Maddox v. Claytor Brief for Defendant-Appellee

    UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT 

NO. 84-SG06

GEORGE A. MADDOX, et al.,
Plaintiffs-Appellants, 

v.
W. GRAHAM CLAYTOR, Secretary of 
the Navy, Defendant-Appellee.

On Appeal from The United States District Court 
for the Middle District of Georgia 

Albany Division

BRIEF FOR DEFENDANT-APPELLEE

JOE D. WHITLEY 
United States Attorney 
JOHN L. LYNCH Assistant U.S. Attorney 
Middle District of Georgia 
Macon, Georgia
C. JOHN TURNQUISTAssociate General Counsel (Litigation) 
DANIEL E. O'CONNELL, JR.
Senior Trial Attorney 
Litigation Office 
Office of the General Counsel Department of the Navy 
Washington, D.C.

DAVID A. HIGLEY
Staff Judge AdvocateJOHN E. BENGIER
Deputy Staff Judge Advocate
Marine Corps Logistics Base
Albany, Georgia
Attorneys for Defendant-Appellee



SUMMARY OF CONTENTS

♦

1 PAGE
STATEMENT REGARDING PREFERENCE .................... . i i
STATEMENT REGARDING ORAL ARGUMENT ................ . iii
TABLE OF CONTENTS ................................ . iv
TABLE OF CITATIONS ................................ . viii
COUNTERSTATEMENT OF THE ISSUES.................. . . 1
COUNTERSTATEMENT OF THE CASE ...................... 3
I. Course of Proceeding and Disposition In i3
II. Statement of the Facts ...................... 6
III. Standard of Review .......................... 30
SUMMARY OF THE ARGUMENT . . . . '................ • 31
STATEMENT OF JURISDICTION ........................ 32

V  •

<

l



UNITED STATES COURT OF APPEALS 
FOR THE ELEVENTH CIRCUIT 

NO. 84-8006

GEORGE A. MADDOX, et al.,
Plaintiff s-Appe Hants, 

v.
W. GRAHAM CLAYTOR, Secretary of 
the Navy, Defendant-Appellee.

On Appeal from The United States District Court 
for the Middle District of Georgia 

Albany Division

BRIEF FOR DEFENDANT-APPELLEE

STATEMENT REGARDING PREFERENCE

This case does not fall within any of the categories of 
cases listed in Appendix One to the Rules of the United States 
Court of Appeals for the Eleventh Circuit. Therefore, pursuant 
to Eleventh Circuit Rule 11, this case is not entitled to pre­
ference in processing and disposition.



STATEMENT REGARDING ORAL ARGUMENT 
Appellee respectfully suggests that oral argument is 

not necessary in this case. The following factors support 
defendant's position. First, the evidence in this case is not 
complex, convoluted or unusual. There is, therefore, no require­
ment to orally clarify subtle organizational history, employment 
practices or promotion system features. Second, the findings of 
fact in this case are clear, concise, logical (in terms of both 
structure and organization), and straightforward. There is no 
need to orally explain what the district court meant or found or 
held. Third, the court below established no new principle of 
law; nor did the district court misapply applicable law.

Moreover, the principal issue in this, appeal involves the 
factual findings of the District Court relative to the statistical 
evidence presented. The opinion issued by the Court included a 
comprehensive discussion of these findings findings which are 
subject to the "clearly erroneous" standard of review. Further, 
those legal issues raised by plaintiffs are covered by clear 
precedent so that there appears to be little need for further 
discussion.

Undeniably, oral argument is often useful. In view of the 
foregoing observations pertinent to this case, however, defendant 
believes oral argument is not necessary. Concern for judicial 
economy as well as a desire to minimize further costs associated 
with this proceeding support resolution of this appeal without 
oral presentations.

Therefore, Appellee respectfully suggests that this case 
is appropriate for disposition on the Court's non—argument calendar.

111



*

TABLE OF CONTENTS

SUMMARY OF CONTENTS ................................ i
STATEMENT REGARDING PREFERENCE ...................... ii
STATEMENT REGARDING ORAL ARGUMENT .................. iii
TABLE OF CITATIONS.................................. viii

COUNTERSTATEMENT OF THE ISSUES.........."........... 1
r~

COUNTERSTATEMENT OF THE C ASE .........................  ,3
I. COURSE OF PROCEEDING AND DISPOSITION IN

THE COURT BELOW ................................. 3
A. Administrative complaints and subsequent

law s u i t ..................................  3
B. Class certification ......................... 3
C. Trial......................................  4
D. Request of the trial court for post-trial

submissions!.......................   5
. - . Ok- rE. Decision of the trial court .......... • • b

II. STATEMENT OF THE F A C T S ......................... 6
V •A. Functions and characteristics of 
the work force at the Marine Corps
Logistics B a s e ...........................  6
1. The Marine Corps Logistics Base:

1952 through 1975 ...................... 6
2. 1976: Functional transfer and

Realignment of the Base's
responsibilities ......................  7

3. The wide range of disparate job, categories in the work force at
the Marine Corps Logistics Base ........ 8

B. Regulations, procedures, and standards 
governing the Merit Promotion System
at the Marine Corps Logistics B a s e .........  10

Page

IV



Page

*

1. Regulatory standards a n d .............. 10
requirements 1 2
a. Regulatory Structure ..............  10
b. Filling of vacancies..............  11
c. Qualification standards ............  11
d. Classification & job grading

standards.............. '........... 12
e. Merit promotion requirements . . . .  13

2. The operation of the merit promotion 
system at the Marine Corps Logistics
B a s e ..................................  13
a. Issuance of the meritpromotion announcement ............  13
b. Personnel office's determination.

of eligibility....................  14
c. The rating p r o c e s s ................  15
d. Selection from the list of ' *certified candidates........'V. . . 16

C. The statistical evidence
V • 1. Defendant's analysis of the filling 

of job openings ...................... • 18
2. Defendant's regression and logit 

analyses of promotions .............. • 20
3. Plaintiffs' statistics .............. 23

a. Descriptive statistics .......... • 23
b. Statistical analyses submitted by

affidavit ........................ 24
(1) Promotion out of g r a d e ........  24
(2) Merit promotion analysis . . . .  25

v



*

-p-a9-e

D. The anecdotal evidence................... 26
1. Claims of named plaintiff Maddox . . . .  26
2. Claims of name plaintiff A b a d ........ 27
3. Testimony of unnamed class members . . .  28

III. STANDARD OF R E V I E W ............................ 30
SUMMARY OF THE ARGUMENT................ ’...........  31

r~
STATEMENT OF JURISDICTION......................  82
ARGUMENT............................ ...............  3 2
I. THE TRIAL COURT DID NOT ERR IN ITS FACTUAL 

FINDINGS THAT PLAINTIFFS' STATISTICS: (1)
FAILED TO ACCOUNT FOR QUALIFICATIONS REQUIRED 
FOR MANY JOBS AT THE BASE: (2) FAILED TO 
CONTROL FOR THE WIDE RANGE OF DISPARATE JOB 
OCCUPATIONS AT THE BASE; AND (3) FAILED TO 
MAKE COMPARISONS WITH A LABOR MARKET REFLECTING 
THE ACTUAL EMPLOYMENT PRACTICES OF THE BASE.
NOR DID THE THE'TRIAL COURT ERR IN ITS FACTUAL 
FINDING THAT DEFENDANT'S MORE REFINED STATISTIC^ 
INDICATED NO DISCRIMINATION IN PROMOTIONS . . 3 2

A. To be found meaningful, statistical evi­
dence must account for certain fundamental 

v principles which are necessary to establishTitle VII liability...................... 32
1. When Addressing a Wide Range of Disparate 

Occupational Categories Siich As Those At 
The Base, Meaningful Statisitcal Comparisons 
Must Account For Both Qualifications And Job 
Categories. The Trial Court Did Not Err In 
Its Factual Finding That Plaintiffs Failed
To Account For Either.................... 33 2

2. The Trial Court Did Not Err In Its Factual
Finding That-Plaintiffs' Relevant'Labor 
Market Comparisons Were Erroneously Restricted 
To The Base's Work Force —  A Labor Market That Did Not Accurately Reflect The Sources 
From Which The Base Filled Vacancies . . .  34

vi



Page
B. The Trial Court Did Not Err In Its Factual 

Findings The Plaintiffs' Statistical 
Analyses Each Suffered From One Or More 
Of The Following Fatal Defects: (1) Failure
To Account For Required Qualifications;
(2) Failure To Control For Job Occupations; 
and (3) Failure To Make Comparisons With An 
Appropriate Labor Market .................. 35

1. Plaintiffs descriptive statistics . . .  36
2. Plaintiffs' statistics on promotions

out of grade . . .................. .. 37
3. Plaintiffs' analyses of merit

promotions ..........................C. Defendants' Statistical Analyses, WhichAccount For Job Categories And Qualifications, 
Either Separately Or In Combination, And Which 
Employ A Labor Market Reflecting The Employment Practices Of The Base, Indicate No Classwide Dis­
crimination In Promotions . . . "•..........

X. Job vacancy analysis ................
V , .2. Promotion analysis ..................1 .. ‘ i t

D. There Is No Evidence That Identifies Any-si*- 
Specific And Discrete, Objective Selection 
Criterion Which Adversely Affected Black 
Applicants So As To Require Validation . . . .

V •

II. THE INDIVIDUAL CLAIMS OF NAMED PLAINTIFFS 
ABAD AND MADDOX WERE PUT AT ISSUE IN THE 
COMPLAINT AND WERE DECIDED BY FACTUAL 
FINDINGS OF THE TRIAL COURT; a'S A MATTER 
OF LAW, THESE NAMED PLAINTIFFS ARE NOT 
ENTITLED TO RELITIGATE THESE CLAIMS ..........

III. THE UNNAMED CLASS MEMBERS WHO TESTIFIEDAT TRIAL ARE NOT ENTITLED TO RELIEF BASED 
ON THEIR TESTIMONY, PARTICULARLY IN LIGHT 
OF THE FINDING OF NO CLASSWIDE DISCRIMINA­
TION ..............................' • •

IV. AS A MATTER OF LAW, A CAUSE OF ACTION 
CANNOT BE BASED ON THE TITLE VII REQUIREMENT TO DEVELOP AN AFFIRMATIVE 
ACTION PLAN ........................

CONCLUSION 50

vi l



TABLE OF CITATIONS

CASES PAGE

Bullard v. Webster, 679 F.2d 92 ( 1 9 8 2 ) .................47
Carpenter v. Stephan F. Austin State University,

706 F. 2d (5th Cir. 1983 ) [  ̂ i 7~. i i • •  ........43
Carroll v. Sears Roebuck Co.f 708 F.2d (5th Cir 1983) . . 35
Casillias v. United States Navy, ,■ F.2a ____ • .--34 FEP Cases 1493, 1497-98 (9th Cir. 1984) .......... } 44
Clark v. Chrysler Corp., 673 F.2d 921 (7thCir. 1978) (cert, denied) 103 S. Ct. 1 6 1 ..............48
Cooper v. Federal Reserve Bank of Richmond,
No. 83-185 (Supreme C t . ) ............................. 49

Delaware State Colleqe v. Ricks, 449 U.S. 250
(1980)    39

*Dickerson v; United States Steel Corp., 582 F.2d 
827 (3d Cir. 1978) ............................

* Eastland v. Tennessee Valley Authority, 704 Fi2d'8±3 
(11th Cir. 1983 ) .......................... ..

EEOC v. Radiator Speciality Co., 610 F.2d
(4th Ci,r. 1979 ) ............................

*Ferguson v. Veterans Administration, 723 F.2d 871 
(11th Cir. 1984 ) .............. - ...........

*Fisher v. Procter & Gamble Manufacturing Co., 
fit 3 F.2d 527 (5th Cir.~1980) cert, denied,
449 U.S. 1115 (1981) ........................

Gay v. Waiters1 and Dairy Lunchmen's Union, 694
F. 2d 531 (9th Cir. 1 9 8 2 ) .......... . . . .

Gay v. Waiters Union Local 30, 489 F. Supp. ,
283, 311 n.37 (N.D. Cal. 1980) aff'd., 694 
F.2d 531 (9th Cir. 1982) ....................  *

*General Telephone Company of the Northwest, Inc. 
v. EEOC, 446 U.S. 318 (1980) ................

34, 38 

49

34

41

41

46

VI 11



*

CASES PAGE

General Telephone Company of the Southwest v.
Falcon, 457 U.S. 147 (1982)........ ,.......... . 46

Harris v. Ford Motor Co., 651 F.2d (8th Cir. 1981) . . .  44
*Hazelwood School District v. United States, 433 U.S.

299 (1977) ..........................................  32, 40
Heagney v. University of Washington, 642 F.2d 1157

(9th Cir. 1981 ) ......................................  37 , 44
Hebert v. Monsanto Co., 682 F.2d 1111 (5th Cir.

19 82 ) ............ ............................. . • • 4 0
^International Brotherhood of Teamsters v. United States,

431 U.S. 324 (1977) . . ............................ -,32, 40
Laffey v. Northwest Airlines, Inc., 567 F.2d429 (D.C. Cir. 1976) cert. denied, 439 U.S. 998

(1978 ) ..............................................  39
Mortensen v.' Callaway, 672 F.2d (10th Cir-. 1982) . . . . 44 •
Moore v. Hughes Helicopter, 708 F.2d (9th Cir. 1983) . . 44, 34, 38
Movement for Opportunity and Equality v General

Motors Corp., 622 F.2d 1235 (7th Cir. 1980 ) ■. • f • • .40
• > ‘ ■

Novack, In the Matter of, 639 F.2d 1274 (5th
Cir. 1982 ) (Unit ..................................... 47

O'Brien Sky Chefs, Inc., 670 F.2d 864
(9th Cir. 1982 ) .....................................  37

Pack v. Energy Research & Development Administration,
566 F. 2d 1111 (9th Cir. 1977)........................  37

Page v. Bolger, 645 F.2d 227 (4th Cir.) (en banc),cert, denied, ,454 U.S. 892 (1981)....................  44 , 49
Piva v. Xerox Corporation, 654 F.2d 591 (9th Cir.

