Petitioners Brief — Detroit Edison Co. v. National Labor Relations Board

Supreme Court brief1979

Ask Donna

What actually matters in this document.

Text

F |

JUN

F ‘ Supreme Court, U. S.

LED

10 1978

DAK, JR., CLERK

IN THE

Supreme Court of the United States

OcrosnerR Teno, 1977

No. 77-968

Tae Detroit Eptson Company, Petitioner

NationaL Lasor Resations Boarp, Respondent

BRIEF OF PETITIONER

JoHN A, MCGUINN

FarRMER, SHIBLEY, McGuinn & FLoop

1120 Connecticut Avenue, N.W.

Washineton, D.C. 20036

LEON S. COHAN

2000 Second Avenue

Detroit, Michigan 48226

Passs or Bryson S. ADAMs PRINTING, INC., WASHINGTON, D. C.

Ct ers oe

~ ane

TABLE OF CONTENTS

Page

SE IE so cu deuuvceeeueoscey sea vdsawes conus 1

PEE Zi Kevvedceriwnnenesccndicuadsakannne es 2

SURED TWUUIIIGED on cccccccce wc ctcccsccedesepes 2

Staturory Provisions INVOLVED ............0.ee0e0: 3

IE Weikds osc avdeddeuneeeedesenenetinedanas 3

NE i heed tele a hi ied ind pales 3

EEE Fh. Sidedescennsniences cbbiaeonsen 6

pS nT ne rT rey ree 8

The 1972 Collective Bargaining Negotiations .... 14

ID ich eekbeeteehacedadavsecrccnss 15

The Unfair Labor Practice Case ............... 20

The Decision of the Court of Appeals .......... 22

Dwersane OF ARGUMENT 2... cc ccccccescccccscccccs 24

FUP ET CT CET ETC OT ETT ECE TET OT eer TE TTOr? 30

I. Disclosure To The Union Of Job-Related Apti-

tude Tests, Test Scores Linked With The

Names Of The Applicants And Test Sheets Of

yy Would Destroy The Future Va-

Be Ge Se BED ve dodecsccvcdedoncdccse: 30

The Violation of EEOC Guidelines on Test-

ing Inherent in Disclosure of the Test Ma-

nc ccemneweuceseuess 31

2. The Inefficacy of the “Restriction” Ap-

proved by the Court of Appeals in Prevent-

ing the Dissemination of the Test Materials

to Unauthorized Persons ................ 39

ii Table of Contents—Continued

Page

II. Disclosure Of The Test Materials To The

Union Has No Probably Relevance Or Useful-

ness To The Union In Carrying Out Its Statu-

tory Duties And Responsibilities In Processing

The Promotion Grievances Through Arbitra-

BN eae 8 er ie de Se ae hed bad 43

1. The Irrelevance of the Tests ............. 44

2. The Irrelevance of the Test Scores Linked

with the Names of the Applicants ........ 49

I ss fe ol ee oh ee eae eecuedees c 57

TABLE OF CITATIONS

CaskEs:

Albemarle Paper Co. v. Moody, 422 U.S. 405 (1975). .25, 35

Amalgamated Utility Workers v. Consolidated Edison

ig Os OE CUUNE Sccrccccvenvicewcesessss 42

Davis v. Washington, 348 F.Supp. 15 (D.C. 1972), 512

F.2d 956 (D.C.Cir. 1975), reversed Washington v.

Davis 426 U.S. 229 (1976).............. 25, 35, 40, 41

Emporium Capwell Co. v. Western Addition Com-

munity Org., 420 U.S. 50 (1975) .............. 30, 57

General Electric Co. v. NLRB, 466 F.2d 1177 (6th Cir.

DD. peda dualesccusaks tases scenes sees 23, 53, 54

Griggs v. Duke Power, 401 U.S. 424 (1971) ...25, 32, 35, 37

International Telephone & Telegraph, Federal Divi-

sion, 22-CA-499, 46 LRRM 1387 (1960)........ 25, 38

Kirkland v. Department of Correctional Services, 520

ee Ge CE GE, SUED wicceestccovccsocs 25, 35, 36

Kroger Co. v. NLRB, 399 F.2d 455 (6th Cir. 1968) 23, 29, 55

Munn v. Illinois, 94 U.S. 113 (1877) ¢............... 37

NLRB v. Acme Industrial Company, 385 U.S. 432

BE re rere 27, 29, 30, 43, 56

NLRB v. Truitt Mfg. Co., 351 U.S. 149 (1956)... . 28, 54, 55

Southern Steamship Co. v. NLRB, 316 U.S. 31 (1942) 30, 57

United Steelworkers of America v. Warrior & Gulf

Nav. Co., 363 U.S. 574 (1960) .................. 56

Table of Citations—Continued iii

Page

Statutes & Reauiations:

United States Code, Title 29, §158(a)(5) ............ 3

United States Code, Title 42, § 2000e-2(h) .......... 2,3

MISCELLANEOUS:

American Psychological Association’s Code of Ethics

‘*Standards for Educational and Psychological

Tests and Manuals”, Standards I5 and J2...... 23, 24,

Psychological Aptitude Tests and the Duty to Supply

Information, 91 Harvard Law Review 869, (1978) 43

IN THE

Supreme Court of the United States

OctToBER TERM, 1977

No. 77-968

THE Detroit Epison Company, Petitioner

NaTIONAL LaBor RELATIONS Boarp, Respondent

BRIEF OF PETITIONER

OPINIONS BELOW

The opinion of the Court of Appeals, dated August

10, 1977, is reported at 560 F.2d 722, and appears in

the Appendix to the Petition for a Writ of Certiorari*

at pp. la-12a, The Court’s denial on November 22, 1977,

of the petitioner’s motion for rehearing appears in the

Petition Appendix at p. 13a. The decision and order of

the National Labor Relations Board, enforced by the

Court of Appeals, is reported at 218 NLRB No. 147

(1975), and appears in the Petition Appendix at pp.

14a-17a. The decision of the Administrative Law Judge

in this case appears in the Petition Appendix at pp.

18a-59a. The original opinion and award of the arbi-

’ The references to the Appendix to the Petition for Certiorari

will be designated as ‘‘P.A.’’ in this brief, while references to the

Appendix prepared after the petition was granted will be desi-

nated as ‘‘A’’.

trator in the related arbitration proceeding, dated De-

cember 3, 1973, appears in the Petition Appendix at

pp. 60a-76a, and the supplemental opinion and award of

the arbitrator in that proceeding, dated October 22,

1974, appears in the Petition Appendix at pp. 77a-87a.

JURISDICTION

The judgment of the Court of Appeals for the Sixth

Circuit was entered on August 10, 1977. A timely peti-

tion for rehearing en banc was denied on November 22,

1977, and the petition for certiorari was filed within

90 days of that date. This Court granted the p:»tition

on March 27, 1978. This Court’s jurisdiction is being

invoked under 28 U.S.C. § 1254(1).

QUESTIONS PRESENTED

1. Whether the judicial interpretation of Section

8(a) (5) of the National Labor Relations Act requiring

employers to furnish unions representing their employ-

ees with tests and test scores conflicts with Title VII of

the Civil Rights Act of 1964 (42 U.S.C. 2000e-2(h) )

and the guidelines of the Equal Employment Opportu-

nity Commission on testing (29 C.F.R. Part 1607),

which require, inter alia, that such tests and test scores

not be divulged to unauthorized persons.

2. Whether Section 8(a) (5) of the National Labor

Relations Act requires employers to furnish unions

representing their employees with copies of tests and

test scores used by employers in the selection of candi-

dates for jobs requiring the aptitude to learn specific

skills in cireumstances where (1) the employer gave or

offered to give the Union sufficient information to de-

termine the validity of the tests, (2) the uncontradicted

ame

expert testimony of the professional psychologists was

that disclosure of the actual test questions to the Union

would not be useful in determining the validity of the

tests, and (3) the arbitrator involved in the grievance

giving rise to this cise, in construing the applicable

collective bargaining agreement, found that production

of the actual test battery was irrelevant to the arbitra-

tion proceeding and award.

STATUTORY PROVISIONS INVOLVED

United States Code, Title 29:

‘Section 158(a) It shall be an unfair labor prac-

tice for an employer—

(5) To refuse to bargain collectively with the

representatives of his employees, subject to the

provisions of section 9(a).”’

United States Code, Title 42:

‘Section 2000e-2(h) ... nor shall it be an unlaw-

ful 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 dr action upon the results is not

designed, intended or used to discriminate because

of race, color, religion, sex or national origin.”’

The EEOC Guidelines on Employee Testing are

found at 29 C.F.R. Part 1607, and are reproduced in

the Petition Appendix at pp. 88a-100a.

STATEMENT

Background

The Detroit Edison Company is a public utility en-

gaged in the generation and distribution of electric

power in the State of Michigan (P.A. 19a). Since about

1943, the Utility Workers Union of America (Local

223), AFL-CIO, has represented certain employees of

the Company. At the time of the hearing in this case,

that Union represented employees in approximately

28 different appropriate units throughout the Company

(P.A. 19a). One of these 28 units was a unit of operat-

ing and maintenance employees at the Company’s then

newly constructed Monroe Power Plant at which the

Union was certified as exclusive bargaining agent on

April 1, 1971 (P.A. 19a-20a). The Company and Union

thereafter agreed that the Monroe Plant employees

would be covered by the terms and conditions of a pre-

exis*ing agreement between the parties effective from

June 16, 1969, to June 12, 1972 (P.A. 20a).

In late 1971, the Company determined that six posi-

tions in the classification of Instrument Man B should

be filled at the Monroe Power Plant. The job desecrip-

tion of the Instrument Man job reveals that an incum-

bent on this job ‘‘installs, maintains, repairs, calibrates,

tests and adjusts precision instruments in power

plants’’ (A. 358.) This includes automatic control sys-

tems which control boilers and their associated equip-

ment and the maintenance of both mechanical and elec-

trical tripping and alarm devices (A. 358-359). The

job description also states that, ‘‘ Fairly frequent op-

portunity to cause extremely serious operating condi-

tions from misjudgment or erroneous action exists”’

(A. 359-360). As Mr. Kanous testified, the Instrument

Man is a ‘“‘key”’ job in the plant and is ‘‘extremely im-

portant to the plant’s operation, both in staying in op-

eration and operating efficiently when it is in opera-

tion’’ (A. 210), and that he has the responsibility ‘‘to

actually keep that system on the line or put it back

. - eat a

essentially without assistance’’ (A. 212). He further

testified that the Instrument Man ‘‘can create situa-

tions which are extraordinarily hazardous to other em-

ployees or the equipment that he’s working on”’ (A.

310). For the foregoing reasons, as found by the Ad-

ministrative Law Judge, the Company considers this

to be a critical job ‘‘because it is involved with mainte-

nance of plant instrumentation, vital to the operation

of the plant... .’’ (P.A. 20a).’

