Louisville Black Police Officers Organization Inc. v. City of Louisville Brief
Public Court Documents
May 13, 1978
Cite this item
-
Brief Collection, LDF Court Filings. Louisville Black Police Officers Organization Inc. v. City of Louisville Brief, 1978. 6a7cc5f2-bb9a-ee11-be36-6045bdeb8873. LDF Archives, Thurgood Marshall Institute. https://ldfrecollection.org/archives/archives-search/archives-item/5562b87f-d778-4b86-ad05-163fe1181ddb/louisville-black-police-officers-organization-inc-v-city-of-louisville-brief. Accessed April 28, 2025.
Copied!
INDEX
Table of Authorities ........................................ iv
Table of Abbreviations ...................................... xii
STATEMENT OF THE CASE ........................................ 1
SUMMARY OF THE ARGUMENT ...................................... 4
ARGUMENT ....................................................... 6
I. DEFENDANTS HAVE ENGAGED IN A CONSPICUOUS
AND LONGSTANDING PATTERN OF DISCRIMINATION
AGAINST BLACKS IN RECRUITMENT, TESTING,
SELECTION, AND HIRING FOR JOBS AS POLICE
OFFICERS............................................. 6
A. Standards of Proof ........................... 6
B. Prior to the Commencement of This Action
in 1974, Defendants Engaged in a Clear
Pattern of Discrimination Against Blacks, and
Defendants Also Engaged in Practices
Which Were Not Job Related and Which Had
a Discriminatory Effect on Blacks ........... 11
1. Statistical Evidence..................... 11
2. Recruitment and Selection Practices ..... 18
3. History of Racial Discrimination ......... 33
C. After the Commencement of This Action in
1974, Defendants Corrected Some of Their
Discriminatory Practices but Persisted
in Others ..................................... 35
1. Statistical Evidence .................... 35
2. Recruitment and Selection Practices ...... 39
II. DEFENDANTS VIOLATED THE STATUTORY RIGHTS OF
PLAINTIFFS AND THEIR CLASS BY USING TEST 165.1
IN A MANNER WHICH HAD A SUBSTANTIAL ADVERSE
. IMPACT ON BLACK APPLICANTS AND DID NOT VALIDLY
MEASURE THEIR QUALIFICATIONS FOR THE JOB ......... 46
PAGE
PAGE
A. The Applicable Law .......................... 46
B. As Used by Defendants, Test 165.1
Had a Substantial Adverse Impact on Black
Applicants .................................... 50
C. Defendants Have Not Demonstrated That
Test 165.1 Is Manifestly Related to
Performance of the Job of a Louisville
Police Officer ................................ 53
1. Test 165.1 has not been shown to be
content valid for use in selecting
police officers .......................... 55
a. Test 165.1 purports to measure
intellectual constructs rather
than observable work behaviors ..... 56
b. Test 165.1 is not a sample or
approximation of job behavior but
merely a verbal representation
of some parts of a highly physical
and personal job ..................... 62
c. . Test 165.1 involves knowledges,
skills, and abilities which new
police officers are expected to
learn in recruit school or on
the job ............................... 66
d. Test 165.1 represents neither a
critical work behavior nor work
behaviors which constitute most
of the important parts of the job ... 69
2. Test 165.1 has not been shown to have
criterion-related validity for use in
selecting police officers ................ 75
a. Test 165.1 involves knowledges,
skills, and abilities which
incumbent police officers have
' learned in recruit school or on
the job ............................... 76
IX
PAGE
b. The sample subjects were not
representative of actual applicants
for the job .......................... 79
c. Test 165.1 has not been shown to
measure fairly any differences in
the job performance of blacks and
whites ................................ 82
d. The concurrent validity study produced
an odd patchwork of results which did
not demonstrate that Test 165.1 is
valid for use in selecting police
officers in every city ............. 88
3. The validity studies offered by defendants
do not support the use of Test 165.1
in selecting Louisville police
officers .................................. 91
4. Defendants substantially increased the
adverse impact of Test 165.1 by setting
an arbitrarily high passing point and
by improperly using the test to rank
applicants ................................ 95
D. Alternative Selection Procedures with Less
Adverse Impact Would Properly Serve the Defendants'
Legitimate Interest in the Selection of Capable
Police Officers ................................ 100
III. THE COURT HAS THE POWER AND THE DUTY TO FASHION
RACE-CONSCIOUS NUMERICAL HIRING RELIEF WHICH
WILL MAKE THE LOUISVILLE POLICE FORCE MORE
REPRESENTATIVE OF THE COMMUNITY IT SERVES ....... 104
IV. PLAINTIFFS ARE ENTITLED TO AN INTERIM AWARD OF
REASONABLE ATTORNEYS' FEES ....................... 118
CONCLUSION ................... 121
- iii -
TABLE OF AUTHORITIES
Cases: Page
Afro American Patrolmens League v. Duck, 503 F . 2d
294 (6th Cir. 1974) ................................ 12,15,39
Albemarle Paper Company v. Moody, 422 U.S. 405 (1975) . 9,21,46,48,
49,76,79,80,82,
84,85,91,100,
103,104
Alexander v. Gardner-Denver Co., 415 U.S. 36 (1974) ... 10
Associated General Contractors of Mass., Inc. v.
Altshuler, 361 F.Supp. 1293 (D. Mass. 1973),
aff'd. 490 F .2d 9 (1st Cir. 1973), cert, denied,
416 U.S. 957 ( 1974 )................................. 105
Arnold v. Ballard, 390 F.Supp. 723 (N.D. Ohio 1975),
aff'd 12 FEP cases 1613 (6th Cir. 1976), vac.
and rem. on other grounds, 16 FEP cases 396
(6th Cir. 1976) .................................... 40,105,108,114,
116,117
Baker v. Columbus Municipal Separate School District,
462 F . 2d 1112 (5th Cir. 1972) ..................... 8
Boston Chapter, NAACP, Inc. v. Beecher, 504 F . 2d
1017 (1st Cir. 1974), cert, denied, 421 U.S.
910 (1975) .......................................... 24,47,54,64,69,
81,82,95,99,
105,116
Bradley v. Richmond School Board, 416 U.S.
696 (1974 )..................................... ..... 118,119,120
Bridgeport Guardians, Inc. v. Bridgeport Civil
Service Commission, 482 F.2d 1333 (2nd Cir.1973),
cert, denied, 421 U.S. 991 ( 1975) ................. 24,46,49,52,
99,105,106,108,
114,116
Brown v. Board of Education 347 U.S. 483 (1954) ........ 35
Castaneda v. Partida, 430 U.S. 482 ( 1977) ..... ■........ 16,17,25,45
Carter v. Gallagher, 452 F.2d 315 (8th Cir. )• (en banc) ,
cert, denied, 406 U.S. 950 ( 1972) ................. 105,106,116
Castro v. Beecher, 459 F .2d 725 (1st Cir. 1972)..;.... 47
- iv -
PAGE
Chandler v. Roudebush, 13 CCH E.P.D. <[11,574
(C.D. Cal. 1977 ).................................... 119
Christiansburg Garment Co. v. EEOC, 434
U.S. 412 ( 1978) .................................... 118
Commonwealth of Pennsylvania v. O'Neill, 14
CCH E.P.D. 117699 (E.D. Pa. 1977) ................. 119
Contractors Association of Eastern Pennsylvania v.
Secretary of Labor, 442 F .2d 159 (3rd Cir.
1971), cert, denied, 404 U.S. 854 ( 1971) .......... 106
Crockett v. Green, 534 F .2d 715 (7th Cir. 1976) ........ 15,106,116
Davis v. County of Los Angeles, 566 F .2d 1334
(9th Cir. 1977), cert, granted, 46 U.S.L.W.
3780 (June 19, 1973) ................................ 10,15,21,46,
105,106,108,
116
Dothard v. Rawlinson, 433 U.S. 321 ( 1977 ) .............. 9,46
Douglas v. Hampton, 512 F .2d 976 (D.C. Cir. 1975) ..... 48
Dozier v. Chupka, 395 F.Supp. 836 (S.D. Ohio 1975) .... 21
EEOC v. Detroit Edison Co., 515 F .2d 301 (6th Cir.
1975 ) vac, and rent, on other grounds, 431
U.S. 951 ( 1977) ..................................... 15,47,99,,105,
107,116
Ensley Branch, NAACP v. Seibels, 14 FEP cases 670
(N.D. Ala. 1977) ... ................................ 49
Erie Human Relations Commission v. Tullio, 493 F .2d
371 (3rd Cir. 1974 ) ................................. 105,106,108 ,
116
Feeney v. Massachusetts, ___ F.Supp.___, 17 FEP
Cases 659 (D. Mass. 1978) .............. ............ 8
Firefighters Institute for Racial Equality v. City
of St. Louis, 549 F.2d 506 (8th Cir. 1977),
cert, denied, 434 U.S. 819 ( 1977 ) ................. 47,73,74,100
Fitzpatrick v. Bitzer, 427 U.S. 445 (1976 ) ............. 118
Foley v. Connelie, 55 L.Ed.2d 287 (1978) .............. 109,112
v
PAGE
Furnco Construction Corp. v. Waters, 46 U.S.L.W..
4966 (June 29, 1978)................................ 38
Gregory v. Litton Systems, Inc., 472 F.2d 631
(9th Cir . 1972)..................................... 21
Griggs v. Duke Power Co., 401 U.S. 424 (1971) .......... 8,9,21,
46,48,49,
103
Harkless v. Sweeny Independent School District,
554 F.2d 1353 (5th Cir. 1977), cert. denied,
434 U.S. 966 (1977)................................. 8,35
Hazelwood School District v. United States, 433
U.S. 299 (1977) ..................................... 9,12,13,14,16,
17,39
Hutto v. Finney, 46 U.S.L.W. 4817 (June 23, 1978)..... 118
International Brotherhood of Teamsters v. United 9,12,13,14,15,
States, 431 U.S. 324 (1977) ........................ 17,35,38,39,
115,120
James v. Stockham Valves & Fittings Co., 559
F.2d 310 (5th Cir. 1977), cert, denied
434 U.S. 1034 ( 1978) ............................... 118,119,120
Jenkins v. United Gas Coro., 400 F .2d 28
(5th Cir. 1968 ) .......................... ......... 37,40 '
Johnson v. Pike Corp., 332 F.Supp. 490 (C.D.
Cal. 1971)............................ '.............. 21
Johnson v. Ryder Truck Lines, Inc., 15 CCH
E.P.D. H7969 (W.D.N.C. 1977).'.... .............. 119
Johnson v. Ryder Truck Lines, Inc., 575 F.2d 471
(4th Cir. 1978) ......................... ;.......... 10,21 '
Jones v. Tri-County Electric Cooperative, Inc.,
512 F . 2d 1 (5th Cir. 1975 ) ........................ 40
Kirkland v. New York State Dept, of Correctional
Services, 374 F.Supp. 1361 S.D.N.Y. 1974,
aff'd in pertinent part, 520 F.2d 420 (2nd
Cir. 1975), cert, denied, 429 U.S. 974 (1976) .... 47,73,
97,99
vi
PAGE
League of United Latin American Citizens v.> City
of Santa Ana, 410 F.Supp. 873 (C.D. Cal.1976) ..... 12,14,25,40,
79,81,107,
116
Lewis v. Philip Morris, Inc., 13 CCH E.P.D. 1(11,350
(E.D. Va. 1976)..................................... 119
Local 53, Asbestos Workers v. Vogler, 407 F .2d
1047 (5th Cir. 1969) ............................... 106
Long v. Ford Motor Co., 496 F .2d 500 (6th Cir.1974) .... 10
Louisiana v. United States, 380 U.S. 145 (1965) ...... 104
McBride v. Delta Air Lines, Inc., 551 F .2d 113
(6th Cir.), vac. and rem. on other grounds,
434 U.S. 916 (1977) ................................ 7
McLaughlin v. Florida, 379 U.S. 184 ( 1964 ) .......... 35
Mills v. Electric Auto-Lite Co., 396 U.S.
375 (1970) .......................................... 119
Morrow v. Crisler, 479 F .2d 960 (5th Cir.1973),
mod. on reh. en banc on other grounds,
491 F . 2d 1053 ( 5th Cir. 1974 )...................... 12
Morrow v. Crisler, 491 F.2d 1053 (5th Cir. 1974)
(en banc), cert, denied, 417 U.S. 895 (1974) .... 106
NAACP v. Allen, 493 F .2d 614 (5th Cir. 1974) .......... 105,106,108,
114,116
NAACP v. Lansing Board of Education, 559 F .2d 1042
(6th Cir. 1977) .................................... . ' 7,10,19^
Officers for Justice, NAACP v. Civil Service
Commission of San Francisco, 371 F.Supp.
1328 (N.D. Cal. 1973 ) ....................,....... . . 15,108,114
Parham v. Southwestern Bell Telephone Co., 433
F.2d 421 (8th Cir. 1970)........................... 38
Patterson v. American Tobacco Co., 9 CCH
E.P.D. 1(10,039 (E.D. Va. 1975) ............ ....... 119
Vll
PAGE
Pennsylvania v. Flaherty, 404 F.Supp. 1022
(W.D. Pa. 1975) ..................................... 15
Regents of the University of California v. Bakke,
46 U.S.L.W. 4896 (June 28, 1978) .................. 34,85,86,88,
106
Reitman v. Mulkey, 387 U.S. 369 (1967) ................. 35
Rice v. Gates Rubber Co., 521 F .2d 782
(6th Cir. 1975)..................................... 38
Rich v. Martin Marietta Corp., 522 F .2d 333
(10th Cir. 1975) ................................... 38
Rios v. Enterprise Association Steamfitters
Local 638, 501 F.2d 622 (2nd Cir. 1974) ........... 106
Rogers v. International Paper Co., 510
F.2d 1340, vacated & remanded on other qrounds,
423 U.S. 809 (1975) .... ......................... 85,S7
Rowe v. General Motors, 457 F .2d 348 (5th Cir. 1972) ... 26,45
Sangmeister v. Woodard, 565 F .2d 460 (7th Cir. 1977) ... 8
Sherrill V. J.P. Stevens & Co., 551 F .2d 308, 13
E.P.D. 1111,422 (4th Cir. 1977) ..................... ... 106
Shield Club,v. City of Cleveland, 13 FEP cases 1373
(N.D. Ohio 1976') ... ................................. 8
Shield Club v. City of Cleveland, 13 FEP cases 1394
(N.D. Ohib 1976 ) . .'.................................. 3
Sims v. Sheet Metal Workers Local 65,. 489 F . 2d
1023 (6th Cir. 1973 ) ........... ..................... 105
Southern Illinois Builders Association v. Ogilvie,
471 F . 2d 680 (7th Cir. 1972) ................. ..... 106
Stamps v. Detroit Edison Co., 365 F.Supp. 87 (E.D.
Mich. 1973) ............... '......... ................. 15,99,107,
116
Stewart v. General Motors Corp., 542 F.2d 445
(7th Cir. 1976), cert, denied, 433 U.S. 919
(1977) ............................................... 25,45
- viii -
PAGE
Strauder v. West Virginia, 100 U.S. 303 (1879)
United Jewish Organizations of Williamsburgh v.
Carey, 430 U.S. 144 (1977) ...............
United States v. City of Chicago, 549 F .2d 415
(7th Cir. 1977), cert, denied, 434 U.S.
875 (1977) .................................
United States v. City of Chicago, 573 F.2d 416
(7th Cir. 1978) .....................................
U.S. v. Georgia Power Co, 474 F.2d 906
(5th Cir. 1973) ....................................
United States v. Ironworkers Local 86, 443 F .2d
544 (9th Cir.), cert, denied, 404 U.S. 984 (1971)..
United States v. Local 38, IBEW, 428 F .2d 144
(6th Cir.), cert. denied, 400 U.S. 943 ( 1970).....
United States v. Local 212, IBEW, 472 F .2d 634
(6th Cir. 1973 ) .............. ......................
United States v. Masonry Contractors Association,
497 F . 2d 871 (6th Cir. 1974) ................ .......
United States v. N.L. Industries, Inc., 479 F .2d
354 (8th Cir. 1973) .................................
United States v. .Texas Education Agency, 564 F . 2d
162 (5th Cir. 1977) .................................
United States v.' Wood Lathers Local 46, 471 F . 2d 408
(2nd Cir.)., cer t. denied, 412 U.S. 939 (1973) ....
Village of Arlington Heights v. Metropolitan
Housing Development Corp., 429 U.S. 252 (1977) ....
Vulcan Society v.. Civil Service Commission, 490 F .2d
387 (2nd Cir. 1973) .............. ..................
Vulcan' Society v. N.Y. Civil Service Commission,
360 F.Supp. 1265 (S.D.N.Y. 1973), aff'd in
relevant part, 490 F.2d 387 (2nd, Cir. 1973) ......
35
35
21,47,48,52,
54,73,83,85,
105,106,116,
117
9,54,73
48,85,95
106
105
105
105
106
7
106
6,7,8,35
54
52,73,75
IX
PAGE
Wallace v. Debron Corp., 494 F .2d 674 (8th Cir.
1974) .............................................. 21
Washington v. Davis, 426 U.S. 229 (1976) ............... 6,7,8,10,12,
25
Western Addition Community Organization v. Alioto,
360 F.Supp. 733 (1973), appeal dismissed,
514 F.2d 542 (9th Cir. 1975), cert, denied,
423 U.S. 1014 ( 1975) ................................ 99
Yick Wo. v. Hopkins, 118 U.S. 356 ( 1886 ) ............... 35
Constitutional Provisions, Statutes and Regulations
United States Constitution, Fourteenth Amendment ...... passim
Title VII of the Civil Rights Act of 1964, 42 U.S.C.
§§2000e et seq....................................... passim
Civil Rights Attorneys' Fees Awards Act of 1976,
Pub. L. No. 94-559, 90 Stat. 2641,
codofied in 42 U.S.C. §1988 ........................ 118,119
42 U.S.C. §1981 ........................................... passim
42 U.S.C. §1983 ........................................... passim
Uniform Guidelines on Employee Selection
Procedures (1978), 43 Fed. Reg. 38290
(Aug. 25. 1978) .......................... 25,47-103
Equal Employment Opportunity Commission Guidelines
on Employee Selection Procedures,
29 C.F.R. §1607 .................................... 48-103
Federal Executive Agency Guidelines on Employee
Selection Procedures, 41 Fed. Reg. 51734
(1976) ............................................. 48-103
Federal Rules of Civil Procedure, Rule 23 ............ 1
x
PAGE
Legislative History
H.R. Rep. No. 92-238, 92nd Cong.
1st Sess. (1971) .................................... 108,109
H.R. Rep. No. 94-1558, 94th Cong. 2nd Sess. (1976) .... 120
S. Rep. No. 92-415, 92nd Cong. 1st Sess. (1971) ........ 108,109
S. Rep. No. 94-1011, 94th Cong. 2nd Sess. (1976) ...... 118,119
118 Cong. Rec. 790 (1972) ................................ 108
Other Authorities
American Psychological Association, Standards for
Educational and Psychological Tests (1974) ......... 48-103
APA Division of Industrial-Organizational Psychology,
Principles for the Validation and Use of
Personnel Selection Procedures (1975) ............. 49-103
Barrett, Content or Construct Validity: What's the
Difference? (1976) .................................. 53,62
Cooper and Sobol, Seniority and Testing Under Fair
Employment Laws: A General Approach
to Objective Criteria of Hiring and Promotion,
82 Harv. L. Rev. 1598 ( 1969) ....................... 25
Edwards, Order and Civil Liberties: A Complex Role
for the Police, 64 Mich. L. Rev. 47 (165) ......... 110
Edwards, The Police on the ’Urban Frontier
(New York 1968) ..................................... 113
Gastwirth and Haber, Defining the Labor Market for
Egual Employment Standards, 99 Monthly
Labor Review 32 (March 1976 )................ ....... 107
Kerner Commission, Report of the National Advisory
Commission on Civil Disorders, (Bantam edition:
1968) ................................................. 110,111,
113
An Overview of the 1978 Uniform Guidelines on
Employee Selection Procedures, 4 3
Fed. Reg. 38290 ..................................... 57
President's Commission on Law Enforcement and
Administration of Justice, Task Force Report:
The Police (GPO 1967)............................. 110,111,113
xi
PAGE
Legislative History
H.R. Rep. No. 92-238, 92nd Cong.
1st Sess. (1971) ................................... 108,109
H.R. Rep. No. 94-1558, 94th Cong. 2nd Sess. (1976) .... 120
S. Rep. No. 92-415, 92nd Cong. 1st Sess. (1971) ....... 108,109
S. Rep. No. 94-1011, 94th Cong. 2nd Sess. (1976) ...... 118,119
118 Cong. Rec. 790 (1972) .............................. 108
Other Authorities
American Psychological Association, Standards for
Educational and Psychological Tests (1974) ........ 48-103
APA Division of Industrial-Organizational Psychology,
Principles for the Validation and Use of
Personnel Selection Procedures (1975) ............. 49-103
Barrett, Content or Construct Validity: What's the
Difference? (1976) ................................. o3,62
Cooper and Sobol, Seniority and Testing Under Fair
Employment Laws: A General Approach
to Objective Criteria of Hiring and Promotion,
82 Harv. L. Rev. 1598 (1969) .... ................. 25
Edwards, Order and Civil Liberties: A Complex Role
for the Police, 64 Mich. L. Rev. 47 (165) ........ 110
Edwards, The Police on the 'Urban Frontier
(New York 1968) .................................... 113
Gastwirth and Haber, Defining the Labor Market for
Equal Employment Standards, 99 Monthly
Labor Review 32 (March 1976 )....................... 107
Kerner Commission, Report of the National Advisory
Commission on Civil Disorders, (Bantam edition:
1968) ............................................... 110,111,113
An Overview of the 1973 Uniform Guidelines on
Employee Selection Procedures, 43
Fed. Reg. 38290 .................................... 57
President's Commission on Law Enforcement and
Administration of Justice, Task Force Report:
The Police (GPO 1967)............................ 110,111,113
xi
TABLE OF ABBREVIATIONS
"Thornberry, Vol. Ill,
6/22/77 at- Witness, trial transcript volume
number, date and page of
testimony,
"Richmond Dep.,
5/11/77 at
"DX ____ at
"PX at
"Uniform Guidelines,§
"EEOC Guidelines,
29 C.F.R. §
Witness, deposition, date, and
page of testimony.
Defendants' exhibit number ____
and page number.
Plaintiffs' exhibit number ____
and page number.
Uniform Guidelines on Employee
Selection Procedures (1978),
43 Fed. Reg. 38290 (Aug. 25,
1978) ,
EEOC Guidelines on Employee
Selection Procedures, 29 C.F.R.