1981)................................................ 37
Pope v. City of Hickory, 679 F.2d (4th Cir. 1982) . . . .  44
*Pouncy v. Prudential Ins. Co., 499 F. Supp. 427 (S.D.Tex. "f9lT0) aff'd", 668 F.2d““795 (5th Cir. 1982) . . . .  40, 43, 44
*Rivera v. City of Wichita Falls, 665 F.2d 531

(5th Cir. 1982) (Unit A ) .............................. 33, 34
IX



CASES PAGE
Sledge v. J. P. Stevens & Co., 585 F.2d 625

(9th Cir. 1978)(cert. denied) 440 U.S. 981 . . .
lSte. Marie v. Eastern Railroad Association, 650

F. 2d 395 (2d Cir. 1981).............. ..........
United Air Lines, Inc. v. Evans, 431 U.S. 553

(1977) ........................................
U.S. Industries, Inc., 726 F.2d (5th Cir. 1984) . .
*Valentino v. United States Postal Service, -674 F.2d 

56 (D.C. Cir. 1982)................ . . . . .- • r-

*Wilkins v. University of Houston, 654 F.2d 
388 (5th Cir.) reh'g denied, ~662 F.2d 1156
(1981), vacated _____ U.S. _____, 103 St.Ct. 34 remanded on other grounds, 695 F.2d
134 (5th Cir. 1983) ............................

. 47

. 40

. 39 

. 44

. 33,

33,

.Vi l f.

34, 35

34

x



STATUTES AND REGULATIONS
Federal Personnel Manual
FPM Ch. 001 ..........
FPM Ch. 002 ..........
FPM Ch. 271 § 2-2 (July 1969 )...... 12

FPM Ch. 315 § 5-2 (April 27, 1981).. 11

FPM Ch. 335 § 13

FPM Ch. 335 § 1-3 '(May 7, 19811:---- 13

FPM Ch. 335 §
t 13

FPM Ch. 338 § .3-1 (May 7, 1981) ... \ 12

FPM Ch. 338 § 3-2a (May 7, 1981)... 12

FPM Ch. 338 § 3-3b (May 7, 1981)... 12

FPM Ch. 511 § 3-1 (July 1969) .... 12

FPM Ch. '511 § 3-2 (July 1969) ..... 12

FPM Ch. 511 §
v ■3-i (July- 1969 ) .... i . A

12

FPM Ch. 511 § 3-4 (July 1969) .... 12
FPM Ch. 511 § 4.1 (November 26, 1975) .

y •

FPM Ch. 532 § S6-1 (December 30, 1982 ). 
FPM Ch. 532 § 6-4a (December 30, (1982) 
FPM Supp. 532-1 § S6-6 (April 14, 1980)

Code of Federal"Regulations
5 C.F.R. § 315.501 (1984) ............
5 C.F.R. § 335.102(a)(1984) ..........
5 C.F.R. § 335.103(1984) .............



United States Code, Title 5
5 U.S.C. § 5105 ............................... .....
5 U.S.C. § 5107 ............. .... 'l.......... ......
5 U.S.C. § 5346 ....................................
42 U.S.C. § 2000e-5(f)(1)(1976)

PPM Regulations
0PM Handbook § X-118 ................ ..............
0PM Handbook § X-118C ....... ................. •
0PM Internal Guidelines ...........................

Books & Treatises
Baldus & J. Cole Statistical Proof of Discrimination 

§ 8.02 (1981 & 1983 Supp.) ......................
H. Theil., Principles of Econometrics 632-35 (1971)..

. A .. '4
•> ‘ •A ■

10
10

10

12

12

12

2 1 , 2 2 , 

22

y • y *

xii



COUNTERSTATEMENT OF THE ISSUES

ISSUE I. Whether the trial court was clearly erroneous in 
its factual finding that the statistical evidence did not establish 
classwide racial discriinination in promotions given that.
(1) the statistics submitted by plaintiffs did not control for 
the wide range of disparate job categories at the Marine Corps 
Logistics Base and did not account for employee qualifications;
(2) plaintiff's use of the Base's work force as the relevant 
labor market was incorrect because many positions were ordinarily 
filled from outside the Base and because of an unusually large 
amount of outside hiring necessitated by a Department of Defense 
ordered realignment of the Base's functional responsibilities;
(3) no discrimination was indicated by the defendant's more 
refined statistics which accounted for the disparate job categories 
and employee qualifications, either separately or in combination; 
and (4) there is no evidence in the record which identifies any 
specific and discrete, objective selection criterion, that adversely 
affected black applicants so as to require validation.

ISSUE II. Whether, as a matter of law, Named Plaintiffs 
Maddox and Abad are entitled to relitigate their individual claims 
of discrimination in subsequent proceedings given that: (1) the
individual claims of the named plaintiffs were put at issue in the 
complaint; (2) after reviewing the evidence submitted by plaintiffs 
at trial, the trial court issued a final judgment finding that 
Named Plaintiffs Maddox and Abad were not entitled to relief; 
and (3) as a matter of federal class action law, Named Plaintiffs

-1-



Maddox and Abad were required to present claims representative 
of the class which they sought to represent.

ISSUE III. Whether unnamed class members, who testified at 
the Phase I proceedings in this class action and whose testimony 
was found by the trial court to be "persuasive on the issue of 
discrimination", are entitled to relief based on this testimony 
given that: (1) the trial court expressly stated that, in crediting
this testimony, it was not finding that an individual case of 
discrimination had been established; (2) plaintiffs never requested
the trial court to grant relief based on this testimony; and (3)

tas a matter of federal class action law in bifurcated proceedings, 
relief for unnamed class members is a matter reserved for Phase 
II proceedings which are held if classwide liability is found in 
Phase I.

ISSUE IV. Whether, as a matter of law, a cause of action 
can be based on the Title VII requirement that federal government 
activities develop affirmative action plans —  given that:
(1) there is clear precedent of this Court holding that a cause 
of action cannot be predicated on the requirement to develop an 
affirmative action plan; (2) plaintiffs have given no rationale 
why this Court's precedent should not be applied in this case; 
and (3) the record clearly shows that the Marine Corps Logistics 
Base did in fact develop the required affirmative action plans.

-2-



COUNTERSTATEMENT OF THE CASE
I. COURSE OF PROCEEDINGS AND DISPOSITION IN THE COURT BELOW

A. Administrative Complaints and Subsequent Law Suit 
Named Plaintiff Maddox made initial contact with an EEO 

Counselor at the Marine Corps Logistics Base on April 28, 1977, 
concerning a class complaint of discrimination. Subsequently 
Plaintiff Maddox filed a formal administrative class discrimi­
nation complaint on June 1, 1977. R.E. 26. Named Plaintiff 
Abad filed a formal administrative individual complaint of 
discrimination on March 21, 1977, while Named Plaintiff Shepard 
filed a formal administrative individual complaint of discrimi­
nation on May 5, 1977. R.E. 27. All three complaints were 
pending more than 180 days without final resolution when Named 
Plaintiffs Maddox, Abad and Shepard filed suit on January 6,
1978, in the United States District Court for the Middle District 
of Georgia. The complaint included individual claims of the 
named plaintiffs as well as class claims. R.E. 10-26.

B. Class Certification
An order preliminarily certifying the class was issued by 

the trial court on December 6, 1979 with the following class 
definition: _

The class is defined as all past, present and future 
black civilian employees and applicants for employment 
at the Marine Corps Logistics Support Base, Atlantic.

R. 5'48. The trial court modified the conditional class certifi­
cation on March 18, 1980 and redefined the class as follows.

-3-



All present, past, and future employees of the 
Marine Corps Logistics Base, Albany, Georgia, who 
since January 28, 1977, have been unlawfully 
discriminated against by employment practices of 
the Marine Corps “Logistics Base. Specifically 
excluded are employees of the "tenant" activities 
on the base which are not within the control of 
the defendant.

R. 600. In an opinion dated November 4, 1983, the trial court
gave final definition to the class as:

[A]11 past, present, and future employees of the base, 
who since January 28, 1977, have been unlawfully dis­
criminated against through promotion denials.

Op. 4 (emphasis added). (
C. Trial
Both the individual claims of the named plaintiffs and class 

claims in this action were tried by the Honorable Wilbur D.
Owens, Jr., United States District Judge, in a non-jury trial 
held on April 6 through 9, 1981. Judge Owens heard the testimony 
of forty-four witnesses, received hundreds of pages of exhibits 
into evidence, considered various stipulations of fact (R. 863—66, 
898-900, 930, & 1C14-18), and received into evidence copies of 
110 depositions taken by both plaintiffs and defendant. At the 
trial, Judge Owens also requested the submission on an additional 
report by defendant's expert, Dr. Charles W. Upton, and the sub­
mission of proposed findings of fact and conclusions of law by 
the parties.^/
1/ Proposed Findings""~of Fact and Conclusions of Law were submitted 
by defendant on August 14, 1981 and by plaintiffs on August 26, 
1981. A report on the additional analyses by Dr. Upton which the court requested was submitted on August 26, 1981. The parties 
replied to each other's Proposed Findings of Fact and Conclusions 
of Law —  plaintiffs on September 11, 1981 and September 23,
1981 and defendant on September 15, 1981.

-4-



D. Request of the Trial Court for Post-Trial Submissions
Over one year after the completion of the trial/ by memorandum 

to counsel dated August 25, 1982, Judge Owens requested that 
both parties submit revised Proposed Findings of Fact and Conclusions 
of Law. Furthermore, Judge Owens indicated that he would admit 
into evidence affidavits covering any of plaintiffs statistical 
evidence that was not admitted during trial • "̂ / The parties complied 
with this request.£/

E. Decision of the Trial Court }
On November 4, 1983, Judge Owens issued findings of fact 

and conclusions of law in a comprehensive 26 page opinion.

27 Despite the fact that,_~one year prior to _ the trial, defendant served interrogatories requesting identification and 
related information regarding plaintiffs' experts ,(R. 599 ), 
plaintiffs did not notify defendant of their experts (other than 
their computer analyst who did testify) until about one week before trial (R. 795-96, 802-804). Because of these violations 
of the Federal Rules of Civil Procedure, the testimony of these 
experts was excluded. R. 816. The trial court had warned plaintiffs on September 17, 1980, that they must be ready for 
trial by February 1981 or face dismissal for failure to prosecute.
R. Vol. 12 at 17. Since four of defendant's attorneys who had 
been preparing the defense of this case for several years were 
scheduled to leave government service in the summer of 1981, 
further delay in scheduling of the trial was not possible.
R. 739, R. Vol. 11 at 18.
3/ As a result of this request, defendant submitted revised Proposed 
Findings of Fact and Conclusions of Law on November 29, 1982. 
Plaintiffs submitted an affidavit from their expert, Dr. Richard 
Drogin, covering his analyses and a new Proposed Finding of Fact 
and Conclusion of Law on December 14, 1982. On March 14, 1983, 
defendant submitted a reply objecting to the admission of the affidavit of plaintiffs' expert and including two affidavits from 
defendant's experts. Additional responses were submitted by 
plaintiffs on July 18, 1983 and on August 22, 1983 and by 
defendant on August 9, 1983.

-5-



Judge Owens granted individual relief to Plaintiff Shepard^/ 
and denied individual relief to Plaintiffs Abad and Maddox.
Further, class relief was denied. Final judgment was entered on 
November 4, 1983. It is from this judgment that appeal has been
taken.
II. STATEMENT OF THE FACTS

A. Functions and Characteristics of the Work Force At the 
Marine Corps Logistics Base ____________________
1. The Marine Corps Logistics Base: 1952 through 1975.

Pursuant to an authorization of Congress, the Marine Corps
iLogistics Base was originally established in 1952 as the Marine 

Corps Depot of Supplies.^/ As of the end of 1975, the Base 
was composed of predominately wage grade (i»£.»_ t blue collar) 
jobs. These wage grade jobs included unskilled, semi-skilled 
and skilled positions in various trades, crafts, and labor categories 
In early 1975, the work force numbered approximately 1700 employees 
with about 1100 in wage grade positions and about 600 in general 
schedule (i.e. , white collar) positions. The size of the work 
force was relatively stable from 1952 through the end of 1975.

4y Pursuant to the judgment, Mr. Shepard^s remedies have been 
accorded and his claims are no longer in issue in this case.
5/ An aspect of this case not at issue on this appeal was previously before this Court. In re Maddox, No. 81-7287 (copy of 
this Court's decision at R. 1020). See R. 774-75, 815; TR. I 2, 
192-93; and TR. II 72-73.
6/ The original purpose of the Base was to overhaul and repair 
equipment and to provide storage of military supplies and war 
reserve material. Prior to 1975, the Base was organized into 
seven operational divisions: Comptroller; Facilities andServices; Headquarters Battalion; Material; Personnel; Plans and 
Systems; and Repair.

-6-



At the end of 1975 the proportion of blacks in the work force 
was about 30%. TR. Ill 48-49, 51, 105. R. 1016-17.

2. 1976: Functional Transfer and Realignment of the
Base's Responsibilities.