The posted requirements for this job were (1) satis-

factory high school credits for 2 years of math and 1

year of science, (2) a satisfactory physical examina-

tion, (3) a minimum of ‘“‘recommended’’ in the re-

quired Instrument Man aptitude test administered by

the Company, and (4) a satisfactory attendance rec-

ord (P.A. 20a).

Ten Monroe plant employees bid for the job, but all

were rejected because they failed to achieve a ‘‘recoin-

mended’”’ score on the battery of aptitude exams. Over

30 employees from ether plants covered by the Union

contract were then considered. One incumbent instru-

ment man from another location was selected without

the necessity of ta cing the test, and the remaining five

positions were filed by the five most senior applicants

who achieved a ‘‘recommended”’ score on the aptitude

tests (P.A. 21a).

* Likewise, the arbitrator in the arbitration proceeding concern-

ing the Union’s grievance over the Company’s failure to promote

the Monroe Plant applicants, found that ‘‘There is no cuestion

that the Instrument Man job is one of critical importaive and

requires long training’’ (P.A. 71a).

The Grievance

On January 17, 1972, the Union filed a grievance

claiming a violation of Article VIII, Section 38 of the

collective bargaining agreement over the failure to

award the Instrument Man job to the Monroe Plant

bidders (A. 120). Article VIII, Section 38, of the

agreement provided that:

‘fa. If a vacancy occurs in any one of the bargain-

ing units covered by this Agreement and such va-

cancy is not filled by promotion or transfer within

the group, a notice of such vacancy shall be posted

for a period of five (5) days in the bargaining unit

in which it occurs and a copy sent to a Union di-

visional officer designated by the respective divi-

sion.

b. If such a vacancy is not filled by a member of

the bargaining unit, notice of the vacancy shall be

posted in other bargaining units of the same de-

partment for a period of five (5) days. If the rea-

sonable qualifications and abilities of the employ-

ees being considered as a result of this posting are

not significantly different, total length of service

in the Company shall govern. ‘Significant differ-

ence’ shail be ‘head and shoulders difference’, and

such factors as advance licenses or step-up experi-

ence shall not of themselves amount to significant

differences.’’ (A. 117-118).

The grievance was processed through each of the

contractually agreed upon steps of the grievance pro-

cedure (A. 141-154)

The Company’s final answer in the grievance proce-

dure, prior to the invocation by the Union of the right

of arbitration, denied the grievance, but ended with the

following recommendation:

= nme notre alll

aque et Or Pe ds

“Tt was evident that the Union does not have a

complete understanding of test validation, estab-

lishment of standards, development of test bat-

teries and similar information relating to the Com-

pany testing program. To that end, I am directing

that a meeting for Union Officers and Chairmen

be arranged to enhance their understanding of the

Company’s objectives in the testing area. A mu-

tually acceptable date will be worked out prior to

November 1, 1972.’’ (A. 153).

Thereafter the grievance continued toward arbitra-

tion after the local union president termed the offer

quoted above to be ‘‘completely unacceptable to the

Union”’ (A. 154). On March 5, 1973, the Union for the

first time made the following requests for information:

‘‘1. The actual battery of tests used in the cap-

tioned matter.

2. The method of scoring or grading and the ac-

tual criteria for finding ‘acceptability’ or ‘rec-

ommended’ or ‘Not Recommended’.

3. A report on the test validation.

4. A report by the National Compliance Com-

pany which, as I understand it, consulted

upon, and reviewed the Company’s whole test-

ing system or apparatus.”’ (A. 121).

Company counsel responded to this letter on March

15, 1973, arranging a meeting between Union repre-

sentatives and Company representatives to discuss and

explain the entire testing procedures, but declined to

release the actual test battery (A. 122).

Thereafter, on April 2, 1973, Company counsel (Mr.

Houghton), its director of union relations (Mr. Ruch)

and its industial psychologist (Dr. Roskind) met with

Mr. Lewis for the Union, explained the entire testing

program and procedures to him, and supplied him

with the Company’s validation study for the Instru-

ment Man B test battery (A. 343-353),° as well as a

validation study of the same test battery by an outside

consultant, the National Compliance Company (A. 304-

378).

The Company declined to give the Union the actual

test batteries, test papers or test scores of the appli-

cants who took the test.

Two days after the April 2, 1973, meeting with the

Company, the Union filed charges with the Labor

Board giving rise to this case (A. 104-106).

The Tests

The Company has used aptitude tests in an attempt

to predict job performance since the late 1920’s or

early 1930’s (A. 75, 178-179). These tests do not meus-

ure current knowledge and skills relevant to the job,

but rather the person’s ability to learn such knowledge

and skills. In 1958, a test battery was developed for the

job of Instrument Man (A. 75-76; 179). In the 1956-

1957 period, after the Company reviewed earlier studies

by industrial psychologists of what instrument men

actuall y do and checked with other sectors of industrial

experience where similar jobs were performed, it ap-

peared that there were four tests that had some possi-

bility of predicting job suecess of the Detroit Edi-

son Instrument Man. Out of these four, three were

$A test ‘‘battery’’ is two or more tests used in combination to

arrive at a selection decision (A. 224).

2 ln BERR rt 8 enters

picked for a validation study (A. 188-189).‘ These

were the Personnel Test, the Minnesota Paper Form

Board (MPFB) and the Engineering and Physical

Science Aptitude Test (EPSAT) (A. 189). At that

point, incumbent instrument men (1) were given the

three tests on an experimental basis and (2) independ-

ently ranked by their supervisors in terms of their

ability to do the various job functions of the Instru-

ment Man job. A positive correlation between demon-

strated ability to perform the Instrument Man job and

higher scores on the three tests was shown to exist

(A. 190).

In the 1969-1970 period, the Instrument Man bat-

tery of tests was again validated by the Company’s

industrial psychologists because ‘‘the technical engi-

neers of the power plants recognized the need for some

improvement in the selection test battery’? (A. 344).

The steps taken in that revalidation study were the

following:

1. The plant engineers were instructed to arrive

at a series of specific job functions of the Instrument

Man, which were a representative sample of the tasks

performed and which ranged from comparatively sim-

ple to complex (A. 233). Twenty-three (23) such tasks

were selected and they are specified in the validation

report (A. 352-353).

2. An evaluation procedure was devised whereby

each of the 43 incumbents on the job, who had taken

the Instrument Man battery of tests, would be rated

*A validation study is a systematic research study to determine

whether or not there is a relationship between test performance

and job performance.

by their supervisors on each of the 23 objective job

tasks as to whether each could perform the task either

(a) without assistance, (b) with minor assistance, or

3. Using these parameters, at least two super-

visors independently rated each of the incumbents on

each of the 23 items. Where there was no supervisory

consensus on the incumbent’s performance in the 23

areas being studied,*the results of such evaluations

were discarded. This happened in the case of ten of

the incumbents (A. 234-235; 346), leaving 33 incum-

bents in the study.

4. The performance figures were then correlated

with the test scores. Using the three tests, the multiple

correlation between tests and performance was +.73

(A. 347). The correlation coefficient measures how suc-

cessful the test battery is in predicting job success. If

the correlation were +1.0, the test would be perfectly

predictive of job success (the high scorers would all be

high performers). If the correlation were —1.0, the

test would be perfectly predictive in the opposite direc-

tion (the high scorers would all be poor performers).

If the‘correlation were 0, the test battery would be in-

dicative of neither success nor failure on the job. The

test scores would have no predictive value and their

relationship to either job success or failure would be

one of pure chance.

Taking the Personnel Test out of the battery, the

correlation coefficient was only reduced to +.71 (A.

347). Since it added so little to the validity of the test

battery, the Personnel Test was removed as a result of

the 1970 revalidation (Jbid).

The uncontradicted testimony of the professional in-

dustrial psychologists was that a correlation coefficient

of +.71 was unusually high in terms of predicting suc-

cessful job performance. Thus Doctor Roskind, the

Company’s Director of Industrial Psychology, testi-

fied that ‘‘you very, very seldom find a correlation of

a test ... anywhere near as high as .7’’ (A. 238). Mr.

Kanous, the Company’s Director of Em,'oyee Train-

ing, testified that a -+.71 correlation was ‘‘very high’’,

and that generally, useful correlation coefficients

would be in the +.3 to +.6 range (A. 206).

In addition, as a result of the 1970 revalidation, the

previous three-tiered scoring of (1) ‘‘not recommend-

ed’’, (2) ‘‘aeceptable’’, and (3) ‘‘recommended”’ was

dropped in favor of a two-tiered system of ‘‘not rec-

ommended’’ and ‘‘acceptable’’, and a slightly higher

cut-off score of 10.3 was determined to be appropriate

(P.A. 64a). The cut-off score of 10.3 was derived from

a scatter plot (A. 349) showing both supervisory eval-

uation of individuals in the study and their test scores.

As testified to by Doctor Roskind:

‘‘After you’ve placed all this information on the

scatter plot, what you do is place the cutoff score

so that it will operate most effectively and you

will make the least number of mistakes in predict-

ing poor performance and in predicting good per-

formers on the job’’ (A. 99).

The scatter plot showed, and the Arbitrator found,

that 13 of the 16 above-average perfcrmers on the job

scored 10.3 or better and that 14 of the 17 below aver-

age performers on the job scored less than 10.3 (P.A.

74a; A. 349). Stated another way, those scoring 10.3

or better had an approximately 80% chance of be-

coming successful instrument men, and those scoring

below 10.3 had an approximately 80% chance of be-

coming unsuccessful instrument men (A. 238).

The content of the Minnesota Paper Form Board

and the EPSAT was not changed in any way as a

result of the 1970 revalidation (A. 236).

The Minnesota Paper Form Board Test is a single

undivided aptitude test that has been used for many

years to predict the ability to visualize in three-dimen-

sional space (A. 193-195).

The Company introduced into the record of the arbi-

tration and of this case ten samples of the kind of ques-

tions appearing in the MPFB (A. 400-402).°

The EPSAT is an aptitude test which is divided into

six parts:

1. Mathematics. This measures the applicant’s fa-

cility with addition, subtraction, multiplication, divi-

sion, fractions, ratios and proportions, and a small

amount of algebra and geometry (A. 199). Examples

of the actual type of question posed in the mathematics

section of the EPSAT were introduced in the record

of the arbitration and of this case and can be found

at A. 384-385.

* These sample questions, as wel! as the sample questions depict-

ing the six part EPSAT Test, which the Company furnished the

Union, were not items taken from the actual tests, but rather from

@ special test, the Multi Aptitude Test (MAT). This test was

specifically developed for training psychology students and others

and is used to demonstrate the majority of questions which appear

in the MPFB and EPSAT. In cases where the MAT did not con-

tain questions similar to several sections of the EPSAT, the Com-

pany obtained additional questions from the public domain or

developed questions to illustrate the type of items found in the

actual test (A, 192-193 ; 215-216 ; 379-426).