§ 1607.
"FEA Guidelines,§
"APA Standards, if at
I f Division 14
Principles,at If
"Test 165.1"
Federal Executive Agency Guidelines
on Employee Selection Procedures,
41 Fed. Reg. 51734 (1976).
American Psychological Association,
Standards for Educational and
Psychological Tests (1974).
APA Division of Industrial-Organi
zational Psychology, Principles
for the Validation and Use of
Personnel Selection Procedures
(1975).
Multijurisdictional Police Officer
Examination, Test No. 165.1.
X ll
of that organization, and Gary Hearn, both of whom are black
officers employed in the Louisville Division of Police; and
Ronald Jackson, James Steptoe, and Len Holt, three black
applicants for such positions. Plaintiffs allege that the
City of Louisville, the Louisville Civil Service Board, the
Director of Civil Service, the Chief of Police, and other
defendants have engaged and are engaging in discrimination
against black persons on the basis of race or color in recruit
ment, testing, selection, hiring, assignment, promotion, dis
cipline, and other employment practices. The Fraternal Order
of Police, Louisville Lodge No. 6, and its president have
intervened as defendants.
On June 27, 1975, this Court entered an order determining
that the action was maintainable as a class action under Rule
23, Fed. R. Civ. P. This order, as amended on April 22, 1977,
provides as follows:
IT IS ORDERED AND ADJUDGED that the
plaintiff Louisville Black Police Officers
Organization, Incorporated, and the plain
tiff Shelby Lanier, Jr., be and they are
hereby designated as representatives of a
class which is composed of all persons
who are black and who are now or have been
police officers employed by the City of
Louisville and who allege that the rules,
regulations and practices of the defendants
have discriminated against black police
officers on the basis of their race with
regards to assignment, promotion and dis
cipline of personnel.
IT IS FURTHER ORDERED AND ADJUDGED
that the plaintiffs Ronald Jackson, James
Steptoe, Len Holt, and Gary Hearn be and
they are hereby designated as representa
tives of a class which is composed of
2
black persons who have sought to obtain
employment with the Louisville Police
Department and who allegedly have been
denied such employment on the basis of
arbitrary, capricious and racially dis
criminatory practices on the part of
the defendants. Said class also consists
of all black applicants for positions
with the Louisville Police Department
who will in the future seek jobs with the
Police Department, and who may be denied
employment because of the allegedly
racially discriminatory and arbitrary
practices complained of in the complaint.
Pending before this Court, following approximately five
weeks of trial on the merits between March and September of
1977, are the issues concerning discrimination in recruitment,
entry-level testing, selection, and hiring. See Order entered
March 2, 1977. Also pending before the Court is the motion far
a preliminary injunction filed by plaintiffs in April 1977 with
respect to the defendants1 use of the Mulitjurisdictional
Police Officer Examination, Test 165.1. See Order entered
June 17, 1977. Plaintiffs submit this post-trial brief in
support of their motion for a preliminary injunction and in
support of their proposed findings of fact, conclusions of law,
and order and judgment.
SUMMARY OF THE ARGUMENT
Defendants have engaged in a conspicuous and longstanding
pattern of discrimination against blacks in recruitment, tasting,
selection, and hiring for jobs as police officers, in violation
of Title VII of the Civil Rights Act of 1964, as amended, 42
U.S.C. §1981, and 42 U.S.C. §1983 and the Fourteenth Amendment.
Defendants have a long history of racial segregation and dis
crimination against blacks in their police employment practices.
Prior to the commencement of this action in 1974, defendants
intentionally discriminated against blacks and defendants engaged
in practices which were not job related and which had a dis
criminatory effect on blacks. After this lawsuit was filed,
defendants corrected some of their discriminatory practices but
persisted in others. (Section I).
Defendants violated Title VII and §1981 by using the Multi-
jurisdictional Police Officer Examination, Test No. 165.1, in a
manner which had an extreme adverse impact on black applicants
and did not validly measure their qualifications for the job.
Defendants did not demonstrate that Test 165.1 has either con
tent validity or criterion-related validity for use in selecting
Louisville police officers, and defendants accentuated the adverse
impact of the test by using it in an improper manner. Alternative
selection procedures with less adverse impact on blacks were and
are available to defendants. (Section II).
4
As a result of defendants 1 extensive and longstanding pattern
of discrimination against blacks, the Louisville police force is
not representative of the substantial black population in the
community it serves. This Court has the power and the duty to
require the defendants to hire qualified blacks as police officers
on an accelerated basis until the effects of the past discrimina
tion have been eliminated. (Section III).
Although this lawsuit has been before the Court for more
than four years, many questions remain to be decided. Once the
Court determines that defendants have engaged in unlawsul dis
crimination, plaintiffs should be granted an interim award of
attorneys' fees to prevent financial hardship during the con
tinuation of this lengthy and costly litigation. (Section IV).
5
ARGUMENT
I. DEFENDANTS HAVE ENGAGED IN A CONSPICUOUS
AND LONGSTANDING PATTERN OF DISCRIMINATION
AGAINST BLACKS IN RECRUITMENT, TESTING,
SELECTION, AND HIRING FOR JOBS AS POLICE
OFFICERS.
A. Standards of Proof
The Supreme Court has held that proof of a racially dis
criminatory intent or purpose is necessary to show a violation
of the Equal Protection Clause- Washington v. Davis, 426 U.S.
229 (1976); Village of Arlington Heights v. Metropolitan
Housing Development Corp., 429 U.S. 252, 265 (1977). "This is
not to say that the necessary discriminatory racial purpose
must be express or appear on the face of the statute, or that
a law's disproportionate impact is irrelevant . . . . Neces
sarily, an invidious discriminatory purpose may often be
inferred from the totality of the relevant facts, including the
fact, if it is true, that the law bears more heavily on one
race than another." Washington v. Davis, supra at 241-42. it
need not be shown that racial discrimination was a dominant or
primary purpose for the challenged action; rather, "proof that
a discriminatory purpose has been a motivating factor in the
decision" is sufficient. Arlington Heights, supra at 265-66.
Determining whether invidious discriminatory purpose was a
motivating factor requires "a sensitive inquiry into such cir
cumstantial and direct evidence of intent as may be available,"
including such factors as the impact of the challenged practice
6
and its historical background. Id. 266-68. As Justice Stevens
has noted,
Frequently the most probative evidence of
intent will be objective evidence of what
actually happened rather than evidence describ
ing the subjective state of mind of the actor.
For normally the actor is presumed to have
intended the natural consequences of his deeds.
This is particularly true in the case of
governmental action which is frequently the
product of compromise, of collective decision
making, and of mixed motivation. Washington v.
Davis, supra at 253 (Stevens, J., concurring).
The Sixth Circuit has held that the showing of discrimina
tory intent or purpose required by Washington v. Davis and
Arlington Heights may be inferred "from a pattern of official
action or inaction which has the natural, probable and foresee
able result of increasing or perpetuating school segregation."
NAACP v. Lansina Board of Education, 559 F.2d 1042, 1047-48
(6th Cir. 1977). The circuits uniformly have adopted this
objective standard for ascertaining segregative intent. See
United States v. Texas Education Agency, 564 F.2d 162, 168 (5th
Cir. 1977), and cases cited therein. The Sixth Circuit has
recognized that the objective standard applies to employment
discrimination cases as well: "a pervasive pattern of dis
criminatory effects may support an inference of intentional
discrimination underlying the individual charge of discrimina
tory firing." McBride v. Delta Air Lines, Inc., 551 F.2d 113,
115 (6th Cir.), vacated and remanded on other grounds, 434 U.S.
916 (1977). Proper findings of unconstitutional discrimina
tory purpose have been made and upheld where statistical
7
evidence of a disproportionate impact has been coupled with
other objective evidence of discrimination in employment.
See Harkless v. Sweeny Independent School District, 554 F.2d
1353, 1356-58 (5th Cir.), cert. denied, 434 U.S. 966 (1977);
Baker v. Columbus Municipal Separate School District. 462 F.2d
1112, 1114 (5th Cir. 1972); Feeney v. Massachusetts, ____ F. Supp.
____, 17 F.E.P. Cases 659 (D. Mass. 1978) (three-judge court);
Shield Club v. City of Cleveland, 13 F.E.P. Cases 1373 and 1394
(N.D. Ohio 1976). Where the disproportion itself is sufficiently
dramatic, that fact alone "may for all practical purposes demon
strate unconstitutionality. . . . " Washington v. Davis, supra
at 242. See also, Sanomeister v. Woodard, 565 F.2d 460, 467
JJ(7th Cir. 1977).
This inquiry into intent and purpose may be relevant but
is not required to show a violation of Title VII of the Civil
Rights Act of 1964, as amended, 42 U.S.C. § 2000e et sea.
"Congress directed the thrust of [that] Act to the consequences
of employment practices, not simply the motivation." Griggs v.
Duke Power Co., 401 U.S. 424, 432 (1971) (emphasis in original);
Washington v. Davis, supra at 246-47. A prima facie violation
of Title VII may be established either by evidence of disparate
treatment or by evidence of disparate impact. Disparate treat
ment is shown where there is evidence, for example, that an
1 / The Washington v. Davis-Arlington Heights standard also
applies to claims under 42 U.S.C. § 1983 for deprivation of
the rights secured by the Equal Protection Clause. See, e.g.,
Harkless v. Sweeny Independent School District, supra.
8
employer treats blacks less favorably than whites- In such
cases, as in cases under the Fourteenth Amendment, proof of
discriminatory motive is critical, but motive can be inferred
from the fact of differences in treatment. International
Brotherhood of Teamsters v. United States, 431 U.S. 324, 335
n.15 (1977). And, as in cases under the Fourteenth Amendment,
gross statistical disparities alone may justify the inference
of a discriminatory motive and thus establish a prima facie
disparate treatment violation. Hazelwood School District v.
United States, 433 U.S. 299, 307-308 (1977); Teamsters, supra
at 339.
Title VII claims of disparate impact, cn the other hand,
need not be supported by any proof of discriminatory motive.
Teamsters, supra at 335-36 n.15; Griggs, supra at 432. See
also, United States v. City of Chicago, 573 F.2d 416, 420-23
(7th Cir. 1978). To establish a prima facie disparate impact
case, a plaintiff need only show, for example, that a facially
neutral test or other selection practice selects applicants for
hire or promotion in a significantly disproportionate pattern.
Once this is shown, the burden shifts to the employer to prove
that the practice is job related. If the employer meets this
burden, the plaintiff may then show that other selection devices
without a similar discriminatory effect would also serve the
employer's legitimate interests. Dothard v. Rawlinson, 433 U.S.
321, 329 (1977); Albemarle Paper Co. v. Moody, 422 U.S. 405, 425
(1975).
Although the Supreme Court has recognized that Title VII
9
and 42 u.S.C. § 1981 embrace "parallel or overlapping remedies
against discrimination," Alexander v. Gardner-Denver Co., 415
U.S. 36, 47 and n.4 (1974), that Court has not yet expressly
decided whether the standards of proof are the same under both
statutes. In this circuit, however, the law is that Title VII
principles as to the order and allocation of proof "apply with
equal force to a § 1981 action," Long v. Ford Motor Co., 496
F.2d 500, 505 n.ll (6th Cir. 1974), and that a prima facie viola
tion of § 1981 may be established by proof of either disparate
treatment or disparate impact. Id. at 506. Other circuits
have concluded subsequent to the Supreme Court's decision in
Washington v. Davis, supra, that the standards of proof under
§ 1981 remain identical to those under Title VII. See Johnson
v. Ryder Truck Lines, Inc., 575 F.2d 471, 474 (4th Cir. 1978);
Davis v. County of Los Angeles, 566 F.2d 1334, 1338-40 (9th
Cir. 1977), cert, granted, 46 U.S.L.W. 3780 (June 19, 1978).
Cf. Kinsey v. First National Securities. Inc.. 557 F.2d 830
2/838 n.22 (D.C. Cir. 1977).
v
2/ The Sixth Circuit, citing Washington v. Davis, supra, 426 U.S.
at 247-48, has also noted that "[tjhe more rigorous 'discrimina
tory effect' test is still applicable to causes of action based
on statutory rights rather than on constitutional grounds, for
example, those granted under Title VII of the Civil Rights Act
of 1964." NAACP v. Lansing Board of Education, supra, 559 F.2d
at 1046 n .3.
10
B. Prior to the Commencement of This Action in
1974, Defendants Engaged in a Clear Pattern
of Discrimination Against Blacks, and Defen
dants also Engaged in Practices Which Were
Not Job Related and Which Had a Discriminatory Effect on Blacks.
1. Statistical Evidence
From 1940 to 1950, the population of the City of Louisville
was approximately 15% black (see plaintiffs’ proposed finding 4),
but only 4-6% of the officers on the police force were black.
See plaintiffs' proposed finding 6. This stark disparity became
worse through the ensuing years and persisted even after this
action was filed in March 1974 (see plaintiffs' proposed findings
4-6) :
Black % of Black % ofLouisville LouisvillePopulation Year Police Officers
1960: 17.9% 1964 6.1%
1965 6.3%
1966 6.4%
1967 6.3%
1968 6.6%
1969 6.3%
1970: 23.8% 1970 6.3%
1971 6.1%1972 5.6%1973 5.6%1974 5.6%1975 7.0%
1977 7.4%
These statistics show the kind of imbalance which
is often a telltale sign of purposeful
discrimination; absent explanation, it
is ordinarily to be expected that non-
discriminatory hiring practices will in
time result in a work force more or less
representative of the racial and ethnic
composition of the population in the com
munity from which employees are hired.
11
Evidence of longlasting and gross dis
parity between the composition of a w r k
force and that of the general population
thus may be significant. . . . Teamsters .
supra, 431 U.S. at 340 n.20.
The extreme and longstanding disparities demonstrated in
this record are sufficient, standing alone, to establish a prima
facie case of racially motivated disparate treatment. Teamsters,
supra, 431 U.S. at 339; Hazelwood, supra, 433 U.S. at 307-308.
See also, Morrow v, Crisler, 479 F^2d 960, 961-62 (5th Cir. 1973),
modified on rehearing en banc on other grounds, 491 F.2d 1053
(5th Cir. 1974). This is the sort of "seriously disproportionate
exclusion of Negroes . . . [which] may for all practical purposes
demonstrate unconstitutionality because- . . . the discrimination
is very difficult to explain on nonracial grounds." Washington v .
Davis, supra, 426 U.S. at 242. Former Director of Safety James
Thornberry explained that it was in fact based on racial grounds :
"there was a good deal of prejudice in the Police Department
against the use of black police at that time [the 1950s]."
Thornberry, Vol. Ill, 6/22/77 at 404.
The disparities shown here demonstrate that this prejudice
continued to dominate the defendants1 employment practices well
after the 1950s and at least until this lawsuit was filed in 1974.
See Afro American Patrolmens League v. Duck, 503 F.2d 294, 299
2 J(6th Cir. 1974); League of United Latin American Citizens v . 7
7 / The Sixth Circuit in Duck affirmed a finding of discrimination
in police hiring practices based in part on evidence that the
minority population of Toledo was 16% but the minority representa
tion in the Toledo police department was only half that figure,
8.2%. 503 F.2d at 299. The disparities in Lousiville have always
been substantially greater. See plaintiffs’ proposed findings 4-6.
12
City of Santa Ana, 410 F. Supp. 873, 896-98, and cases cited
therein. From 1964 through 1973, only 11 of 328 new officers
accepted into recruit school classes, or 3.4%, were black (see
plaintiffs' proposed finding 8; PX 7):
Number of Whites Number of Blacks
Graduation
Year
Accepted Into
Classes
Accepted
Classes
1964 25 2
1965 15 0
1966 38 2
1967 29 2
1968 16 0
1969 24 1
1970 24 0
1971 36 0
1972 28 2
1973 82 2
There is a significant disparity between the number of black
officers hired during this period and the number one would expect
to have been hired based on any reasonable standard of comparison.
The Supreme Court has recognized that comparisons between the per
centage of minorities in an employer's work force and the percentage
of minorities in the general population of either the city or the
surrounding metropolitan area may be highly probative evidence of
intentional discriminaton where the job skill involved is "one that
many persons possess or can fairly readily acquire." Hazelwood,
supra, 433 U.S. at 308 n.13; Teamsters, supra, 431 U.S. at 337
n.17. Such comparisons may have less probative value when special
13
qualifications are required to fill particular jobs. Hazelwood.
supra at 308 n.13. The police officer's job, like the truck
driver's job in Teamsters and unlike the public school teacher's
job in Hazelwood, does not require an applicant to possess
advanced degrees or other specialized training or experience prior
to employment; instead, new police officers are given extensive
training after they have been hired. See Nevin, Vol. iv,, 6/5 3/77
at 506-510, 523-54. Accordingly, as in Teamsters. comparisons be
tween the percentage of blacks hired as Louisville police officers
and the racial composition of the general population are highly
probative. See also. League of United Latin American Citizens v .
City of Santa Ana, supra, 410 F. Supp. at 891.
Whether the geographic area for comparison is defined by the
Louisville city limits or by the far broader boundaries of the
_4/Louisville Standard Metropolitan Statistical Area (SMSA), the
evidence of purposeful discrimination is clear. In 1970, blacks
accounted for 12.2% of the population of the SMSA and 23.8% of the
population of the City of Louisville. But from 1964 through 1973,
only 3.4% of the officers accepted into recruit school were black.
This stark disparity between the proportion of blacks hired and
the proportion one ordinarily would expect to have been hired from
either the City or the SMSA is "a telltale sign of purposeful dis
crimination." Teamsters, supra, 431 U.S. at 340 n.20. 4
4_/ Plaintiffs submit that the proper geographic area to use in
determining the existence of a prima facie case, as well as in
setting an appropriate percentage goal for affirmative hiring relief
(see Section III, infra), is the City of Louisville. The courts,
including the .Sixth Circuit, "have consistently looked to the city,
i.e., the geographic area served by police and fire departments, in
considering the existence of a prima facie case." League of United
Latin American Citizens v. City of Santa Ana, supra, 410 F.Supp. at
14
Even if, as claimed by defendant's expert Dr. Michael Spar, the
appropriate population for comparison included all persons in the
1970 civilian labor force throughout the SMSA who were between the
ages of 20 and 34 and who were high school graduates or above in
educational level (see DX 28; Spar, Vol. Ill, 6/22/77 at 354-
55), the disparities would remain highly significant. The black
proportion of the group defined by Dr. Spar was 10%. Id_. Thus,
10% or 33, of the 328 officers accepted into recruit classes from
1964 through 1973 would be expected to be black if officers were
hired on a nondiscriminatory basis. Cf. Teamsters, supra, 431,
U.S. at 340, n.20. However, only 11 blacks were actually hired
4/ Cont'd.
896 and cases cited therein. See also, Afro American Patrolmens
League v. Duck, supra, 503 F.2d at 299; Davis v. County of Los
Angeles, supra, 566 F.2d at 1337; Pennsylvania v. Flaherty, 404
F.Supp. 1022' (W.D. Pa. 1975); Officers for Justice v. Civil Service
Commission, 371 F.Supp. 1328, 1330-31 (N.D. Cal. 1973). Cf.
Stamps v. Detroit Edison Co., 365 F.Supp. 87, 111 (E.D. Mich. 1973),
aff'd in part and rev'd on other grounds, sub nom. EEOC v. Detroit
Edison Co.. 515 F.2d 301 (6th Cir. 1975), vacated and remanded on
other grounds, 431 U.S. 951 (1977); Crockett v. Green, 534 F.2d
715, 718 (7th Cir. 1976). The defendants in this case have argued
that the Court should ignore the substantial weight of precedential
authority and look instead to the entire SMSA. However, the
evidence shows that police officers are required to live within 20
miles of police headquarters at 7th and Jefferson Streets in
Louisville (see PX 63-69); that they are required to secure Kentucky
operator's licenses (see PX 63-69) and thus, in effect, to be
residents of the State of Kentucky (see KRS 186.412-186.414;
Mitchell, Vol. Ill, 9/28/77 at 503, 507); and that in recent years
42-51% of the candidates on eligible lists have been residents of the
City of Louisville and another 31-40% have been residents of other
parts of Jefferson County. (Lee, Vol. IV, 7/14/77 at 634-36)-.
On these facts, the ’Louisville SMSA — which includes Clark and
Floyd Counties in Indiana, and Bullitt and Oldham Counties as well
as Jefferson County in Kentucky (see Vahaly, Vol. Ill, 6/22/77 at
313-15)— clearly is far too wide an area to be a proper basis for
comparison.
15
during this period. PX 7. The statistical analysis adopted by
the Supreme Court in Castaneda v. Partida, 430 U.S. 482, 496-97
n.17 (1977), shows that there is a difference of more than 4
standard deviations between the expected number and the actual
mumber of blacks hired based on Dr. Spar's view of the relevant
5/labor market. Because "a fluctuation of more than two or three
standard deviations would undercut the hypothesis that decisions
were being made randomly with respect to race," Hazelwood, supra,
433 U.S. at 311 n.17, this statistical comparison indicates the
existence of intentional discrimination against blacks in hiring
during the period in question. _Id. at 308-09 and n.14.
Moreover, if the comparison population is more appropriately
defined as the 1970 population of the City of Louisville, the dis
parity is even more pronounced. Since the black proportion of
this group was 23.8%, see PX 111, nondiscriminatory hiring practices
should have resulted in the hiring of 78, not 11, black officers
from 1964 through 1973. The Castaneda analysis shows that there
is a difference of more than 8 standard' deviations between the
actual and the expected numbers of blacks hired during this period
£/based on the City of Louisville data. Thus, under either view of
5/ The Castaneda statistical model measures fluctuations from the
expected value in terms of the standard deviation, which is defined
as the square root of the product of the total number in the sample
(here, 328) times the probability of selecting a black (33/328 = 0.1006)
times the probability of selecting a white (295/328 = 0.8994). Thus,
the standard deviation based on Dr. Spar's view of the relevant labor
market is 5.45. The difference between the expected and observed
numbers of blacks hiring during this period is 4.04 standard
deviations ([33-11/5.45 - 4.04). 430 U.S. at 496-97 n.17.
6/ The standard deviation is the square root of the product cf the
total number in the sample (328) times the probability of selecting
a black (78/328 - 0.2378) times the probability of selecting a white
(250/328 - 0.7622). Thus, the standard deviation is 7.71. There is
a difference of 8.69 standard deviations between the expected and
observed numbers of blacks hired ([78-11]/7.71 = 8.69). 430
U.S. at 496-97 n.17.