On July 1, 1975, pursuant to a Department of Defense 
Directive, employees of the Marine Corps' Supply Activity in 
Philadelphia were notified that all of the positions supporting 
the inventory control point function would be transferred to the 
Base in Albany during 1976. ,

The Supply Activity work force was overwhelmingly general 
schedule (of a total of 827 positions only 6 were wage grade).
The Philadelphia work force was composed of some 525 whites and 
300 blacks. All employees were given the opportunity to transfer 
to Albany; only 280 took advantage of the opportunity. Nearly 
one-half of the white employees elected to do so; only 28 (10 
percent) of the black employees did so. The transfer was effected 
in April of 1976. TR. Ill 48-51. The Philadelphia to Albany 
transfer dramatically changed the composition of the base's work 
force and caused the percentage of blacks at the base to drop 
from 30% to 26% or 27%. Op. 5. (R.E. 41); TR. Ill 51. As a 
result of the Philadelphia transfer, the Base was required to 
fill nearly 600~new vacancies.V The Base recruited from all

7/ As a result of the transfer, the character of the Base's 
work- force was changed to such an extent that the general schedule positions previously in a minority became a majority 
of the positions in the work force. After the transfer, the 
Marine Corps Logistics Base consisted of twelve divisions, 
adding Supply Operations, Technical Operations, Procurement (now 
Contracts), Provisioning, and Logistics Systems Support to the 
prior identified divisions.

-7-



DOD activities employing individuals with requisite skills. The 
positions filled were mid— to high-level positions in skills 
found predominantly in the federal service. Because of the 
nature of the positions filled (professional/technical and 
managerial/ administrative), more white employees were being 
hired than black employees. This was of special concern to the 
Base Commanding General.£/ Op. 5-6 (R.E. 41-42) TR. Ill 52-58.

3. The Wide Range of Disparate Job Categories inthe Work Force at the Marine Corps Logistics Base.
There is a wide range of disparate occupations in the work 

force at the Marine Corps Logistics Base. Each different civil 
service job whether in the general schedule or in the wage grade 
pay plan is designated by a series code that is assigned by the 
Office of Personnel Management (OPM).£/ R. 1017 (Stipulations 
21, 22, 28). During the 1976 through 1979 time period, there

8/ The Base made special efforts to recruit blacks for^positions 
by contacting blacks in the Albany area, obtaining applications 
from them, and forwarding these applications to the Office of 
Personnel Management for rating and ranking. Through a work- 
trainee or inverted register, which granted credit for less 
education and experience, 24 spaces were set aside for such 
individuals. The Base initiated recruitment at predominantly 
black colleges such as Albany State College, Savannah State 
College, and Fort Valley State College for the Navy's co-op program (noncompetitive appointment for students). The Base 
hired twenty-three blacks through this special_recruitment effort. 
Federal law, however, precluded the Base Civilian Personnel Officer 
from hiring directly from the non-federal work force, except for 
Veterans Readjustment Act appointments. TR. Ill 52-55, 59,
65-66, 68-69. See R. 1017. Individuals who are not federal 
employees must be selected from registers competitively 
established by OPM. See FACTS, infra, at p. 11.
9/ General schedule occupations are generally considered "white 
collar" in nature (e.g. , professional, technical, managerial, 
administrative, clerical). Wage grade positions are generally 
referred to as "blue collar" (various trades, crafts, and labor 
occupations).

-8-



were employees in 181 different occupations (series codes) at 
the Marine Corps Logistics Base. D. Ex. 4a. Among the
most populous occupation series at the Base were the following:

Computer Specialist
Electronic MechanicHeavy Duty Equipment MechanicInventory Management Specialist
Production DispatcherGeneral Accounting Administrator
Clerk Steno and Reporter
General Administrator
Supply Cataloger
Equipment SpecialistMobile Equipment Metal Mechanic

Stockman
Packer
LaborerRoad Sweep Operator 
Box MakerAccounting Technician 
Supply Technician 
Supply Program Manager 
Clerk Typist 
Secretary 
Painter

1979 & 1978 Affirmative Action Plans introduced into evidence
by Plaintiffs (1978 Plan at C-26-28; 1979 Plan at C-25-27).
Among the other diverse occupations on the Base were the following:

Safety ManagerPersonnel Management Specialist 
Personnel Staffing Specialist Industrial Engineering Technician 
Production Control Specialist 
Quality Assurance Specialist Electronic Equipment Installer 
Contract and Procurement 

Specialist
Budget Administrator 
Freight Rate Specialist Optical Instrument Repairer 
Machine Tool Mechanic 
Boiler Plant Operator

Engineering Technician
Electronics Engineer
Electronics Technician
Industrial EngineerTransportation Specialist
Computer Operator
Management Analyst
AccountantGeneral Engineer
Electrician
Mason
Welder
Plumber
Carpenter

D. Ex. 1, 2, 4a (series codes in Ex. 4a, titles for series in
Ex. 1 and 2). The wage grade pay plan consists of several levels. 
The lowest level, WG-2, comprises laborers, janitors, and custodial
workers. Levels WG-4 through WG-6 are semi-skilled jobs such as
warehousemen, packers, and craters. Skilled jobs start at WG 5
10/ d"I Ex. 4a Indicates a total of 581 records which is the sum 
of the different grade levels for each of the job series. For 
example, Series 00212 appears four times: once each for grades 7, 9, 
11, 12. D. Ex. 4a at 1. Series 00212 is the identification code for Personnel Staffing Specialists. D. Ex. 2 at 5 of the attachment.

-9-



helper and progress to journeyman WG-9 through W3-11. The general 
schedule is also divided into grades and, at Albany, includes 
grades 2-15. As with the wage grade pay plan, the general schedule 
grades reflect varied skill levels and degrees of responsibility.
TR. Ill 104-105.

B. Regulations, Procedures, and Standards Governing theMerit Promotion System at the Marine Corps Logistics Base
1. Regulatory Standards and Requirements .-L̂ /

a. Regulatory Structure. 0PM issues the basic rules
and regulations governing the federal civil service. Implementing
requirements and regulations are published in a multi-volume set
of instructions entitled the Federal Personnel Manual (FPM) and
supplemental volumes thereto.il/ TR. I 23; TR. Ill 67. The
Department of the Navy issues regulations entitled "Civilian
1X7 There is an extensive statutory basis for the comprehensive 
regulations, procedures, and standards governing the Base s employment practices relative to federal civil service employees.
The basic statutory provisions covering general schedule 
employees are contained in Chapter 51, Title 5 of the United 
States Code with the applicable pay rates defined in Subchapter 
III of Chapter 53, Title 5 of the United States Code. For wage grade employees the basic statutory provisions are contained in 
Subchapter IV, Chapter 53, Title 5 of the United States Code.
■ The statutes require that 0PM establish classification standards 

for positions under the general schedule. 5 U.S.C. $-5 Agencies are then required to follow these standards for general 
schedule positions. 5 U.S.C. § 5107. For wage grade posjt.onB,
0PM is required to establish job grading standards which federal 
agencies must follow. 5 U.S.C. § 5346.
12/ There are currently 26 supplements to the FPM. FPM Ch. 001 
TTntroduction) at 1-2 (November 12, 1982). Updates to the FP^ 
and its supplements are periodically issued by 0PM. Further- 
system of FPM Bulletins and FPM Letters is used for the ear y 
dissemination of policy changes or for material of a temporary 
nature. FPM Ch. 001 (Introduction) at 2; see, e^. , D. tx. /.
0PM organizes its regulations and instructions in ĵ-ne su jec matter groups. FPM Ch. 001 (Introduction) at 2. The contents 
of the FPM are organized under these groups. FPM Ch. 002 at w  
(September 24, 1982).

-10-



Personnel Instructions" which further supplement the Federal
Personnel Manual. TR. I 23. Additionally, Department of the 
Navy activities, such as the Marine Corps Logistics Base, issue 
local instructions governing civil service employment practices.
R. 1018 (Stipulation 32); Lee Dep. 25-27; Deiter Dep. 12. It is 
the responsibility of the local civilian personnel office to 
insure compliance with this comprehensive set of regulations 
and standards. TR. Ill 119; R. 1015-16 (Stipulations 13-16).

b. Filling of Vacancies. Federal agencies such' as
the Department of the Navy have direct authority to fill vacancies 
by the promotion, transfer, or reassignment of individuals with 
civil service status. 5 C.F.R. § 315.501 and § 335.102(a)(1984);
FPM Ch. 315 § 5-2 (April 27, 1982). Within the Department of 
the Navy, this authority is delegated to the civilian personnel 
offices which service the various activities. With certain ex­
ceptions, federal agencies do not have authority to directly fill 
vacancies with individuals from outside the federal civil service. 
These individuals must be selected from registers established by 
OPM. TR. Ill 65. Based on the needs of federal agencies, OPM 
opens announcements for selected jobs, accepts applications, rates 
and ranks applicants, and establishes registers. TR. Ill 56-68;
TR. IV 135-52.

c. Qualification Standards. When vacancies are filled 
internally, federal agencies are required to ensure that individuals 
meet the qualifications standards for the job and at the specific 
grade level at which it is filled. For general schedule positions,

-11-



extensive qualification standards for each job series are published
in OPM Handbook X-118, "Qualification Standards for Positions
under the General Schedule." TR. Ill 84; Bass Dep. 13; Johnson
Dep. 11; FPM Ch. 338 §§ 3-1, 3-2a, 3-3b (May 7, 1981). For wage
grade positions, qualification requirements for each job series
are published in OPM Handbook X-118C "Job Qualification System
for Trades and Labor Occupations," which is further supplemented
by OPM's "Internal Qualification Guides for Trades and Labor
Jobs" and by detailed examining guides for the various job series.
R. 1017 (Stipulation 29). Deiter Dep. 17. 1

d. Classification & Job Grading Standards.13/
For each general schedule position, the Base was required by OPM
regulations to develop a written position description. See Bass
Dep. 20; FPM Ch. 511 § 4-1 (November 26, 1975). Personnel specialists
assigned to the Base are then required to compare the position
description against OPM issued classification standards and
thereby to determine the appropriate grade level and position
series TR. I 20; R. 1015-16 (Stipulation 14); see Lee Dep. 131-32,
159-61; Deiter Dep. 16-17, 39-40; FPM Ch. 511 §§ 3-1, 3-2, 3-3
(July 1969). For wage grade positions the Base was required by
OPM regulations to develop a written job description. FPM Supp.
532-1 § S6-6 (April 14, 1980). Base personnel specialists are
13/ Qualification standards relate to the capabilities of the 
Tndividual employee in terms of required knowledges, skills, and 
abilities. Classification standards cover the duties, responsi­
bilities, and qualifications required for positions under the 
general schedule. Job grading standards are equivalent to classi­
fication standards but apply to wage grade jobs. Since classi­
fication and qualification standards are interrelated, OPM 
develops them in coordination. FPM Ch. 271 § 2-2 (July 1969);
FPM Ch. 511 § 3-4 (July 1969).

-12-



then required to compare the job description against OPM job 
grading standards and thereby to determine the appropriate grade 
level and job series. TR. I 20; FPM Supp. 532-1 §§ S6-1, S6-4a 
(December 30, 1982).

e. Merit Promotion Requirements. Federal agencies 
are delegated authority to effect promotions by the provisions of 
5 C.F.R. § 335.102(a)(1984). Promotions can only be effected 
pursuant to an agency adopted merit promotion program that conforms 
with OPM merit promotion standards and requirements. 5 C.F.R. § 
335.103 (1984 ); FPM Ch. 335 §§ 1-1, 1-3, 1-4 (May 7, 1981)(. The 
Department of the Navy's Merit Promotion Program is promulgated 
as a Civilian Personnel Instruction and was included in local 
Base instruction covering merit promotions. R. 1018 (Stipulation 32 
Navy and Marine Corps activities, such as the Base, issue local 
merit promotion instructions which implement and supplement the 
Navy's Merit Promotion Program. Id.; TR. Ill 106-107, 119.

2. The Operation of the Merit Promotions System at the 
Marine Corps Logistics Base____________ __________
a. Issuance of the Merit Promotion Announcement.

When a vacancy exists within the work force, the supervisor 
having cognizance over the open position initiates a Standard 
Form 52 requesting that the position be filled. This request is 
forwarded via the division director to the Civilian Personnel 
Office. At the personnel office, the classification branch 
verifies the existence of an approved position or job description 
for the vacancy. TR. Ill, 80; Lee Dep. 32; Bass Dep. 20.

If merit promotion is chosen as the source to fill the 
position, a staffing specialist within the Civilian Personnel



Office analyzes the position, identifies the job elements to be 
used to rate and rank applicants, establishes a crediting plan 
(sets points to be allowed on the job elements for experience, 
etc.), and publishes the vacancy announcement. Lee Dep. 43-48;
Bass Dep. 12.

The Merit Promotion Announcement describes the duties and 
location of the position, the area of consideration and the 
experience requirements listed in the OPM qualification standards.

7 .TR. Ill, 89. This announcement is then reviewed by the division 
requesting the merit promotion and the head of the staffing 
division of the Civilian Personnel Office. (Bass Dep. 15, 16).

Based on the type of position and the grade level and in 
order to attract a response from a sufficient number of qualified 
applicants, the Merit Promotion Announcement is disseminated 
within either the Base, the Department of the Navy, or the 
Department of D e f e n s e . T R .  Ill 21, 101, 103; Bass Dep 17.
Merit promotion announcements are posted on official bulletin 
boards within each division for a minimum of ten days. See also 
R. 1018 (Stipulation 34).

b. Personnel Office's Determination of Eligibility.
Employees submit their applications with supplemental information,
stating specific experiences they have on each job element. After
14/ The Base recruits for professional and technical positions 
on Base and through DOD.activities worldwide. The Base recruits 
for managerial and administrative type positions on Base, DOD- 
wide, and federal service-wide. The Base entry-level clerical 
work force comes from registers provided by the Office of 
Personnel Management. Mid—level clerk positions at the Base 
are filled by internal promotions. TR. Ill; 103-04. The 
higher the grade level of the position to be filled the greater 
the likelihood that off Base candidates were sought. Bass Dep. 17.