: Ai

2. Formulation. This measures the facility with

which a person can take a verbal problem and express

it in a formula-type way (A. 199). Examples of the

actual type of question posed in the formulation sec-

tion of EPSAT were introduced in the record of the

arbitration and of this case at A. 403.

3. Physical science comprehension. This measures

aptitude for solving physics-type problems including

application of physical concepts such as reflectivity, re-

tention of heat, displacement and the like (A. 201).

Examples of the kind of question posed in this section

of EPSAT were introduced into the record of the arbi-

tration and of this case at A. 403.

4. Arithmetic reasoning. This measures aptitude

to reason in a deductive way (A. 202-203). Examples

of the kind of question posed in this section of EPSAT

were introduced in the arbitration and in this record

at A. 405-413.

5. Verbal comprehension. This measures the abil-

ity to use language effectively in a physical science

context (A. 203). Examples of the kind of question

posed in this section of EPSAT were introduced in

the arbitration and in this record at A. 380-381.

6. Mechanical comprehension. This measures the

ability to apply mechanical principles of levers, gears,

center of gravity, etc. (A. 203). Examples of the kind

of question pesed in this section of EPSAT were in-

troduced in the arbitration and in this record at A. 391-

393.

The Instrument Man battery of tests was again

studied—together with other job test batteries—in

1972 by an independent organization known as the

National Compliance Company (NCC). The conclu-

sion of the NCC study was that:

‘It has been reported above that the total test bat-

tery is statistically significantly related to the two

criterion variables for the job of Instrument Man.

The test battery appears to be effectively selecting

men who will perform better as Instrument Men’’

(A. 367)

The Company administers the tests to applicants

with the express commitment that each applicant’s test

scores will remain confidential (A. 85, 445). The Com-

pany insures test and test score security by keeping

the tests and test scores under lock and key in the

offices of the Company’s professional industrial psy-

chologists who have an ethical commitment under the

American Psychological Association’s Code of Ethics °

not to disclose the tests or test scores to unauthorized

persons (A. 82-83, 438). Under this policy the Com-

pany’s industrial psychologists do not disclose tests or

report actual test scores even to members of manage-

ment (A. 77, 83, 91, 127).

The 1972 Collective Bargaining Negotiations

The Company and Union engaged in collective bar-

gaining negotiations in 1972 to negotiate a successor

agreement to the 1969 agreement. At the second meet-

ing in these negotiations on April 20, 1972—exactly

four months after the Union filed the grievance con-

cerning the Instrument Man test—the Union proposed

the following to the Company:

*The APA Code of Ethics is expressly incorporated in the

statutory codes of over twenty states, including the State of Michi-

wan. Mich. Stat. Ann, § 14.677(1)(b) (1976).

‘‘The Union shall be furnished copies of all tests

given to employees. Union shall have the right to

review all test results. All tests shall be job re-

lated and only for the job or position applicable”’

(A. 164).

The Company rejected this proposal on May 9, 1972,

and advised the Union that acceptance of its proposal

‘‘would destroy the testing program’’ (A. 165). On

May 18, 1972, when the subject came up again, the

Company again rejected the Union’s demand for the

tests and test scores on the basis that nothing in the

contract gave the Union a right to access to this infor-

mation, but offered to have its industrial psyeholo-

gists discuss with the Union the integrity of its testing

program (A. 168). The Union responded that it was

interested in hearing the Company’s explanation of

its testing program after the conclusion of negotia-

tions. There was no change made in the 1972 contract

as a result of the Union’s demand for the tests and test

scores. The Company did hold a meeting with the Union

in the Fall of 1972 to explain its testing program in

accordance with the contractual commitment that it

made in the negotiations (A. 168-169).

The Arbitration

After filing the charge with the Labor Board lead-

ing to this ease on April 4, 1973, the Union in prepara-

tion for the arbitration sent a memo to the Arbitrator

on May 23, 1973, requesting the Arbitrator to direct

the Company to furnish the Union with copies of the

actual tests, or to subpoena the tests (A. 123-124).

On the same day the hearing opened before Arbitra-

tor Jones. The original arbitration hearing took place

on May 23, 24, 30 and 31, 1973.

In the course of this hearing, as noted in part above,

the Company furnished the following information:

1. Its 1970 validation study.

2. The 1972 NCC validation study.

3. Explanations of the MPFB and each of the six

parts of EPSAT.

4. Representative sample questions showing the

type of question appearing on the MPFB and on each

part of the six-part EPSAT.

Also during the course of the five-day hearing, the

Company offered to turn over to the Union test scores

of any employee who would sign a waiver releasing the

Company psychologist from his pledge of confidential-

ity (A. 7).’ Finally, the Company disclosed test scores

of applicants taking the test battery without tying the

score to the applicant’s name (A. 279-280).

Nevertheless, on June 2, 1973, two days after the

close of the original hearing, Mr. Lewis for the Union

wrote the Company stating:

‘Inasmuch as I am writing you, and after much

deliberation, I find I cannot withdraw my request

for the actual tests, scores, and weights. While the

Company has made samples for me, and I appre-

ciate that, and while it gave scores without re-

vealing names, I have the problem of knowing the

weights in the 6 parts of the EPSAT test—the

total weight of EPSAT—then the weight of Min-

nesota Form Board—then how the various parts

fit together to form the possible apex of 15.3 in

"The Union flatly refused to determine whether the employees

it represented would want their raw scores divulged to the Union

(A. 44).

ee ee

the upper limit, quite apart from the 10.3 cut off

seore.’’ (A. 125-126).

This request was answered by the Company on July

10, 1973, in which the Company responded affirma-

tively and in detail to the Union’s requests concerning

the battery weights and scoring mechanisms while

again declining to give the Union the test battery and

the scores linked with the names of the examinees (A.

127-131).

After the close of the initial hearing, the Arbitrator

ruled on the Union’s request and held that he lacked

the power to require the Company to produce a copy

of the tests and the scores achieved by applicants taking

the tests (P.A. 67a). At the same time, the Arbitrator

invited the Union to produce case citations demon-

strating that such information should be furnished, but

the Union declined to do so (Ibid.).

In view of the dual pendency of both the Union’s

charge before the Labor Board and the case before the

Arbitrator, the Company and Union on August 6, 1973,

jointly advised the Arbitrator of the situation, and

concluded that the arbitration proceeding should not

await a final decision by the Board or its enforcement

by a Court of Appeals (A. 138-139).

The Arbitrator thereafter closed the hearing in the

ease on August 18, 1973, and rendered his decision on

December 3, 1973, finding that the Company did not

violate the Agreement by establishing an acceptable

score on the tests as a qualification for the job, but di-

recting the Company to re-examine employees with

scores between 9.3 and 10.3 to determine whether their

lower scores were offset by relevant educational or job

experience (P.A. 76a).

In compliance with the award, the Company re-

examined three of the Monroe applicants who had

scored between 9.3 and 10.3 on the tests and subse-

quently promoted one such employee who had a back-

ground which might tend to offset his failure to achieve

the 10.3 score, but again rejected two others (P.A.

39a). The Union protested the failure to promote the

Monroe applicants who were rejected and a supple-

mental hearing was held before the same Arbitrator

on July 18, 1974, to determine whether the Company

had complied with the original award. The transcript

of that hearing is also included in the record in this

case (A. 331-342). In a supplemental decision the Arbi-

trator found that the Company had complied with the

award (P.A. 77a-87a).

In the course of the Arbitrator’s original decision

based on four days of hearing on the issue of testing,

he made several findings not disputed by the Admini-

strative Law Judge, the Board, or the Court of Ap-

peals, which are relevant to the disposition of this case:

1. He found that the Company had the right under

the contract to establish minimum reasonable qualifi-

cations for the job of Instrument Man (P.A. 70a-71a).

2. He found that:

‘*There is no doubt that the Company has estab-

lished and maintained the right to use tests as a

measure of an employee’s qualifications—the

Union concedes this point.’’ (P.A. 71a).

3. He found that:

‘(T]he test [battery] is what the Company claims

it to be—a reliable valid test with a correlation

coefficient of .72. There is also no question that it

is a ‘fair test’ in the sense that its administration

and scoring have been standardized.’’ (P.A. 72a-

73a).

4. He found that:

‘*There is no doubt that as tests go, this particular

test has a high degree of validity. The way in which

it was developed and is used also overcomes many

of the objections raised by arbitrators and indus-

trial relations practitioners; that is, the test was

developed especially for this job (job reiated), and

has been properly validated. In addition, and very

important, the instant test is not used to compare

one employee against another in terms of scores

attained . . . This problem is eliminated by re-

porting scores in terms of ‘not reconumended’ and

‘accepted’.’’ (P.A. 73a).

5. Addressing himself particularly to the Union’s

attempt to attack the test by attacking individual sam-

ple questions or items on the test as being themselves

not related to the work of the Instrument Man, the

Arbitrator stated:

‘*The Union questioned the validity of the test by

reference to the type of questions asked. It was to

further this line of questioning that the Union de-

sired a copy of the test. But this was simply a

questioning of the face validity of the test. Such

questions, even if taken from the test itself, prove

nothing. Face validity does not prove or disprove

the validity of the test in determining the aptitudes

necessary for successful job performance.’ (P.A.

72a).

6. Additionally and most significantly, he found

that:

‘**In short, the Arbitrator does not believe that the

Union’s position was damaged in any way by lack

of access to the test’’ (P.A. 72a).

In conclusion, the Arbitrator found that the attain-

ment of a 10.3 score created a presumption of ‘‘signifi-

eant difference’ as provided for in Article VIII, Sec-

tion 38, of the parties’ agreement, over those failing to

obtain such a score unless those scoring between 9.3

and 10.3 had relevant educational or job experience

which might tend to offset their failure to achieve the

10.3 score.

The Unfair Labor Practice Case

The hearing in the unfair labor practice case took

place on September 23, 1974. As noted above, the en-

tire 690-page transcript of the original arbitration

case, as well as the transcript of the supplemental hear-

ing before the Arbitrator, were introduced in the rec-

ord before the Administrative Law Judge.

In the proceeding before the Administrative Law

Judge the Company made two further and significant

efforts to be responsive to the Union’s requests and the

General Counsel’s Complaint. At the outset of that

hearing, the Company’s counsel stated:

‘The Company has offered to do two things to

appease the Union in this case. One, our position

is we would be willing to disclose these tests to a

qualified industrial psychologist for his perusal,

for a determination on behalf of the Union. We

have also indicated to Mr. Lewis, the spokesman

for the Union, that we would permit him to take

the test himself’’ (A. 6)

In addition, at the hearing two professional psycholo-

gist witnesses—one of them an outside consultant, Dr.