16
the relevant comparison population, the disparity exceeds that
which the Supreme Court found indicative of intentional discrimina-
2/txon in Hazelwood, supra. 433 U.S. at 308-09 and n.14, 311 n.17.
The resulting inference of discrimination in hiring is further
supported by applicant flow data, which provide "very relevant"
additional proof of intentional discrimination. Id. at 308 n.13.
8/Here, the only available applicant flow data for this period show
that 20% of the 413 persons who applied between July and December
1973 were black (PX 71, Books 6-7), but only 2 of the 84 officers
appointed in 1973 were black (PX 7). See plaintiffs' proposed
finding 23. This is far removed from the pattern one would expect
to result from nondiscriminatory hiring practices. Cf. Teamsters.
supra, 431 U.S. at 340 n.20.
7/ The result is the same if the relevant labor market is defined
as all 1970 residents of the City of Louisville who were 25 years
of age or older and were high school graduates with no further
education or who were between the ages of 18 and 24 and (1) had
completed four years of high school and were not enrolled in school
or (2) were enrolled in their fourth year of high school. Since the
black proportion of this population was 20.4%, see PX 111, nondis
criminatory hiring practices should have resulted in the hiring of
70 black officers instead of 11 from 1964 through 1973. There is
a difference of almost 8 standard deviations between these figures.
See Castaneda, supra. 430 U.S. at 497-97 n.17.
8/ Defendants did not keep records of the race of applicants
prior to July 1973. See PX 71, Books 1-6.
17
2. Recruitment and Selection Practices
This extreme departure from the expected hiring pattern
was a direct result of both intentional discrimination in re
cruitment and hiring and the use of discriminatory tests and
other discriminatory selection practices. It is clear from
the record that, at least until this action was filed in 1974,
there was a strict upper-limit quota on the number and percentage
of black officers allowed to be on the force at any one time.
See plaintiffs' proposed findings 6-8. From 1964 through 1973,
black officers never accounted for more than 6.6% of the police
force (PX 38), and between 15 and 82 new white officers but
never more than 2 new black officers were accepted into the
recruit school classes which graduated in each of those years
(PX 7) .
These restrictions were accomplished in part by the defen
dants' negative reputation in the black community for discrimina
tion against black applicants and black police officers. See
plaintiffs' proposed findings 20-21. This reputation deterred
many blacks from seeking jobs on the force. Id. The defendants
and their predecessors were aware that in the black community
there was "considerable peer pressure against young men . . .
becoming policemen." Burke, Vol. Ill, 6/22/77 at 384; Burton,
Vol. Ill, 6/22/77 at 397-98. The defendants also claimed that
between 1965 and 1972 there were always vacancies in the Division
-18-
of Police and there was a continual shortage of applicants to
fill those vacancies. Richmond Dep., 5/11/77 at 87, 98-99.
But after a short-term effort by James Thornberry in the early
1950s which was allowed to succeed only in a very limited way,
see Thornberry, Vol. Ill, 6/22/77 at 404-406, defendants took
no active role in attempting to recruit black applicants until
after the filing of this lawsuit. See plaintiffs' proposed
finding 22. This failure or refusal to recruit black applicants
for jobs as police officers constituted "a pattern of official .
inaction which [had] the natural, probable and foreseeable
result" of perpetuating the exclusion of blacks from the police
force. NAACP v. Lansing Board of Education, suora, 559 F.2d
at 1047-48.
When blacks overcame these obstacles and attempted to apply
for jobs as police officers, they were met with other discrimina
tory barriers to employment. The selection procedures which
defendants used prior to the commencement of this action in 1974
are set forth in plaintiffs' proposed finding 24. Whien the
defendant Civil Service Board reviewed these procedures in
January 1975, it made the following findings: the receptionist
in the Civil Service office was making all decisions as to the
right of an applicant to fill out an application, without ever
conducting an initial interview to determine whether the appli
cant met the minimum qualifications with respect to such
-19 ~
characteristics as "speech defect, " "marked deformity, " vision,
education, age, and military discharge and Selective Service
status; agency requisitions were not being processed in accordance
with Civil Service rules and regulations; applicants who were
certified by Civil Service for employment were sometimes dis
qualified by the agencies without any written reasons; proper
eligibility lists were not maintained; a backlog of vacancies
had developed, and open-continuous testing had to be used for
several months to eliminate the backlog; proper job analysis
procedures were not followed; the written test was not validated
and was deficient in many respects; the oral interviews for
police officer were unstructured and subjective, and they were
not validated; test weights were set in an arbitrary manner;
the physical fitness standards were not valid and they did not
necessarily measure physical stamina or physiological ability
to tolerate stress; the Division of Police conducted the back
ground investigation and made recommendations to disqualify
applicants which usually were accepted by the Civil Service
Director without information as to whether the reason for dis
qualification was job-related; and the practice of giving a
"training and experience" rating gave an extra advantage to
persons with "inappropriate" training and experience which was
not required to do the job and also benefited applicants who
received high scores on the unvalidated written test. See
plaintiffs' proposed finding 25.
-20-
The disproportionate exclusion of blacks from the force
on these grounds bearing no demonstrable relationship to job
performance violated Title VII and 42 U.S.C. §1981- See, e.g.,
Griggs v. Duke Power Co., supra, 401 U.S. at 432; Albemarle
Paper Co. v. Moody, supra, 422 U.S. at 425; Davis v. Countv
of. Los Angeles, supra, 566 F.2d at 1340-42; Johnson v. Ryder
Truck Lines, Inc., supra, 575 F.2d at 474; Dozier v. Chupka,
395 F.Supp. 836, 850-52 (S.D. Ohio 1975). Moreover, defendants
disqualified applicants on the basis of several specific
criteria which have been shown to have an adverse impact on
blacks and which were not job-related, such as (1) juvenile and adult
arrest records, PX 63-67, 79, 85; Richmond Dep., 5/23/77 at 312 —
see United States v. City of Chicago, 549 F.2d 415, 432 (7th Cir.^
cert. denied. 434 U.S. 875 (1977); Gregory v. Litton Systems, Inc.,
472 F.2d 631 (9th Cir. 1972); (2) maximum weight standards, PX
63-67 — see VonderHaar, Vol. Ill, 9/28/77 at 431; and (3) financial
condition, Richmond Dep., 5/11/77 at 76 — see United’ States
v. City of Chicago, supra, 549 F.2d at 432; Wallace v.
Debron Corp., 494 F.2d 674 (8th Cir. 1974); Johnson v. Pike
Corp., 332 F.Supp. 490 (D.D. Cal. 1971).
The written tests which defendants used during this period
also had an adverse' impact on blacks and were not job-related.
"Test for Policeman (10-D)," Px 19-20, was used until 1971
despite the fact that there had been serious doubts about its
security since at least 1965. Richmond Dep., 5/11/77 at 19,
5/23/77 at 98-99; Olges, Vol. II, 7/12/77 at 190-91. The defendant
- 2 1 -
Civil Service Board found that parts of this test were not job-
related or did not apply to the job of a Louisville police officer.
Id; Richmond Dep., 5/23/77 at 149. See plaintiffs' proposed
finding 26. Similarly, "Examination for Police Patrolman No.
0044, " PX 18, which was used from 1971 to 1975, was "out of gear"
according to former Director of Safety James Thornberry, Vol. Ill,
6/22/77 at 411; "they were vocabulary-type tests, I.Q. tests,
AGC-type things, .. . which are not particularly job-related." Id_. at
413. See plaintiffs proposed finding 27.
When Darryl Olges became Chief Examiner in May 1975, he
performed a study of tests and other civil service selection
procedures which confirmed Thornberry's view. Olges, Vol. II,
7/12/77 at 132-85; 190-99. The defendant Civil Service Board
found in 1975 that many of its written tests were outdated and
no longer applicable to the job duties in question; that item
analyses had not been performed; that "[a]11 questions were
related to a candidate's knowledge, rather than his skill or
ability in performance, and a definite advantage was extended
to those who had the benefit of advanced education"; that "[a]11
written examination procedures were posited on the applicant's
ability to reduce his skill or his performance to a mental
exercise and be able to relate that with paper and pencil";
that the persons grading examinations had knowledge of the names
and personal information concerning candidates, and that test
- 22 -
security — which was "of utmost importance for fairness and
non-discriminatory practices" — continued to be a serious
problem; that examination grade scales were in many instances
outdated and not adjusted for current relevance; that in most
instances heavy weight was given to written tests based on
academic materials; and that test weights were set in an
arbitrary manner. DX 75, "Narrative" at 8, 12, 14. The Board
further found that Police Patrolman Examination No. 0044 had
been scored and used as a ranking device with a weight of 65%
of the total examination process without any available rationale
for the cut-off score of 52, and that "[t]he written examination
as it stood prior to August 1st [, 1975] was not validated." DX 75,
"Louisville Civil Service Board Selection Procedures and Recruit
ment Program, Book I" — "Written Entrance Test" at 1-2. The
written test also had been administered in a manner which per
mitted candidates to memorize the items; an item analysis was
finally performed and it showed that many of the items did not
adequately differentiate between candidates; parts of the written
test were not sufficiently related to the content of the police
officer's job; and the test placed too much emphasis on reading
and mathematics skills and was "approximately 80-90% invalid."
Olges, Vol. II, 7/12/77 at 182-85, 188-99; Vol. I, 7/11/77 at 155; DX 38
The available evidence demonstrates that these written tests
had a substantial adverse impact on blacks. For example, in 1975
-23-
only 12 of 342 white applicants (3.5%) failed Examination No.
0044, PX 18, but 14 of the 93 black applicants (15.1%) failed
this test. DX 75, "Statistical Data, Book Three" — "General
Statistical Summary, Sworn Personnel, " at f6. Because
defendants did not keep records of the race of applicants prior
2/to July 1973, and because the records which were kept between
July 1973 and December 1975 are unreliable and in many instances
10/
illegible, the record does not permit a precise mathematical
computation of the adverse impact which these written tests
had on black applicants prior to the institution of this action.
However, the record shows that tests of this kind traditionally
have an adverse impact on minorities, Barrett, Vol. IV, 7/14/77
at 556, and it has long been widely recognized that black and
other minority persons typically perform below the norm for
whites on such culturally biased paper-and-pencil tests of
generalized intelligence or aptitude. Boston Chapter, NAACP,
Inc, v. Beecher, 504 F.2d 1017, 1021 (1st Cir. 1974), cert.
denied, 421 U.S. 910 (1975); Bridgeport Guardians, Inc, v.
Members of Bridgeport Civil Service Commission, 482 F.2d 1333, 1340
(2d Cir. 1973) cert, denied, 421 U.S. 991 (1975); League of United
9/ See PX 71, Books 1-6.
10/ See DX 75, "Statistical Data, Book Three" — "Statistical
Variables" at 1-2. See also, Affidavit of Joshua Tankel, filed
herewith.
-24-
Latin American Citizens v. City of Santa Ana, supra, 410 F.Supp.
at 902. See Cooper & Sobol, Seniority and Testing Under Fair
Employment Laws; A General Approach to Objective Criteria of
Hiring and Promotion, 82 Harv. L. Rev. 1598, 1640 (1969).
In view of the gross underrepresentation of blacks on the Louis
ville police force and in view of the defendants' failure to keep
adequate records concerning the impact of its tests, it can properly
be inferred that these tests had an adverse impact on black
applicants. See §4d , Uniform Guidelines on Employee Selection
Procedures (1978), 43 Fed. Reg. 38290 (Aug. 25, 1978).
The Civil Service Board's review of its pre-1975 selection
procedures disclosed not only that many of its practices and
criteria were not demonstrably related to successful job per
formance, but also that many applicants and potential applicants
were being rejected on the basis of arbitrary, subjective
determinations and improper procedures on the part of defendants
and their employees and agents. See plaintiffs' proposed finding
25. As the Supreme Court has held in the context of jury selec
tion, "a selection procedure that is susceptible of abuse or
is not racially neutral supports the presumption of [intentional]
discrimination raised by the statistical showing. Washington v .
Davis, 426 U.S., at 241. . ..." Casteneda v. Partida, supra,
430 U.S. at 494. The same principle has been applied in employ
ment discrimination cases. See, e.g., Stewart v. General Motors
-25-
Corp., 542 F .2d 445, 450 (7th Cir. 1976), cert. denied, 443
U.S. 919 (1977); Rowe v. General Motors, 457 F.2d 348, 358-59
(5th Cir. 1972).
The testimony of a number of individual witnesses in this
case demonstrated that this potential for abuse was realized
when blacks sought to become police officers. For example,
when Norma Boyd tried to apply for the job over a period of years,
white receptionists in the Civil Service office repeatedly re
fused to give her an application form for a variety of spurious
reasons: first, in 1971, she was told that a GED certificate was
inadequate, Boyd, Vol. I, 4/25/77 at 39; second, in 1972, she
was told that a year of college was required of all applicants,
id. at 43; and third, in 1973, she was told that she was too
short, id. at 44. In fact, in 1971 and 1972 a GED certificate
was acceptable (PX 21E; Richmond Dep., 5/23/77 at 139), and a
year of college was not required (PX 21c). And the defendant
Civil Service Board had specifically directed in September
1972— six months before Boyd was denied an application on theii/ground that she was too short — that persons who did not appear
to meet the height qualifications must nevertheless be given an
application. PX 70. Finally in 1974, three years after her
11/ Officer James Brown reported a similar experience. In 1972
or 1973, after he had been on the police force for two to three
years, Brown decided to go to the Civil Service office in
civilian clothes and ask for an application, "[t]o see where
in the whole scale of things were blacks being cut loose, where
-26-
initial attempt, Boyd was given an application form by a black
receptionist. Boyd, Vol. I, 4/25/77 at 48.
In addition, black applicants were subjected to unexplained
delays in the processing of their applications, and they were
disqualified on the basis of inaccurate information which the
defendants refused to correct. David Lyons first applied
in 1969. He filled out an application and passed the written
test, the records check,the physical fitness test, and the medical
examination. Lyons, Vol. IV, 4/28/77 at 657-59. Although most
applicants were then certified immediately to the Division of
Police and were hired shortly thereafter (Lee, Vol. II, 6/21/77
at 260-61; Richmond Dep., 5/11/77 at 43-44, 84-85), it was not
until April 1970, four to five months after he had applied, that
Lyons received a letter from the Civil Service Board informing
him that he had been disqualified by the character or background
investigation due to his military record. Lyons, Vol. IV,
4/28/77 at 659, 670. When Lyons went to the Civil Service office
to question his disqualification, he learned that it was based
on speculation by Civil Service as to why he had been granted an
honorable discharge for the "convenience of the government."
1V Cont'd
they were being shoved out." Brown, Vol. IV, 4/28/77 at
580-82. When he asked for the application, the woman at the
desk refused to give it to him on the ground that he was too
short to qualify for the job of police officer. Id. at 582-83.
Brown is six feet tall. Id. at 582.
-27-
Id. at 660-61. No one in the Civil Service office made any
attempt to determine the reason for this discharge from any
official source. Instead, even after Lyons himself had secured
the reason and advised the defendants of the basis for the dis-J2/
charge, they refused to accept his explanation and correct the
erroneous disqualification. I<2. at 661. Only after Lyons spoke
to Senator Cook and the Senator intervened on his behalf was
Lyons allowed to become a police officer. Id_. at 662-63.
Ronald Jackson, another black applicant and a named plain
tiff in this action, was also initially disqualified on the
basis of inaccurate Civil Service Board information. He first
applied to be a police officer in 1973, with three years of
college work in law enforcement and one year of experience in
the Army Military Police. Jackson, Vol. II, 4/26/77 at 255,
257. He passed the written test and then took the medical
examination. Id. at 25 8-59. About a week later, after not hearing
anything further regarding his application, Jackson inquired at
the Civil Service office and was told that he had failed the
eye examination. Id. at 260. Although defendants' files
variously indicated that Jackson's vision was 20/100 in each eye
12 / in fact, the discharge was granted for the"convenience
of the government" because, when Lyons returned from overseas
with only about 20 days left to serve, it was easier for the
Army to discharge him early than to reassign him to another
unit. Id. at 661.
-28-
(DX 15, Richmond letter dated Nov. 13, 1973) or 20/70 in each
eye (DX 15, Lawwill letter dated Dec. 7, 1973), Jackson has
never had any serious problems with his eyes and had never been
disqualified from any job because of his eyesight. Id_. at 261.
After consulting the Louisville Urban League, Jackson was re
examined by an opthamologist who determined that his uncorrected
visual acuity was 20/50-1 in each eye separately and 20/40-1
with both eyes together. DX 15, Schiller letter dated Sept. 24,
1974. Based on this examination, the Civil Service Board finally
decided in September 1974 that Jackson met the visual standards
and that he should be contacted to complete his processing for
entrance into the police department. DX 15, Richmond letter
dated Sept. 27, 1974. However, Jackson instead was given a
temporary job as a jail guard, and he had no further contact
from the defendants concerning his application for a job as a
police officer until he attended a meeting arranged by the
Urban League in the summer of 1975 to discuss the cases of a
number' of black applicants who had been screened out by the
selection process as he had been. Jackson, Vol. II, 4/26/77 at
268-70.
The way was apparently clear for Jackson to become a police
officer as of September 1974, but his processing was unaccountably
delayed until well after Jack Richmond had left as Civil Service
Director. It was not until July 15, 1975, almost ten months
later, that the new Director, Jeanette Priebe,apologized to
-29-
Jackson because he had been "put to so much inconvenience"
and informed him that he would be the first applicant referred
to the police department when it next started hiring. DX 15,
Priebe letter dated July 15, 1975. During this delay, many
other applicants were hired. See PX 53. Jackson was finally
admitted to recruit school in November 1975, almost two years
after he had first applied. Jackson, Vol. II, 4/26/77 at 273.
Gary Hearn, another named plaintiff in this action,also
was denied a job as a police officer for more than a year on
the basis of erroneous information which the defendants re
peatedly refused to correct. Hearn was allowed to become a
police officer only after he had undertaken extensive efforts,
which included the securing of a court order, to force the
defendants to disregard their inaccurate information. See
plaintiffs' supplemental post-trial brief.
For a large part of the period in question, 1965 through
late 1974, Jack Richmond was the Director of Civil Service.
Richmond and the employees who worked under him had virtually
unlimited discretion over the operation of the entire civil
service selection and referral process. People who attempted
to apply for jobs were refused applications at the whim of the
receptionist; requisitions were not processed in accordance
with Civil Service rules and regulations; eligibility lists
were not maintained; proper job analysis procedures were not
followed; unvalidated written tests and other non-job related
-30-
criteria were used; subjective and unstructured oral interviews
were given. DX 75, "Narrative" at 2-14. See plaintiffs' pro
posed finding 25.
Throughout Richmond's tenure, applicants were not certified
or appointed in rank order from eligibility lists; instead,
whenever the Division of Police notified Richmond that it had
vacancies to fill, applicants were put through Richmond's pro
cedures, and those who survived were referred by Richmond for
appointment. Richmond Dep., 5/11/77 at 43-44, 84-85; Lee/
Vol. II, 6/21/77 at 260-61; Arnold, Vol. IV, 9/29/77 at 767,
778-82rcoleman, Vol. IV, 9/29/77 at 658; DX 75, "Narrative" at 4.
Richmond "had a lot to do with who was on that Police Department.
He was involved in the initial process all the way to the end."
Coleman, Vol. IV, 9/29/77 at 629-30. Richmond had the sole
authority until 1972, and substantial authority thereafter until
his departure in 1974> to determine who passed and who failed
the subjective and unstructured oral interview examination for
police officer. Richmond Dep., 5/11/77 at 79, and 5/23/77 at
247-48; Olges, Vol.II, 7/12/77 at 183-84; Arnold, Vol. IV, 9/29/77
at 768; Coleman, Vol. IV, 9/29/77 at 630; DX 75, "Narrative"
at 8-9 and "Louisville Civil Service Board Selection Procedures
and Recruitment Program, Book I" — "Oral Interview" at 1.
This procedure made Richmond the final judge of such subjective
factors as a candidate's "appearance," "voice and speech,"
-31-
sincerity," and "judgment."mental alertness," "stability," "
Richmond Dep., 5/23/77 at 231-46. The result of this con
centration of power in Richmond's hands was candidly, if
somewhat gingerly, described by former Director of Safety
James Thornberry:
[W]e had some difficulty with the gentleman
that was at that time the head of Civil Service
[Richmond], and we had to go to a great deal of
trouble, particularly on getting people past the
physical examination; and there were certain
other things . . ., they always ran a police
check and that sort of thing . . . . [Richmond]
was an ex-Army man with some rather rigid ideas
about things; and I heard as many complaints from
would-be white applicants as black, but the re
sult was that a whole lot of people were being
hurt, and . . . whether it was intentional or not,
I certainly wouldn't say, but generally, it seemed
to come down the hardest on the black applicants.
So consequently we had to do a lot of fighting
and scratching to get some of them past that
business. Thornberry, Vol. Ill, 6/22/77 at
409-410.
See also, Coleman, Vol. IV, 9/29/77 at 634-35;Arnold, Vol. IV,
9/29/77 at 771.As the record demonstrates, prior to the filing of this
lawsuit in 1974, very few black applicants were able to fight
and scratch their way past "that business." The defendants
selected police officers largely on the basis of tests and
other criteria which had an adverse impact on blacks and were
not job-related, and on the basis of arbitrary and subjective
criteria and procedures which were not racially neutral, were
susceptible to abuse, and were in fact abused by defendants.
The result was a strict limitation on the number and percentage
of black officers on the Louisville police force.
-32-
3. History of Racial Discrimination
The intentionally discriminatory nature of the defendants1
pre-lawsuit recruitment and selection practices is brought into
sharp focus by the City's long history of racial segregation and
discrimination in police employment practices. When the City of
Louisville established a police force in the 1820s, controlling
the black population and keeping black slaves from escaping across
the Ohio River were among its primary functions. Keil, Vol. IV,
9/29/77 at 737-38. When a limited number of blacks finally were
allowed to become officers on the force, they were restricted
primarily to the task of policing the black community. See plain
tiffs' proposed findings 9-12. Until well into the 1960s, all
black uniformed patrol officers were assigned to walking beats
i!/within rigidly defined boundaries of the second district, and
later the fourth district, without regard to their desires,
qualifications, or length of service. See plaintiffs' proposed
findings 9-10. As former Director of Safety James Thornberry-
testified, there was "a good deal of prejudice in the police
department against the use of black police at that time [the 1950s]
They thought that the only place black police officers could be
assigned was in the Second District." Thornberry, Vol. Ill,
6/22/77 at 404-405.