-14-



the closing date listed on the announcement has passed, no more 
applications are accepted. The Civilian Personnel Office compares 
all applications to the OPM qualification standards to determine 
which applicants meet the basic eligibility requirements. For 
employees who meet the basic eligibility requirements, the current 
supervisor is requested to provide an appraisal evaluating the 
employee on the specific elements in the merit promotion announcement.
TR. Ill 84.

c. The Rating Process. Each qualified applicant is 
then rated and ranked according to his or her potential ability 
to perform the duties of the position. TR. Ill 82-88, 90. The 
crediting plan is the "yardstick" used to measure this potential.
Bass Dep. 25. The crediting plan is developed by .a staffing 
specialist based upon the OPM qualification standards, the OPM 
classification or job grading standards, and the position or job 
description. Bass Dep. 25; Deiter Dep. 28-35. For each job 
element of any given position, the crediting plan defines what 
numerical point value (from one to four) will be assigned for 
any level of prior effort, training or experience demonstrated 
by the applicant. Lee Dep. 49; TR. Ill 91; R. 1018 (Stipulation 
36). The guidelines in the X-118 and X-118C set forth what 
level of experience is required to receive specific point ratings.
Deiter Dep. 35; TR. Ill 84, 91.

For every Merit Promotion Announcement, a separate ranking 
panel is then established for the purpose of comparing each 
eligible employee's application with the crediting plan.
Lee Dep. 49; TR. Ill 91. The ranking panel consists of two or

-15-



more technical or subject matter experts who are familiar with 
the position being filled. Johnson Dep. 12, Bass Dep. 21.
Rules regarding the selection of panel members are set forth in 
the Base's Civilian Personnel Manual.̂ / Bass Dep. 22.
After all of the applicants have been rated and the results 
reviewed by the staffing specialist, the applicants are listed 
on a register in sequential order from the highest to the lowest 
aggregate score. TR. Ill 86. Those applicants who averaged 
three points per job element are classified as being highly 
qualified. Bass Dep. 27. The named of the top five highly 
qualified employees are then placed in alphabetical
order on a certificate which is forwarded to the selecting official 
for his or her consideration. TR. Ill 96; Johnson Dep. 17.

d. Selection From the List of Certified Candidates.
The selecting official must either personally interview the five 
employees or appoint an advisory selection panel to conduct the 
interviews and to recommend the name of an employee to fill the 
position. TR. Ill 87; Bass Dep. 36.

]_5/ The panel Ti then provided both written and oral instructions 
regarding the rating procedure. Each member has the Standard 
Operating Procedure provided to them. J ohnson Dep. 13. Under 
the direction of a staffing specialist, the members of the panel, 
first individually and then collectively, compare an employee's_ experience as reflected in his or her application with the crediting 
plan. In performing this task, the level of experience^disclosed 
in the application is correlated to the proper point value designated 
in the crediting plan for that level of experience. TR. Ill 93.Each panel member independently rates each candidate against all 
of the elements in the crediting plan. Subsequently, the panel 
members compare ratings, and discuss any differences. After discussion, the differences are reconciled by consensus of the 
panel members. Finally, a staffing specialist reviews the entire 
work product of the panel. Lee Dep. 84—85; Bass Dep. 25—26,
Deiter Dep. 40. -16-



When a black employee is listed on the certificate, the 
Civilian Personnel Office encourages that a black employee also 
be appointed to the advisory selection panel and, where applicable, 
notifies the selection official of affirmative action goals and 
areas of underrepresentation.̂ -6/ TR. Ill 75 , 76 , 94; Johnson 
Dep. 19. Finally, the selecting official selects the most highly 
qualified applicant(s) for the position(s).

e. Expert Evaluation of the Base's Merit Promotion 
Program. Defendant's entire selection process was reviewed by 
Dr. Charles T. Kenny, an expert in developing and implementing 
personnel selection systems.^ / TR. Ill 126. When Dr. Kenny 
was asked to explain the "similarities or dissimilarities . . .
between the promotion program aboard the Base and others you have 
seen," he stated:

I have never seen a promotion system that was as constructed 
and complex as this one . . . there are so many processes
that have to be gone through that the net effect would 
be to minimize the subjectivity that's involved in 
promotions as much as possible. I have not seen a promotional system anywhere in private industry that 
comes anywhere near putting as many checks and balances 
and steps into a system so as to minimize sub­
jectivity as much as possible. TR. Ill 162.

16/ The trial record reflects the fact that, when black employees were interviewed, a black was often a member of the 
advisory panel. See, e.g., TR. I 118, 125, 126, 152,^171,^185, 
TR. II 55, 77 , 95T96, 108, 192, 219; TR. Ill 9, 24; TR. IV 
72-73, 116-117.
17/ Dr. Kenny is an industrial psychologist and President 
St Criterion Development Associates, a consultant firm.
Dr. -Kenny received his doctorate in psychology in 1969 from 
Clark University. TR. Vol. Ill, 124-25. Dr. Kenny has taught statistics and research methodology at the University of Tennessee 
and at Memphis State University. Dr. Kenny is an expert in 
three areas relevant to this case: (1) the implementation ofpersonnel selection procedures and personnel systems; (2) analysis 
of employment statistics; and (3) work force analysis. TR. Ill 
125-28. -17-



c. The Statistical Evidence 18//
1. Defendant's Analysis of the Filling of Job Openings. 

This analysis was performed by defendant's expert, Dr. Charles 
T. Kenny. Dr. Kenny's "job opportunity analysis" is based on 
the fact that vacancies at the Base can be filled either by 
promotion of present Base employees or by accessions to the 
Base work force by hiring from outside. Therefore, to most 
accurately assess the treatment of blacks when vacancies »are 
filled, it is necessary to analyze both promotions and acces­
sions.^/ TR. Ill 154-56. In each of the census categories, 
for calendar years 1976, 1977, and 1978, Dr. Kenny compared the 
proportion of blacks filling vacancies (either by promotion or 
accession) at the Base with the proportion of blacks in the

18/ The primary source of employment data for the various "statistical analyses presented in this case was the litigation 
database. The litigation database is a computer tape containing 
work histories for each employee of the Marine Corps Logistics 
Base. These work histories included entries for each personnel 
action that was effected while the individual was employed at 
the Base. This tape was constructed and checked under the direction of personnel specialists employed at the Base. A copy 
of the tape was_furnished to plaintiffs for_their analyses. The 
parties stipulated to the accuracy of the Litigation Database.
R. 864. (Stipulation 6).
19/ Underlying all of Dr. Kenny's analyses is his initial 
step whereby all the Base occupations were translated into the 
appropriate census job categories. D. Ex. 1, 2, 4. The parties 
stipulated to the accuracy of defendant's exhibits 1 and 2 R.865. The nine census job categories are: Professional/Technical,Managerial/Administrative; Sales; Clerical; Craft Worker; Operatives
Transport; Laborers; and Service. See R. 864. (Stipulation 5)

-18-



relevant labor market.̂ 0/ to obtain the proportion of blacks 
available in each category for each year, a weighted average of 
the proportion applicable to promotions (The Base's work force) 
and the proportion applicable to accessions (either the DOD work 
force or the Albany SMSA) was obtained. The weight applied to 
each labor market was the proportion of job filled by either 
promotions or accessions.^ / TR. 156 — 58; D. Ex. 24a. Dr.
Kenny's results are as follows:22/

CENSUS CATEGORY
DISPARITY (NUMBER OF STANDARD DEVIATIONS)

1976 1977 1978
Profess ional/Technical
Managerial/Administrative
Sales WorkerClerical Worker
Craft Worker
OperativeTransport Worker
Laborer
Service Worker

+3.53
+0.87
-0.46-1.28
-0.60
-0.12
-0.45
+0.14
-1.39

+4.70 
+0.02 
-0.74 
-1.09 
-1.76 
-1.65 
-0.77 
-2.12 ' 
-2.04

1 +5.56 
-1.69 
0.00 
+0.26 -0.41 
-2.85 
-1.17 
+0.70 
-1.03

D. Ex. 24; TR. 158-59. Positive disparities are favorable to
blacks and negative disparities are unfavorable. The only results
exceeding three standard deviations are in favor of blacks in the
Professional/ Technical category. Dr. Kenny concluded that the
"207 The labor market used for promotions was the the Base’s work 
force. For accessions in the Professional/Technical and in the 
Managerial/Administrative Categories, Dr. Kenny determined the 
relevant labor market to be the overall Department of Defense 
work force. For accessions in the remaining seven census categories, 
the labor market used was that for the Albany Standard Metropolitan 
Statistical Area (SMSA).
21/ For example, if 1/3 of the vacancies were filled by promotions, the weight of 1/3 would apply to the proportion of 
blacks in the Base's wotk force and the weight of 2/3 would apply to the proportion of blacks in the labor market applicable
to accessions.
22/ When comparing the proportion of those blacks available in 
the relevant labor market with those blacks filling vacancies,
Dr. Kenny used either the standard deviation (z) test or the exact binomial calculation when there were small numbers involved.
TR. Ill 142.

-19-



statistical data does not show unfavorable treatment of blacks 
in the filling of vacancies.^/ TR. Ill, 159-60.

2. Defendant's Regression and Logit Analyses of Promotions 
Defendant's regression and logit analyses were performed 

by Dr. Charles W. Upton, an associate professor in the Graduate 
School of Management, Rutgers University.24/ D. Ex. 25. Dr.
Upton analyzed promotions at the Base using the litigation data­
base. TR. Ill 216-218. In his fipst set of analyses,^/
Dr. Upton used multiple linear regressions to analyze promotions, 
while accounting for such variables such as education level,
23/ Dr. Kenny provided a second set of analyses in which "the racial composition of the Base was compared with relevant 
labor markets. Again the Base occupations were grouped into the nine census categories and comparisons were made for four calendar 
years: 1976, 1977, 1978, 1979. After a review of the operation
of the personnel system at the Base, Dr. Kenny determined that the most appropriate labor market for the Professional/Technical 
category and for the Managerial/Administrative category was the 
DOD labor force. Fop the other seven categories, Dr. Kenny determined that the Albany SMSA was the most appropriate labor 
market. For these "static" analyses Dr. Kenny compared the pro­
portion of blacks in the Base work force versus the proportion in the labor market using the standard-deviation (z) test or, 
when the ̂ number were small, the exact binomial test. Dr. Kenny 
concluded these results did not indicate a pattern of adverse 
treatment of blacks in the employment practices of the Base.Dr. Kenny also presented results using a variety of other possible 
relevant labor markets. The results -for these other analyses 
were similar to the results for the labor markets determined 
by Dr. Kenny to be most appropriate. TR. Ill 128-54; D. Ex. 1-3, 
3a, 4, 5, 7, 8, 8a-8c, 9, 12-23, 23a.
24/ In 1969, Dr. Upton received his doctorate in economics 
from Carnegie-MelIon University. Dr. Upton is an expert 
in econometrics, a discipline which applies statistical 
techniques to economic data. TR. Ill, 213. ,
25/ As a preliminary matter, Dr. Upton conducted a "naive 
analysis" in which qualifications were disregarded. TR. Ill 
219; Ct. Ex. 2 (See TR. IV 26-27). This [analysis was "naive’’ 
in the sense that it did not "take into account any information. 
TR. IV 26.

-20-



length of time on the job, and time since last promotion.^/
TR. Ill 219-22; Dr. Upton's objective was to determine the 
comparative treatment of blacks while accounting for these factors. 
Dr. Upton used two different regression models in this first set 
of a n a l y s e s . D. Ex. 25 at 26; TR. Ill 230. For the first 
model the regression was run for all employees while for the 
second the regression was run for white employees only.

In his second set of analyses, logit analyses, Dr. Upton used 
the same independent variables that were used for the regressions. 
TR. Ill 232-33;- D. Ex. 25 at 10-11. The same two models as used 
for the regression anslyses were used again for the logit analyses. 
TR III 236-37. Dr. Upton made calculations for various groupings 
of employees and for various time periods. Dr. Upton s results 
for the two regression models and for the two logi't models for 
all employees (general schedule plus wage grade) for three time
periods are as follows:

DISPARITY (NUMBER OF STANDARD DEVIATIONS)
REGRESSION LOGIT

PERIOD USING CO­EFFICIENTS 
FOR ALL EMPLOYEES

USING CO­EFFICIENTS 
FOR WHITE EMPLOYEES

USING CO­EFFICIENTS 
FOR ALL EMPLOYEES

USING CO­EFFICIENTS 
FOR WHITE EMPLOYEES

1973-79
1975-79
1977-79

-0.66 
-0.5 8 
-0.79

-0.27
-0.18
-0.21

-0.87
-0.88
-0.18

-0.63
-0.09
-0.46

26/ Regression analysis is discussed in D. Baldus & J. Cole, 
Statistical Proof of Discrimination § 8.02 (1980 and Supp.
1983 ) .
27/ Each of these models involved five steps: Step 1: For allemployees in each group, the regression was run using the dichotomous 
variable of promotion or no promotion during the time period as the independent variable. As a result, the coefficients were 
obtained for each dependent variable in the regression equation.
Step 2: For each black employee, the probability of promotion
(FOOTNOTE 27 CONTINUED ON NEXT PAGE.)