Marvin Dunnette—testified without contradiction that

disclosure of the actual items on the test battery to the

Ein sat time a

Union would not be useful in determining the validity

or fairness of the test battery (A. 50-51, 71-72, 79).

The Board’s Administrative Law Judge in a deci-

sion rendered on January 29, 1975, ordered that the

battery of tests and actual test sheets not be disclosed

to the Union directly, but rather to a qualified psychol-

ogist of the Union’s choice who would be free to con-

sult with the Union and show the Union the tests and

test sheets to the extent necessary to process a Union

grievance over promotion to the Instrument Man classi-

fication, but not to permit the Union to copy the test

or otherwise disclose the test to those who may have

taken the test in the past or might take it in the future.

He further ordered that the Company disclose to the

Union directly the test scores achieved by each spe-

cifically named employee to whom the test battery was

administered.

Inasmuch as the Company at the outset of the unfair

labor practice hearing had offered to show the battery

of tests to a qualified psychologist of the Union’s choice

‘to appease’’ the Union, the Company did not except

to that portion of the decision of the Administrative

Law Judge, but did except to his proposed order that

the raw test scores of employees to whom the battery

of tests had been administered be turned over directly

to the Union. The Union excepted to the requirement

that the test battery be disclosed to a qualified psychol-

ogist of the Union’s choice rather than to the Union

directly.

The Board in a two to one decision upheld the Union

exception and ordered that all the test materials in-

volved be disclosed to the Union directly. The Board

concluded that:

‘*While we are making the tests available directly

to the Union, we shall, in order to preserve their

future utility, impose the same restrictions upon

their use by the-Union as recommended by the

Administrative Law Judge’’ (P.A. 16a).°

Former Member Kennedy would have affirmed the de-

cision of the Administrative Law Judge in its entirety,

noting that:

‘There is no professional obligation on the part

of the Union not to publicize the tests or their re-

sults. I do not see how the Board can enforce its

exhortation not to copy or disclose the tests’’

(P.A. 17a).

The Decision of the Court of Appeals

The majority decision of the panel of the Court of

Appeals enforced the Board’s order in all respects.

In answer to the Company’s concern about dissemi-

nation of the tests by turning them over to the Union

directly, the majority merely referred to the provision

* Specifically, the restrictions imposed by the Administrative

Law Judge were the following:

‘(T]he Union shall have the right to see and study the tests,

and to use the tests and the information contained therein to

the extent necessary to process and arbitrate the grievances,

but not to copy the tests or otherwise use them, for the purpose

of disclosing the tests or the questions to employees who have

in the past or may in the future take these tests, or to anyone

(other than the arbitrator) who may advise the employees of

the contents of the tests. After the conclusion of the arbitration

proceeding, or if no request is made to reopen the arbitration

hearing within 90 days after the psychologist receives the

battery of tests, all copies of the battery of tests shall be

returned to Respondent’’ (P.A, 55a-56a),

Cteweiwe 8 et ee + et nn,

in the Board decision quoted above purporting to im-

pose a restriction on the dissemination of the test ma-

terials. In answer to the Company’s concern that turn-

ing raw test scores, linked with the names of the persons

taking the tests, and actual test papers of the appli-

eants directly to the Union would involve the Company

in a breach of an express pledge given to each of the

persons taking the tests that the results of the test

would not be disseminated, and subject such persons to

needless embarrassment, humiliation and harassment,

the majority of the Court simply cited its decision in

General Electric Co. v. NLRB, 466 F.2d 1177 (6th Cir.

1972), holding that wage information received by the

Company from other companies on a confidential basis

and used as a part of a wage survey to determine rates

of pay for employees of the Company must be disclosed

to the Union.

Finally, in answer to the concern of both the Com-

pany and the American Psychologicai Association

(herein referred to as APA) as amicus curiae that dis-

closure directly to the Union of the test materials would

involve the Company psychologists in a violation of

their professional ethical code, the majority of the

Court simply found that ‘‘the principles which under-

lie the National Labor Relations Act are paramount in

this case’’ (P.A. 8a).

Circuit Judge Weick filed a dissenting opinion in

which he concluded that the Board had not followed the

standard used in Kroger Co. v. NLRB, 399 F.2d 455

(6th Cir. 1968), of recognizing and adequately protect-

ing each of the conflicting interests involved. He found

that the Board order recognized only the interest of the

Union and that its order ‘‘constituted a gross abuse of

discretion’’ (P.A. 12a).

Judge Weick pointed out the several Company ef-

forts to accommodate the Union’s interest in the Com-

pany’s test program and concluded that such informa-

tion was ‘sufficient to permit the union to process ade-

quately the grievance before the Arbitrator, or to per-

form its duties under the collective bargaining agree-

ment’’ (P.A. lla). To the majority opinion’s claim that

dissemination of the test materials was adequately pro-

tected by the Board’s Order, Judge Weick found such

reasoning to be-‘‘really naive’’ and quoted, with ap-

proval, former Member Kennedv’s analysis in support

of this conclusion (P.A. lla).

In addition, Judge Weick found that disclosure of

the test materials would involve the Company psycholo-

gists in a violation of the Code of Ethies of the APA,

which has been recognized by the statutes of the State

of Michigan, and that disclosure of the test scores and

test papers linked to the name of the person taking the

test constituted an invasion of a confidential and priv-

ileged relationship between the psychologists and the

examinees.

SUMMARY OF ARGUMENT

1. Disclosure to the Union of the Test Materials Would Destroy

the Validity and Future Usefulness of the Tests.

The disciosure of tests and test scores to unauthor-

ized, non-professional persons is in contravention of

Standards I5 (tests) and J2 (test scores) of the Ameri-

ean Pychological Association’s ‘‘Standards for Edu-

cational and Psychological Tests and Manuals’’, which

prohibit disclosure of aptitude tests and test scores to

unauthorized persons because prior knowledge of spe-

cific test items by individuals who are to take the test

in the future destroys the validity of the test. These

)

A re BM PN pl Ne tet

APA standards have in turn been incorporated into

the Equal Employment Opportunity Commission’s

‘*Guidelines on Employee Selection Procedures’’ pro-

mulgated under Section 703(h) of the Civil Rights Act

of 1964, which permits employers to give and act upon

the results of professionally developed ability tests,

provided that the tests are not used to discriminate.

The EEOC Guidelines on Testing, and particularly

their incorporation of the Ai’A Standards have, in

turn, received the explicit approval of this Court on at

least three occasions. Griggs v. Duke Power Co., 401

U.S. 424, 433-34 (1971); Albemarle Paper Co. V.

Moody, 422 U.S. 405, 430-31 (1975); Washington v.

Davis, 426 U.S. 229, 250, fn. 16 (1976).

The Court of Appeals for the Second Circuit has held

that production of aptitude tests to members of an af-

fected class of minority employees who would be taking

the test in the fviure in competition with others would

not be approved because advance inspection of the tests

by the class members would place all others at a com-

petitive disadvantage. Kirkland v. Department of

Correctional Services, 520 F.2d 420 (2d Cir. 1975).

Because of the ‘‘unique character’’ of aptitude tests,

the Labor Board in its forty-three year history of ad-

ministration and interpretation of Section 8(a) (5) has

never before required employers to furnish unions with

test questions, and on at least one occasion its General

Counsel has even refused to issue a complaint seeking

the production of test questions on the basis that ad-

vance inspection of the test would permit their content

to be widely disseminated and thus impair the useful-

ness of the test. International Telephone & Telegraph,

Federal Division, Case No. SR 657, 46 LRRM 1387

(1960).

The invention by the Labor Board of a ‘‘restriction’’

seeking to limit the Union’s dissemination of the tests

to other employees who may take the tests in the future

is unworkable and unenforceable. Once control of the

tests passes to a person or persons with no professional

obligation against disclosing the tests to others, the

security and thus the validity of the tests have been

hopelessly compromised in accordance with the APA

Standards. The ‘‘restriction’’ places an impossible bur-

den on the Company of seeking to police the ‘‘restric-

tion’’. Even if the Company were able to detect a de-

liberate violation of the ‘‘restriction’’, it has no inde-

pendent right to seek contempt proceedings in the

Court of Appeals since that right resides solely with

the Labor Board. It is also highly doubtful whether a

Union which violates the ‘‘restriction’’ could be found

in contempt at all since the order of the Labor Board

runs against the Company and not the Union, which

was not even a party to the enforcement proceeding in

the Court of Appeals. Even if the Company were able

to detect a deliberate violation of the ‘‘restriction’’ and

even if the ‘‘restriction’’ were in some way enforce-

able against the Union, there is no adequate remedy

for the violation since the Company’s tests would be

invalidated and it would take years to validate a new

test battery.

Even if the Union did not deliberately or inadvert-

ently violate the ‘‘restriction’’, the disclosure of the

tests to ‘‘the Union’’ invalidates the tests since ‘‘the

Union’’—however that term is construed—consists of

employees who change on a regular basis due te Union

elections and who well might take the tests in the fu-

ture. Additionally, disclosure of the test questions cculd

result without a violation of the ‘‘restriction’’ from a

ee ee

public arbitration or court proceeding in which the

questions were revealed.

For ail these reasons it is simply naive, as dissenting

Judge Weick found, to believe that a hortatory ‘‘re-

striction’’ will prevent intentional or inadvertent dis-

closure of the tests to those who might take the tests in

the future.

2. Disclosure to ihe Union of the Test Materials Is of No Probable

Relevance or Usefulness to the Union in Pursuing Its Statutory

Functions.

Even under the broad ‘‘discovery type standard’”’ ap-

proved by this Court in NLRB v. Acme Industrial

Company, 385 U.S. 4382 (1967), the Company had no

obligation to furnish the Union the test battery and

test scores linked to the names of the examinees, be-

cause the materials have no probable relevance or use-

fulness to the Union.

In the course of this proceeding the Company fur-

nished the Union with what dissenting Judge Weick

found was a ‘‘wealth of material’’ on its tesung pro-

gram. This included furnishing the Union with two

validation studies done on the test battery, written ex-

planations of the tests and scoring weights of the

tests, generous samples of test questions on each of

the tests and subparts of the EPSAT Test, and the

test scores of all applicants without linking the appli-

cant’s name to the score. In addition, the Company

offered to disclose to the Union the score of any appli-

cant willing to have his score divulged and to disclose

the test battery to a qualified psychologist of the

Union’s choice.

In light of the above information, the three profes-

sional psychologists testifying in this case concluded

without contradiction that disclosing the test battery

itself to the Union would not be relevant or useful to

the Union in determining the validity of the test bat-

tery. In addition, the Arbitrator who decided the un-

derlying arbitration case concluded that providing the

Union with the actual test battery would ‘‘prove noth-

ing’’ and that ‘‘the Union’s position was not damaged

in any way by lack of access to the test’’.