Moreover, black uniformed patrol officers were not permitted
to ride in patrol cars until 1964; blacks were .excluded from
_ / Black officers, were restricted to the area bounded by
6th Street on the east, 14th Street on the west, Jefferson Street
on the north and Esquire Alley or Broadway on the south. See
plaintiffs' proposed finding 9.
-33-
homicide, burgulary, and other special squads; blacks were re
quired to attend racially segregated daily meetings to receive
their orders; black recruits received inferior physical training
in racially segregated facilities; and black officers were not
given the same training opportunities as whites. See plaintiffs'
proposed findings 12, 14-15, 17-18.
Blacks were hired and promoted specifically to fill "black
jobs" in segregated units. See plaintiffs' proposed findings 7-8.
Until this lawsuit was filed in 1974, strict limits were maintained
on the number of black officers hired in any one year and on the
percentage of blacks on the force at any given time. See plain
tiffs’ proposed findings 6-8. Three positions specifically for
black sergeants were created in approximately 1944, and there
have never been more than three black sergeants at any time since;
there have never been more than two black lieutenants at any one
time; there has never been a black captain in the history of the
Division of Police. See plaintiffs' proposed finding 16. At
least until the 1960s, the only way a black officer could become
a sergeant or a lieutenant was for one of the three black sergeants
or a black lieutenant to die, retire, or be fired. Id.
These practices were based on racial prejudice against black
officers, see Thornberry, Vol. Ill, 6/22/77 at 404-405, and they
violated the Fourteenth Amendment under "the cardinal principle
that racial classifications that stigmatize — because they are
drawn on the presumption that one race is inferior to another or
because they put the weight of government behind racial hatred
and separatism — are invalid without more." Regents of the
University of California v. Bakke, 46 U.S.L.W. 4896, 4920 (June 28,
-34-
1978) (opinion of Brennan, White, Marshall, and Blackmun, j j .).
See United Jewish Organizations of Williamsburgh, Inc, v. Carey.
430 U.S. 144, 165 (1977) (opinion of White, Rehnquist, and Stevens,
JJ.); Reitman v. Mulkev. 387 U.S. 369, 375-76 (1967); McLaughlin
v . Florida, 379 U.S. 184, 191-92 (1964) ; Brown v. Board of Educa
tion, 347 U.S. 483, 494-95 (1954); Yick Wo v. Hopkins, 118 U.S.
356, 374 (1886); Strauder v. West Virginia. 100 U.S. 303, 308
(1879). This pattern of racial stigmatization provides strong
additonal evidence of intentional discrimination against blacks
in recruitment and hiring, "particularly [since] it reveals a
series of official actions taken for invidious purposes."
Arlington Heights, supra, 429 U.S. at 267. See also, Harkless
v. Sweeny Independent School District, supra, 554 F.2d at 1357.
It is clear that, prior to the commencement of this action,
"racial discrimination was the [defendants'] standard operating
procedure — the regular rather than the unusual practice,"
Teamsters, supra, 431 U.S. at 336, in violation of the Fourteenth
Amendment, §§ 1981 and 1983, and Title VII.
C . After the Commencement of This Action in
1974, Defendants Corrected Some of Their Discriminatory Practices but Persisted
in Others.
1. Statistical Evidence
After this lawsuit was filed, the arbitrary and subjective
procedures which for so many years had been manipulated to exclude
blacks from the police force were briefly turned to their temporary
-35-
benefit. From 1974 to the time of trial, the following numbers
of black and whites were appointed as police officers each year
(see plaintiffs' proposed finding 31):
Year
Number of
Whites
Appointed
Number of
Blacks
Appointed
1974 78 21
1975 31 11
1976 11 0
1977 28 1
The limited gains in 1974 and 1975 were not the result of
any voluntary changes in the defendants' attitudes or practices.
For example, only three blacks initially had been selected for
the recruit class which was scheduled to begin in June 1974.
Coleman, Vol. IV, 9/29/77 at 631, 637. The members of this re
cruit class were selected during what Louis Coleman, Housing and
Urban Affairs Director of1the Louisville Urban League, described
as "the Jack Richmond era." Id. at 636-37. See pp. 30-32, supra.
The Urban League, which had been actively involved in recruiting
minorities for Louisville civil service jobs since 1971 and speci
fically for jobs as police officers since 1973, Coleman, Vol. IV,
9/29/77 at 614-15, reviewed the composition of the June 1974 re
cruit class and "saw that 11 good black candidates were for some
reason left out of that class." Id. at 637. After the Urban League
protested the exclusion of these black applicants at a meeting with
-36-
the Mayor and the Director and the Assistant Director of Safety,
their disqualifications were reversed and all 11 of them were placed
in the recruit class. Id. at 631, 637-38. Without the vigorous
intervention of the Urban League, there would have been only 3
blacks instead of 14 in that class. Id. Furthermore, the
defendants were acutely aware of the effect which their hiring
practices during this period might have on the Cou: t 's view of
this litigation, and they acted accordingly. For instance, as
soon as the Chief of Police learned that some minority applicants
had done well enough on the August 1975 written test to be ranked
high on the eligibility list,
The chief said he would go to the Mayor
with these figures and he felt that there
may be a good chance of getting a training
class started before the court case comes
up. He said that normal attrition would
be the basis of his request. . . . PX 107.
See Wilds, Vol. IV, 7/14/77 at 677.
The courts have repeatedly recognized that " [s}uch actions
in the face of litigation are equivocal in purpose, motive and
permanence." Jenkins v. United Gas Corp., 400 F. 2d 28, 33
(5th Cir. 1968). The hiring of 32 blacks as officers in the
two year period following the filing of this lawsuit, after hiring
_ 14/
14/ Richmond apparently did not attend this meeting. The tran
script of Coleman's testimony reads as follows: "And Jack Richmond
was [not?}, at that particular meeting. He did not show up, and I
remember that quite vividly, and we were able to look at the dis- .
crepancies, why these blacks were not entered into this class, and
those discrepancies were resolved. . . . " Coleman, Vol. IV, 9/29/77
at 638.
-37-
only 11 blacks in the previous ten years, does not provide the
defendants with any defense to the claims of discrimination stated
in the plaintiffs' EEOC charges or in their complaint in this
action. See Rice v. Gates Rubber Co., 521 F.2d 782,785 (5th
Cir. 1975); Rich v. Martin Marietta Corp., 522 F.2d 333, 346.
and n.10 (10th Cir. 1975); Parham v. Southwestern Bell Telephone
Co., 433 F.2d 421, 425-26 (8th Cir. 1970). Indeed, the post
filing conduct of the defendants in this case "tend[s] to show
the existence of prior discrimination and an effort to repair the
harm after discovery." Rich v. Martin Marietta Coro., supra, 522
F.2d at 346. Moreover, as the Supreme Court has recognized,
even a racially balanced work force
cannot immunize an employer from liability
for specific acts of discrimination. . . .
"[A defendant's] later changes in its hir
ing and promotion policies could be of
little comfort to the victims of the earlier
post-Act discrimination, and could not erase
its previous illegal conduct or its obliga
tion to afford relief to those who suffered
because of it." Furnco Construction Corp. v .
Waters, 46 U.S.L.W. 4966,4969-70(June 29, 1978),
quoting Teamsters, supra, 431 u.S. at 341-43.
In this case, the hiring of more black officers immediately
following the institution of this lawsuit did not signal any per
manent improvement in the defendants' practices. Nearly half of
the blacks who were hired in 1974 and 1975 proved to be only
temporary additions to the police force. By December 1976, 83%
of the whites who had been hired during this period (91 of 109)
but only 53% of the blacks (17 of 32) were still on the force.
See plaintiffs' proposed finding 31. And after 1975, the defendants'
hiring of blacks came virtually to a complete halt: out of 40
-38-
police officers hired in 1976 and 1977, only one was black. Id.
Thus, by the time of trial only 53 of the 714 officers on the
force, or 7.4%, were black. PX 38. The disparity between this
figure and the black percentage of the 1970 Louisville population
(23.8%), or any other reasonable standard of comparison, is so
large that it demonstrates a prima facie case of discrimination
in hiring. See Teamsters, supra, 431 U.S. at 339, 340 n.20;
Hazelwood, supra, 433 U.S. at 307-308; Afro American Patrolmens
League v . Duck. supra, 503 F.2d at 299.
2. Recruitment and Selection Practices
As established above, despite their negative reputation in
the black community, defendants made no sustained effort to re
cruit black applicants for jobs as police officers until after
this lawsuit was filed. See plaintiffs' proposed findings 20-22;
see pp. 18-19, supra. Much earlier, however, they had demon
strated their ability to secure black applicants when they felt
that it was necessary to do so. For example, the City had no
apparent difficulty in finding blacks to fill the segregated jobs
in three black platoons which were created in the 1940s, see
Taylor, Vol. I, 3/9/77 at 195, nor in replacing black officers
who had left those jobs with other blacks in the 1950s, see Frac
tion, Vol. I, 3/8/77 at 119.'
In the 1970s, after this lawsuit was filed, the defendants
again perceived a need to find black applicants, and they again
demonstrated their ability to do so. Between January and October
of 1975, 166 of 966 applicants, or 17.2%, were black. Priebe,
39
Vol. I, 9/26/77 at 127-28, 134-35. Between November 1976 and
January 1977, 222 of 944 applicants, or 23.5%, were black. PX
34; Priebe, Vol. I, 9/26/77 at 113-16, 134-35. While these
improvements in defendants' recruiting efforts are desirable and
may even be commendable, "in the face of litigation [they] are
equivocal in purpose, motive and permanence." Jenkins v. United
Gas Corp., supra, 400 F.2d at 33. See pp. 37-38, supra.
The defendants' recent success in attracting black applicants
provides neither a defense to plaintiffs' prima facie case of
discrimination nor a justification for the denial of affirmative
hiring relief. See Arnold v. Ballard, 390 F. Supp. 723, 738
(N.D. Ohio 1975), aff'd, 12 FEP .Cases 1613, 1614 (6th Cir.),
vacated and remanded on other grounds, 16 FEP Cases 396 (6th Cir.
1976); League of United Latin American Citizens v, City of Santa
Ana, supra, 410 F. Supp. at 881; Jones v. Tri-County Electric
Cooperative, Inc., 512 F.2d 1, 2 (5th Cir. 1975).
Blacks who were recruited and applied for jobs as police
officers after the filing of this action continued to face dis
criminatory testing and selection procedures. By January 1975,
defendants were aware of numerous defects in these procedures.
See plaintiffs' proposed finding 25; see pp. 19-23, supra. But
defendants continued until August 1975 to select police officers
in essentially the same way, except that the unstructured and
subjective oral interview was assigned a weight of 20% of the
total score, the inappropriate training and experience rating was
assigned a weight of 15%, and the unvalidated written test was
assigned a weight of 65%. See plaintiffs 1 proposed finding 32.
40
These practices bore no demonstrable relationship to job per
formance. See pp. 19-23, supra.
In August 1975, the defendant Civil Service Board instituted
new selection procedures which eliminated the practice of granting
bonus points for unnecessary training and experience, attempted
to bring some structure to the oral interview process, and re
placed Police Patrolman Examination No. 0044, PX 18, with a new
written examination devised by Darrell Olges, Chief Examiner of
the Civil Service Board, DX 39. See plaintiffs' proposed finding
33. This new test drew upon "assessment center" techniques and
included audio-visual components. Olges, Vol. II, 7/12/77 at
213-25. Only 1.1% (3 of 280) of the white applicants failed this
test while 8.2% (5 of 61) of the black applicants failed. DX 75,
"Statistical Data, Book Three" — "Sworn Personnel" at 5{ 6. But
a sufficient number of blacks scored high enough on the test that
it did not have a substantial adverse impact as it was used
Lsybetween August 1975 and November 1976. Olges, Vol. II, 7/12/77
at 226. See plaintiffs' proposed finding 35.
Despite these reforms, black applicants continued to be sub
jected to unexplained delays and disqualified on the basis of
arbitrary, subjective, and discriminatory criteria. For instance,
even after the Civil Service Board finally acknowledged in September
1374 that plaintiff Ronald Jackson satisfied its vision standards,
the processing of his application was unaccountably delayed for
15_/ Defendants then decided to replace the Olges test with Test
165.1 (PX 93), an unvalidated written examination which had an
extreme adverse impact on black applicants. See Section II, infra.
41
a period of nearly ten months, during which many other applicants
were hired. See pp. 28-30, supra.
Norma Boyd, after finally obtaining an application form on'
her fourth attempt in 1974, see pp. 26-27, supra, passed written
tests and oral interviews in 1974, 1975, and 1977, but was never
hired as a police officer. Boyd, Vol. I, 4/25/77 at 51-52, 57-64.
She took a medical examination in March 1977 and subsequently
received a letter signed by Jeanette B. Priebe, Personnel Director,
stating that "you have been disqualified as an applicant for
Police Officer because you failed the medical examination." PX
40; Boyd, Vol. I, 4/25/77 at 69-70. Boyd telephoned and made two
visits in person to the Civil Service office to learn the reason
for this disqualification, but Priebe and Jerry Lee, Manager of
Placement and Records, told her only that there was some
unspecified problem with her back. Id. at 70-71. Lee also told
her that the letter she had received did not really mean that she
had been disqualified, but only that there was "some question
about the medical." Id. at 72. Applicants seemed to contact
Civil Service much faster, he said, when they received letters
16/falsely informing them that they had been disqualified. id.
Boyd finally learned by contacting Dr. William VonderHaar, who
is in charge of the pre-employment medical examinations for police
and other Civil Service applicants, that the letter had been sent
because he had assumed, based on the fact that she had been given
16/ It is more reasonable to assume, however, that most applicants
would not demonstrate Norma Boyd's persistence but instead would
take such letters to mean what they said.
42
a myelogram by another physician, that she must have some sort
of back injury. Id. at 72-73; VonderHaar, Vol. Ill, 9/28/77 at
452-54. His assumption proved to be incorrect,but Boyd was not
asked to explain the myelogram nor did defendants make any other
effort to determine whether she really had any back injury until
after her persistent attempts to learn why she had been disqualified.
Id; Boyd, Vol. I, 4/25/77 at 73-74. She then passed the medical
examination, id. at 74, but later was told that she had failed
the "stress test." Id. at 75-80. Thus, after Boyd had aoplied or
attempted to apply a total of six times since 1971, id. at 82,
she was rejected in 1977 on the basis of a selection procedure
which had never been used to screen out applicants until November
1976. See DX 75, "Louisville Civil Service Board Selection Pro
cedures and Recruitment Program, Book I" — "Physical Agility
Test" at 2.
Another black applicant, Mary Gaines, applied in the spring
of 1975 and by November of that year had passed the written, oral,
and medical examinations. Gaines, Vol. II, 4/26/77 at 154-56;
Priebe, Vcl. I, 9/26/77 at 22-25. She was then informed by Jerry
Lee that she would be included in the next recruit class. PX 41.
She passed another medical examination in March 1976 and was
scheduled for a hiring interview the following month, but was not
hired at that time due to a hiring freeze. Gaines, Vol. II, 4/26/77
at 164; Priebe, VOl. I, 9/26/77 at 26.
When she was certified for appointment as a police officer in
September 1976, Gaines underwent still another medical examination.
Id. at 26-27; Gaines, Vol. II, 4/26/77 at 166. However, after this thir
43
examination, she was told that she had failed the medical due to
high blood pressure. PX 44; Gaines, Vol. II, 4/26/77 at 174;
Priebe, Vol. I, 9/26/77 at 27. Following additional examinations
by two private physicians and several re-examinations at Louis
ville Memorial Hospital, and after intervention on her behalf by
the Urban League, it was finally decided that Gaines had passed
the medical examination for the third time. Gaines, Vol. II,
4/26/77 at 175-80; Priebe, Vol. I, 9/26/77 at 28. But Gaines,
like Norma Boyd, then failed the "stress test." Gaines, Vol. II,
4/26/77 at 182-83. This was the first time this selection pro
cedure had ever been used to screen out applicants. See DX 75,
"Louisville Civil Service Board Selection Procedures and Recruit
ment Program, Book I" — "Physical Agility Test" at 2.
Black applicants were subjected to other arbitrary medical
standards as well. When Ora Seay and Sandra Richardson applied
in 1975, the defendants in theory had eliminated strict height
and weight requirements (compare PX 21-A with PX 21-B), which
they recognized to be racially discriminatory. VonderHaar, Vol.
Ill, 9/28/77 at 430-31. But Seay failed the medical examination
because she was "obese" at 140-̂ pounds, Seay, Vol. II, 4/26/77
at 230, although she would have been regarded as qualified if
she had weighed 135^ pounds, VonderHaar, Vol. Ill, 9/28/77 at
493. Richardson also was disqualified because her height and
weight "place [d] her in an obesity category with regard to
standards for employment." DX 80. Dr. VonderHaar acknowledged
that a person could lose 2 pounds by going to the bathroom, and
3 pounds overnight. VonderHaar, Vol. Ill, 9/28/77 at 490-91.
44
Indeed, when Seay and Richardson were reexamined following inter
vention on their behalf by the Urban League, Dr. VonderHaar
reported that Seay had become inches taller and 2 pounds
lighter, and that Richardson had become 1 inch taller and four
pounds lighter. DX 80. He testified that the difference of
1-g- inches in the two measurements of Seay's height was "highly
unusual," VonderHaar, Vol. Ill, 9/28/77 at 492, but he persisted
in, and the Civil Service Board followed, his view that both
Seay and Richardson should be disqualified for "obesity." DX
80; VonderHaar, Vol. Ill, 9/28/77 at 490-95. Long after the
strict height and weight requirements were eliminated from the
written job description, defendants continued to use similarly
arbitrary but more vague and subjective standards in practice.
These standards were not job-related, and they operated to dis
criminate against black applicants. See PX 81, Table 45;
VonderHaar, Vol. Ill, 9/28/77 at 431.
Thus, even after the filing of this action, the defendants .
continued to exclude black applicants through the use of "highly
subjective method[s] of selection" which were "susceptible of
abuse." Castaneda v. Partida, supra, 430 U.S. at 495 and n.14.
See also, Stewart v. General Motors C'orp. , supra, 542 F.2d at 450;
Rowe v. General Motors, supra, 457 F.2d at 358—59. They also
persisted in using selection practices which had an adverse impact
on blacks and were not job related. See pp. 40-41, supra, and
pp. 46-103, infra. Therefore, defendants have continued to violate
Title VII, §1931, and §1983 and the Fourteenth Amendment while
this lawsuit has been pending before the court.
45
I I .
rprTTr ctATUTORY RIGHTS OF DEFENDANTS VIOLAT-D ss By using TEST
PLAINTIFFS AND I5P;Ltch HAD a substantial 165.1 H A MANNER WHICH HAD A ^ DID
QC-ifications
for the j o b.
A . ^ Applicable Law
Title vil forbids the use of employment tests that ate dis-
. e££ect unless the employer meets "the but en
Minatory in . . a manifest rela
wwing that any givsn requir . _M r l . nuke Power
ionship to the employment in ques ' 422
O.. ,01 U,. «4. « » ( » n „ a_ « ^ ^ _ _ 29
•s. ,05. « 5 (1975): D ^ d ^ U ^ niscriminatory
. The same standard applies to
lasting practices under .!*».
566 F 2d at 1338-40. See p. 10. SUSS.' supra, 56b £ a
proscribe alsQ ctices
not only overt discrrminatidiscriminatory in
that are iair is'business necessity,
operation. The °ractice which operates to ex
If an employment? be shown to be related to .
elude Negroes ca^£° R a c t i c e is prohioited.job performance h e p ci 431.
nriacrs, supra., 401 u.s- . ,
v, n that a test had a racially n ..Liffa have shown than aonce the plaintiffs h ^ ^ ^ butflen on the defen-
disproportionate^impac ^ ^ the discrislina-
dant to celationship to successful performance
tion bears a demons Guardians^nc-
on the job for which it was . - ^ 1333, 2337
r - . s ^ r t Civil_Serydce_catffiAeai ’
46
(2d Cir. 1973), cert, denied, 421 U.S. 991 (1975). The employer
must demonstrate that the test classifies applicants in accordance
with their capacity to perform the job and not in accordance with
irrelevant factors such as race, test-taking ability, or general
intelligence. See EEOC v. Detroit Edison Co., 515 F.2d 301, 313
(6th Cir. 1975), vacated and remanded on other grounds, 431 U.S.
951 (1977); Firefighters. Institute for Racial Equality v. City
of St. Louis, 549 F .2d 506, 510 (3th Cir.), cert, denied, 434 U.S.
819 (1977); United States v. Citv of Chicago, 549 F.2d 415, 427
(7th Cir.), cert. denied, 434 U.S. 875 (1977); Boston Chapter,
NAACP, Inc, v. Beecher, 504 F.2d 1017, 1019-21 (1st Cir. 1974),
cert. denied, 421 U.S. 910 (1975); Castro v. Beecher, 459 F.2d
725, 732 (1st Cir. 1972).
To carry this burden, the emoloyer is required to prove
17/
that the test has been "validated" in accordance with applic
able legal and professional standards. The Equal Employment
Opportunity Commission, ,the Civil Service Commission, the Depart
ment of Labor, and the Department of Justice have recently
adopted "Uniform Guidelines on Employee. Selection Procedures
17/ "'Validation' is the term of art designating the process of
determining the job-relatedness of a selection procedure."
Kirkland v. New York State Department of Correctional Services,
374 F. Suop. 1361, 1370 (S.D.N.Y. 1974), aff'd in pertinent part,
520 F .2d 420 (2d Cir. 1975), cert, denied, 429 U.S. 974 (1976).
Similarly, plaintiffs' expert, Dr. Richard sJ Barrett., defined
"validation" as "the process by which it is determined that the
test is performing some useful function. . . . The simplest
definition is [a] demonstration that the test is measuring what
it is supposed to be measuring." Barrett, Vol. Ill, 7/13/77
at 382.
47
(1978)," 43 Fed. Reg. 38290 (Aug. 25, 1978) (hereinafter "Uniform
Guidelines"), which set forth the validation requirements of
Title VII and other federal civil rights laws as interpreted by
the issuing agencies. These new guidelines are based upon and
supersede the EEOC "Guidelines on Employee Selection Procedures",
29 C.F.R. § 1607 (D. Greenberg Affidavit, 4/20/77, Ex. L.)
(hereinafter "EEOC Guidelines"), and the Federal Executive Agency
Guidelines, 41 Fed. Reg. 51734 (1976) (D. Greenberg Affidavit,
4/20/77, Ex. M) (hereinafter "FEA Guidelines"), which were in
effect at the time of the trial of this action.