- 21-



D. Ex. 25 at 27, 39, 43, 48. In all cases the disparities are 
less than one standard deviation.28/

At the request of the trial court, Dr. Upton submitted a 
post trial report in which he repeated his analysis while accounting 
for each division and major staff office at the Marine Corps 
Logistics Base. TR. IV 199-201. The results of these analyses 
for 1977 through 1979 are as follows (staff offices indicated by *):

(FOOTNOTE 27 CONTINUED) }was determined by: (1) inserting that employee's values for
each dependent variable; (2) multiplying each value by the appropriate coefficient (determined from Step 1); and (3) summing the results.
Step 3: The expected number of black promotions was then determined
by summing the results for each black employee as determined in Step 2. Step 4: The value of a standard deviation was determined 
by using a procedure that accounted for the estimation error of 
the regression. Step 5: The disparity in terms of the numberof standard deviations was determined by subtracting the expected 
number of black promotions from the actual number and dividing 
the result by the value of the standard deviation as determined 
in Step 4. D. Ex. 25 at 2-10. The logit analyses were basically 
the same as the regressions except, rather than using a dichotmous 
variable as the independent variable, the logit uses the variable 
log (P/(l-P)), where P is the probability of promotion. See D.
Baldus & J. Cole, Statistical Proof of Discrimination § 8.3 
n.55 (1980 and Supp. 1983); H. Theil, Principles of Econometrics 
632-35 (1971).
28/ Dr. Upton presented at trial the results of one further 
regression analysis. TR. Ill 237—39. In this analysis, Dr.Upton used the data for all employees and included an additional 
dependent variable, the race of the employee. Dr. Upton analyzed 
nine different groups for seven different time periods, for a 
total of 63 cases. In 28 cases Dr. Upton found the coefficient 
of the race variable to be positive for blacks and in 35 cases 
the coefficient was negative. Given this relative balance, Dr.
Upton concluded that the difference could easily be attributed 
to chance. TR. Ill 238-39.

-22-



DIVISION OR 
STAFF OFFICE

DISPARITY (NUMBER OF STANDARD DEVIATIONS)
REGRESSION COEFFICIENTS 

FOR ALL EMPLOYEE
REGRESSION COEFFICIENTS 

FOR WHITE EMPLOYEES
WG

POSITIONS
GSPOSITIONS

WGPOSITIONS
GS

POSITIONS
*1. Central Design & Proaram Activity 0.00 + 0.42 0.00 + 0.62
2. Comptroller 0.00 -1.17 0.00 -0.29
3. Contracts 0.00 + 0.14 0.00 +0.45
*4. Dep.Ch.of Staff 0.00 -0.26 0.00 — 0.26
5. Facil. & Serves. -0.47 -1.11 -0.15 -0.70
6. Headquarters + 0.49 0.00 + 0.49 0.00
7. Loqistics 0.00 + 0.81 0.00 + 0.70
8. Material + 0.08 -*0.02 -0.02 -0.19
9. Supply Systems 0.00 0.00 0.00 0.00
10. Personnel + 0.67 +1.45 + 0.31 +0.42
11. Provisioning 0.00 + 0.25 0.00 ,+0.19
12. Repair -0.51 + 0.91 -0.63 +0.40
13. Supply Ops. -0.36 -0.78 -0.28 -0.85
14. Technical Ops. 0.00 +0.65 0.00 + 0.61

TOTAL -0.52 -0.29 -0.43 +0.16
Upton Report, Analysis of Promotions at Mar-ine Corps Logistics_Center,
Albany Phase II, at 33, 34, 36, 37 (July 27, 1981)(filed August 26, 
1981). All of the disparities are less than two standard deviations.

' - ■ AThe largest disparity is favorable to blacks, +1.45,,.standard deviations, 
Although none are significant, 20 of the disparities are favorable 
to blacks a'nd 20 are adverse.

3• Plaintiffs’ Statistics.
a. Descriptive Statistics. Plaintiff's computer 

analyst, Mr. Martin Mader presented a series of computer printouts 
containing data that was derived from the litigation database.
P. Ex. 1-22; TR. Vol. I, 42-82. These printouts contained descriptive 
statistics which provided a variety of static'snapshot comparisons 
between white and black employees of the Marine Corps Logistics 
Base. Appellant's have summarized these descriptive statistics

-23-



in their brief at pages 21 and 22.29/
b. Statistical Analyses Submitted By Affidavit.

Pursuant to an order of the trial court, plaintiffs, over the 
objection of defendants, were permitted to submit additional 
statistical evidence by affidavit. Trial Court Memorandum To 
Counsel (August 25, 1982); R. 1369-75; OP. 10 n. 4 fR.E. 46).
These analyses were performed by Dr. Richard Drogin, a statistician. 
Dr. Drogin performed two sets of analyses: (1) promotions out
of grade; and (2) analysis of the merit promotion process.
R. 1414-33.

(1) Promotion Out of Grade. For both general
schedule and the wage grade positions, Dr. Drogin compared the
proportion of blacks promoted out of a given grade, in a given
year, with the proportion of blacks in that grade at the Base,
at the beginning of the year. Separate comparison were made for
general schedule employees and wage grade employees. However,
for each pay plan, all occupations were lumped together in a
single pool for each grade. Dr. Drogin conducted these analyses
for seven years, 1973 through 1979.^/ Dr. Drogin's model
29/ When all general schedule employees for all occupations were ̂ 
grouped together by grade level, plaintiffs' descriptive statistics 
indicated that a higher proportion of whites than blacks were at 
grade GS-11 and_above. When all wage grade employees_for all 
occupations were grouped together by grade level, plaintiffs descriptive statistics indicated that a higher proportion of 
whites than blacks were at grade WG-10 and above. Further, when 
all general schedule and all wage grade employees were grouped together across all of occupations, the proportion of whites in 
supervisory positions was higher than the proportion of blacks.
Id. Appellants' Brief at 21, 22.
30/ For the general schedule employees, when Dr. Drogin summed 
his results together for all seven years for all grades, his 
model indicated a disparity of 23.1 promotions which equated
(FOOTNOTE 30 CONTINUED ON NEXT PAGE)-24-



contained no information on qualifications and did not control 
for job categories. R. 1416-19; 1427-28.

(2) Merit Promotion Analyses. These analyses were 
based on data coded by plaintiffs from the merit promotion vacancy 
announcement files of the Marine Corps Logistics Base. These 
analyses were also conducted for the seven years, 1973-1979.
For these analyses, Dr. Drogin compared the relative proportions 
of blacks and whites in five categories. These analyses included 
no information on qualifications and lumped together all job
categories. I

Defendant's expert, Dr. Upton, reviewed Dr. Drogin's affidavit 
and provided detailed analyses illustrating that the results 
reported were unreliable because of improper data aggregation.^/
R. 1387-88. In response plaintiffs submitted a reply affidavit 
from Dr. Drogin in which he simply denied that had improperly 
aggregated. R. 1461 (item 9). However, Dr. Drogin did not 
respond with any specific details as to the aggregation procdure

(FOOTNOTE 30 CONTINUED)to -2.0 standard deviations. For wage grade positions, when the 
results were similarly summed together, Dr. Drogin's model indicated 
a disparity of 90.7 promotions which equated to a disparity from 
expected of -7.8 standard deviations. This model failed to account 
for qualifications,and job categories and assumes that the Base is 
the relevant labor market.
31/ Dr. Drogin obtained from his model the following results for 
each of his five categories: (1) for Qualified or Highly Qualified
from Applicants, the disparity was -4.72 standard deviations;
(2) for Highly Qualified from Qualified or Highly Qualified, 
the disparity was —2.64 standard deviations; (3) for Selected 
from Qualified or Highly Qualified, the disparity was -2.69 standard deviations; (4) for Selected from Among Highly Qualified, 
the disparity was —2.47 for standard deviations; and (5) for Selected from Among Applicants, the disparity was -4.32 standard 
deviations. R. 1416-17, 1419-22, 1429-33.

-25-



for the announcements at each grade level which would have been 
necessary to refute Dr. Upton's analyses. Based on this evidence 
in the record, the Trial Court factually determined that Dr. 
Drogin had improperly aggregated the vacancy announcement data.
Op. 11 (R.E. 47).

D. The Anecdotal Evidence
1. Claims of Named Plaintiff Maddox. For reasons known 

only to plaintiffs, Mr. Maddox did not testify during the trial 
of this case. Plaintiffs chose instead to rely on Mr. Marddox's 
deposition. Appellant's Brief at 53. Mr. Maddox claimed that 
he had been discriminatorily denied the opportunity to transfer 
into a wage grade position as an electrical equipment helper at 
the WG-5 level. R.E. 16. Mr. Maddox was hired by the Base as a 
painter's helper, WG-5, on August 12, 1974, pursuant to the 
Base's authority under the Veteran's Readjustment Act. Maddox 
Dep. 6. About five months after he was hired, Mr. Maddox applied 
under a Merit Promotion Announcement for a position as an auto 
electric helper. Id. 8-9.

Along with several others, Mr. Maddox was interviewed for 
the position but was not selected. Two individuals, one white 
and one black, were selected to fill the available positions.
Id. 12. These events transpired in early 1975. 1̂ 3. 8-9.
In April 1977, Mr. Maddox observed that two white males 
were hired into the electric shop in the Base's Repair 
Division.32/ As a result of these observations, he initiated his

32/ At some time subsequent to early 1975, but before April 
1977, Mr. Maddox observed that a white male had been hired from 
the outside into the Base's generator shop. I_d. 16.

-26-



administrative complaint in April 1977. Although since 1975 
Mr. Maddox had never filed another application for a Base job 
in the electrical field, he was of the opinion that the Base 
should have known that he was still interested in electrician 
jobs. Id. 16-17. After the filing of his complaint, Mr. Maddox 
ygs transferred into a position as an electrical equipment helper, 
on July 5, 1977.

Based on a review of Mr. Maddox's deposition, the trial 
court held:

The job for which Mr. Maddox sought a transfer was 
filled by one black and one white, and he was eventually 
transferred into the department where he wanted to be. His 
loose allegations concerning his claim that he was not moved into the department where he wanted to be_as soon as he would 
have liked and concerning his claim that informal channels of 
communciation about job vacancies existed for whites are not sufficient to establish a prima facie case of discrimination.

Op. 25 (R.E. 61). Plaintiffs have not appealed this factual
finding of the trial court.

2. Claims of Named Plaintiff Abad. Mr. Abad transferred 
to Albany from Philadelphia, Pennsylvania, in April 1976. Mr.
Abad claimed he was denied a supervisor's position because of
his race. R.E. 16-17. Mr. Abad admitted he had been told
prior to moving to Albany that, in the event he could not be kept .
in a supervisory position, he would nonetheless be retained in
his field at his current grade, GS-7. TR. Ill 14. Shortly
after Mr. Abad reported to Albany he was interviewed for a supervisory
position at his same grade level, GS-7. Tr. Ill 8. The interviewing
panel consisted of three female employees, one black and two
whites. TR. Ill 8-9. Mr. Abad was told by his EEO counsellor

-27-



that the panel selected .’a white male for the supervisory position 
because of his "tenure in the position" TR. Ill 15; see TR. II 130.

The trial court found that since Mr. Abad's individual 
administrative complaint was not filed until March 21, 1977, his 
claim was time-barred.33/ 0p. 25 (R.E. 61). Plaintiffs’ have not
appealed this factual determination of the trial court.

3 # Testimony of Unnamed Class Members. Twenty— i.our 
unnamed class members testified for plaintiffs. While some of

i

these class members believed they had been denied a promotion on 
the basis of race, none of these witnesses offered any direct 
evidence of discrimination.

The unnamed class members did not present information as 
to the required qualifications for the positions about which 
they testified nor did plaintiffs otherwise present such evidence. 
Generally, the unnamfed class members were unaware pf the quali­
fications of those individuals selected for the positions about 
which they testified, nor did plaintiffs otherwise present such 
evidence. Oftentimes other blacks or minorities were selected for 
the positions about which the unnamed class members testified. See 
e. g. , TR. II 40, 43, 91 , 92, 222, 223.
33“/ The court also noted that, if Mr. Abad' s claim had not 
been time barred, his claim still would have failed because 
"Mr. Abad failed to show that the defendant's articulated, legitimate, nondiscriminatory business reasons [selection o 
an employee with tenure in the job] was pretexual. Op. 2b
(R.E. 61).
34/ Several of the unnamed class members testified that either 
they had not been discriminated against in promotions or that 
they had 'Ho'complaint about their promotions. See, e^. , TR. ll 
135, 189, 190, 206-208. Many of the unnamed class members testifie 
that they had received repeated promotions. Seer e.'JL* • TR* 1 , 172-85, 212; TR. II 37, 217-23. Many of the unnamed class members
(FOOTNOTE 34 CONTINUED ON NEXT PAGE.)-28-



The trial court reviewed the testimony of the unnamed plaintiffs 
and found that:

The testimony of base employees not selected for 
promotion was conclusory. None of the individuals 
gave a descriptive explanation of the application 
and interview process. Although the incidents related 
reflect a series of bad results, that evidence does 
not establish a pattern or practice of discrimination.

Apart from the named plaintiffs, only half of the 
testifying employees were persuasive on the issue of 
discrimination in the promotion process. . . . However, 
repeatedly in those cases, blacks were involved in the 
selection process as a member of a selection panel, a 
staffing specialist, a supervisor, or in one incident 
as a division director who had to approve a selection.