With respect to the test seores linked with the name

of the applicant, there was absolutely no evidence in

the record that such information was of any probable

relevance or usefulness to the Union. Contrary to the

Board’s finding, the Arbitrator did not conclude that

the Union needed all the scores linked with the names

of all the applicants to police the agreement. Rather,

he concluded that the Company had discharged its

obligation by disclosing to the Union the names of the

three applicants who had scored between 9.3 and 10.3.

There are, moreover, compelling reasons against dis-

closure of test scores. All test applicants are given an

express pledge of test score confidentiality before they

take the test and disclosure of actual test scores would

not only breach that pledge but would subject those

scoring low on the test battery to needless embarrass-

ment and humiliation. In addition, there is an emerging

Congressional recognition and protection of the right

of individuals against disclosure of private and per-

sonal information such as their aptitude test scores.

3. A Proper Balance of the Conflicting Interests Involved, as Well

as Overriding Principles of Federal Labor Law. Require That

the Test Materials Not Be Disclosed to the Union.

The Court of Appeals failed to apply the principle

established by this Court in NLRB v. Truitt Mfg. Co.,

ite

ne SRE Plan.

0. ed Maced etn te

a es

ee ee

351 U.S. 149, 153-54 (1956), that ‘‘the inquiry must

always be whether or not under the circumstances of

the particular case the statutory obligation to bargain

in good faith has been met’’, as well as the principle

established by this same Court of Appeals itself in

Kroger Co. v. NLRB, 399 F.2d 455, 457 (6th Cir. 1968),

that ‘‘the critical problem appears to be how to recog-

nize and how adequately to protect each of the con-

flicting interests that are involved here’’.

In the present case, if there were any real analysis

of ‘‘the circumstances of the particular case’’ or any

real balancing of the ‘‘conflicting interests’’ involved,

it would have become obvious that in balancing the lack

of probable relevance or usefulness of the test materials

to the Union with the probability of invalidation of

the Company’s entire testing program by disclosure

of the materials that the balance in the circumstances

of this particular case should have been struck in the

Company’s favor.

Two other policies of federal labor le‘islation sup-

port this balance. First, the policy fav cing final and

binding resolution of contract dispute : by arbitration

as set out by this Court in the Steelworkers Trilogy is

served by not disclosing the test materials since the

Arbitrator ruled that the materials were irrelevant

to his resolution of the dispute. Thus, far from aiding

the arbitral process, as was found to be the case with

the information sought by the Union in NLRB v. Acme

Industrial Company, 385 U.S. 432 (1967), the informa-

tion in the present case has already been found to be

irrelevant to the arbitration process. Secondly, this

Court has had previous occasion to remind the Labor

Board that it should construe the Act in a manner con-

sistent with otl.er equally important Congressional ob-

jectives and should not pursue the policies of the Act

so single-mindedly that it conflicts with other legisla-

tion. Southern Steamship Co. v. NLRB, 316 US. 31,

47 (1942); Emporium Capwell Co. v. Western Addi-

tion Community Org., 420 U.S. 50, 66 (1975). The

novel construction of Section 8(a) (5) of the Act by the

Labor Board here clearly conflicts with the equally im-

portant Congressional objective of assuring that tests

used in job selection decisions be valid and nondis-

eriminatory.

ARGUMENT

Disclosure To The Union Of JobRelated Aptitude Tests, Test

Scores Linked With The Name Of The Applicants And Test

Sheets Of Applicants Would Destroy The Future Validity Of The

Tests

The argument made in this section of the brief as-

sumes, solely for purposes of argument, that disclo-

sure of the test materials to the Union in connection

with the arbitration proceeding had some cegree of

probable relevance and usefulness to the Union in

carrying out its statutory obligations, and would, thus,

arguably be subject to production under the ‘‘discov-

ery-type standard’’ enunciated in NLRB v. Acme In-

dustrial Co., 385 U.S. 432 (1967), were it not for the

unique character of the test materials involved and the

Company’s inability to continue a valid testing pro-

gram under Title VII of the Civil Rights Act of 1964,

while at the same time disclosing the test materials to

the Union. In Part II of this Argument, the issue of

the irrelevance and lack of usefulness to the Union of

the test materials at issue here will be addressed under

the standard set by this Court in the Acme case.

eS ae

1. THE VIOLATION OF EEOC GuipeLines on TESTING

INHERENT IN DISCLOSURE OF THE TEST MATERIALS

TO THE UNION

As shown in the petition for certiorari in this case,

testing is a major employee selection device in both

American industry and in the federal government it-

self (Petition, pp. 9-10). Disclosure to unions of tests

and test sheets of named applicants will result in a

widespread invalidation of testing programs as a selec-

tion device for applicants for employment as well as

for selection of existing employees to fill jobs of ad-

mittedly critical importance, as is the case here. As dis-

senting Judge Weick correctly observed, ‘‘It’s [the

Company’s] testing program has been nullified by the

action of the Board’’ (P.A. 1la-12a).

This invalidation occurs because disclosure of the

test materials to ‘‘the Union’’ naturally leads to dis-

closure and, therefore, foreknowledge of the actual

items on the tests by some or all of the persons who

will take the tests in the future, as will be demonstrated

shortly in this brief.

Assuming, however, for the moment, the premise of

foreknowledge of the actual test items by those who are

to take the test in the future by disclosure of the tests

to ‘‘the Union’’, the conclusion of the practical and

legal invalidation of the testing program is easily es-

tablished and was not really controverted by the Board

or the Court of Appeals.

As a practical matter, it is evident that such fore-

knowledge will result in a situation where the high

scorers on the test will reflect not the aptitude of the

examinees to perform the job, but rather the extent of

their familiarity with the items on the test and their

ability to memorize the ‘‘right’’ answers.

As a legal matter, such foreknowledge invalidates

the test battery under civil rights laws, particularly

Title VII of the Civil Rights Act of 1964. Section

703(h) of that Act permits employers to give and act

upon the results of professionally developed ability

tests, provided that the test or its administration is

not designed, intended or used to discriminate. Under

the mandate of Section 703(h), the EEOC has de-

veloped ‘‘Guidelines on Employee Selection Proce-

dures’’,’ which begin with the proposition that:

‘‘properly validated and standardized employee

selection procedures can significantly contribute

to the implemen. ation of nondiscriminatory per-

sonnel policies ....’’ 29 C.F.R. § 1607.1 (P.A. 88a-

89a).

This Court in Griggs v. Duke Power Co., 401 US.

424 (1971), approved this general philosophy on test-

ing by noting that:

‘*Nothing in the Act precludes the use of testing

or measuring procedures; obviously they are use-

ful. ... Far from disparaging job qualifications

*On December 30, 1977, the EEOC, Civil Service Commission,

Department of Justice and Department of Labor jointly issued a

notice of proposed rulemaking or ‘‘Proposed Uniform Federal

Guidelines on Employee Procedures’’, which are still in the rule

making process. The proposed guidelines state that:

‘‘These guidelines are based upon, combine, and supersede all

previously issued guidelines on employee selection procedures.

These guidelines have been built upon court decisions, the

previously issued guidelines of the agencies, the practical ex-

perience of the agencies, as well as standards of the psycho-

logical profession. These guidelines are intended to be con-

sistent with existing law’’ See. 1(c).

eee

as such, Congress has made such qualifications the

controlling factor, so that race, religion, nation-

ality, and sex become irrelevant. What Congress

has commanded is that any tests used must meas-

ure the person for the job and not the person in

the abstract’’ 401 U.S. at 436. (Emphasis added).

Under the heading of ‘‘Minimum Standards for

Validation”’ of tests, the Guidelines state that evidence

in support of a test’s validity must be based ‘‘on studies

employing generally accepted procedures for determin-

ing criterion related validity, such as those described

in ‘Standards for Educational and Psychological Tests

and Manuals’ published by the American Psychological

Association . . .”’? 29 C.F.R. § 1607.5 (P.A. 92a-93a).”

Standard I5 of the APA Standards referred to in

the EEOC Guidelines cited above and the proposed

joint agency guidelines states the following:

‘*15.The test user shares with the test developer

or distributor a responsibility for maintaining test

security.

Comment: Test security is a problem whenever

a lapse in security can result in changing an indi-

vidual’s score without making a change in his true

‘©The proposed joint agency guidelines referred to above state

the same proposition in the following manner in the section entitled

‘‘General Standards for Validity Studies’’ :

‘‘These guidelines are intended to be consistent with gener-

ally accepted professional standards for evaluating standard-

ized tests and other selection procedures such as those de-

scribed in the Standards for Educational and Psychological

Tests prepared by a joint committee of the American Psycho-

logical Association, the American Educational Research As-

sociation and the National Council on Measurement in Educa-

tion (American Ps; chological Association, Washington, D.C.,

1974 (hereafter ‘APA Standards’)) and standard textbooks

and journals in the field of personnel selection’’. Sec. 5(c).

score. For some kinds of tests a lapse of security

would not be serious. If one is to be tested for

achieved skill, for example, knowing and practic-

ing the test samples might be highly recommended.

In many cases, however, prior knowledge of test

items or scoring procedures could destroy validity.

The problem is not simply one of cheating. Secur-

ity may be compromised where examinees have had

much prior experience with a popular test, have

been taught specific test items, or have heard q lot

about the test’’. Standards for Educational & Psy-

chological Tests, published by the American Psy-

chological Association (1974). z. €7.

It is undisputed on this record that the test battery in

issue is not a test for achieved skill, but rather an ap-

titude test where ‘‘prior knowledge of test items could

destroy validity.’’ It is also undisputed that the indi-

vidual questions on the In..rument Man test battery

are identical each time the test is administered and that

there are no variations or substitutions of test items

(A. 78).

Standard J2 of the same APA Standards cited in the

EEOC and proposed joint agency guidelines states the

following :

‘“‘J2. Test scores should ordinarily be reported

only to people who are qualified to interpret them.

If scores are reported, they should be accompanied

by explanations sufficient for the recipient to in-

terpret them correctly.

Comment: There are difficult problems associ-

ated with the question of who should have access to

test scores within an organization. Certainly, curi-

ous peers should not have access to them. An indi-

vidual who must make the ultimate decision to ad-

mit or to reject or to hire or to reject, or to certify

or not to certify, must have the interpretation. One

useful (and unanswered) question is whether such

a person who lacks the training necessary for the

interpretation of scores should be given that train-

ing or should be given only the interpretations of

seores.’’ Id. at p. 68.

By turning over the test scores to unions and thereby

to the employees represented by the unions, access to

the test scores is thereby given to ‘‘curious peers’’.

In addition, the EEOC Guidelines on testing ex-

pressly provide for test score security. Section 1607.5

(b) (2) states that:

‘*Tests must be administered and scored under con-

trolled and standardized conditions, with proper

safeguards to protect the security of test scores

and to insure that scores do not enter into any

judgments of employee adequacy that are to be

used as criterion measures’’ (P.A. 94).

