v. Fort Worth Bank and Trust (108 S. Ct. 2777 1988). While the full and 

lasting implications of the Court's decision have yet to be determined, it is clear that 
Title VII disparate impact analysis may now be applied to subjective employment 

criteria.—Ed. 

  
 



  

Non-Discriminatory Use 
of Personnel Tests 
CONFERENCE REMARKS 

Donald |. Schwartz! 

My remarks will cover three major areas. These are: (1) the development of 

government regulation of test use; (2) the influence of these regulations on 

testing practices; and (3) new issues in testing and how these issues may impact 

on the role of the government. 

GOVERNMENT REGULATION OF TEST USE: DEVELOPMENT 

The basic federal government policy concerning the nondiscriminatory use of 

tests and other selection procedures derives from Section 703(h) of Title VII of 

the Civil Rights Acts of 1964, as amended. This regulation provides that 

personnel decisions may be based on test scores, unless the test is designed, 

intended, or used to discriminate on the basis of race, sex, or ethnic group 

membership. This policy has been implemented by federal agencies in part 

through the issuance of a number of guidelines. The first set of testing 

guidelines, adopted by the Equal Employment Opportunity Commission 

(FEOC) in 1966, was primarily a general statement that tests which have an 

adverse impact are discriminatory unless they have been professionally 

validated. What constitutes “professional validation” was described in an 

attached policy statement prepared by three consultants to the EEOC, one of 

whom was an industrial psychologist. 

The second set of guidelines, adopted by the EEOC in 1970, was somewhat 

more detailed and incorporated by reference the professional testing 

standards published by the American Psychological Association (APA). In 

addition to the incorporated standards, the 1970 guidelines included a strong 

emphasis on job analysis, the need to avoid bias in criterion measures, and the 

need to validate tests separately for the different race, sex, and ethnic groups. 

In essence, these guidelines drew from professional standards those aspects 

which were perceived as important in avoiding discrimination against race, scx, 

or cthnic groups. 

  

Donald J. Schwartz is a staff psychologist with the United States Equal Employment 

Opportunity Commission (EEOC). These remarks were contributed by the author in his 

private capacity. No official support or endorsement by the EEOC or any other agency of 

the United States government is intended or should be inferred.