Op. 17 (R.E. 53). The trial court found that this anecdotal
testimony either standing alone or in combination with the
statistical evidence in the record was insufficient to make
out the class claim. OP. 18-19 (R.E. 54-55). Plaintiffs have
not appealed this factual finding of the trial court.

(FOOTNOTE 34 CONTINUED)testified that crther blacks served on the advisory selection 
panel for the positions about which they testified. See, e.g., 
TR. I 118, 125, 126, 152, 171, 185? TR. II 55, 77, 95, 96, 108, 
192, 219. Many of the class members also admitted that they had 
never been refused training which they had requested. See e♦g., 
TR 1.124, 168, 213-4; TR II 100, 110, 174, and one turned down 
the opportunity for training that would have materially assisted 
him in obtaining a promotion. TR. II 210-11. Several of the unnamed class members testified that they had black supervisors, 
and one noted that one of the personnel staffing specialist, Ms. 
Mabel J. Bass, was black. See, e.g., TR. I 215; TR. II 93 , 104, 
112, 138. In one case, the selection about which the unnamed class member testified was approved by a black officer—in—charge. 
TR. IV 185-86, 190).

-29-



III. STANDARD OF REVIEW
A. Issue I & Argument I.

The finding of the trial court that the statistical and 
anecdotal evidence failed to show classwide discrimination in 
promotions is subject to review under the "clearly erroneous"
standard. Pullman Standard v. Swint, 456 U.S. 273, ____ (1982);
Eastland v. Tennessee Valley Authority, 704 F.2d 613, 620 (11th 
Cir. 1983).

B . Issue II & Argument II. 1
The question raised by plaintiffs relative to relitigation 

of the individual claims of the named plaintiffs is a question of 
law that was not directly raised in the court below. If the 
Court accepts this issue on appeal, it is reviewable as a matter 
of law. The underlying findings of the trial court relative to 
the individual claims of the named plaintiffs, while not challenged 
on appeal, are findings of fact.

C. Issue III & Argument III.
The question raised by plaintiffs relative to relief for 

unnamed class members is a question of law that was not raised 
with the court below. If this Court accepts this issue on appeal, 
it is reviewable as a matter of law. The underlying findings of 
the trial court relative to the testimony of the unnamed class 
members, while not challenged on appeal, are findings of fact.

D. Issue IV & Argument IV.
The question raised by plaintiffs relative to whether or 

not a cause of action can be based on the Title VII requirement 
for an affirmative action plan is a question of law and is reviewable 
as such.

-30-



SUMMARY OF THE ARGUMENT
The trial court was not clearly erroneous in its factual 

findings that plaintiffs' statistical analyses each suffered from 
one or more of the following fatal defects (1) failure to account 
for required qualifications; (2 ) failure to control for the widely 
diverse job occupations; and (3) failure to use an appropriate 
labor market. Nor did the trial court err in its factual finding 
that defendant's more refined statistics indicated no discrimination

iin promotions. Given these subsidiary findings, the trial court s 
ultimate finding of no classwide discrimination was not clearly 
erroneous.

As required by fundamental class action principles under the 
Federal Rules of Civil Procedure, Named Plaintiffs Abad and Maddox 
placed their individual claims at issue in their class complaint 
and presented evidence on these claims to the trial court. The 
findings of the trial court denying these claims stand as a bar 
to religiation of these individual claims by these named plaintiffs.

The unnamed class members who testified at the trial are not 
entitled to relief based on their testimony for the trial court 
made no findings of discrimination relative to this testimony nor 
was it asked to do so by plaintiffs. Moreover, such findings 
and relief for unnamed class members would not be appropriate as 
a result of Phase I proceedings in a Title VII class action.

-Clear precedent of this court states that a cause of action 
cannot be based on the Title VII requirement to develop an affir-

-31-



mative action plan. Plaintiffs have given no rationale for why 
this precedent should not apply in this case.

STATEMENT OF JURISDICTION
This appeal was taken from a final judgment of the district 

court denying individual relief for named Plaintiffs Maddox and 
Abad and denying class relief. R.E. 36. The jurisdiction of this 
Court rests on 28 U.S.C. § 1291.

ARGUMENT

I. THE TRIAL COURT DID NOT ERR IN ITS FACTUAL FINDINGS THATPLAINTIFFS' STATISTICS: (1) FAILED TO ACCOUNT FOR QUALIFICATIONS
REQUIRED FOR MANY JOBS AT THE BASE: (2) FAILED TO CONTROL FOR
THE WIDE RANGE OF DISPARATE JOB OCCUPATIONS AT THE BASE; AND(3) FAILED TO MAKE COMPARISONS WITH A LABOR MARKET REFLECTING
THE ACTUAL EMPLOYMENT PRACTICES OF THE BASE. NOR DID THE
THE TRIAL COURT ERR IN ITS FACTUAL FINDING THAT DEFENDANT'SMORE REFINED STATISTICS INDICATED NO DISCRIMINATION IN PROMOTIONS.
A. To Be Found Meaningful, Statistical Evidence Must Account 

For Certain Fundamental Principles Which Are Necessary 
To Establish Title VII Liability________________________

It is well accepted that in cases where discrimination is 
at issue "'[s]tatistical analyses have served and will continue 
to serve an important role. . .'" International Brotherhood of
Teamsters v. United States, 431 U.S. 324, 339 (1977) (citations 
omitted). Where statistical disparities are sufficiently 
compelling, they alone can constitute a prima facie case of 
discrimination. Hazelwood School District v. United States, 433 
U.S. 299, 307-08 (1977). The Supreme Court has cautioned, however, 
that statistical analyses "come in infinite variety" and that 
"their usefulness depends on all of the surrounding facts and 
circumstances." Teamsters, supra, 431 U.S. at 339—40. The courts 
have also recognized that due to "the significant role that

-32-



statistics can play in discrimination cases and [because] of 
their inherently slippery nature, it is imperative that they be 
used properly." Wilkins v. University of Houston, 654 F.2d 388,
395 (5th Cir.) reh'g denied, 662 F.2d 1156 (1981), vacated

U.S. ______, 103 S. Ct. 34 remanded on other grounds, 695
F.2d 134 (5th Cir. 1983). The courts have further noted that 
"[t]he successful prosecution of class—wide employment discrimination 
claims demands considered and refined statistical analysi^s. . .". 
Rivera v. City of Wichita Falls, 665 F.2d 531, 547 (5th Cir.
1982)(Unit A). The underlying reason for these concerns is 
clear —  care and refinement are necessary to ensure that the 
statistical models employed are actually comparing similarly 
situated employees and are actually measuring the employment 
practices at issue.

1. When Addressing a Wide Range of Disparate 
Occupational Categories Such As Those At The Base, Meaningful Statisitcal Comparisons 
Must Account For Both Qualifications And Job 
Categories. The Trial Court Did Not Err In 
Its Factual Finding That Plaintiffs Failed 
To Account For Either.____________________

When analyzing the Base's wide range of disparate occupational 
categories, it is particularly important that refined statistical 
evidence be proferred. Eastland v. Tennessee Valley Authority,
704 F.2d 613, 624-25 (11th Cir. 1983); Valentino v. United States, 
Postal Service, 674 F.2d 56, 66, 68 (D.C. Cir. 1982); Wilkins 
v. University of Houston, 654 F.2d 388, 398, 405 (5th Cir. 1981).
It is not sufficient to attempt only to account for qualifications

-33-



without also controlling for occupation category.35/ As the court 
in Eastland noted: " . . .  Eastland's analyses account for many
objective qualifications, but the failure to control for job 
category casts doubt on whether the regressions are comparing 
appropriate groups." 704 F.2d at 625? Accord, Valentino, 674 
F.2d 70-71. Of course, in the instant case, the statistical 
analyses introduced by plaintiffs failed to account for either 
qualifications or job categories. Op. 9, 10 (R.E. 45, 46).

2. The Trial Court Did Not Err In Its Factual Finding That Plaintiffs' Relevant Labor Market Comparisons Were Erroneously Restricted 
To The Base's Work Force —  A Labor Market 
That Did Not Accurately Reflect The Sources 
From Which The Base Filled Vacancies.________

As the trial court correctly held (Op. 13-14 [R.E. 49-51]), 
statistics based on comparisons to the internal work force without 
evidence of qualifications are meaningful only if there is a 
policy of filling upper level positions through promotion of 
employees hired at entry-level and trained by the company in the
requisite skills." Rivera, 665 F.2d at 541 n.16; Fisher_v.
Procter & Gamble Manufacturing Co., 613 F.2d 527, 544 (5th Cir. 
1980), cert, denied 449 U.S. 1115 (1981). The record is 
devoid of evidence showing that, for any major portion of the

35/ The fact that special qualifications are required for 
many of the jobs and positions on the Base is evident from the 
job titles. See FACTS, supra, at p. 9. Therefore, plaintiffs 
were required to present statistics accounting for required 
qualifications. Wilkins, 654 F.2d at 409 n.37. Accord.,
Moore v. Hughes Helicopter, 708 F.2d 475, 482-83 (9th Cir. 
1983); EEOC v. Radiator Speciality Co., 610 F.2d 178, 185 (4th 
Cir. 1979.)

-34-



work force, the Base has such a policy.36/ In fact, the evidence 
shows that for major portions of the higher level positions in 
the work force, just the opposite is true —  the Base recruits 
extensively from the outside.37/ Plaintiffs' erroneous assumptions 
that the Base work force was the relevant labor market either for 
all positions or for any major portion of the positions on the Base, 
is not supported by any evidence in the record.

B. The Trial Court Did Not Err In Its Factual Findings The Plaintiffs' Statistical Analyses Each Suffered 
From One Or More Of The Following Fatal Defects:(1) Failure To Account For Required Qualifications;
(2) Failure To Control For Job Occupations; and
(3) Failure To Make Comparisons With An Appropriate
Labor Market. ___________________________________

Because of the failure of plaintiffs' statistics to account 
for qualifications and to control for disparate job categories, 
these statistics provide no meaningful measure of the effect of 
the Marine Corps Logistics Base's employment practices on similarly 
situated black and white employees. Eastland, supra, 704 F.2d 
at 624-25; Valentino, supra, 674 F.2d at 66, 68, 70-71. Moreover, 
plaintiffs' statistical comparisons that are based on an improper 
labor market are of little or no meaning. These and other defects 
in each of the three sets of analyses introduced by plaintiffs 
are briefly addressed in the arguments that follow.

367 Plaintiff's cite Carroll v. Sears Roebuck Co., 708 F.2d 
183 (5th Cir. 1983) to support their contention that the internal 
work force is appropriate. Appellants' Brief at 36 n.39. In 
Carroll, the court found that "Sears has no written criteria or 
guidelings for promotion, and does not post notices concerning 
specific job openings or promotion opportunities." I_d» 192.The facts in this case are far different. See FACTS, supra, 
at p. 10-17.
T77 See FACTS, supra, at pp. 7-8, 13-14. Federal Government 
employees outside the Base are eligible to compete for positions 
(FOOTNOTE 37 CONTINUED ON NEXT PAGE.)

-35-



1. Plaintiffs Descriptive Statistics. Plaintiffs' 
descriptive statistics cover the distribution of blacks and whites 
by grade level and by supervisory position. These statistics 
fail to account for both qualifications and job category. Plaintiffs 
also provide salary comparisons using two proxies for qualifications: 
level of education; and years of service.38/ However, these 
salary comparisons fail to control for the wide range of disparate 
job categories at the Marine Corps Logistics Base.

Plaintiffs' descriptive statistics are "static statistics" 
that encompass the effects of acts and practices which long predate 
the relevant time period applicable to this case. See Argument, 

at p. n. • Further, even if these descriptive
statistics had addressed only one job category, which of course 
they did not, such "distribution statistics" are of probative 
value in only very limited situations which are not applicable 
(FOOTNOTE 37 CONTINUED)"under Merit Promotion Announcements. Further, most higher level 
positions are announced Navy-wide or DOD—wide. Further, Federal 
employees elsewhere in the Navy can be noncompetitively reassigned 
(without promotion —  a so-called "lateral transfer") to fill a 
vacancy or, similarly Federl employees cam be transferred non­
competitively from other Government agencies. Additionally, 
individuals outside the Base can be selected from OPM registers.
See FACTS, supra, at p. 11.
38/ Plaintiffs derived years of service from the service compu­
tation date information in the litigation database. Service 
computation date encompasses total federal service, including 
military service and is not limited to service at the Marine 
Corps Logistics Base. The most significant defect in this proxy is that no credit is given for non-federal government work experience. 
Thus, for example, an individual with 2 years in the federal service 
and no other experience is equated to another employee with 2 
years in government service and 10 years of work experience 
in the private sector.

-36-



here. Rivera, 665 F.2d at 541 n.16; Fisher, 613 F.2d at 544,39/ 
Plaintiffs' distribution statistics lump together numerous job 
categories, a large portion of which require special qualifications. 
In these circumstances, such distribution statistics are entitled 
to no weight.