This Court on at least three recent occasions has held

that the EEOC Guidelines on Employee Testing are

entitled to ‘‘great deference’’ and to be regarded as

“expressing the will of Congress.’’ Griggs v. Duke

Power Co., 401 U.S. 424, 433-34 (1971); Albemarle

Paper Co. v. Moody, 422 U.S. 405, 430-31 (1975);

Washington v. Davis, 426 U.S. 229, 250 n. 16 (1976).

In the Albemarle case the Court expressly noted with

approval that the EEOC Guidelines ‘‘draw upon and

make reference to professional standards of test vali-

dation established by the American Psychological

Association’’, 422 U.S. at 431.

One Court of Appeals already has held that it is in-

appropriate to disclose tests to employees who may take

the tests in the future in competition with others. In

Kirkland v. Department of Correctional Services, 520

F.2d 420 (2d Cir. 1975), a class of minority applicants

for promotion to a position within the New York State

Department of Correctional Services, filed a civil rights

complaint under 42 U.S.C. 1981 and 1983 challenging,

inter alia, a State Civil Service examination which the

State used to determine the successful applicants for

the job. The Second Circuit agreed with the District

Court that the examination was not job related and had

a discriminatory impact on minorities, but reversed the

District Court’s order requiring that any new test for

the position in question be furnished by the State to

the minority applicants for their prior review. The

Court held:

‘‘The District Court ordered that the new test

prepared by defendants be submitted to the plain-

tiffs for review. We find this requirement difficult

to comprehend. Presumably, this examination will

be taken by members of the plaintiff class in com-

petition with others. Permitting advance review

by plaintiffs would place all others at a competitive

disadvantage. If the District Judge is seeking pro-

fessional assistance from plaintiffs’ expert, his

order should so provide; and proper steps should

be taken to insure confidentiality ’’ 520 F.2d at 427.

From the foregoing, therefore, it is evident that dis-

closure of the actual tests and the test scores of indi-

viduals taking the tests to unions and employees rep-

resented by unions, who have no professional obligation

not to publicize the tests and test scores, will invalidate

the tests both as a practical matter and under Title VII

of the Civil Rights Act. Invalidation of testing as an

employee selection device, both in private industry and

government, may well lead to unqualified persons be-

ing placed on jobs of critical importance, such as the

job in question in the present case, to the detriment of

the employees, employer and the public at large.

The Company involved in this case is a large public

utility—a classic example of a ‘‘business affected with

a public interest’? (Munn v. Illinois, 94 U.S. 113

(1877) ). As such it has a duty to its consumers and the

public to take all reasonable precautions to determine

that only qualified personnel perform important jobs

such as the job involved in this case. Validated, job

related testing is a business necessity essential to carry

out the Company’s duty to place only qualified em-

ployees on admittedly ‘‘critical’’ jobs such as the job

involved here.

The uncontradicted record in this case shows that the

Company spent ‘‘tens of thousands of dollars’? (A.

128) in validating the instrument man battery of tests

to the exacting requirements of the EEOC Guidelines

and the APA Standards.” In addition, the Company’s

own validation study was corroborated by a later vali-

dation study by an outside consultant, the National

Compliance Company. The result of these considerable

and expensive efforts was upheld in the arbitration

award in which the distinguished and experienced

labor-management Impartial Arbitrator Dallas Jones

found that the Company had a contractual right to use

tests as a selection device for the Instrument Man job

and that the tests used had a high degree of validity

(P.A. 72-73a).

In short, the test battery involved in this case

scrupulously adheres to the Court’s seminal observa-

tion in Griggs v. Duke Power Co., 401 U.S. 424, 436

See pp. 8-12 of this brief for a detailed description of how

the Company determined the criterion-related validity of the In-

strument Man test battery.

(1971), ‘‘that any tests used must measure the person

for the job and not the person in the abstract’’.

If, therefore, the extreme and novel judicial construc-

tion of Section 8(a)(5) of the National Labor Rela-

tions Act by the Court below is not reversed, this Com-

pany’s substantial and successful efforts to validate its

Instrument Man test battery under the EEOC Guide-

lines and APA Standards, as approved by this Court,

will have been completely undermined. It is apparent

that any such rule would not be limited to the Detroit

Edison Company, but would adversely affect all testing

programs administered by private and public em-

ployers.

In the over forty years of enforcement of the Na-

tional Labor Relations Act, this is the first case in which

the Board has construed Section 8(a)(5) of the Act to

impose a requirement on employers to disclose to

unions job related aptitude tests and test sheets and

scores by named applicants. Indeed, in the only previ-

ous reported case, the Board’s General Counsel had the

common sense to refuse even to issue a complaint on a

union charge seeking production of actual tests under

Section 8(a)(5) of the Act. International Telephone

& Telegraph, Federal Division, Case No. SR 657, 46

LRRM 1387 (1960). In this case the Board’s General

Counsel agreed with the Company’s contention that the

specific items on the tests had a ‘‘unique character’”’

and that:

‘“‘TWhile it [the Company] would bargain con-

cerning the general subject of the tests, it would

not furnish specific test questions, since advance

inspection thereof would permit their content to

be widely disseminated and thus impair the useful-

ness of the test.’’

Thus, well before the passage of the Civil Rights Act

of 1964 and the EEOC Guidelines on testing, the

Board’s General Counsel understood that tests had a

unique character and that their disclosure to a union

would permit them to be widely disseminated and im-

pair their usefulness.

2. THE INEFFICACY OF THE ‘‘ RESTRICTION’? APPROVED

BY THE CourT oF APPEALS IN PREVENTING THE

DISSEMINATION OF THE TEST MATERIALS TO Un-

AUTHORIZED PERSONS

The Board’s opposition to the petition for certiorari

does not appear to challenge the proposition that tests

and test sheets and scores that are divulged to unauthor-

ized parties result in invalidation of the tests. Rather,

the Board claims that the ‘‘restriction”’ invented by the

Board and approved by a majority of the Court of

Appeals will prevent unauthorized disclosure of the

tests * to those who have taken the test in the past or

who may take it in the future. This was the view that

dissenting Judge Weick termed ‘‘really naive” (P.A.

lla).

The reasons why this view is naive are the following:

1. A violation of the copying and dissemination

provisions in the Board opinion (no protective provi-

sion appears in the Board’s Order) would be most diffi-

cult, if not impossible, for a company to enforce and

police once it has given a union the test materials as

required by the Board order. At that point it has lost

“*It is to be noted that the ‘‘restriction’’ is applicable only to

the test battery and not to the test scores. Under the Board order

as enforced by the Court of Appeals, the Union could publish the

scores achieved by the examinees and distribute them generally.

control over the materials, and ‘‘test security’’ in ac-

cordance with Standard I5 of the APA Standards

quoted above and the EEOC Guidelines incorporating

those standards has been breached.

2. There is a built-in ambiguity with respect to

exactly what specific individuals comprise ‘‘the Union”’

to which the Board and Court have ordered the Com-

pany to divulge the tests and test scores. Is ‘‘the

Union”’ the individuals who comprise the executive

officers of the Union, its Executive Board, its grievance

committee, or is it the entire membership of this pre-

sumably unincorporated association? Obviously argu-

ments could be made for any one or a combination of

the above alternatives. Even if it were confined to the

executive officers, Executive Board or grievance com-

mittee of the Union, these individuals are also mem-

bers of the bargaining unit and change on a regular

basis as the result of internal Union elections. The

knowledge thus gained by these individuals pursuant

to the Court enforced Board Order could be used by

them to gain an unfair competitive edge on others not

privy to the test or test scores, and thus invalidate the

test, even though the Union officials themselves had not

copied or otherwise disseminated the test.

3. As organizations properly responsive to their

members’ desires and free from professional restric-

tions against disclosure, unions are in a uniquely poor

position to prevent intentional or inadvertent dis-

semination of test materials, particularly in view of

the ubiquitous copying machine. Inadvertent disclo-

sure by non-psychologists can be illustrated by the case

of Davis v. Washington, 348 F.Supp. 15 (D.C. 1972),

where Judge Gesell entered an order sealing the Civil

Service examination in question ‘‘because it is given

on a regular basis to many federal job applicants”’,

only to have the dissenting judge on the Court of Ap-

peals append the entire examination to his public

opinion. See Davis v. Washington, 512 F.2d 956, 967

D.C.Cir. 1975), reversed, Washington v. Davis, 426

U.S. 229 (1976). The record in this case provides

an excellent example of how disclosure of the test

items to employees could oecur without an express vio-

lation of the ‘‘restriction’’. When the Union represen-

tative was given sample items of each of the tests in-

volved in the arbitration proceeding, he read into the

record the actual sample test item and asked the Com-

pany psychologist and employee witnesses Low those

questions were relevant to the Instrument Man job

(See, e.g., A. 261, 286-289, 293, 328-330). Presumably,

if the Union were given the actual test items the Union

representative would in like marner read the actual

test items or some of them into the record for an em-

ployee response concerning relevance. Since the arbi-

tration record is transcribed, and since all Union mem-

bers presumably have access to such transcripts, dis-

closure of actual test questions may well occur in this

manner even though the ‘‘restriction’’ may not have

been literally breached.

4. This record clearly shows a hostility to the en-

tire concept of testing as an employee selection device

by this Union and a preference for promotion by sen-

iority alone (A. 160-169), despite the contractual

agreement of the parties to promote on the basis of

seniority only where there is no ‘‘head and shoulders’”’

difference in the qualifications and abilities of the ap-

plicants. Thus, disclosure of the test materials to the

Union is not a disclosure to an impartial body with no

motive to discredit testing as a selection device, but

rather is a disclosure to an avowed enemy of the con-

cept of testing. It requries a complete suspension of

ordinary common sense to assume that furnishing a

Union, which is hostile to testing, the ammunition to

destroy test validity will not result directly or in-

d:.ectly, inadvertently or intentionally in the Union’s

use of the ammunition.

5. Even if the Union were found to violate the

‘‘restriction’’ contained in the Board’s opinion, the

Company remedy is speculative at best. In the first

place, only the Board’s General Counsel can initiate

contempt proceedings, and the initiation of such pro-

ceedings is solely within his discretion. The Company

has no independent right to initiate such proceedings.

Amalgamated Utility Workers v. Consolidated Edison

Company, 309 U.S. 261, 264 (1940). It is also difficult

to conceive of a contempt proceeding being brought

in the Court of Appeals against a union that (1) has

not been found guilty of an unfair labor practice,

and (2) was not a party to the proceedings in the

Court of Appeals. Finally, it would appear that in-

advertent disclosure would be completely beyond the

contempt powers of the courts.