The Supreme Court has held that the EEOC's interpretation
of the test validation requirements of Title VII is "entitled to
great deference." Albemarle Paper Co. v. Moody, 422 U.S. at
430-31; Griggs v. Duke Power Co.. 401 U.S. at 433-34. The EEOC
and other agencies in the Uniform Guidelines, § 5C, and the *
Supreme Court in Albemarle, 422 U.S. at 431, have recognized
that the American Psychological Association’s "Standards for
Educational and Psychological Tests" (1974) (PX 98)(hereinafter
"APA Standards") set forth generally acceptable professional
standards for evaluating standardized tests and other selection
procedures. See also EEOC Guidelines, 29 C.F.R. § 1607.5(a);
FEA Guidelines, § 5(b). The EEOC Guidelines and the APA Standards
have often been applied by the courts. See, e.g., Albemarle,
supra. 430-35; United States v. City of Chicago, supra, 549 F.2d
at 429-34; Douglas v. Hampton, 512 F.2d 976, 986 (D.C. Cir. 1975);
United States v. Georgia Power Co.. 474 F.2d 906, 913 (5th Cir.
48
1973). Courts have also found guidance in the FEA Guidelines and
in the APA Division of Industrial-Organizational Psychology's
"Principles for the Validation and Use of Personnel Selection
Procedures" (1975) (PX 101) (hereinafter "Division 14 Principles").
See. e.g., Enslev Branch, NAACP v. Seibels, 14 FEP Cases 670, 67313/
(N.D. Ala. 1977). Dr. Richard S. Barrett, who testified as an
expert for the plaintiffs in this case, and the expert witnesses
who testified for the defendants, acknowledged that the EEOC Guide
lines, the FEA Guidelines, the APA Standards, and the Division 14
Principles provide appropriate standards for evaluating the
validity of employment tests. Barrett, Vol. Ill, 7/13/77 at 380- 19/ 20/
82; Tyler, Vol. IV, 6/23/77 at 607-608? Crosby, Vol. II, 7/12/77
18/ or . Barrett has been recognized as "a nationally known
industrial psychologist with special expertise in testing."
Bridgeport Guardians, Inc, v. Bridgeport Civil Service Commission,
supra, 354 F. Supp. 778, 790. Courts have repeatedly agreed with
Dr. Barrett's analysis of the inadequacies of various testing
procedures. See, e.g., Griggs v. Duke Power Co., 401 U.S. 424
(1971) ? Albemarle Paper Co. v. Moody, 422 U.S. 405 (1975). PX 105
lists the cases in which he has testified.
19/ Dr. Thomas A. Tyler, who testified as an expert for defendants,
is employed as the Director of Testing of the International Per
sonnel Management Association, Tyler, Vol. IV, 6/23/77 at 530-31,
which participated in the development of Test 165.1 and which has
a financial interest in the outcome of this litigation because it
derives income from renting’the test to localities for use in
selecting police officers. Id. at 532-34.
20/ Dr. Andrew Crosby, another of defendants' experts, is employed
by the International Association of Chiefs of Police, Crosby, Vol.
II, 7/12/77 at 240-41, which participated in the development of
Test 165.1 and which also has a financial interest in the outcome
of this litigation in that it will profit from the use of the
test by many jurisdictions. Id. at 260-61.
49
at 261-63; Helms, Vol. IV, 7/14/77 at 617.
2 1 /
B. As Used by Defendants, Test 165.1 Had
Substantial Adverse Impact on Black Applicants.
On January 28, 1977, the defendants administered the "Multi-
jurisdictional Police Officer Examination, Test No. 165.1" (here
inafter "Test 165.1") to more than 600 applicants for the job
of police officer. PX 33; Gavin-Wagner, Vol. II, 7/12/77 at 327-
328. Test 165.1 (PX 93) was developed by the Educational Testing
Service at the request of the International Association of Chiefs
of Police and the International Personnel Management Association.
DX 31, at 1. The test is a written examination consisting of
ten sets of 15 multiple-choice items, for a total of 150 items.
PX 93. The defendants set the passing point at 128 correct
answers out of the 150 items, Gavin-Wagner, Vol. Ill, 7/13/77
at 361-62, and the defendants gave the score on Test 165.1 a
weight of 75% and the score on an oral interview a weight of
25% to determine a ranking for each passing applicant on an
eligibility list, id. at 366-67; Priebe,'Vol. II, 9/27/77 at
248-49.
2V Dr. Wayne Helms, another expert witness for defendants,
testified that he had been employed by the defendant Louisville
Civil Service Board not only to testify in this case but also
to conduct a concurrent validity study of Test 165.1 in Louis
ville, to automate personnel functions concerning testing, and
to inquire into the use of assessment center techniques. Helms,
Vol. IV, 7/14/77 at 593-95. It now appears that he never performed
the concurrent validity study. See plaintiffs' motion to reopen
and supplement the record, 8/15/78; defendants' response to motion
to reopen'and supplement the record, 8/31/78.
50
Of the 480 whites who took the test, 374 (78%) passed; of
the 135 blacks who took the test, only 49 (36.3%) passed. PX 33.
Thus, white applicants passed at more than twice the rate of
blacks, and although nearly 22% of the applicants who took the
test were black, less than 12% of the persons who passed were
black. Only 5 of the 135 black applicants, or 3.7%, scored high
enough to be among the top 100 on the eligibility list (PX 52, 35)
and thus had a realistic chance of being appointed (see plaintiffs'
proposed finding of fact 41), while 95 of the 480 white applicants,
or 19.8%, ranked among the top 100. PX 33, 35. Although one of every
five applicants was black, only one of every twenty candidates
likely to be appointed from the eligibility list was black, id.,
and only one of twenty-nine actually appointed was black. PX 15;
PX 52. These results are summarized in the following table
showing the initial number and percentage of black applicants
and the number and percentage of black applicants remaining
after each stage of the selection process (from PX 15, 35, 52;
Priebe, Vol. I, 9/26/77 at 104-106):
Stage of Selection
Process
Number
of Black Applicants
Total
Number of
Applicants
Percentage
of Black
Applicants
Application 222 944 23.5%
After
Disqualification 207 883 23.4%
After Written Test 49 42 7 11.5%
After Oral Interview 48 401 12.0%
On Eligibility List 48 401 12.0%
In Top 100 of
Eligibility List 5 100 5.0%
Certified 2 56 3.6%
Appointed 1 29 3.4%
_ 51
These very substantial disparities are clearly sufficient
to establish a prima facie case of unlawful discrimination. See
Bridgeport Guardians. Inc, v. Bridgeport Civil Service Commission,
supra, 482 F.2d at 1335 (whites passed at 3.5 times the rate of
blacks and Hispanics); Vulcan Society v. New York City Civil
Service Commission, 360 F. Supp. 1265, 1268-69 (S.D.N.Y.), aff1d
in pertinent part, 490 F.2d 387 (2d Cir. 1973) (whites scored
high enough to be appointed at 2.3 times the rate for blacks and
Hispanics); United States v. City of Chicago, supra, 549 F.2d at
429 (7.07% of white candidates but only 2.23% of black candidates
had a practical chance of being promoted). Under the Uniform
Guidelines, a selection rate for any racial group which is less
than four-fifths (80%) of the rate for the group with the highest
rate generally is regarded as evidence of adverse impact. Uniform
2 2/Guidelines, § 4D. Here, even assuming that all of the top 100
candidates on the list were to be appointed, the overall selection
rates would be 13.2% (95/722) for white applicants but only 2.3%
(5/222) for black applicants, and the selection rates on Test
165.1 would be 63.2% for whites (427/676) but only 23.7%
(49/207) for blacks. PX 35. These disparities are far in
excess of those required by the Uniform Guidelines to establish
adverse impact.
22/ See also/FEA Guidelines, § 4b.
52
c. Defendants Have Not Demonstrated That Test
165.1 Is Manifestly Related to Performance
of the Job of a Louisville Police Officer.
There are three generally recognized strategies for evaluat
ing the validity of a test: (1) criterion-related validation,
which requires "empirical data demonstrating that the selection
procedure is predictive of or significantly correlated with
important elements of job performance," Uniform Guidelines,
§§ 5B, 14B; APA Standards at 26-28, 33-45; Division 14 Principles
23/
at 3-9; Barrett, Vol. Ill, 7/13/77 at 382-90; (2) content valida
tion, which requires "data showing that the content of the selec
tion procedure is representative of important aspects of per
formance on the job for which the candidates are to be evalatea,"
Uniform Guidelines, §§ 5B, 14C; APA Standards at 28-29, 45-46;
Division 14 Principles at 9-11; Barrett, Vol. Ill, 7/13/77 at
24/
390-404; and (3) construct validation, which requires "data
showing that the procedure measures the degree to which candi
dates have identifiable characteristics which have been determined
to be important in successful performance in the job for which
the candidates are to be evaluated," Uniform Guidelines, §§ 5B,
14D; APA Standards at 29-31, 46-48; Division 14 Principles at
9; Barrett, Vol. Ill, 7/13/77 at 390-404. "Face validity"
23/ See also, EEOC Guidelines, 29 C.F.R. § 1607.5; FEA Guidelines,
§ 12b.
24/ See also, EEOC Guidelines, 29 C.F.R. § 1607.5(a); FEA Guide
lines, § 12c.
25/ see also, EEOC Guidelines, 29 C.F.R. § 1607.5(a); FEA Guide
lines, § 12 d.
53
— i.e., the use of items in Test 165.1 which appear to relate
in some way to police work — is not an acceptable substitute
for a proper validation study: "The writing of items in terms
used in a particular job . . . may give an appearance of
relevance while contributing nothing to content validity or indeed to
any other useful validity information . . . " APA. Standards at
29; Barrett, Vol. Ill, 7/13/77 at 404-405. See United States v.
Citv of Chicaqo, 573 F.2d 416, 427 (7th Cir. 1978); Boston 26/
Chapter, NAACP, Inc, v. Beecher, supra, 504 F.2d at 1021-22.
Equally unacceptable, in the face of conflicting evidence, is
the testimony of defendants' experts that Test 165.1 is valid.
United States v. Citv of Chicaqo, supra, 549 F.2d at 434;
Vulcan Society v. Civil Service Commission, supra, 490 F.2d at
397. Similarly,
Under no circumstances will the general
reputation of a test or other selection pro
cedure, its author or its publisher, or
casual reports of its validity be accepted in
lieu of evidence of validity. Specifically
ruled out are: Assumptions of validity based
on a procedure's name or descriptive labels;
all forms of promotional literature; data
bearing on the frequency of a procedure’s usage;
testimonial statements and credentials, of sellers,
users, or consultants; and other non-empirical
26/ "A test fashioned from materials pertaining to the job (here,
from a preliminary fire fighters' manual) superficially may
seem job-related. But what is at issue is whether it demonstrably
selects people who will perform better the required on-the-job
behaviors after they have been hired and trained. The crucial
fit is not between test and job lexicon, but between the test and
job performance." Boston Chapter, NAACP, Inc, v. Beecher, supra,
504 F .2d at 1021-22 .
54
or anecdotal accounts of selection procedures
or selection outcomes. Uniform Guidelines,
§ 9A.27
Defendants here claim that Test 165.1 has been shown to be
job-related by both content and criterion-related validity
2 8/strategies. Neither claim will withstand scrutiny. Moreover,
even if Test 165.1 were shown to be valid for some uses, it
would not be permissible here: defendants have not demonstra
ted its validity for selecting police officers in Louisville;
defendants have used the test in an inappropriate and arbi
trary manner; and defendants have not availed themselves of
alternative selection procedures with equal or greater validity
and less or no adverse impact.
1. Test 165.1 has not been shown to be content
valid for use in selecting police officers.
In order to demonstrate the content validity of Test 165.1,
defendants are required to show "that the behavior(s) demon
strated in the selection procedure are a representative sample
of the behavior(s) of the job in question . . . ," or that the
test measures a representative sample of an operationally
27/ See also, EEOC Guidelines, 29 C.F.R. § 1607.8; FEA
“Guidelines, § 7.
28/ Defendants make no claim of construct validation. Indeed,
"[i]n view of the lack of a substantial literature extending
the concept of construct validity to employment practice, no
principles for its use are presented" by Division 14, beyond noting
that "obtaining support for the relevance of a construct to a particular jot, and of the validity with which a particular se
lection procedure measures that construct, is both an extensive
and arduous undertaking, involving more than a single criterion-
related validity study." Division 14 Principles at 9. See
Helms, Vol. IV, 7/14/77 at 609-610.
55
defined knowledge, skill, or ability which "is used in and
is a necessary prerequisite to performance of critical or
important work behavior(s)." Uniform Guidelines, § 14C(4).
Content validity is not an appropriate strategy for demon
strating the validity of tests which purport to measure
traits or constructs, nor is it an appropriate strategy for
tests involving knowledges, skills, or abilities which an
employee will be expected to learn after being hired. Id.,
§ 14C(1). In addition, the defendants must establish that
the test represents either a critical work behavior or work
behaviors which constitute most of the important parts of
the job. Id., § 14C(8). Defendants have not satisfied any
of these requirements, and their claim of content validity
therefore must be rejected.
a. Test 165.1 purports to measure intellec
tual constructs rather than observable
work behaviors.
The primary focus of content validation is upon whether
"the behavior(s) demonstrated in the selection procedure are
a representative sample of the behavior(s) of the job in
question . . ." Uniform Guidelines, § 14C(4)(emphasis added).
The APA Standards state that, "to demonstrate the content
validity of a set of test scores, one must show that the
behaviors demonstrated in testing constitute a representative
sample of behaviors to be exhibited in a desired performance
domain." Id. at 28 (emphasis added). As Dr. Barrett testified,
"[t]he significant words in that statement are behavior[s].
Content validity is concerned with observable behavior . . . "
56
Barrett, Vol. Ill, 7/13/77 at 391. Thus, a good typing test
has content validity for a typing job. Id. As one moves away
from such direct comparisons between test behaviors and work
behaviors, content validity becomes a more questionable strategy.
Content validation is not an empirical process but rather is
"primarily a judgmental process concerned with the adequacy of
a test as a sample of specified activities." Division 14 Prin
ciples at 9. It is for this reason that the emphasis is on
observable work behaviors or work products, and that in order
to show content validity, the gap between the test and perform
ance on the job must be a small one. See "An Overview of the
1978 Uniform Guidelines on Employee Selection Procedures," 43
Fed. Reg. 38290, 38295 at 5 8. See also, Division 14 Principles
29 /
at 10; Barrett, Vol. Ill, 7/13/77 at 391-98. As stated in
the Uniform Guidelines,
29/ A content domain should be defined "principally in terms
of activities or consequences of activities which can either
be observed or be reported by the job incumbent. One can add
to this nucleus, without straining credulity, statements of
specific items of knowledge, or specific job skills, prerequi
site to effective activity. It is a much larger 'inferential
leap,1 however, to move from observation to inferences concern
ing underlying psychological constructs such as empathy,
dominance, dexterity, leadership skill, spatial ability, etc.
Such constructs suggest hypotheses to be tested in criterion-
related or other empirical research. It is therefore inappropriate
to define job domains in such terms if one's purpose is to develop
and justify a test solely on the basis of that domain." Division
14 Principles at 10.
57
The closer the content and the context
of the selection procedure are to work samples
or work behaviors, the stronger is the basis
for showing content validity. As the content
of the selection procedure less resembles a
work behavior, ... the less likely the selection
procedure is to be content valid, and the
greater the need for other evidence of validity.
Id. § 14C(4). .
See also, R. Barrett, "Content or Construct Validity: What's
the Difference?" (1976) (PX 106).
Tests which purport to measure certain kinds of "abilities"
may be justified by content validity. Uniform Guidelines,
§ 14C(1). However, the ability being measured must be "opera
tionally defined in terms of observable aspects of work behavior
of the job"; the test must measure and be a representative
sample of the ability; and the ability must be "used in and [be]
a necessary prerequisite to performance of critical or important
30/
work behavior (s). " Id.-' § 14C(4). In addition, a test which
purports to measure such an ability "should either closely
approximate an observable work behavior, or its product should
closely approximate an observable work product." Id.
Test 165.1 does not satisfy these requirements. The
Educational Testing Service report by Michael Rosenfeld and
Richard F. Thornton, entitled "The Development and Validation of
a Multijurisdictional Police Examination" (DX 31), states that
30/ See also, FEA Guidelines, § 12c (1).
58
the test which they developed was designed to measure only
"intellectual abilities," DX 31 at 17-18, and that twelve of
these intellectual abilities were judged to be important to the
job of a police officer: verbal comprehension, serial recall,
paired associate memory, memory for relationships, memory for
ideas, semantic ordering, induction, problem sensitivity,
flexibility of closure, spatial orientation, spatial scanning,
and visualization. Id_. at 19-20. Multiple-choice items which
purported to measure each of these twelve abilities were devel-
31/
oped and included in a 180-item test. I_d. at 20-23.
Dr. Rosenfeld, one of the principal authors of the test and the
validation report, testified that the "abilities" which the
test was designed to measure did not concern an applicant's
"present observable competence to perform a function," but
rather " [t]he capacity of an individual to learn to perform, and
to perform certain activities." Rosenfeld Dep., 5/26/77 at 151-
52. Thus, the "abilities" which Test 165.1 purports to measure
are not the operationally defined abilities for which content
validation may be appropriate under the Uniform Guidelines,
32/
§ 14C(1) and (4);• indeed, the Uniform Guidelines specifically
31/ Test 165.1, which was used in Louisville, had only 150 items
and did not include any items purporting to measure "flexibility
of closure" or "serial recall." Tyler, Vol. IV, 6/23/77 at 515-16.
32/ See also, FEA Guidelines, § 12c(1).
59
define "ability" as " [a] present competence to perform an
observable behavior or a behavior which results in an observable
33/
product." Id., § 16A. Rather, they are "constructs"— i.e.,
ideas "developed or constructed as a work of informed, scientific
imagination."- APA Standards at 29. See also, Division 14 Prin
ciples at 16. As such, they are not the appropriate subject of
content validation. Barrett, Vol. Ill, 7/13/77 at 413-21.
Although defendants have argued that the "intellectual
abilities" which Test 165.1 purports to measure differ in some
way from constructs, their experts were unable to articulate
any such difference. Terry Talbert, while maintaining that
these "abilities" were not constructs, testified that "visualiza
tion is an ability area that may or may not be defined as a
construct, but I really couldn't say," Talbert, Vol. V, 6/24/77
at 732, and that both the "semantic ordering" items and the
"problem sensitivity" items on Test 165.1 were matters of "common
sense." Id. at 718, 722. Dr. Andrew Crosby, another of defend
ants' expert witnesses and a research scientist with the
International Association of Chiefs of Police who participated in
the development of Test 165.1, attempted to explain the difference
as follows (Crosby, Vol. II, 7/12/77 at 268):
A. Well, as I said, one way of possibly dis
tinguishing is the ’— is of making visible
33/ See also, FEA Guidelines, § 14(a).
60
or exhibiting the ability of a person's
construct. That is like with a known,
that is the only ... [ellipsis in original]
Q. That is the only difference you can think
of?
A. That's the clearest way to describe the
difference I can think of.
Q. Can you think of any other way to describe
the difference?
A. No, not at the moment.
Dr. Crosby had difficulty describing the difference
because there is no difference. Visualization, semantic order
ing, problem sensitivity, spatial orientation, spatial scanning,
and the other "intellectual abilities" which Test 165.1 is
alleged to measure are not "operationally defined in terms of
observable aspects, of work behavior of the job." Uniform Guide
lines, § 14C(4). Instead, they are theoretical concepts based
on inferences about what "goes on inside the mind." Crosby,
Vol. II, 7/12/77 at 278-79, 302-303; Barrett, Vol. Ill, 7/13/77
at 394-97, 413-19. The Uniform Guidelines expressly identify
certain of the "intellectual abilities" purportedly measured by
Test 165.1 as examples of constructs which are not the appropriate
subject of a content validation strategy;
A selection procedure based upon inferences
about mental processes cannot be supported
solely or primarily on the basis of content
validity. Thus, a content strategy is not appro
priate for demonstrating the validity of selection
61 “
procedures which purport to measure traits or
constructs such as intelligence, aptitude, per
sonality, common sense, judgment, leadership
and spatial ability. Uniform Guidelines,
§ 14C (1) . M /
Such procedures must instead be empirically validated by a
criterion-related or construct validity strategy. Division 14
Principles at 10; Barrett, Vol. Ill, 7/13/77 at 414.
b. Test 165.1 is not a sample or approxi
mation of job behavior but merely a
verbal representation of some parts of
a highly physical and personal job.
As indicated above, a demonstration of content validity
entails a showing "that the behavior(s) demonstrated in the
selection procedure are a representative sample of the behavior(s)
of the job in question. ..." Uniform Guidelines, § 14C(4).
Therefore, to be content valid, "a selection procedure measuring
a skill or ability should either closely approximate an observable
work behavior, or its product should closely approximate an
35/
observable work product." Id_. Dr. Barrett testified to the
same effect. Barrett, Vol. Ill, 7/13/77 at 390-98. See also,
R. Barrett, "Content or Construct Validity: What's the Differ
ence?" (1976) (PX 106).
It is clear that the items on Test 165.1 are not samples
34/ See also, FEA Guidelines, § 12c(l).
35/ See also, FEA Guidelines, § 12c(4).
62
or approximations of behaviors which police officers would
perform in the course of their duties. No officer spends his
or her time on the job sitting at a table and choosing among
four possible synonyms for the word "apprehended" (DX 31, at
108; see Test 165.1, PX 93, items 1-15), or answering multiple-
choice questions about whether forms were properly filled out
(Test 165.1, PX 93, items 121-135), or choosing the common
theme in a series of pictures (Test 165.1, PX 93, items 106-
120). As Dr. Wayne Helms, one of defendants' experts, admitted,
there are no items on Test 165.1 which constitute a representa
tive sample of the job behaviors, skills, and prerequisite job
knowledges required in performing the duties of a Louisville
police officer. Heims, Vol. IV, 7/14/77 at 622-23. Indeed,
there are "not intended to be [any such items]. It is not a job
sample test. It is a test of abilities. ..." Id_. Dr. Thomas
Tyler, another expert witness for defendants, agreed that the
test items were not samples of the job. Tyler, Vol. IV,
6/23/77 at 650-53.