2. Plaintiffs' Statistics On Promotions Out of Grade. 
Plaintiffs submitted two analyses of promotions out of grade, one 
for general schedule positions and the other for wage grade 
positions. For both cases, all occupations were lumped together 
in single pools at each grade level.££/ Moreover, no attempt was

39/ When special skills are required, statistical data must 
account for required qualifications. Pack v. Energy Research 
& Development Administration, 566 F.2d 1111, 1113 (9th Cir.
1977) (The fact that 95% of the professional employees below 
GS-11 were female while only 2% of the employees at GS-11 and 
above were female did not establish a prima facie case because 
there was no evidence of qualifications for the higher level 
positions); Piva v. Xerox Corporation, 654 F.2d 591, 596-97 (9th 
Cir. 1981) (distribution statistics unpersuasive because sales 
personnel required advance sales skills). Meaningful statistics 
must account for the nature of the work performed and must focus on whether entitled promotions were denied. Heagney v. University 
of Washington, 642 F.2d 1157, 1164 (9th Cir. 1981). Only when a 
position involves skills that many persons can easily learn and when no qualification criteria have been established for filling 
such positions, then plaintiffs can support their claim with 
data demonstrating a statistically signficiant disparity between 
representation in lower positions and that in higher paying 
positions. O'Brien v. Sky Chefs, Inc., 670 F.2d 864, 867 (9th 
Cir. 1982).
40/ Plaintiffs note that more than half of the disparity indicated 
by their model for wage grade positions 46.4 out of 90.7 for 1973-1979) occurred at the WG-5 level. Appellants’ Brief at 24.
As the record indicates.the WG-5 level is at or near the top of 
the .grade range for many semi-skilled wage grade jobs while, on the 
otherhand, is the entry level for many of the highly skilled 
technical jobs in the Base's work force. See FACTS, supra, at p. 9-10. Lumping together all employees at the WG-5 level regardless 
of job occupation clearly does not provide results that compare similarly situated employees. Additionally, of course, qualifications 
of those in the highly skilled technical jobs such as electronics 
mechanic or electrician must be addressed in any meaningful 
comparisons. -37-



made to account for qualifications. Such analyses are simply 
meaningless. As the trial court noted:

This additional evidence does nothing to bolster 
the inadequate descriptive statistics . . . submitted
by the plaintiff. Op. 10 (R.E. 46).
3. Plaintiffs' Analyses of Merit Promotions. Plaintiffs'

third set of analyses covered merit promotion data which plaintiffs
coded from the vacancy announcement files at the Marine Corps
Logistics Base. Like plaintiffs' first two sets of analyses,
this last set is also fatally flawed by the failure to properly
account for job categories and qualifications.^/

As the trial court noted, "the only somewhat meaningful"
comparison made by plaintiffs was the comparative rates of black
and white selections from among those highly qualified and from
amony those both qualified and highly qualified. However, plaintiffs'
analysis suffers from other defects that render these results
meaningless. Based on a review of plaintiffs' analysis of merit
promotion data, defendant's expert, Dr. Upton determined that
41/ The Base's determination of those highly qualified was based 
oii the application, by a panel of subject matter experts, of a 
crediting plan developed from the extensive OPM mandated qualification, classification, and job grading standards applicable 
to each specific job occupation. See FACTS, supra, at pp. 10-16. 
Under these circumstances, meaningful comparisons of those highly 
qualified from all applicants required that qualifications be 
accounted for. JWilkins, 654 F.2d at 409 n.37; Accord., Moore,
708 F.2d at 482-83; Radiator Speciality Co., 610 F.2d at 185.

"Computer Specialist" is one of the most populous occupations 
on the Base. See FACTS, supra, at p. 9. the OPM qualification standards for computer specialists has specific provisions relative 
to types of experience required for these positions. X-118 
Qualification Standards, Computer Specialist Series, at 2-4 
(1968). Similarly, this standard includes specific provisions 
for substituting relevant education for experience. Id. at 4-5. 
Including relevant computer related experience and education in 
statistical comparison would provide results that would have a meaningful relationship to the treatment of similarly situated 
individuals.

-38-



Dr. Drogin had improperly aggregated the vacancy announcement in
such a fashion as to render the results unreliable.^2/ "This
'aggregation' renders the negative numbers found by Dr. Drogin
deceptive." Op. 11 (R.E. 47). There is no evidence in the record
to support the contention that this was a clearly erroneous
finding.43/ Moreover, as the trial court noted (Op. 15 [R.E.
4 2/ Dr . Drogin '”s bald denial of improper aggregation procedures 
simply wasn't a sufficient response in the face of Dr. Upton's 
substantial analysis. Plaintiffs bore the-burden of proof and 
had the opportunity to respond. Tfyeir response simply was not sufficient. The trial court had an opportunity to observe Dr.
Upton's credibility and evaluate his competence on the witness 
stand. Plaintiffs were well aware of this. It was because of their own egregious pre-trial tactics that plaintiffs forfeited 
their right to present Dr. Drogin at the trial. See PROCEEDINGS, 
supra, n.2 at p. 5. The second chance granted to them by the trial court, sua sponte and over defendant's objection, was much 
more than they were entitled to after such outrageous pre-trial 
behavior. , ,
43/ It is important to note that results reported by plaintiff 
are based to a large extent on time—barred acts and practices.
For this comparison, for the years 1973 through 1979, Dr. Drogin's 
flawed model indicates a total disparity of 9.96 selections 
which equated to 2.47 standard deviations. Of this disparity of 9.96 selections indicated by Dr. Drogin's model ,'h'owever, about 
half were found by these analyses to have occurred prior to 1977 
(disparity of 4.74 selections in 1976 alone.) which predates 
the relevant time period in this cases. R. 1432 (Attachment to 
Aff. ‘of Dr. Drogin). Therefore, even Dr. Drogin's flawed analysis 
indicates a very small loss of selections —  an average of ap­
proximately 1.5 selections per year for the three actionable years, 1977 through 1979. Title VII embodies strict time limitations 
that govern the filing of an administrative charge of discrimination.
E.g., Delaware State College v. Ricks, 449 U.S. 250 (1980);
United Air Lines, Inc, v. Evans, 431 U.S. 553 (1977). The_rule is no different-for class action plaintiffs than for individual 
plaintiffs; a class representative cannot resurrect the individual 
claims of class members that are time-barred. See, e.g. , Laffey 
v. Northwest Airlines, Inc., 567 F.2d 429, 472-76 (D.C. Cir.
197 6")“'"“cert, denied, 4 3? U.S. 998 (1978). In the instant case 
the informal administrative complaint upon which the class 
allegations rest was filed on April 28, 1977. R. 578. Under then applicable Civil Service Regulations^ all claims preceeding 
the date of this informal complaint by mope than 90 days, i.e., 
claims prior to January 28, 1977 were, as a matter of law, time- 
barred. 5 C.F.R. § 713.602. Recognizing the importance of the 
time limitations in defining statistical evidence in this case.
As stated by the Supreme Court in Evans, supra, 431 U.S. at 558: 
(FOOTNOTE 43 CONTINUED ON NEXT PAGE.)

-39-



51]), even these flawed results of Dr. Drogin did not exceed 
three standard deviations.44/

C. Defendants' Statistical Analyses, Which Account For 
Job Categories And Qualifications, Either Separately 
Or In Combination, And Which Employ A Labor Market 
Reflecting The Employment Practices Of The Base,
Indicate No Classwide Discrimination In Promotions.

Defendant's statistical evidence is substantially more 
probative than that submitted by plaintiffs because defendant's 
statistics account for job categories and qualifications, either 
separately or in combinations while plaintiffs accounted Jfor

(FOOTNOTE 43 CONTINUED)[a] discriminatory act which is not made the basis for a 
timely charge is the legal equivalent of a discriminatory 
act which occured before the statute was passed. It may 
constitute relevant background evidence in a proceeding 
in which the status of a current practice is at issue, but 
separately considered, it is merely an unfortunate event in 
history which has no present legal consequences.

Thus, just as an employer may defeat a Title VII class charge by 
showing "that the claimed discriminatory pattern is a product of^ pre-Act hiring rather than unlawful post-Act discrimination. . .",
Teamsters, supra, 431 U.S. at 360; Hazelwood, supra, 433 U.S. at 
310, too may an employer defeat a Title VII class charge by showing that the claimed discriminatory pattern is a product of 
acts which occurred outside the relevant charging period i »e•• prior to the period within the reach of the administrative charge. 
Ste. Marie v. Eastern Railroad Association, 650 F.2d, 395, 401-02; 
(2d Cir, 1981); Movement for Opportunity and Equality v. General 
Motors Corp., 622, F.2d 1235, 1245 (7th Cir. 1980); Pounqy 
v. Prudential Ins. Co., 499 F. Supp. 427, 444-45 (S.D. Tex. 1980), 
af f rd", 668 F. 2d 79 5 (5 th Cir. 1982), Accord Hebert v. Monsanto 
Co.,682 F.2d 1111, 1123 (5th Cir. 1982). Clearly then the 
Trial Court was correct in its finding that the years directly 
at issue in this case are 1977, 1978, and 1979. Op. 4 (R.E.40). The 1979 ending for the relevant time frame in this case 
was based on the discovery cutoff of December 31, 1979. R. 600.
44/ The Ninth Circuit has recently held that a statistical 
disparity sufficient to independently support an inference of
(FOOTNOTE 44 CONTINUED ON NEXT PAGE.)

-40-



neither in their inferential statistics.
1. Job Vacancy Analysis. The "job opportunity analysis" 

performed by defendant's expert, Dr. Charles Kenny, is more 
probative than any of plaintiffs' statistics because Dr. Kenny 
both accounted for job categories and more accurately modeled 
the relevant labor markets.^5/

Dr. Kenny controlled for the wide range of disparate jobs 
at the Base by grouping the occupations in the nine census job 
categories. See FACTS, supra, at pp. 18-20. Given that vacancies 
were filled both by accessions (hiring) and promotions, Dr.
Kenny's procedure accurately accounted for both sources by weighing 
together the actual contribution from each in the correct relative 
proportions. For promotions, Dr. Kenny used the work force 
proportions in the particular census job category as the relevant 
labor market. For accessions, Dr. Kenny used the DoD labor 
force statistics for the Professional/ Technical category for 
the Managerial Administrative category. The Albany SMSA was 
used for the other seven census categories. The trial court

(FOOTNOTE 44 CONTINUEDintentional discrimination must be significantly greater than 
three standard deviations. Gay v. Waiters' and Dairy Lunchmen s 
Union, 694 F.2d 531, 551, 52 (9th Cir. 1982). In reaching this 
conclusion, the_Court was making reference to a statistical dis­
parity adverse to the minority which covered a five-year period 
and which had a value of 3.07 in terms of standard deviations.
Gay v. Waiters Union Local 30, 489 F. Supp. 283, 311. n.37 
(N.D. Cal. 1980 ) af f 1 d. ,~~694 F.2d 531 (9th Cir. 1982 ).
45/ Although Dr. Kenny does not include qualifications in his 
analyses, the significant refinements that he does include 
make his analysis substantially more probative than those 
submitted by plaintiffs which fail to address both qualifications 
and job categories.

-41-



found that:
These comparisons are particularly appro­

priate because of the choice of relevant labor 
market . . . .  Dr. Kenny's comparisons reflect 
the fact that the base may have to look beyond 
its own work force to find qualified people for 
certain jobs.

Op. 21 [R.E. 57]. See also Op. 13 [R.E. 49]. The only results 
found by Dr. Kenny to exceed three standard deviations were in 
favor of blacks in the Professional/Technical category. Dr.
Kenny's analyses, even without considering qualifications,

j
demonstrate no pattern of discrimination against blacks in the 
filling of job vacancies at the Marine Corps Logistics Base.

2. Promotion Analyses. At trial, defendant's expert, Dr. 
Charles W. Upton, presented a series of regression and logit 
analyses to analyze promotions while accounting for qualifications. 
Dr. Upton's analyses revealed no significant disparities between 
the actual and expected number of black promotions. See FACTS, 
supra, at pp. 20-22. The trial court was accurate in its assessment 
that:

Both the linear and logit models are more sophisticated than the naive analysis conducted 
by Dr. Drogin and initially conducted by Dr. Upton.

Op. 23 [R.E. 59). The results of Dr. Upton's sophisticated

46/ Plaintiffs' attack on Dr. Upton's choice of variables is 
unpersuasive. See Appellants' Brief 45-46. Dr. Upton used reasonable proxies for experience and education and excluded 
"tainted variables." See Op. 23 n. 15 [R.E. 59]. In these 
circumstances a meaningful challenge to Dr. Upton's results required that plaintiffs show the impact of the improvements 
that they suggest by rerunning regressions with these improvements 
incorporated. See generally, D. Baldus & J. Cole Statistical 
Proof of Discrimination § 8.02 (1981 & 1983 Supp.).

-4 2-



analyses clearly indicate no discrimination against blacks in 
promotions.^2/

D. There Is No Evidence That Identifies Any Specific 
And Discrete, Objective Selection Criterion Which 
Adversely Affected Black Applicants So As To 
Require Validation.__________

Plaintiffs assert that defendant's had an obligation to 
conduct validation studies of the selection standards used for 
promotions. Plaintiff's Brief at 47-48. The duty to validate 
does not arise until there is evidence that a neutral employment
standard has a statistically significant impact on a protected

Igroup. Carpenter, v. Stephen F. Austin State University, 706
F.2d 608, 621 (5th Cir. 1983). The evidence must identify a
specific and discrete, objective selection criterion which
adversely affects the protected class.^8/ Pouncy v. Prudential
Insurance Company of America, 668 F.2d 795, 800-802 (5th Cir. 1982).
TT7 Although the "trial court did not find it necessary to make a 
detailed review of defendant's statistics in light of plaintiff's 
unpersuasive statistics. Op. 23 [R.E. 59] the trial court's factual 
findings are further bolstered by the post-trial report submitted 
by Dr. Upton. For these regression analyses, Dr. Upton grouped 
promotions by pay plan within each division at the Marine Corps 
Logistics Base. These groupings serve as a proxy for job category 
given that there is less disparity in job categories by pay plan 
within division than there is for the base as a whole. These 
analyses of Dr. Upton are the only statistical analyses in the 
record that account for both job categories and qualifications.
These results clearly demonstrate that, when job category is accounted for and when qualifications are considered, there are 
no significant disparities between the actual and the expected 
number of black promotions. See FACTS, supra, at p. 23.
48/ There is no such evidence in the record. There are 181 
different job occupations in the Base’s work force. See FACTS, 
supra, at p. 9. There is no evidence demonstrating an adverse 
impact for any one of these occupations or for any closely related 
group of occupations. Plaintiffs undifferentiated statistical 
data, even if they were reliable, would not be sufficient to 
generate the requirement to conduct formal validation studies 
relative to the 181 different sets of standards.