6. Even if the Union were liable in contempt for

a violation of the ‘‘restriction’’, there is no adequate

remedy for the violation, since the tests will have been

hopelessly invalidated and it would take years to

validate a new test.* The only sanction against the

Union would be punitive, not remedial—a very small

source of comfort to a company whose testing pro-

gram has been gutted, and of absolutely no comfort

13 The evidence in this case showed that it took seven years to

validate the power plant operator test battery (A. 309-310).

45,

to the public whose interests could well be jeopardized

by the presence of unqualified persons on jobs of criti-

cal importance in the generation of electrical power."

Disclosure Of The Test Materials To The Union Has No Probable

Relevance Or Usefulness To The Union In Carrying Out Iis

Statutory Duties And Responsibilities In Processing The Pro-

motion Grievances Through Arbitration

As noted at the outset of the first argument, made

above, the Board had a preliminary burden of proving

by substantial evidence on the record considered as a

whole that disclosure of the tests and test scores linked

to the name of the applicant was of probable rele-

vance and usefulness to the Union in carrying out its

statutory duties and responsibilities. This ‘‘discovery-

type’’ standard was approved by this Court in NLRB

v. Acme Industrial Company, 385 U.S. 432 (1967),

where the Court found that the Company violated

Section 8(a)(5) of the Act by refusing to disclose to

the Union information concerning removal of machin-

ery from the plant that had been the subject of a

Union grievance and proposed arbitration proceeding.

Even applying the broad standard of Acme to the

‘‘unique’’ information sought in the present case, there

was no showing of probable relevance or usefulness to

the Union in obtaining the tests and test scores linked

to the names of the applicants.

‘* For many of the foregoing reasons a commentator on this case

concluded that the ‘‘restriction’’ imposed by the Board and ae-

cepted by the Court of Appeals would be ineffective in preventing

dissemination of the tests. Psychological Aptitude Tests and the

re | to Supply Information, 91 Harvard Law Review 869, 875-876

Any analysis of this subject must begin with the

series of substantial Company efforts to be completely

responsive to the Union’s test information requests

in a mutually satisfactory form without invalidating

the testing program itself. These efforts included fur-

nishing the Union with (1) the two validation studies

done on the test battery, (2) written explanations of

the tests and scoring weights of the tests, (3) some

146 samples of the types of questions found in the

MPFB and each of the six subparts of the EPSAT,

(4) the test scores of all the examinees without link-

ing the examinees’ name to the scores, and offering to

the Union (1) to release the score of any named ex-

aminee who consented to such release, (2) to let the

Union’s representative take the test, and (3) to turn

over the test battery to a qualified industrial psycholo-

gist of the Union’s choice.

In short, the Company gave or offered to give the

Union all the information it possessed concerning its

instrument man test battery except for the actual tests

and test sheets and scores linked with the names of the

applicants. This is what dissenting Judge Weick found

was the ‘‘wealth of information’’ on testing which he

found ‘‘was, in my judgment, sufficient to permit the

Union to process adequately the grievance pending be-

fore the Arbitrator, or to perform its duties under the

collective bargaining agreement’’ (P.A. lla).

1. THE IRRELEVANCE OF THE TESTS

The three professional industrial psychologists who

testified in either the arbitration case or Labor Board

hearing each testified without contradiction that dis-

closure of the actual tests to the Union directly would

not be helpful in determining the validity of the test

An nen ane ela

battery, and that the validity of the test battery could

best be determined by disclosure of the validation

studies performed in 1970 and 1972, which were pro-

vided to the Union (A. 50-51, 71-72, 79-80, 191). No

expert witness for the Union or the Board’s General

Counsel ever disputed this fact.

The Impartial Arbitrator came to the same conclu-

sion, holding that production of the actual test battery

to the Union would ‘‘prove nothing’’, and that the

Union’s position in the arbitration case was not ‘‘dam-

aged in any way by lack of access to the test’’ (P.A.

72a). Even the NCC, which revalidated the test in 1972

did not need or utilize the actual test battery in its

validation study (A. 100).

The irrelevance of the actual tests themselves to the

Union’s statutory responsibility in processing the pro-

motion grievances through arbitration is further un-

derlined by what the Unien did when it was furnished

sample questions of each of the tests involved. The

only use made of the sample items—and presumably

the only use to be made of the actual test items—was

to pose the samples to the Company’s industria! psy-

chologist, to incumbent Instrument Men and appli-

cants for that job and ask them whether the sample

items were ‘‘connected with or relevant to the instru-

ment man job’’ (See, e.g., A. 261, 286-289, 293, 328-

330). That series of questions and answers, however, is

irrelevant since it is admitted, as found by the Ad-

ministrative Law Judge, that:

‘‘These tests are not designed to test the skills or

knowledge of the applicant which may be related

to the job, but the capacity or aptitude of the

employee to be trained and to do the job efficient-

ly” (P.A. 22a).

Thus, for example, it is evident that a Company In-

strument Man may not need as a part of his substan-

tive knowledge in performing the Instrument Man job

to know which way an airplane will turn if one of its

two engines is out (see sample 9 on the Mechanical

Comprehension subpart of EPSAT, A. 393). A lay-

person might well conclude that the answer to that

question was not ‘‘connected or relevant to’’ the instru-

ment man job since Instrument Men do not work on

airplanes. Since, however, the test is not designed to

measure job knowledge, but rather to determine apti-

tude to perform the job, there is no necessary direct

relationship of each test item with the day-to-day work

of the Instrument Man. Thus, disclosure of the test

items to the Union to obtain lay opinions about the

relevancy of the specific test items to the Instrument

Man job when no such direct relevance necessarily

exists is neither relevant nor useful to the Union in the

exercise of its statutory rights.

Despite the foregoing, the Administrative Law

Judge, with the Board’s approval, held that while pro-

duction of the actual tests might not be necessary to

determine the test’s validity in predicting employee

success on the job, it might be necessary for the fol-

lowing reasons:

‘‘To inform the employees, or their representa-

tives, whether the tests are truly job related or

contain objectionable distortions . . . whether they

have built-in bias and are, in fact, discriminatory,

or whether, in sum, they tend to undercut Respond-

ent’s contract commitment to promote by seniority

where there is no ‘significant difference’ between

the ‘reasonable qualifications and abilities’ of the

applicants for promotion’ (P.A. 47a).

But the conjecture of the Administrative Law Judge

that the battery may contain ‘‘distortions’’, ‘‘built-in

bias’’ and be ‘‘discriminatory’’—whatever those gen-

eralizations mean—is at war with the uncontradicted

record that there is a demonstrated high correlation be-

tween the good performers on the tests and good per-

formers on the job and the decision of the Arbitrator

that the test battery was a ‘‘reliable valid test’’ which

had a “‘high degree of validity’’. If the tests had ‘‘dis-

tortions’’, ‘‘built-in bias’’ or were ‘‘discriminatory”’,

that obviously could not be the case.

If the problem posed by the Administrative Law

Judge is one of advising the employees or their repre-

sentative of the nature of the items on the Instrument

Man test battery so that they might better know the

‘‘eriteria required for promotion’ (P.A. 48a), the

Company has already done this by explaining to the

Union in some detail the nature of both the MPFB and

the six-part EPSAT, together with the furnishing of

146 sample questions closely corresponding to the kind

of actual questions on each portion of both tests.

The Court of Appeals simply brushed off the Com-

pany’s relevance argument in the following manner:

‘“Without actually contesting the finding that the

tests, answer sheets and scores are relevant, De-

troit Edison argues, in effect, that the circum-

stances of this case are such as to require that the

tests and answer sheets be delivered only to a quali-

fied pychologist’’ (P.A. 5a) (Emphasis added).

The italicized portion of the above quote is simply

erroneous factually. In its brief to the Court of Ap-

peals, as well as in its petition for rehearing en banc,

the Company argued strenuously under separate head-

ings that the tests and test scores linked with the name

of the applicants were irrelevant (Co.Br. to Sixth Cir-

euit Court of Appeals, pp. 26-33).

The Board seeks to avoid the relevance argument on

its merits by claiming that the Company’s failure to

file an exception to the decision of the Administrative

Law Judge ordering the Company to disclose the tests

to a qualified psychologist of the Union’s choice consti-

tuted an admission by the Company of the probable

relevance and usefulness of the information to the

Union.

This stilted reasoning overlooks the fact that the

decision of the Administrative Law Judge on the issue

of the tests (as opposed to the test scores) was no more

than what counsel for the Company offered to do at

the outset of the hearing as yet another compromise

expressly ‘‘to appease’ the Union. There was, there-

fore, no practical reason for the Company to file an ex-

ception to the order to disclose the tests to a profes-

sional psychologist. The Board has now seized on the

last of a series of compromises offered by the Company

simply ‘‘to appease’ the Union, which was accepted

by the Administrative Law Judge without Company

exeeption, as an admission by the Company of rele-

vance. Worse still, the Board now argues that when it

upset that compromise and ordered the Company to

disclose the tests to the Union directly, it thereby fi-

nessed the Company’s further right to argue relevance

to the courts. The absurd result of the Board’s reason-

ing is that either the Company should not have offered

the compromise or it should have filed an exception to

this aspect of the decision which was deemed favorable

to its interests on the possibility that the Board might

upset the compromise and order disclosure of the tests

to the Union directly.

2. THE IRRELEVANCE OF THE Test Scores LINKED

WITH THE NAMES OF THE APPLICANTS

It will be recalled that during the arbitration pro-

ceeding the Company furnished the Union with the

raw test scores achieved by each applicant who took the

tests, but did so without linking the name of the appli-

cant with his score.” The Company did this because it

agreed with the Union that the scores themselves might

be relevant. If, for example, twenty persons took a test

with a 10.3 cutoff score, it might be relevant for the

Union to know if a large majority of the twenty were

clustered just above or just below the 10.3 score (there-

by permitting an argument that there was very little

difference in the applicants). Other arguments based

on the scoring spread could be made. What this record

totally lacks, however, is any evidence supporting the

relevance of linking the applicant’s name with the

score he achieved.

The only reference to the alleged relevance of the

scores linked with the names of the applicants in the

decision of the Administrative Law Judge was the

following:

‘Contrary to Respondent’s contention, the arbi-

trator, who passed on the merits of the grievances,

stated, during the reopened hearing, after his ini-

tial decision, that, in his opinion, the Union could

not adequately ‘police’ the contract without the

** In addition, the Company offered to release the test scores of

any employee who indicated his willingness to have his test score

divulged.

information. The arbitrator, in fact, directed Re-

spondent to reconsider its selection process, in part,

based upon the scores achieved by the applicants”’

(P.A. 49a).

The Board accepted this finding without comment, even

though the Company filed a specific exception to the

relevance finding, and, as noted above, the Court of

Appeals erroneously found that the Company did not

‘‘actually’’ contest the relevance finding on the test

scores (P.A. 5a).