As the Chief of Police testified, the job of a Louisville
police officer is physically demanding, Nevin, Vol. Ill,
6/22/77 at 450, and "most of the work that he does involves
very difficult and discriminating judgments concerning social
issues. ..." Ijd. at 451. The developers of Test 165.1 also
recognized the clear importance to the job of attitudes,
63
personality dimensions, and physical performance, and they
acknowledged that their test did not measure these aspects of
the job. DX 31, at 17-18. But the written multiple-choice
items on Test 165.1 place a premium on reading and verbal
skills rather than providing adequate representations of the
highly physical and personal job of a police officer. As the
Second Circuit stated concerning a written test for firefighters,
The questions deal with fire fighting,
yet there is a difference between memorizing
(or absorbing through past experience) fire
fighting terminology and being a good fire
fighter. If the Boston Red Sox recruited
players on the basis of their knowledge of
baseball history and vocabulary the team
might acquire authorities like the late John
Kiernan but no one who could bat, pitch or
catch. The test does not examine traits seem
ingly more relevant to a fire fighter's perform
ance such as agility, stamina, quick thinking
under pressure, poise, mechanical aptitude and
the ability to work with others. Experts for
both sides agreed that verbal memory is not a
very important attribute for the job. And
unlike the motor vehicle rules covered in a
driver's test, it seems unessential whether
the candidate absorbs the tested vocabulary
before or after acceptance. Nomenclature and
similar matters can be mastered during training
and on the job. Testing them before acceptance
puts a premium on ability to memorize terms that,
at the time, contain only abstract meaning.
Boston Chapter, NAACP, Inc, v. Beecher, supra,
504 F .2d at 1023.
In the case of Test 165.1, the inherent shortcomings of
verbal tests for action-oriented jobs are further emphasized by
the use of a pre-examination booklet. DX 33. Candidates are
64
required to study and memorize materials in the booklet prior
36/
to the administration of the test, which contains many items
concerning the details of those materials. Without the mater
ials in the pre-examination booklet, approximately half of the
questions in the test could not be answered correctly. DX 31,
at 52. The use of this booklet makes Test 165.1 "very much a
test of an academic nature; that is ... it is a test where the
people go in and read the book, remember what is in it, and be
able to answer questions about it." Barrett, Vol. Ill, 7/13/77
at 424.
Dr. Barrett's testimony as to the inappropriate academic
nature of the test was confirmed by two analyses performed by
the defendants. One demonstrated that applicants with education
beyond the level required for the job were more likely to suc
ceed on the test than applicants with a high school' diploma or
G.E.D. DX 58; Gavin-Wagner, Vol. II, 7/12/77 at 346-48. The,
i
other showed that scores on Test 165.1 and the oral interview
were not significantly correlated with overall training scores
36/ The test developers state that the booklet "should be
distributed sufficiently in advance of the test (at least
three weeks) to provide adequate opportunity for preparation."
DX 31 at 52. In Louisville, the booklet was distributed two
weeks prior to administration of the test. Gavin-Wagner,
Vol. II, 7/12/77 at 327-28.
65
butin the police academy (Helms Dep., 11/4/77 at 24-27),
that there was a significant positive correlation between
scores on Test 165.1 and scores at the end of police training
on the "KLEC" examination, another academically-oriented written
test. Id. at 6-9, 21-23. In other words, applicants who had the
reading and verbal skills to obtain high scores on Test 165.1
tended to receive high scores again on another written test
administered at the end of their police recruit training, but
their overall training scores in their academy courses were not
significantly different from those of applicants who did not
perform as well on Test 165.1. Both of these analyses indicate
that Test 165.1, as used by the defendants, tended to select
applicants on the basis of verbal and academic skills which were
beyond the level required for successful performance as a police
officer.
c. Test 165.1 involves knowledges, skills,
and abilities which,new police officers
are expected to learn in recruit' school
or on the job.
Plaintiffs1 and defendants' experts agreed that content
validity is not an appropriate strategy for demonstrating the
37/
37/ Dr. Wayne Helms, who performed this analysis, made this
computation only on the basis of combined scores on Test 165.1
and the oral interview. Helms Dep., 11/4/77 at 26-27. The
score on Test 165.1 accounted for 75% of the combined score.
Id. at 7-8.
66
validity of a test of knowledges, skills, or abilities which
an employee will be expected to learn after being hired.
Barrett, Vol. Ill, 7/13/77 at 421-23; Crosby, Vol. II, 7/12/77
at 287. Their testimony is fully supported by the applicable
legal and professional standards: Content validity may justify
a test of an operationally defined knowledge, skill, or ability
which is "a necessary prerequisite to successful job performance";
content validity is "not an appropriate strategy when the
selection procedure involves knowledges, skills, or abilities
which an employee will be expected to learn on the job."
Uniform Guidelines, § 14C(1) (emhasis added). See Division 14
38/Principles at 10. More generally, test users "should avoid
making employment decisions on the basis of measures of know
ledges, skills, or abilities which are normally learned in a
brief orientation period, and which have an adverse impact."
3 9/
Uniform Guidelines, § 5F.
i
One of the defendants' experts. Dr. Wayne Helms, admitted
that "some of the items [on Test 165,1] ... at least on the face"
test for knowledge, behaviors, and 'skills which are taught to
new employees in the police recruit school and learned through
experience on the job. Helms, Vol. IV, 7/14/77 at 626. Col. John
38/ See also, EEOC Guidelines, 29 C.F.R. § 1607.5(a); FEA
Guidelines, § 12c(l).
39 / See also, FEA Guidelines, § 5e.
67
Nevin, the Chief of Police and a veteran of almost 20 years on
the force, testified that there were courses in the recruit
training school which were designed to teach new officers vir
tually all of the subjects covered by Test 165.1: vocabulary
used in police work (see Test 165.1, PX 93, items 1-15), city
geography (see items 16-30), identification of suspects by
40/
pictures (see items 31-45 and 136-150), patrol procedures
(see items 46-60 and 61-75), filling out automobile accident
reports (see items 76-90), and writing various reports and forms
(see items 121-135) are all taught to new recruits after they
have been hired. Nevin, Vol. IV, 5/23/77 at 506-510. Chiaf
Nevin found only two subjects on Test 165.1 which were not
taught in specific courses in recruit school: (1) making judg
ments as to which of two actions an officer should take first in
a given situation (see items 91-105), which new officers learn
in part in recruit school "and they learn the biggest part out
on the street after they get out and making decisions with their
partner ...," id_. at 508-509; and (2) choosing the common theme
in a series of'three cartoon pictures (see items 106-120), which
is not related to anything a police officer does either in recruit
40/ Chief Nevin testified at one point that he did not know
whether there was specific training in recruit school as to
how to identify suspects from wanted posters (Nevin, Vol. IV,
6/23/77 at 510), but he also testified that there are courses
in recruit school regarding.the identification of suspects by
pictures (id_. at 507) .
68
school or on the job. Id. at 509.
Thus, even if the items on Test 165.1 were a representa
tive sample of job behaviors rather than verbal formulations
designed to measure certain constructs, content validity would
not be an appropriate strategy because the test involves know
ledge, skills, and abilities which the new officers will be
expected to learn in recruit school and on the job. Barrett,
Vol. Ill, 7/13/77 at 421-23. As the court said in Boston
Chapter, NAACP, Inc, v. Beecher, supra,
it seems unessential whether the candidate
absorbs the tasted vocabulary [and other
materials] before or after acceptance. Nomen
clature and similar matters can be mastered
during training and on the job. Testing them
before acceptance puts a premium on ability
to memorize terms that, at the time, contain
only abstract meaning. 504 F.2d at 1023.
Chief Nevin himself recognized precisely the same point (Nevin,
Vol. IV, 6/23/77 at 486):
I don't think it's necessary to test an
applicant for knowledge of specific police
techniques any more than it would be fair to
test an applicant for law school on knowledge
that' the law school is going to provide them.
d. Test 165.1 represents neither a critical
work behavior nor work behaviors which
constitute most of the important parts
of the job.
A test which is supported on the basis of content validity
"may. be used for a job if it represents a critical work behavior
(i.e., a behavior which is necessary for performance of the job)
69
3
or work behaviors which constitute most of the important parts
of the job." Uniform Guidelines, § 14C(8). The APA
Standards similarly provide that an employer "cannot justify
an employment test on grounds of content validity if he cannot
demonstrate that the content universe includes all, or nearly
all, important parts of the job." APA Standards at 29. See
also, Division 14 Principles at 10..
As demonstrated above. Test 165.1 does not represent any
critical work behavior of a Louisville police officer; indeed,
it is not a sample of any job behaviors at all, but merely
purports to measure certain "intellectual abilities" which are
claimed to be related to the job. (See pp. 56-62, supra.)
Moreover, Test 155.1 does not even purport to measure most of
the important parts of the job. As the developers of the test
candidly admitted,
The focus of this project was the develop
ment of a test measuring intellectual abilities.
Specifically excluded were affective measures
which assess attitudes and personality dimen
sions, and psychomotor measures which refer to
indices of physical performance. The exclusion
of affective and psychomotor dimensions should
not be interpreted to mean that these dimensions
are regarded as unimportant. On the contrary,
their importance is clearly recognized. DX 31,
at 17.
41/ See also, EEOC Guidelines, 29 C.F.R. § 1607.5(a); FEA
Guidelines, § 12c(3).
70
Defendants' experts agreed that Test 165.1 does not
include all or nearly all of the important parts of a police
officer's job but is "limited to the intellectual or cognitive
domain." Rosenfeld Dep., 5/26/77 at 35. See also, Crosby,
Vol. II, 7/12/77 at 281-82; Helms, Vol. IV, 7/14/77 at 621.
According to one of defendants' experts, Test 165.1, the 150-
item test which was given in Louisville, does not even include
two of the twelve intellectual abilities which the test develop
ers regarded as "necessary" (DX 31, at 19-20) for learning and
performing the tasks of a police officer. Crosby, Vol. II,
7/12/77 at 284-86. Moreover, the intellectual abilities which
are included are arbitrarily assumed to be of precisely equal
importance to the job. The test consists of ten equally weighted
sections of fifteen equally weighted items, each section pur
porting to measure an intellectual ability; "problem sensitivity"
and "semantic ordering," for example, are given equal weight on
the test without any consideration of whether they are equallv
important on the job. As Dr. Crosby testified, "There is no
deliberate attempt to weight parts of the job." _Id. at 280-81.
Thus, there is no indication that the 'test is a representative
sample of the job based on "an analysis of the important work
behavior(s) required for successful performance and their rela
tive importance. ..." Uniform Guidelines, § 14C(2). See also,
APA Standards, 5 E12.4 at 46.
71
Even if Test 165.1 were a proper measure of relevant
intellectual abilities, it would not be measuring the most
important parts of a police officer's job. According to
defendants' expert Terry Talbert, the test does not even argu
ably measure the ability to communicate effectively under
stress, to make oral presentations, to write clearly and con
cisely, to operate an automobile safely and in accordance
with traffic laws, and to be firm but courteous in treatment
of the public. Talbert, Vol. V, 6/24/77 at 776-80. All of
these abilities have been identified by the defendants as
important to the job of a Louisville police officer. Olgas,
Vol. II, 7/12/77 at 204, 215-17; DX 38, "Weighting Forms";
PX 100. The test clearly does not measure listening ability,
energy, stress tolerance, risk-taking, tenacity, physical
condition, human relations, salesmanship, initiative, leadership,
and other factors which defendants have found to be "critical"
to the job. Olges, Vol. II, 7/12/77 at 205-206, 216-17; DX 33,
"Desirable Knowledges, Abilities & Skills— Corresponding Dimen
sions." As the Chief of Police testified, the job is highly
physical, and "modt of the work ... involves very difficult and
discriminating judgments concerning social issues. ..." Nevin,
Vol. Ill, 6/22/77 at 450-51. Because the test does not include
these crucial aspects of the job, it is defective in the same
72
manner as the hypothetical test for airline pilots described
by defendants' expert Talbert:
For instance, where we test people to be
airline pilots, we would probably be falling
short, if you will forgive the pun, if we
tested them only on their ability to take the
plane off. If we never tested them on their
ability to land the plane, we might be in
trouble. ... Talbert, Vol. V, 6/24/77 at 774.
The courts have repeatedly held that tests are not content
valid where, as in the case of Test 155.1, they fail to test or
improperly weight important parts of the job. United States v.
City of Chicago, 573 F.2d 416, 425-26 (7th Cir. 1978); Fire
fighters Institute for Racial Equality v. City of St. Louis,
supra, 549 F.2d at 511-12; United States v. City of Chicago,
supra, 549 F.2d at 434; Kirkland v. New York State Department
of Correctional Services, supra, 374 F. Supp. at 1378; Vulcan
Society v. New York City Civil Service Commission, supra, 360
F. Supp. at 1274.
Defendants have suggested that this flaw in Test 165.1 is
not fatal to their claim of validity because some of the aspects
of the job which are not reflected in the test are included in
subsequent stages of the selection procedure, such as the oral
interview and the medical examination. See Gavin-Wagner, Vol.
II, 7/12/77 at 341-46. However, this argument ignores the manner
in which defendants used the test in selecting among the more
than 900 applicants for the job of police officer. After initial
73
disqualifications for failure to satisfy minimum requirements
— which did not, in the period from November 1976 to January
1977, have an adverse impact on black applicants— the very
first step in the selection process was the administration of
Test 165.1 and the resulting elimination from further consider
ation of 158 blacks who received scores below the passing point
set by defendants. Priebe Dep., 2/17/77 at 15-21; PX 35. The
test also excluded from any realistic chance for appointment
another 43 blacks who received scores which were too low to
place them in the top 100 of the eligibility list. PX 35. Thus,
201 of the 207 blacks who took the test were effectively screened
out before they ever reached the oral interview, the medical
42/
examination, or any other stage of the selection procedure.
"The many who are not picked ... obviously have no opportunity
to compete or to raise their rank by a demonstration of their
ability." Firefighters Institute for Racial Equality v. City
of St. Louis, supra, 549 F.2d at 512-13. Accordingly, Test
165.1, as used by defendants, cannot be justified on the basis
of content validity.
42/ This use of Test 165.1 is especially questionable in view
of the testimony of defendants' expert Talbert, based on his
study of a number of police selection procedures in the Atlanta
area, that a physical fitness test was superior to a written
test, ' level of education, age, experience, and all other cri
teria studied in predicting supervisory ratings. Talbert,
Vol. V, 6/24/77 at 701-704, 763-65.
74
2. Test 165.1 has not been shown to have
criterion-related validity for use in
selecting police officers.
Criterion-related validation requires a demonstration by
empirical data "that the selection procedure is predictive of
or significantly correlated with important elements of job
performance." Uniform Guidelines, § 5B. This demonstration
may be in the form of a predictive validity study or, in
appropriate circumstances, a concurrent validity study. In
Vulcan Society v. New York City Civil Service Commission, supra,
the court described these two methods of criterion-related
validation as follows:
Predictive validation consists of a
comparison between the examination scores
and the subsequent job performance of those
applicants who are hired. If there is a
sufficient correlation between test scores
and job performance, the examination is
considered to be a valid or job-related
one. Concurrent validation requires the
administration of the examination to a
group of current employees and a comparison
between their relative scores and relative
performance on the job. 360 F. Supp. at 1273.
Defendants contend that Test 165.1 has been shown to be
valid by the concurrent criterion-related validity study which
is described in M. Rosenfeld and R. Thornton, "The Development
and Validation of a Multijurisdictional. Police Examination,"
43/DX 31. In this study, a 180-item test was administered to
44/
samples of incumbent police officers at four sites after
43/ Test 165.1, which consists of 150 items, was not the subject of
this study, although items which were later included in Test 165.1
were used in the test which was studied. Rosenfeld Dep., 5/26/77 at
57, 124-25.
44/ The four sites are identified in the deposition of Dr. Michael
Rosenfile, 5/27/77 at 128. Louisville was not involved in the study.
75
those officers had been rated for job performance by some of
their supervisors. The results were then statistically
analyzed to determine whether there were significant correla
tions between the test scores and the supervisory ratings. DX
31 at 30-45. This study does not demonstrate the validity of
Test 165.1: concurrent validation is not an appropriate
strategy for a test of knowledge, skills, or abilities which
employees learn on the job, see Uniform Guidelines, § 14B(4);
the sample subjects used in the study were not representative
of the actual applicants for the job, id; there are indications
of differential validity and there has been no adequate investiga
tion of test fairness, id. § 14B(8); and the study discloses an
"odd patchwork of results," Albemarle Paper Co. v. Moody, supra,
422 U.S. at 432. A predictive validity study of the relation
ship between applicants' scores on Test 165.1 and their sub
sequent performance as Louisville police officers is both feasible
and necessary to demonstrate the validity of this test. See
PP- 93-95, infra.
a. Test 165.1 involves knowledges, skills,
and abilities which incumbent police
officers have learned in recruit
school or on the job.
In any criterion-related validity study, predictive or con
current, the sample subjects should be representative of the
candidates normally available in the relevant labor market for
the job in question. Uniform Guidelines, § 14B(4). See pp. 79-82.
76
infra. In a concurrent validity study, "the user should take
into account the extent to which the specific knowledges or
skills which are the primary focus of the test are those which
employees learn on the job," Uniform Guidelines, § 14b (4), and
"the psychologist should be particularly aware of . . . [the]
effect of learning on the job . . .," Division 14 Principles at
5-6. See also APA Standards, 5 E6.1.1, at 37. Moreover, regard
less of the nature of the validity study, "[i]n general, users
should avoid making employment decisions on the basis of measures
of knowledges, skills, or abilities which are normally learned
in a brief orientation period, and which have an adverse impact."
45/
Uniform Guidelines, § 5F. See pp. 66-6 9, supra..
The effect of on-the-job training and experience is a par
ticularly significant concern in a concurrent validity study
because the sample subjects— whose performance provides the
empirical data for the study — are incumbent employees who have
been trained and have gained experience on the job. A concurrent
validity strategy may be appropriate for a selection criterion
such as height, which does not change after an employee is placed
on the job; but it is decidedly inappropriate for validating a
test which involves knowledges, skills, or abilities which are
learned after an employee is hired. In the latter situation, a
concurrent validity study could show whether there is a relation
ship between the test scores of experienced employees and their
performance on the job, but it would not show whether the test is
valid or useful in determining which untrained applicants.will be
45/ See also, FEA Guidelines, § 5e.
77
better able to learn to perform the necessary duties in a
training program and on the job. Barrett, Vol. Ill, 1/1.2/11
at 384-85, 426-27. As the developers of Test 165.1 recognized
in their report,
changes in employee ability and performance
which have occurred as a result of training,
education and job experience mean that the
abilities of incumbents may not be comparable
in many ways to those of applicants. In
general, the longer the time between the test
and the criterion, the lower the correlation.
DX 31 at 29.
The record in this case contains ample evidence that Test
165.1 involves knowledges, skills, and abilities which officers
are taught in recruit school and learn on the job. See pp. 67-69,
supra. Yet, despite the test developers' forthright acknowledg
ment of the resulting problem, DX 31 at 29, they failed to
include in their validation report the information required by
the APA Standards, «[ E7.4.4, at 39, as to the extent of training
and the length of experience of the incumbent police officers who
4 f i / ~
were the subjects of their concurrent study. Without this
46/ The report states, however, that the "[l]ength of service
of police officers participating in the concurrent validity
studies was significantly correlated with ratings on certain
criterion dimensions in all sites." DX 31 at 35. This statement
suggests that, to the extent test scores were correlated with
performance ratings, the concurrent validity study may merely
show that incumbents who had acquired more knowledge in their
training and experience tended to get higher scores on the test,
without showing anything about the validity of the test for
inexperienced applicants. Since the report does not disclose
data on the length of service of each incumbent in the samples,
the quoted statement can be neither confirmed nor refuted.
78
information, and in the absence of any convincing evidence that
training and experience are unrelated to test performance,
the fact that a correlation exists between
experienced individuals1 performance on
the test and their job performance does
not permit a leap to the conclusion that
inexperienced individuals1 performance on
the test . . . will correlate with their
ability to perform the job. . . . [F]or
purposes of this test the sample of ex
perienced workers is so radically different
from inexperienced applicants that conclu
sions based on one sample's experience cannot
be properly attributed to the other.
* * *
This court, therefore, cannot assume that
because there is a relationship between test
performance and current job performance for
persons with job experience that a similar
relationship obtains for applicants without
job experience. League of United Latin
American Citizens v. City of Santa Ana, supra,
410 F. Supp. at 903-904 (emphasis in original).
See also, Albemarle Paper Go. v. Moody, supra, 422 U.S. at 434-35.
b. The sample subjects were not representa
tive of actual applicants for the job.
iDifferences in training and experience are not'the only
factors which bear on the appropriateness of the sample in a
criterion-related validity study. In addition, "the sample sub
jects should insofar as feasible be representative of the candi
dates normally available in the relevant labor market for the
jobs . . . in question, and should insofar as feasible include
the races, sexes and ethnic groups normally available in the
79
The APArelevant job market." Uniform Guidelines, § 14b (4).
Standards state that it is "essential" for evidence of validity
to be obtained "for subjects who are of the same age or in the
same educational or vocational situation as the persons for whom
the test is recommended." APA Standards, J E6.1.1, at 37. See
also, Division 14 Principles at 6. The Supreme Court relied on
the APA Standards in holding that a concurrent validity study
was "materially defective" where it "dealt only with job-experienced,
white workers . . . [while] the tests themselves are given to
new job applicants, who are younger, largely inexperienced, and in
many instances nonwhite." Albemarle Paper Co. v. Moodv, supra,
422 U.S. at 435.
As defendants' expert Thomas Tyler admitted, the sample sub
jects in the concurrent validity study in this case were essentially
different from the applicant population which took Test 165.1 in
Louisville. Tyler, Vol. IV, 6/23/77 at 656. First, as demon
strated above, there were substantial and significant differences
in training, job experience, and vocational situation between
the incumbent police officers in the samples and the inexperiencedi
applicants to whom Test 165.1 was administered in practice. See
pp. 67-69, supra. Second, although the test developers failed
to report the age of the incumbents in the samples as well as
their length of service, it can be inferred that these 'experienced
officers were substantially older than applicants for the job in
Louisville, who were only required to be 21 years of age. Priebe
Dep., 2/17/77 at 11. See Tyler, Vol. IV, 6/23/77 at 656. Third,
_4/
47/ See also, EEOC Guidelines, 29 CFR § 1607.5(b)(1); FEA
Guidelines, § 12b(4).
80
although 20 percent of the Louisville applicants from November
1976 through January 1977 were women, Priebe Dep., 2/17/77 at
80—81, not one woman was included in any of the concurrent validity
samples. DX 31 at 31.
Fourth, the incumbents in the samples were not even Louisville
police officers, but were officers in departments in four other
cities which were not demonstrably comparable to Louisville in
size, geographic region, racial composition, or any other relevant
48/
characteristic. DX 31 at 31. Rosenfeld Dep., 5/26/77 at 128-30.