-43-



The results of plaintiff's statistical analyses of merit promotions, 
even if they were reliable, would not have triggered any reguirement 
to validate because they did not account for required qualifications.^/ 
See, e.g., Moore, 708 F.2d at 481-483.
II. THE INDIVIDUAL CLAIMS OF NAMED PLAINTIFFS ABAD AND 

MADDOX WERE PUT AT ISSUE IN THE COMPLAINT AND WERE 
DECIDED BY FACTUAL FINDINGS OF THE TRIAL COURT?
AS A MATTER OF LAW, THESE NAMED PLAINTIFFS ARE NOT 
ENTITLED TO RELITIGATE THESE CLAIMS________________
Plaintiffs appear to allege that the Individual claims

of the named plaintiffs were not .before the trial court a/id that
only issue before the trial court was that of classwide
discrimination.^/ Appellant's Brief 2 (Issue IV), 53—55. This
allegation is in direct conflict with the complaint which clearly
placed the’individual claims of the named plaintiffs at issue.

J97 Although it is not necessary for this Court to reach the 
question in this case, there are a number of decisions holding 
that disparate impact analysis is inappropriate when dealing 
with decisions made by rating panels. Heagney, 642 F.2d at 
1163? Mortensen v. Callaway, 672 F.2d 822, 824 (loth Cir. 1982); 
Pope v. City of Hickory, 679 F.2d 20, 22 (4th Cir. 1982);^
Pouncy,” 668 F.2d at 800; and Harris v. Ford Motor Co. , 651F72d 60 9 611 (8th Cir. 1981). However, there is some uncertainty.
Eastland, 704 F.2d at 619-20; Moore, 708 F.2d 481-82; Page v.
U.S. Industries, Inc., 726 F.2d 1038, 1045-48 (5th Cir. 1984).

Plaintiffs' assertion of excessive subjectiveness in the 
Base's promotion system clearly conflicts with the record.
See Facts, supra, at pp. 10-17. Specific objective standards 
are used whenever reasonably possible; detail written procedures 
are provided; and substantial checks and balances are included.
The exercise of~judgment and discretion in the environment is 
not inappropriate. See, Page, 726 F.2d at 1053. .See also,
Casillias v. U.S. Navy, _____ F.2d ____, 34 FEP Cases 1493,
1497-98 (9th Cir. 1984).
50/ Plaintiffs have not appealed the factual finding of the 
Trial court relative to claims of Named Plaintiffs Abad and 
Maddox. If they had, the clearly erroneous standard would have
applied. '

-44-



R.E. 11-12, 15-18, 26-28. As was stated in the complaint,
"[plaintiffs bring this action for themselves and on behalf of
other persons similarly situated, pursuant to Rule 23(b)(2) of
the Federal Rules of Civil Procedure." R.E. 11. Further,
plaintiffs assert that the named plaintiffs . . . [did not have]
the opportunity to have their individual claims litigated and
decided." Appellants Brief 55 n. 59. This assertion is clearly
in conflict with the facts in this matter. At the very opening
of the trial, the trial court noted to plaintiffs that their
trial brief suggestion of limiting the proceedings to the issue
of classwide liability was not appropriate. TR. I 4-5. Plaintiffs
response indicated that what they intended was a Phase I proceeding
on the question of liability. TR. I 5. At stage-Phase I, the
individual claims of the named plaintiffs are presented as well
as the classwide claims. The issue of relief for individual
class members is then reserved for Phase II proceedings if classwide
liability is found,51/ As plaintiffs note in their brief, both Named
Plaintiff Abad and Named Plaintiff Shepard testified at the trial.̂ — /

TR. Ill 2-21 & 21-34. For reasons known only to themselves,
plaintiffs chose not to put Named Plaintiff Maddox on the stand
but instead to rely on his deposition which had been admitted
51/ When considering this issue, it is important to distinguish the 
claims of the named plaintiffs, and evidence in support thereof, 
from the testimony of unearned class members and any underlying individual claims that such class members may have. The question 
of the impact of a Phase I determination on the individual claims 
of unnamed class members is a far different matter and is addressed 
in Argument III, infra.
52/ Moreover, as a result of the evidence presented, the trial court 
found for Named Plaintiff Shepard and relief was subsequently 
granted. If the claims of the named plaintiffs were not before 
the trial court, such relief was clearly not appropriate.

-45-



into evidence by the trial court. Appellants' Brief 53.
Plaintiffs have misconstrued the role of the named plaintiffs 

in a class action. In essence, plaintiffs are arguing that the 
individual claim of the named plaintiffs are no different than 
those of unnamed class members.^/ Plaintiffs cite no cases 
supporting their assertions in this matter and enunciate no 
principles or policies supporting their position. There is no 
justification for giving named plaintiffs "two bites of the 
apple". The normal res judicata effects should be given >to 
the judgment of the trial court relative to the individual claims
of the named plaintiffs;54//

If judgment is on the merits, res judicata 
is an absolute bar in another case on the same 
cause of action between the same parties not 
only on the claims presented but also as to 
every ground as to recovery that might have 
been presented.

53/ In a "pattern and"practice" suit brought by the Equal Employment 
Opportunity Commission, pursuant to 42 U.S.C. § 2000e-5(f)(1), 
the class action provisions of Rule 23 of the Federal Rules of 
Civil Procedure are not applicable. General Telephone Co. v.
EEOC, 446 U.S. 318, 323 (1980). However, a class action brought 
by private parties must comply with the requirements of Rule 23:

Title VII, however, contains no special authori­
zation for class suits maintained by private parties.
An individual litigant seeking to maintain a class 
action under Title VII must meet "the prerequisites 
of numerosity, commonality, typicality, and adequacy 
of representation" specified in Rule 23(a). These 
requirements effectively "limit the class claims to 
those fairly encompassed by the named plaintiff's 
claims."

General Telephone Co. v'. Falcon, 457 U.S. 147, 156 (1982) (citations 
omitted)". Acceptance of plaintiffs' position would eliminate 
the distinction between an EEOC "pattern and practice suit and 
a class action brought by private parties.
54/ Plaintiffs appear to be asking this Court to rule on a 
question that has never been litigated below and from which 
(FOOTNOTE 54 CONTINUED ON NEXT PAGE.)

-46-



Bullard v. Webster, 679 F.2d 92, 93 (5th Cir. 1982).
The Fourth Circuit has held that an adverse finding as to 

named plaintiffs during Phase I proceedings acts as a bar with 
respect to the grounds upon which their claims were adjudicated. 
Sledge v. J. P. Stevens & Co., 585 F.2d 625, 637 (4th Cir. 1978), 
cert, denied 440 U.S. 981. Plaintiffs have presented no reasons 
why the named plaintiffs should not be barred from relitigating 
individual claims that were adjudicated by the trial court in 
this case.££/

IIII. THE UNNAMED CLASS MEMBERS WHO TESTIFIED AT TRIALARE NOT ENTITLED TO RELIEF BASED ON THEIR TESTIMONY, 
PARTICULARLY IN LIGHT OF THE FINDING OF NO CLASSWIDE 
DISCRIMINATION._____________________________________
Plaintiffs assert that the trial court erred in not awarding

relief to those unnamed class members who testified and who were
found by the trial court to be "persuasive on the issue of

(FOOTNOTE 54 CONTINUED)there is no final judgment on which an appeal can be based 
the res judicata effect of the judgment rendered against named 
plaintiffs by the trial court. Normally, a question not litigated 
below is not considered on appeal. In re Novack, 639 F.2d 1274, 
1276-77 (5th Cir. 1982) (Unit B). This issue would be presented 
in a more appropriate appellate context if the res judicata 
effect of the judgment against named plaintiffs in this case had 
been considered in a court below. In the event that the named 
plaintiffs chose to file any further actions relative to their 
claims in this case, such a determination would be made and 
would be subject to appeal.
55/ The Court in Sledge was concerned that named plaintiffs^ 
be given the benefit of any finding of classwide discrimination. 
585 F.2d at 637-38. Of course, that is not a concern here 
because class relief was denied.

-47-



discrimination in the promotion process.56/ Appellant's Brief 2,
55. As an initial matter it must be noted that plaintiffs' 
arguments on this issue are contradictory for they further assert 
"that class member testimony was presented for illustrative 
purposes, and that . . . [these] class members . . . [did not
have] the opportunity to have their individual claims litigated 
and decided". Appellants' Brief 55 n.59.

The legal issue raised by plaintiffs is a matter that was 
addressed by the Third Circuit based on a certified question from 
the district court in Dickerson v. United States Steel Corp., 582 
p,2d 827 (3rd Cir. 1978). The court held that it would not be 
appropriate to grant relief in these circumstances to "an individual 
class member, who has not presented any evidence of having 
satisfied the jurisdictional prerequisites in his own name and 
who is not a named plaintiff in the class action". _Id. at 834;
Accord Clark v. Chrysler Corp., 673 F.2d 921, 929 (7th Cir. 1982) 
cert, denied, 103 S. Ct. 161. Plaintiffs' have provided no

■56/ plaintiffs fail to cite the trial court's footnote on this 
point:

This is not to say that each of those individuals 
could make out an individual case of discrimination 
but merely that the court finds merit in their testimony in determining the class question of discrimination.

Op. 17 n. 11 (R.E. 53). Moreover, this matter is not properly 
before this Court because a review of the complaint (R.E. 10 2b), 
plaintiffs' trial brief. (R. 867-97), plaintiffs’ opening statement 
(TR. I 12-17), and plaintiffs' post trial submission (R. 1021 l'6 & R. 1312-67) demonstrates that plaintiffs never requested the 
trial court to provide relief to unnamed class members based on their testimony during the proceedings which plaintiffs characterized 
as "Phase I" proceedings TR. I 5.

-48-



rationale for not following this precedent.57/
IV. AS A MATTER OF LAW, A CAUSE OF ACTION CANNOT BE BASED ON THE 

TITLE VII REQUIREMENT TO DEVELOP AN AFFIRMATIVE ACTION PLAN
Plaintiffs assert "that the defendant's failure to develop

and implement an affirmative action plan violates [Title VII]".
Appellants' Brief 2. Appellants have failed to address the
recent decision of this Court that is depositive on this issue:
" We hold, however, that absent a showing of discrimination,
there is no Title VII cause of action for the failure to implement
or utilize an affirmative action program." Ferguson v. Veterans

I

Administration, 723 F.2d 871, 1872 (11th Cir. 1984); Accord 
Page v. Bolger, 645 F.2d 227, 233-34 (4th Cir.) (en banc), cert, 
denied, 454 U.S. 892 (1981). In view of these judicial pronouncements, 
it is difficult to perceive any justification for appellants' 
ascribing neglect to the Trial Court on this issue.

57/ Whether individual class members may now proceed individually 
or whether they are barred by the finding of no classwide liability 
is an issue now before the Supreme Court in Cooper v. Federal 
Reserve Bank of Richmond, No. 83—185. The Solicitor General in an 
amicus brief in Cooper,""argued that individual class members should not be barred. If the Supreme Court adopts the Government's 
position, the unnamed class members can proceed individually, if 
they so desired.
58/ Nor do the facts support appellants on this issue. Appellant's 
assertion "that the defendant's fail[ed] to develop, implement 
or maintain an affirmative action" (Appellants^ Brief ii) is in 
direct conflict with their admission that "plaintiffs introduced 
into evidence each of the Base's affirmative action plans for 
the relevant period of time" (Appellants' Brief 18).

-49-



CONCLUSION
The District Court engaged in a comprehensive analysis of all 

aspects of plaintiffs' claims. Plaintiffs have shown no clear 
error in the District Court's finding. There is none.

Plaintiffs failed to prove the allegations of discrimination 
both with respect to the individual claims of Named Plaintiffs 
Abad and Maddox and with respect to the class claim. The legal 
issues raised by plaintiffs are not supported by either precedent

i

or principle. Therefore, defendant respectfully requests that 
this Court affirm the judgment of the District Court.

Respectfully submitted,
JOE D. WHITLEYid States At

L. LYNCHlistant U.S. AttdrneyDistrict of Georgia 
Macon, Georgia

C. JOHN TURNQUISTAssociate. General CounselA(Litigation)

yANlEL E% O'CONNELL’, JR. 
Senior Trial Attorney 
Litigation Office 
Office of the General Counsel 
Department of the Navy 
Washington, D.C.
DAVID A. HIGLEY
Staff Judge Advocate
JOHN E. BENGIERDeputy Staff Judge Advocate
Marine Corps Logistics Base
Albany, Georgia
Attorneys for Defendant-Appellee

-50-



CERTIFICATE OF SERVICE

were
I hereby certify that copies of Defendant-Appellee's Brief 
served on counsel for appellant on this _L2i_Kday of June

1984 by United States mail, postage prepaid, as follows:
Gail J. Wright, Esquire 
99 Hudson Street 
16th Floor New York, NY 10013
C. B. King, Esquire 502 South Monroe Street 
Drawer 3468Albany, Georgia 31706
Herbert E. Phipps Suite 226, Albany Towers 
235 Roosevelt Avenue 
Albany, Georgia 31706

DANIEL E. O'CONNELL, JR. 
Attorney for Appellee