The reference in the Administrative Law Judge’s

opinion to the Arbitrator’s statement in the reopened

arbitration proceeding is highly misleading. The state-

ment referred to was not a ruling, but rather the be-

ginning of a colloquy between counsel and the Arbi-

trator directed toward ascertaining whether the Com-

pany had properly implemented the original award di-

recting that those scoring between 9.3 and 10.3 be re-

examined. The Arbitrator was inquiring how those in-

dividuals could be re-examined if their scores were not

divulged. Counsel for the Company agreed that the

names of the three individuals who scored between

9.3 and 10.3 should be disclosed, and stated that he

proposed to call the responsible Company witnesses who

would testify concerning the three named individuals

who scored between 9.3 and 10.3. He then stated that he

believed that this was ‘‘sufficient proof’’ (A. 338). The

Arbitrator responded :

‘*T think I would agree with that, Mr. Houghton.

If we could have the Detroit Edison employee who

is responsible for that activity testify that these

names do go with these scores, then I will have to

accept that as reliable evidence’’ (A. 338)

The names of the individuals scoring between 9.3 and

10.3 were then disclosed (A. 341-342). In his supple-

mental decision, the Arbitrator did not rule that the

Union was entitled to the test scores of all named ap-

plicants. He ruled that:

‘The Company, for the purpose of the instant

hearing, did disclose the names of those individuals

at the Monroe Power Plant who scored between

9.3 and 10.3. The actual score received by each in-

dividual was not disclosed; however, this is not of

such importance as to preclude a review of the

Company’s action in implementing the Award.

Regardless of the score an individual attained in

the range between 9.3 and 10.3, it is, under the

terms of the Award, his qualifications other than

test score which became of decisional importance”’

(P.A. 8la) (emphasis added).

It is, therefore, plainly erroneous to state that the

Arbitrator held that all scores linked with the names of

all the applicants had to be disclosed to the Union. All

he held was that the Union was entitled to know the

names of the applicants scoring between 9.3 and 10.3,

and the Company complied with this holding.”

trator ’s opinion and award. The Arbitrator’s written holding that

providing the Union with the test battery would ‘‘prove nothing”’

and that the Union’s position in the case was not damaged in any

way by lack of access to the tests is simply ignored, while control-

ling weight is given to some ambiguous remarks made on test scores

at the beginning of a colloquy which ultimately resulted in the

Arbitrator’s concurrence that the Company had discharged its

obligations on disclosing test scores linked with the names of the

applicants by having its industrial psychologist testify to the names

of the three individuals who scored between 9.3 and 10.3 on the

test battery.

The irony of the Board’s misplaced reliance on re-

marks made by the Arbitrator in colloquy with coun-

sel is that if he really meant that the scores achieved

by each named applicant be disclosed to che Union, it

would tend to undercut one of the main reasons he

used in finding that the tests were proper. Thus, he

ruled that:

‘‘In addition, and very important, the instant test

is not used to compare one employee against ano-

ther in terms of scores attained . .. This problem

is eliminated by reporting scores in terms of ‘not

recommended’ and ‘accepted’ ’’ (P.A. 73a).

The record is thus bereft of any evidence support-

ing the relevance of disclosing test scores linked with

names of the applicants to the Union.

The Company nevertheless submitted evidence why

the test scores achieved by named individuals should

not be disclosed to the Union. In addition to the breach

of APA Standards and the consequent invalidation of

the tests under the EEOC Guidelines by disclosure of

test sheets and scores of named applicants discussed in

Part I of the Argument, the record in this case showed

that the Company gave a specific oral promise of con-

fidentiality to all applicants prior to testing. That

promise was that:

‘‘Your test scores are confidential. We will tell the

Employment Department generally how you did

on these tests and they may talk to Company de-

partments about your being put in a certain job,

but your vest scores are kept only in our files’’

(A. 445).

This principle of confidentiality is not—as intimated

by the Board—a device to foil the Union in its attempt

to process employee grievances. Rather, it is an emerg-

ing and far-reaching principle now firmly implanted

in the federal law that there is a right of privacy for

those taking psychological aptitude tests. Thus, for ex-

ample, for students in federally assisted educational

institutions there is an express statutory prohibition

against the release of education records of such stu-

dents without consent. 20 U.S.C. § 1232(g)." While it

is true that employees of the Company are not students

in federally assisted educational programs, the point

is that the Company’s concern is not a parochial one,

involving its defenses to Union grievances over promo-

tions, but parallels a broad Congressional concern

shown for the disclosure of the same type of test scores

in a federally assisted educational context.

In addition, the Company provided evidence that

disclosure of test scores linked with the names of the

applicants in the past resulted in harassment of the

lower scoring employees who were called ‘‘stupid”’,

‘‘dummy”’ and the like, and who actually left the Com-

pany because of such harassment (A. 84).

This evidence was brushed aside by the Board and

Court of Appeals. Both held that the decision in the

Court of Appeals in General Electric Co. v. NLRB,

466 F.2d 1177 (6th Cir. 1972), was controlling (P.A.

7a, 53a). In the General Electric case, the Company de-

clined to provide the Union with information obtained

in a survey of wages paid by various companies in

the survey. The results of the survey were used to

'* See also 5 U.S.C. § 552(b) (6), where an exemption from dis-

closure was created by Congress in the Freedom of Information

Act for ‘‘personnel and medical files and similar files the disclosure

of which would constitute a clearly unwarranted invasion of per-

sonal privacy’’.

determine the rates of pay of the General Electric

employees. The Company argued that it should not be

required to disclose the wage data obtained by named

companies in the survey because it had obtained the

information from these companies on a confidential

basis. The Sixth Circuit ruled that:

‘*Employers cannot be allowed to collect wage in-

formation on a pledge of confidentiality to parties

outside the bargaining unit under these circumstan-

ces, then turn around and deny the Union the use

of that data based on its voluntary pledge’’ 466

F.2d at 1185.

The harm of disclosure of wage information gained

on a confidential basis from employers outside the bar-

gaining unit cannot be equated with the harm of dis-

closure of test scores of employees within the unit,

which were gained on a confidential basis.

The Administrative Law Judge, with the Board’s

approval, appeared to recognize that there might be

a harm or, as he termed it, ‘‘odium’’ in test score dis-

closure, but claimed that such an odium is inherent

in all testing programs (P.A. 53a). That observation

simply misses the point. There is no particular odium

involved in generally failing to achieve a 10.3 cutoff

score. That only means that the applicant scored from

0 to 10.2. But if scores linked with the names of appli-

ecants are disclosed, and a particular applicant has

scored particularly poorly in relation to the other ap-

plicants, a totally unnecessary ‘‘odium’’ is attached

to that applicant, and the record disclosed past inci-

dents of harassment of such individuals.

aa ad * + *

There is a clear relationship between the two argu-

ments made in this brief. This Court in NDRB vy.

15)

Triatt Mfy. Co., 351 U.S. 149 (1956), in assessing a

company refusal to provide a union with information

relating to the financial condition of the company in

the face of a company claim of inability to afford

union economic demands, held that:

‘*Kach case must turn upon its particular facts.

The inquiry must always be whether or not under

the circumstances of the particular case the statu-

tory obligation to bargain in good faith has been

met’’ 351 U.S. at 153-154.

The Sixth Circuit Court of Appeals itself in Kroger

Co. v. NLRB, 399 F.2d 455 (6th Cir. 1968), emphasized

that a union’s right to information under Section 8(a)

(5) of the Act is not a per se or absolute right, but

rather that it may involve a balancing of interests.

‘To us the critical problem appears to be how to

recognize and how to adequately protect each of

the condicting interests that are involved here”’

399 F.2d at 457.

Any such balancing of interests or examination of

the circumstances of the particular case here must in-

volve a determination of (1) the probable relevance

and usefulness to the Union of the testing information

involved—or the degree of such probable relevance and

usefulness—in the light ./ the wealth of information on

testing already disclosed by the Company to the Union,

and (2) the invalidation or potential for invalidation

of the Company’s testing program under the APA

Standards and EEOC Guidelines by disclosure of the

test materials to the Union.

The less relevant and useful the requested informa-

tion is to the Union in fulfilling its statutory require-

ments, as shown in Part II of this Argument, the more

the Company is entitled to protection against dissemi-

nation of the tests and test scores to prevent test in-

validation, as shown in Part I of this Argument. Like-

wise, the less relevant and useful the test materials are

to the Union, the less need there is for the Board and

Court of Appeals to construct and seek to enforce a

‘‘restriction’’ of highly dubious efficacy.

In addition, any such balancing must take into ac-

count the strong federal labor policy favoring collec-

tive bargaining and arbitration as the preferred means

of settling labor disputes arising under a collective bar-

gaining agreement, as enunciated by this Court in

United Steelworkers of America v. Warrior & Gulf

Nav. Co., 363 U.S. 574 (1960). That policy has been

completely ignored by the Board and the Court of Ap-

peals in this case. The Arbitrator expressly found that

the Union’s lack of access to the actual tests did not

damage its position in arbitration in any way and that

disclosure of the tests to the Union would prove noth-

ing. Contrary to the finality of arbitration awards en-

dorsed by this Court in the Steelworkers Trilogy, the

Board and Court of Appeals have simply brushed aside

the conclusion of the Arbitrator on the irrelevance of

the tests. Far from serving the complementary role to

arbitration envisaged by this Court in NLRB v. Acme

Industrial Co., 385 U.S. 432 (1967), the Board’s proc-

esses in this case have been used to undermine the Ar-

bitrator’s role and decision. The decision also serves to

undermine the collective bargaining process itself. The

Union sought to obtain the tests and test scores in the

1972 negotiations. The Company successfully resisted

these demands, only to have them now imposed by the

Labor Board.

Pa.

Finally, in any balancing of the interests involved,

this Court has had previous occasion to remind the

Labor Board that it ‘thas not been commissioned to

effectuate the policies of the Labor Relations Act so

single-mindedly that it may ignore other and equally

important Congressional objectives’’, and that it must

accommodate the statutory scheme of the National

Labor Relations Act to other statutory schemes.

Southern Steamship Co. v. NLRB, 316 US. 31, 47

(1942). See also Emporium Capwell Co. vy. Western

Addition Community Org., 420 U.S. 50, 66 (1975).

Here the Board, with the approval of the Court of Ap-

peals, has sought to enforce the duty-to-furnish-infor-

mation requirement of Section 8(a)(5) of the Act so

single-mindedly—even though the information is of no

relevance or barely marginal relevance—that it has

totally failed to accommodate the equally important

Congressional objective of insuring that tests used in

employment selection decisions be valid and job related.

CONCLUSION

For the above reasons the Company respectfully re-

quests that the Court reverse the decision of the Court

of Appeals.

Respectfully submitted,

JOHN A. McGuinn

FARMER, SHIBLEY, McGuinn & FLoop

1120 Connecticut Avenue, N.W.

Washington, D.C. 20036

Lron S. COHAN

2000 Second Avenue

Detroit, Michigan 48226

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.