Fifth, as the test developers acknowledged, DX 31 at 28-29, there
was a significant preselection problem because the sample subjects
had already demonstrated their test-taking ability. It is dif
ficult to draw meaningful conclusions from such a sample about
the relationship between test scores and job performance of those
who were excluded. Barrett, Vol. Ill, 7/13/77 at 385. See Boston
Chapter, NA&CP, Inc, v. Beecher, supra, 504 F.2d at 1025-26;
League of United Latin American Citizens v. City of Santa Ana,
supra, 410 F. S'upp. at 903-904. Sixth, there was a significant
difference between the motivation of the sample subjects, whose
43/ These sites are identified in the deposition of Dr. Michael
Rosenfeld, supra, at 128, under a stipulation of .confidentiality'.
Rosenfeld testified that he did not know how many officers were
on the S ite 1 force; that he did not know, the exact number of
officers on the Site 2 force but "Suspect [ed] several hundred,"
and that he did not know how many or these were black and how many
were white; that he did not know how many officers were on the
Site 3 force; and that several hundred or several thousand officers
were on the Site 4 force, and that he did not know what proportion
of these officers was black. Id. at 128-29.
scores on the test would have no effect on their jobs, and
actual applicants, whose test scores would determine whether
they would become police officers. Barrett, Vol. Ill, 7/13/77
at 428. See APA Standards, 5 E6.1.2, at 37; Division 14
Principles at 6.
Finally, while more than 23 percent of the applicants for
the job of police officer in Louisville were black, PX 35, the
samples in the two sites on which the defendants and the test
developers based their claim of concurrent validity — Sites 1 and
3, see DX 31 at 46-47 — contained, respectively, 137 whites, 1
"other," and 10 blacks (Site 1), and 57 whites, 26 Hispanics, 1
"other," and no blacks (Site 3). See DX 31 at 31. Thus, black
incumbents made up only 4.3% of this total sample, and blacks
and Hispanics together constituted only 15.5% of this sample.
In view of the severe adverse impact which defendants' use of
Test 165.1 had on black applicants, the virtual exclusion of
blacks from the samples on which defendants rely is fatal to
their claim of'Validity. See Albemarle Paper Go. v. Moody, suora.
422 U.S. at 435; Boston Chapter, NAACP, Inc, v. Beecher, supra,
504 F .2d at 1026.
c. Test 165.1 has not been shown to measure
fairly any differences in the job
performance of blacks and whites.
"When members of one race, sex or ethnic group character
istically obtain lower scores on a selection procedure than
members of another group, and the differences in scores are not
reflected in differences in measures of job performance, use of
82
the selection procedure may unfairly deny opportunities to mem
bers of the group that obtains the lower scores." Uniform Guide
lines, § 14B(8)(a). Where, as here, a test has an adverse impact
on a racial group which is a significant factor in the relevant
labor market,
The user generally should investigate the
possible existence of unfairness for that
group if it is technically feasible to do
so. The greater the severity of the adverse
impact on a group, the greater the need to
investigate the possible existence of unfair
ness. Id., § 14B(8)(b).
See also APA Standards, 5 E9, at 43-44. If unfairness is demon
strated by evidence that the job performance of members of a
particular group is better or worse than their test scores would
indicate in comparison to the job performance and test scores of
members of another group,
The user may either revise or replace the
selection instrument in accordance with
these Guidelines, or may continue to use
the selection instrument operationally
with appropriate revisions in, its use to
assure compatibility between the prob
ability of successful job performance and
the probability of being selected. Uniform
Guidelines, § 14B(8)(d). 49/
In this case, there are substantial grounds to question
whether the criterion measures against which the test scores are
alleged to be validated reflect any real differences in job per
formance. The merit of any criterion-related validity study
"depends bn the appropriateness and quality of the criterion
chosen. . . . " United States v. City of Chicago, supra, 549 F.2d
49/ See also, EEOC Guidelines, 29 C.F.R. § 1607.5(b)(5); FEA
Guidelines, § 12b(7).
83
at 431, quoting APA Standards at 27. The criteria chosen here
were supervisory ratings of job performance, DX 31 at 33, which
are a matter of particular concern because of "the possibility
of bias in subjective evaluations." Uniform Guidelines, § 14B(2)
In fact, in this validity study there were "extreme criterion
problems in several of the sites." Rosenfeld Dep., 5/26/77 at 133
34. Dr. Rosenfeld, who conducted the study, testified that the
supervisory ratings were affected by racial and ethnic bias at
three of the four sites studied id. at 49, and that appropriate
steps to eliminate this bias were not taken, id. at 137-38. The
only site where he found no rater bias was the site with the
smallest number and proportion of minority officers — Site 1,
where only 10 of 148 officers in the sample were black. id; DX
31 at 31, 41. As Dr. Rosenfeld stated,
In evaluating a test, if you discover
that the criterion measure against which
you're evaluating the test has problems
associated with it, it makes it extremely
difficult to evaluate the test, because' the
measure against which you're relating it is
inappropriate to use in some respects.
Rosenfeld Dep., 5/26/77 at 136.
Under these circumstances, there is "simply no way to determine
whether the criteria actually considered were sufficiently re
lated to the [defendants’] legitimate interest in job-specific
ability to justify a testing system with a racially discriminatory
impact." Albemarle Paper Co. v. Moody, supra, 422 U.S. at 433
(emphasis in original).
50/
50/ See also, EEOC Guidelines, 29 C.F.R. § 1607.5(b)(4); FEA
Guidelines, § 12b(2).
84
However, even assuming that the supervisory ratings measured
some real differences in job performance, the adverse impact of
Test 165.1 is so extreme that defendants are required to investigate
test unfairness if it is technically feasible to do so. See
51/Uniform Guidelines, § 14B(8)(b). This requirement of "differential
validation" for minority and nonminority candidates has been re
peatedly approved and applied by the courts. See Albemarle Paper
Co. v. Moody, supra, 422 U.S. at 435; United States v. City of
Chicago, supra, 549 F.2d at 430; Rogers v. International Paper Co.,
510 F.2d 1340, 1350 (8th Cir.), vacated and remanded on other
grounds, 423 U.S. 809 (1975); United States v. Georgia Power Co.,
474 F .2d 906, 913-14 (5th Cir. 1973). If the required investiga
tion discloses that lower test scores for black candidates cor
respond to higher test scores for white candidates in predicted
job performance, then Title VII requires that "employers, to
ensure equal opportunity, have to adopt race-conscious hiring
practices" to take.account of the differential validity of the
test. Regents of the University of California v. Bakke, 46
U.S.L.W. 4896, 4921 n.37 (opinion of Brennan, White, Marshall,
and Blackmun, JJ.). An employer who uses such a test is required
to select lower-scoring black candidates over higher-scoring
whites because, in these circumstances, the black candidates
actually have a higher prediction of job success. _Id. Title VII
requires employers "to use test criteria that fairly reflect the
qualifications of minority applicants vis-a-vis nonminority
51/ See also, EEOC Guidelines, 29 C.F.R. § 1607.5(b)(5); FEA
Guidelines, § 12b(7).
85
applicants, even if this means interpreting the qualifications
52/
of an applicant in light of his race." Id. at 4921.
Dr. Rosenfeld testified that he did not feel the samples in
the four concurrent validity sites were sufficiently large for a
differential validity study. Rosenfeld Dep., 5/26/77 at 130.
When asked how these samples were selected, he replied that it
was through a contact person at each site, and that "we tried to
indicate that we wanted as good a racial mix as we could get in
the department, and we tried to get as close as we could to a
hundred or more candidates in each site; and any more we could
gat above a hundred was, was very acceptable to us." Id. at 131.
There was no attempt to include specific numbers of minority
officers in the samples for the purpose of conducting a dif-
ferntial validity study. Id.- 129-30. The reasons given for
this failure to inquire into the question of test fairness —
particularly when they are offered on behalf of a major test
developer such as the Educational Testing Service, with an annual
budget of approximately seventy-five million dollars at its
disposal as well as the resources of the International Personnel
Management Association and the International Association of Chiefs
of Police, see Rosenfeld Dep., id. at 3-7, 11-12 and Ex. 1-3, 3
— are not sufficient to demonstrate that such an inquiry was
52/ m his opinion in the Bakke case, Justice Powell also stated
that, at least in academic selection procedures, a racial
classification which counter-balances inaccuracies in predicting
performance may be an appropriate way to make a "fair appraisal
of each individual's academic promise in the light of some
cultural bias in grading or testing procedures." 46 U.S.L.W.
at 4906.
86
technically infeasible" within the meaning of the applicable
guidelines and court decisions. Uniform Guidelines, §§ 14B(8)(e),
j y16U. See Rogers v. International Paper Co., supra, 510 F.2d at
1350. Indeed, because fairness studies generally require sub
stantial numbers of employees in the jobs being studied, the
primary obligation to conduct such studies rests upon "users or
groups of users with a large number of persons in a job class,
or test developers. . . . " Uniform Guidelines, § 14B(8) (emphasis
54/
added). This obligation was not satisfied here.
Despite their failure to conduct an adequate study of fair
ness, the test developers concluded that, in at least one of the
concurrent validity sites, the results "mask what appears to be
differential validity by race." DX 31 at 47. The results in
this sample suggest that, while blacks get lower scores on the
test than whites, see PX 35, this difference is not reflected
in the performance ratings which black and white officers receive
from their supervisors on the job. An independent concurrent
53/ See also, EEOC Guidelines, 29 C.F.R. § 1607.5(b)(5); FEA
Guidelines, §§ 12b(7)(v), 14(j).
54/ Because of the small number of black officers in the Louis
ville Division of Police — 53 blacks of a total of 714 officers
at the time of trial, PX 38 — it probably would nor be tech
nically feasible at present to conduct a study of fairness involving incumbent Louisville police officers alone. However, the ex
clusion of blacks from the force is the result of a long history
of both international discrimination and facially neutral prac
tices with discriminatory effects. See pp. 11-45, supra.
Therefore, even if Test 165.1 were otherwise valid, it could not
lawfully be used as it has been used in Louisville because "the
technical infeasibility resulted from discriminatory employment
practices which are demonstrated by facts other than past failure
to conform with requirements for validation of selection procedures."
Uniform Guidelines, § 14B(8)(f).
87
study of Test 165.1 in Delaware also made a preliminary finding,
based on a small sample, that blacks had significantly lower
scores than whites on the total test and the subtests, but that
there were no significant differences between the performance
ratings of blacks and whites with the exception of a difference
of 0.5 of a point on two out of twelve performance dimensions.
DX 32 at 31. Under these circumstances, racial adjustments —
such as setting the passing point lower for blacks than for
whites, or regarding a score of 50 by a black as equivalent to a
score of 60 by a white — must be made in the use of the test to
reflect the actual meaning of the scores. Uniform Guidelines,
§ 14B(8)(d). See Bakke, supra, 46 U.S.L.W. at 4921 and n.37. In
view of the substantial evidence of bias in the performance
measures, the existing indications of differential validity, and
the test developers' failure to undertake an adequate study of
test fairness. Test 165.1 has not been shown to measure rairly
any differences in the performance of blacks and whites as police
officers.
d. The concurrent validity study produced
an odd patchwork of results which did
not demonstrate that Test 165.1 is valid
for use in selecting police officers in
every city.
In a criterion-related validity study, the degree of the
relationship between test scores and criterion measures should be
examined and computed "using professionally acceptable statistical
procedures," Uniform Guidelines, § 14B(5), and "[u]sers should
avoid reliance upon techniques which tend to overestimate validity
88
findings as a result of capitalization on chance unless an appro-
.55/
priate safeguard is taken." Id., § 14B(7). See APA Standards,
15 E8-E8.5, at 40-43.
The developers of the test in this case reported varying
results in evaluating the relationship between job performance
ratings and scores on their 180-item test in the four sample
sites: in Site 1, total test score had significant positive cor
relations with 2 of the 15 rating dimensions (DX 31 at 175); in
Site 2, total test score was not correlated with any rating
dimensions (id. at 176); in Site 3, total test score had signifi
cant positive correlations with 11 of the rating dimensions (id.
at 179); and in Site 4, total test score had a significant nega
tive correlation with 1 rating dimension — i.e., higher test
scores were correlated with lower performance ratings in that
.56/dimension (id. at 182). These conclusions were not based on
55/ See also, FEA Guidelines, § 12b(6).
56/ The test developers included these findings in an appendix
to their report. DX 31, Appendix K at 174-85. In the body of
the report, they stated that total test score had significant
positive correlations with 10 of the 15 rating dimensions in
Site 1, no correlations with any rating dimensions in Site 2,
significant positive correlations with all of the rating dimen
sions in Site 3, and significant negative correlations with 3
of the rating dimensions in Site 4. DX 31 at 45. These con
clusions were based on improper tests of statistical significance
on correlation coefficients which had been corrected for the
restriction in range resulting from the fact that the samples
consisted of trained and experienced police officers. Id* at
40-44. The APA Standards state that it is "essential" that
significance tests be made with uncorrected correlation co
efficients. APA Standards, 1 E8.2.1, at 41. See Barrett, Vol.
IV, 7/14/77 at 582.
89
scores on Test 165.1 (the 150-item test which was used in
Louisville), Rosenfeld Dep., 5/26/77 at 58, 125, and the reported
correlation coefficients were not recomputed on the basis of Test
12/165.1,Tyler, Vol. IV, 6/23/77 at 656.
The defendants and their experts have selectively cited the
partial positive correlations obtained in Sites 1 and 3 as
evidence of validity, while ignoring the lack of any correlations
in Site 2 and describing the evidence of negative correlations in
Site 4 as "puzzling because it doesn't make sense." Tyler,
Vol. IV, 6/23/77 at 622. Although the same criterion measures
were used in all four sites, defects in those measures have been
used as the stated justification for disregarding the undesired
results in two of the sites without any explanation as to why the
same defects would not also call into question the hoped-for
results in the other two sites. DX 31 at 46-47; Rosenfeld Dep.,
5/26/77 at 178. As Dr. Barrett stated, "it is necessary . . .
to evaluate all the data that is available, and that means we
have to evaluate the good and poor alike and come to some con
clusion about what they mean." Barrett, Vol. Ill, 7/13/77 at
43 5..
Taking all four of.the sites here as a whole, total test
score had significant positive correlations with only 13 out of
60 rating dimensions. DX 31 at 174-85. When all of the data
57/ Defendants offered another concurrent validity study, DX 32,
which was based on Test 165.1. However, defendants set the pass
ing score too high and used the test as a ranking device, con
trary to the findings of this study. Id. at 33-39. See pp. 96-98.
infra.
90
are evaluated, the most that can be said is that they suggest
positive relationships between test scores and some aspects of
job performance in some places, negative relationships in other
places, and no relationship at all in still other places. The
study does not indicate the category in which Louisville belongs
— whether applicants with higher scores would perform better as
police officers (cf. Sites 1 and 3), whether applicants with
lower scores would perform better as police officers (cf. Site
4), or whether the test scores have no relationship whatsoever
with job performance (cf. Site 2). This mixture of positive,
negative, and zero correlations is comparable to the "odd patch-
work of results" which the Supreme Court found insufficient in
Albemarle Paper Co. v. Moody, supra. 422 U.S. at 432. This study
does not demonstrate that Test 165.1 is valid for use in select
ing police officers in all jurisdictions throughout the country;
rather, "the study's checkered results appear to compel the
opposite conclusion." Id. See Barrett, Vol. Ill, 7/13/77 at 58/
429-36.
3. The validity studies offered by defendants
do not support the use of Test 165.1 in
selecting Louisville police officers.
"Users may, under certain circumstances, support the use of
selection procedures by validity studies conducted by other users
58/ It is undisputed that the validation study report, DX 31, con
tains numerous computational and other errors. Tyler, Vol. IV,
6/23/77 at 611-13; Rosenfeld Dep., 5/26/77 at 171-75; Barrett, Vol.
Ill, 7/13/77 at 436-43. These errors cast further doubt on the
adequacy and accuracy of the research performed and the conclusions
reached. See Uniform Guidelines, § 5E; Barrett, Vol. Ill, 7/13/77
at 443.
91
or conducted by test publishers or distributors and described in
test manuals." Uniform Guidelines, § 7A. Such studies provide
acceptable evidence of validity if (1) they "clearly demonstrate[ ]
that the selection procedure is valid"; (2) the user's job con
sists of "substantially the same major work behaviors" as the
job studied; and (3) the studies include "a study of test fair
ness for each race, sex and ethnic group which constitutes a
significant factor in the borrowing user's relevant labor market
for the job . . . in question." Id., § 7B. In addition, the
user may not rely upon such studies if they contain "variables
. . . which are likely to affect validity significantly. . . . "
59/Id., § 7D. See APA Standards, 5 E5.2.2, at 36; Division 14
Principles at 13.
The defendants here rely primarily upon the study conducted
by the Educational Testing Service (ETS), DX 31, as support for
60/
their claim of validity. The standards for such reliance have
not been satisfied. As established above, the ETS study does not
meet the requirements of the applicable legal and professional
standards, and it does not demonstrate that Test 165.1 is valid
for use in selecting police officers anywhere. Uniform Guide
lines, § 7B(1). Moreover, there is evidence that a police officer
in Louisville does not necessarily perform the same work behaviors
59/ See also, EEOC Guidelines, 29 C.F.R. § 1607.7; FEA Guidelines,
§ 6.
60/ Defendants also presented an independent study involving
Delaware police departments. DX 32. However, defendants did not
use Test 165.1 in accordance with the findings of this study.
See pp. 96-98, infra.
92
as the police officers who were the subjects of the study (see
Talbert, Vol. V, 6/24/77 at 735-36; Crosby, Vol. II, 7/12/77 at
290-91), and that there may be significant differences from
place to place in the characteristics of applicants even where
61/
the major work behaviors are the same. See Barrett, Vol. Ill,
7/13/77 at 506-508; Talbert, Vol. V, 6/24/77 at 739. Uniform
Guidelines, §§ 7B(2), 7D. Finally, the ETS study did not
thoroughly investigate the question of test fairness; however,
there are substantial indications that different scores on Test
165.1 do not fairly reflect differences in job performance be
tween blacks and whites. See pp. 84-88, supra. Uniform
Guidelines § 7B(3). Any one of these deficiences is enough to
render the ETS study unacceptable as support for the claim that
the test is valid for use in Louisville. Uniform Guidelines,
§ 7B.
An empirical study of Test 165.1 should have been conducted
in Louisville to determine whether the test was valid for use
here. The developers of the test themselves recommended that,
" [w]here feasible, criterion-related validity studies should
be conducted by individual jurisdictions using the test." DX 31
at 53; Rosenfeld Dep., 5/26/77 at 203. Such a study clearly was
feasible in Louisville: the Chief Examiner of the defendant
Civil Servipe Board testified that in July 1977 — approximately
fil/ Defendants' expert Talbert admitted that it is necessary to
demonstrate similarities in the demographic characteristics of
the two applicant populations in order to generalize validity
from the study site to the user's site. ■ Talbert, Vol. V, 6/24/77
at 739-40. No such demonstration has been made here.
93
six months after Test 165.1 was administered to applicants in
Louisville and four months after the trial of this action com
menced — the Board finally entered into a contract with Dr.
Wayne Helms, a psychologist who testified as an expert for defen
dants, to conduct both a local concurrent validity study of the
test and, depending upon the results, a local predictive validity
study as well. Gavin-Wagner, Vol. Ill, 7/13/77 at 358-59. Although
62/it is unclear whether these studies were ever in fact conducted,
this testimony shows that defendants recognized that the studies
were feasible. As noted above, both of the validation strategies
relied upon in this case — content validation (see pp. 66-69,
supra) and concurrent criterion-related validation (see pp. 76-79
supra) — are inappropriate for validating a selection procedure
which, like Test 165.1. involves knowledges, skills, or abilities
which employees learn on the job. The applicable professional
standards indicate that the appropriate method of validating
such a test is by a predictive criterion-related study involving
a comparison between the test scores of applicants and their sub
sequent job performance as employees. Division 14 Principles at
5-6; APA Standards at 26-27. Such studies are frequently done?
they are "not only economically feasible but economically very
desirable." Barrett, Vol. Ill, 7/13/77 at 446-47; Vol. IV,
7/14/77 at 579-80. The courts too have recognized that a predic
tive validity study is "[t]he most accurate way to validate an
62/ Plaintiffs have sought the Court's permission to reopen the
record for the purpose of determining these facts. See plain
tiffs' motion to reopen and supplement the record, 8/15/78;
defendants' response to motion to reopen and supplement the record,
8/31/78.
94
employment test. . . . " United States v. Georgia Power Co..
supra, 474 F.2d at 912; Boston Chapter, NAACP, Inc, v. Beecher,
supra, 504 F.2d at 1025. Where, as in this case, such a study is
technically feasible and where the existing evidence of validity
is so conflicting and insufficient, an internal predictive
validity study is required in order to justify the use of an
examination having the extreme adverse impact of Test 165.1.
4. Defendants substantially increased the
adverse impact of Test 165.1 by setting
an arbitrarily high passing point and
by improperly using the test to rank
applicants.
Even if a test has been shown to be valid for selecting
applicants for a job, it must be used in a manner which is con
sistent with the evidence of validity. When passing or cut-off
scores are used, "they should normally be set so as to be reason
able and consistent with normal expectations of acceptable pro-
£3/
fxciency within the work force." Uniform Guidelines, § 5H.
Further, "[e]videncs which may be sufficient to support the use
of a selection procedure on a pass/fail (screening) basis may be
insufficient to support the use of the same procedure on a ranking
basis. . . . " Id., § 5G. In addition "[s]ole reliance upon a
single selection instrument which is related to only one of many
job duties or aspects of job performance will also be subject to
64/close review." Id., § 14B(6).
63/ See also, FEA Guidelines, § 5f.
64/ See also, FEA Guidelines, § 12b (5) .
95
The record here demonstrates that Test 165.1, even if it
had been shown to be valid, was not used properly in selecting
Louisville police officers. The developers of Test 165.1 made
no specific finding as to an appropriate cut-off score, but they
stated that a score of 139 correct answers (or 11.2%) on the
180-item test which was the subject of their validity study "might
represent a cutting score somewhat higher than would be used in
an actual selection context," and that the use of this cutting
score would not have a substantial adverse impact on blacks.
DX 31 at 48. See Rosenfeld Dep., 5/26/77 at 155-59. The indepen
dent University of Delaware study of the 150-item test which was
used in Louisville concluded that a total test score of 120 (or
80%) was "a technically justifiable cutoff score" because this
"appeared to be the point at which a noticeable demarcation
occurred with respect to officer performance on the job." DX 32
at 35. The defendants, however, set the cutoff score for use in
Louisville at 128 of the 150 items on the test, or 85.3%. Gavin-
Wagner, Vol. II, 7/12/77 at 349; Vol. Ill, 7/13/77 at 361. This
score, which was substantially higher than any passing point sug
gested in either of the validity studies, was chosen because it
was " [a]cceptable in terms of creating a sufficient size list"
for the projected number of vacancies and because the defendants
felt that it would include as many minority applicants as possible
at "what would still be considered an acceptable level." Id. at
361-62. However, only 49 of the 207 blacks who took the test in
Louisville passed at this level. PX 35. If the passing point had
been set at 120 as recommended in the University of Delaware study,
96
approximately 30 additional black applicants would have passed
the test. Gavin-Wagner, Vol. Ill, 7/13/77 at 365. Defendants'
experts testified that there was no basis for believing that
applicants who scored below 128 were not qualified to be Louisville
police officers. Helms, Vol. IV, 7/14/77 at 624; Rosenfeld Dep.,
5/26/77 at 160-65. Thus, defendants set the passing score in a
manner which
subordinates the goal of job-relatedness to
that of administrative convenience . . . [and]
departs from the requirement, imposed by law
[when the procedure has an adverse impact],
that such decisions be made so as to further
the paramount goal of job relatedness. Kirk
land v. New York State Department of Correc
tional Services, supra, 374 F. Supp. at 1377. 65/
As defendants recognized, the effective cutoff score on
Test 165.1 as used in Louisville was in fact far higher than 128
because the defendants elected to assign the test score a weight
of 75% in determining the position of applicants on a ranked
eligibility list. Gavin-Wagner, Vol. Ill, 7/13/77 at 365-66.
Only 50 to 60 of the 401 candidates on the list had a realistic
chance of appointment. Id_. at 362 . Where applicants are ranked
on the basis of test scores, and where those applicants scoring
below a particular level "have little or no chance of being
selected for employment, the higher cutoff score may be appro
priate, but the degree of adverse impact should be considered."
65/ The APA Standards similarly provide that it is "essential"
for a test user to "have a rationale, justification, or explanation
of the cutting scores adopted." APA Standards, 5 14, at 66. The
accompanying comment explains that the validity of a test "should
be determined in light of its actual use. . . . The intent is to
recommend that test users avoid the practice of designating purely
arbitrary cutting scores they can neither explain nor defend."
Id. at 67.
Uniform Guidelines, § 5H. Here the adverse impact on blacks
approached total exclusion, see PX 35, and the defendants' only
basis for using the test as a ranking device was that they had
been told that other jurisdictions were using it in this way and
they
felt that the test was designed to be an
instrument to discriminate between people
who had varying levels of knowledges, skills,
and abilities required to do the work of a
police officer; and following that thought
through then, somebody possessing a greater
degree of these skills would, in fact, would
logically follow that they could be rank ordered
on the list. Gavin-Wagner, Vol. Ill, 7/13/77
at 366-67.
This "logic" is not supported by the evidence. Defendants'
expert, Dr. Crosby, admitted that the ETS study, DX 31, did not
provide any basis for using the test as a ranking device. Crosby,
Vol. II, 7/12/77 at 289. As Dr. Tyler, another of defendants'
experts, stated, "[a] bigger, higher score doesn't mean better.
. . ." Tyler, Vol. IV, 6/23/77 at 669. Moreover, the authors
of the University of Delaware study examined the available data
and concluded that
little justification exists for ranking
individuals on the basis of test scores
above 120. That is, the mean overall per
formance ratings of officers in various
test score ranges above 120 are fairly
similar — little variation exists. That
variation which appears is not of sufficient
magnitude to suggest that individuals with
test scores in the upper score ranges above
120 will have a considerably higher expecta
tion of success than individuals who score
in the lower ranges above 120. DX 32 at 36.
66/ Cf. FEA Guidelines, § 5f.
98
Thus, there has been no showing that either the cut-off
selected by defendants or their use of the test to rank applicants
was supported by the available evidence of validity and utility.
The use of arbitrary passing scores and arbitrary selection
scores based on rank-ordering has repeatedly been held unlawful
where the result is an adverse racial impact. See Boston Chapter,
NAACP, Inc, v. Beecher, supra, 504 F.2d at 1023; Kirkland v. New
York State Department of Correctional Services, supra, 374 F. Supp.
at 1377; Stamps v, Detroit Edison Co., 365 F. Supp. 87, 118 (E.D.
Mich. 1973), aff'd in pertinent part sub nom EEOC v. Detroit
Edison Co., 515 F.2d 301 (6th Cir. 1975), vacated and remanded on
other grounds, 431 U.S. 951 (1977); Western Addition Community
Organization v. Alioto, 360 F. Supp. 733, 738 (N.D. Cal. 1973),
appeal dismissed. 514 F.2d 542 (9th Cir.), cert, denied, 423 U.S.
1014 (1975). Cf. Bridgeport Guardians, Inc, v. Bridgeport Civil
Service Commission, supra, 482 F.2d at 1338.
Here the defendants further emphasized the adverse impact by
using this test at an early stage of the selection process, a
sequence which effectively screened out 201 of the 207 black
applicants who took the test before they ever had a chance to
demonstrate the physical and interpersonal skills which are crucial
to the job of a police officer. See pp. 72-74, supra. In
effect, defendants placed virtually "sole reliance upon a single
selection instrument which is related to only one of many . . .
aspects of job performance. . . . " Uniform Guidelines, § 14B(6).
This use of the test at this stage of the selection process was
99
clearly improper. See Firefighters Institute for Racial Equality£ 2 /
v. City of St. Louis, supra, 549 F.2d at 512.
D . Alternative Selection Procedures with
Less Adverse Impact Would Properly Serve
the Defendants' Legitimate Interest in the
Selection of Capable Police Officers.
"Where two or more selection procedures are available which
serve the user's legitimate interest in efficient and trustworthy
workmanship, and which are substantially equally valid for a
given purpose, the user should use the procedure which has been
demonstrated to have the lesser adverse impact." Uniform Guide-
68/lines, § 3B. See id. §§ 6A, 14B(6), 17(3)(d). Even where an
employer has demonstrated that a test which has an adverse racial
impact is job-related,
it remains open to the complaining party to
show that other tests or selection devices,
without a similarly undesirable racial effect,
would also serve the employer's interest in
"efficient and trustworthy workmanship."
. . . Such a showing would be evidence that
the employer was using its tests merely as a
"pretext" for discrimination. Albemarle
Paper Co. v. Moody, supra, 422 U.S. at 425.
67/ As stated in the University of Delaware study, "The written
test is but one component in the selection process. Up to this point in most selection processes, applicants have already
passed a number of preliminary screening devices (e.g., minimum
requirements, physical agility, etc.). In multiphase selection
systems, the objective is to rationally reduce the number of
applicants to be considered at each successful [sic: read "suc
cessive"] step." DX 32 at 34. The system used in Louisville
did not accomplish this objective in a rational way.
68/ See also, FEA Guidelines, § 3c.
100
Because Test 165.1, as used by defendants, has not been
shown to be job-related, it is unnecessary to engage in exten
sive consideration of alternative selection procedures at the
present stage of this litigation. Id. However, the record
plainly discloses at least two viable alternatives which would
have less adverse impact on black applicants. First, the defen
dants themselves developed a written test (DX 39) which was admin
istered to police applicants in August 1975. This test drew upon
"assessment center" techniques and included audio-visual components.
Olges, Vol. II, 7/12/77 at 218-25. Although only 1.1% (3/280) of
the white applicants failed this test while 8.2% (5/61) of the
black applicants failed (DX 75, "Statistical Data, Book Three"
— "Sworn Personnel" at 1 6), a sufficient number of blacks scored
high enough on the test that it did not have a substantial adverse
impact as it was used at that time. Ogles, Vol. II, 7/12/77 at
226. See stipulations of counsel, Vol. II, 6/21/77 at 273-74;
Vol. I, 7/11/77 at 157; Vol. I, 9/26/77 at 130-32. The Chief of
Police testified that most of the officers hired on the basis of
this test and the other selection procedures used between August
1975 and November 1976 were satisfactory. Nevin, Vol. IV, 6/23/77
at 510. Darrell Olges, the Assistant Director and former Chief
Examiner of the Civil Service Board, testified that there was no
reason to believe that Test 165.1 was any better than the August
1975 test in predicting successful job performance. Olges, Vol.
II, 7/12/77 at 231-32. Thus, the written test which the defendants
developed themselves and used prior to Test 165.1 may well be an
acceptable alternative with less adverse impact.
101
The second obvious alternative would be for the defendant
to continue using Test 165.1 to select police officers but to
modify that use to eliminate the adverse impact on blacks. As
demonstrated above, defendants' present use exacerbates the
adverse impact of the test by setting an arbitrarily high cut-off
score, by ranking candidates on the basis of their test scores
without any adequate justification, and by improperly screening
out virtually all black applicants on the basis of test score
alone at the beginning of the selection process. See pp. 95-100
supra. Simply lowering the cutoff score from the point used in
Louisville (128) to the point recommended by the University of
Delaware validity study (120), DX 32 at 35, would have increased
the number of blacks passing the test from 49 to 78. Gavin-Wagner,
Vol. Ill, 7/13/77 at 364-65. The adverse impact would be further
reduced by not ranking applicants on the basis of their scores on
Test 165.1; as the University of Delaware study concluded, there
is no adequate justification for using the test as a ranking
device. DX 32 at 36. If the test is used instead as a qualifying
device, in conjunction with a variety of other possible selection
procedures, see DX 32 at 36-37, most or all of the test's adverse
impact on blacks may be eliminated. In the absence of evidence
that differences in passing test scores are related to significant
differences in job performance, see DX 32 at 36, 5 3, it would be
professionally proper to eliminate the adverse impact altogether
by appointing otherwise qualified blacks and whites who have
obtained an appropriate passing score on the test in the proportions
in which they appear in the pool of applicants. Barrett, Vol. Ill,
102
7/13/77 at 454-60. Where, as in this case, the test has not
been shown to be manifestly job-related, elimination of the
adverse impact is not only professionally proper but is compelled
by Title VII. Albemarle Paper Co. v. Moody, supra, 422 U.S. at
425; Griggs v. Duke Power Go., supra, 401 U.S. at 431.
103
III. THE COURT HAS THE POWER AND THE DUTY TO
FASHION RACE-CONSCIOUS NUMERICAL HIRING
RELIEF WHICH WILL MAKE THE LOUISVILLE
POLICE FORCE MORE REPRESENTATIVE OF THE
COMMUNITY IT SERVES.
The record demonstrates that the defendants have a long his
tory of racial segregation and overt discrimination against blacks
in their police employment practices; that they engaged in inten
tional discrimination in their recruitment practices prior to the
filing of this action; that both before and after the filing of
this action, they engaged in intentional discrimination in their
selection practices and they used tests and other selection pro
cedures which had an adverse impact on blacks and were not re
lated to job performance; and that immediately before the trial
of this case they selected police officers largely on the basis
of an unvalidated written test which had such an extreme adverse
impact on blacks that it eliminated virtually all black applicants
from consideration. • As a result of this clear and longstanding
pattern of discrimination, the City of Louisville, with a 23.8%
black population in 1970, had a police force which was only 7.4%
black by 1977.
On this record, the Court has "not merely the power but the
duty to render a decree which will so far as possible eliminate
the discriminatory effects of the past as well as bar like dis
crimination in the future." Louisiana v. United States, 380 U.S.
145, 154 (1965); Albemarle Paper Co. v. Moody, supra,422 U.S. at
418. In fulfilling this duty, many courts have held that
numerical race-conscious remedies may, and in some instances must,
be imposed under Title VII, § 1981, and § 1983 and the Fourteenth
104
Amendment to eradicate the effects of such discriminatory police
and fire department recruitment and selection practices. See
Davis v. Countv of Los Angeles, supra, 566 F.2d at 1342-43 (§ 1981)
United States v. City of Chicago, supra, 549 F.2d at 436-37 (Title
VII); Boston Chapter, NAACP, Inc, v. Beecher, supra, 504 F.2d at
1026-28 (Title VII, § 1981, § 1983, Fourteenth Amendment); NAACP
v. Allen, 493 F.2d 614, 617-21 (5th Cir. 1974) (Fourteenth Amend
ment) ; Erie Human Relations Commission v. Tullio, 493 F.2d 371,
373-75 (3rd Cir. 1974) (§ 1981, § 1983, Fourteenth Amendment);
Bridgeport Guardians, Inc, v. Members of Bridgeport Civil Service
Commission, supra, 482 F.2d at 1340-41 (§ 1981, § 1983, Fourteenth
Amendment); Carter v. Gallagher, 452 F.2d 315, 328-31 (8th Cir.)
(en banc), cert. denied, 406 U.S. 950 (1972) (§ 1981, Fourteenth
Amendment). The Sixth Circuit has repeatedly upheld the authority
of the district courts to order such numerical relief. See EEOC
v. Detroit Edison Co., supra, 515 F.2d at 317; Arnold v. Ballard,
12 FEP Cases 1613, 1614, (6th Cir.), vacated and remanded on other
grounds, 16 FEP Cases 396 (6th Cir. 1976); United States v.
Masonry Contractors Association, 497 F.2d 871, 877 (1974); United
States v. Local 212, IBEW, 472 F.2d 634, 636 (1973); Sims v .
Local 65, Sheet Metal Workers, 489 F.2d 1023, 1027 (1973); United
States v. Local 38, IBEW, 428 F.2d 144, 149, cert. denied, 400
U.S. 943 (1970). The courts of appeals of eight other circuits
69/are in accord, and a majority of the Justices of the Supreme
69/ See, e.g..Boston Chapter, NAACP, Inc0 v. Beecher, 504 F.2d
1017 (1st Cir. 1974), cert, denied, 421 U.S. 910 (1975); Associated
General Contractors of Mass., Inc, v. Altshuler, 361 F. Supp. 1293
105
Court recently acknowledged that courts may properly impose
race-conscious remedies for judicially determined violations of
Title VII and the Fourteenth Amendment. Regents of the University
of California v. Bakke, supra. 46 U.S.L.W. at 4919 n. 28, 4922
(opinion of Brennan, White, Marshall, and Blackmun, JJ.); 46
20/U.S.L.W. at 4905-4907 and nn. 40-41, 44 (opinion of Powell, J.).
89/ (Cont'd.)
(D. Mass.), aff1d , 490 F.2d 9 (1st Cir. 1973), cert, denied, 416
U.S. 957 (1974); Rios v. Enterprise Association Steamfitters Local
638, 501 F.2d 622 (2d Cir. 1974)7 Bridgeport Guardians, Inc, v.
Members of Bridgeport Civil Service Commission, 482 F.2d 1333
(2d 1973) 7 United States v. Wood Lathers Local 46, 471 F.2d 408
(2d Cir.), cert. denied, 412 U.S. 939 (1973); Erie Human Relations
Commission v. Tullio, 493 F.2d 371 (3rd Cir. 1974) 7 Contractors
Association v. Secretary of Labor, 442 F.2d 159 (3rd Cir.), cert.
denied, 404 U.S. 854 (1971); Sherrill v. J.P. Stevens & Co., 551
F .2d 308, 13 E.P.D. 5 11,422 (4th Cir. 1977) 7 NAACP v. Allen, 493
F.2d 614 (5th Cir. 1974) 7 Morrow v. Crisler, 491 F.2d 1053 (5th
Cir. 1974) (en banc), cert, denied, 417 U.S. 965 (1974); Local
53, Asbestos Workers v. Vogler, 407 F.2d 1047 (5th Cir. 1969) 7
United States v. City of Chicago, 549 F.2d 415 (7th Cir. 1977),
cert, denied, 434 U.S. 875 (1977); Crockett v. Green, 534 F.2d
715 (7th Cir. 1976) 7 Southern Illinois Builders Association v.
Ogilvie, 471 F.2d 680 (7th Cir. 1972) ; United States v. N. L.
Industries, Inc., 479 F.2d 354 (8th Cir. 1973); Carter v.
Gallagher, 452 F.2d 315, 327 (8th Cir.) (en banc), cert, denied,
406 U.S. 950 (1972); Davis v. County of Los Angeles, 566 F.2d
1334 (9th Cir. 1977), cert. granted, 46 U.S.L.W. 3780 (U.S.,
June 19, 1978); United States v. Ironworkers Local 86, 443 F.2d
544 (9th Cir.), cert denied, 404 U.S. 984 (1971).
70/ Justice Powell believed that such remedies are permissible
if they are supported by appropriate "judicial, legislative, or
administrative findings of constitutional or statutory violations,"
id. at 4906, or by some other compelling governmental interest,
id. at 4907-4908. Justices Brennan, White, Marshall, and Blackmun
held that racial classifications which are designed to remedy the
effects of past societal discrimination are permissible with or
without judicial, legislative, or administrative findings, so
long as they do not stigmatize any discrete group or individual
as inferior, and so long as they are reasonably used in light of
their purpose. Id. at 4920-24. Chief Justice Burger and Justices
Stewart, Rehnquist, and Stevens expressed no view on this subect. Id. at 4933-36.
106
effects of the past and to bar like discrimination in the future,
the Court should require the defendants to appoint a sufficient
number of qualified blacks so that, within a reasonable period
of time, the proportion of black officers on the Louisville
police force will approximate the proportion of blacks in the
21/population of the City of Louisville. Many courts in similar
cases have required that remedial, race-conscious hiring of
In this case, in order to eliminate the discriminatory
71/ Plaintiffs recognize that setting the appropriate numerical
or percentage goal of such affirmative hiring relief is a matter
committed largely to the Court's discretion, and that a number of
factors may be considered in the exercise of that discretion.
See Stamps v. Detroit Edison Co., 365 F.Supp. 87, 122 n.4 (E.D.
Mich. 1973), aff'd in pertinent part sub nom. EEOC v. Detroit
Edison Co., 515 F.2d 301, 317 (6th Cir. 1975), vacated and re
manded on other grounds, 431 U.S. 951 (1977). The Court may,
for example, take additional evidence and attempt to construct
a model of the proper applicant pool and the racial composition
of the work force which would have resulted if the defendants had
not engaged in discriminatory recruitment and selection practices.
See League of United Latin American Citizens v. City of Santa Ana,
supra, 410 F.Supp. at 910-11. Such a model might include a
weighting of the racial compositions of the various geographical
areas represented in the proper applicant pool in order to arrive
at an estimate of the racial composition of the resulting work
force. See Gastwirth & Haber, Defining the Labor Market for
Equal Employment Standards, 99 Monthly Labor Review 32, 34
(March 1976). Although the record as it stands contains some
data which would be relevant in making an estimate of this sort,
see Lee, Vol. IV, 7/14/77 at 634-36, further evidence would be
required to assure its accuracy. On the present record, both
precedent and policy support plaintiffs' position that the black
proportion of the population of the City of Louisville is the most
appropriate goal for the affirmative hiring relief which is
necessary in this case.
10 7
minority police officers and firefighters continue until the
departments involved reflect the racial composition of the
communities they serve. See Arnold v. Ballard, supra. 390
F.Supp. at 736, aff1d . 12 FEP Cases at 1616, vacated and remanded
on other grounds, 16 FEP Cases 396 (6th Cir. 1976); Davis v.
County of Los Angeles, supra. 566 F.2d at 1343-44 and n.3; Bridgeport
Guardians. Inc, v. Bridgeport Civil Service Commission, supra.
482 F . 2d at 1341; NAACP v. Allen, supra. 493 F.2d 614, 621 (5th
Cir. 1974); Officers for Justice v. Civil Service Commission.
371 F.Supp. 1328., 1330-31 (N.D. Cal. 1973). Cf. Erie Human
Relations Commission v. Tullio, 493 F.2d 371, 375 (3rd Cir. 1974).
The need for such relief is particularly acute where, as in
Louisville, the discriminatory exclusion of blacks has impaired
the ability of the police force to provide fair and effective law
enforcement and other police services to its community.
Congress, in extending the coverage of Title VII to public
72 /
employment in 1972, found that state and local governments in
11/ 74/general, and police and fire departments in particular, were
engaging in widespread discrimination against minorities in * S.
72/ Equal Employment Opportunity Act of 1972, Pub. L. No. 92-261,
86 Stat. 103.
73/ See H. R. Rep. No. 92-238, 92d Cong., 1st Sess. 17-18 (1971);
S. Rep. No. 92-415, 92d Cong., 1st Sess. 10 (1971).
74/ See 118 Cong. Rec. 790 (1972), reprinting excerpts from U. S.
Commission on Civil Rights, For All The People . . . By All The
People, at 71 (1969).
108
is particularly acute and has the most
deleterious effect in those government
activities which are most visible to the
minority communities (notably education,
law enforcement, and the administration
of justice) with the result that the
credibility of the government's claim to
represent all the people is negated. H.
R. Rep. No. 92-238, 92d Cong. 1st Sess.
17 (1971). 75/
As the Supreme Court stated recently, the police serve a
basic and pervasive function of government in modern society;
police officers "are clothed with authority to exercise an al
most infinite variety of discretionary powers," and the execution
of those powers "affects members of the public significantly and
often in the most sensitive areas of daily life." Foley v.
Connelie, 55 L.Ed. 2d 287, 293 (1978). Black citizens are
affected even more than whites; blacks tend to be subjected to
overenforcement by the police, Keil, Vol. IV, 9/29/77 at 715,
and blacks perceive that they are disproportionately subjected
to police harassment in the use of unnecessary force, abusive
language, and unnecessary stops and searches. _Id. at 717-18.
employment. Congress recognized that such discrimination
75/ A Senate committee report similarly stated: "The failure
of State and local governmental agencies to accord equal employ
ment opportunities is particularly distressing in light of the
importance that these agencies play in the daily lives of the
average citizen. From local law enforcement to social services,
each citizen is in constant contact with many local agencies. . . .
Discrimination by government therefore serves a doubly destructive
purpose. The exclusion of minorities from effective participation
in the bureaucracy not only promotes ignorance of minority pro
blems in the particular community, but also creates mistrust,
alienation, and all too often hostility toward the entire pro
cess of government." S. Rep. No. 92-415, 92d Cong., 1st Sess.
10 (1971).
109
There is substantial evidence of racial bias on the part of white
police officers in Louisville as in other major cities. See Nevin,
Vol. II, 9/27/77 at 391-410; Keil, Vol. IV, 9/29/77 at 729-34;
Report of the National Advisory Commission on Civil Disorders at
305-306 (Bantam ed. 1968); President's Commission on Law Enforce
ment and Administration of Justice, Task Force Report: The Police
at 164, 172 (GPO 1967).
The high visibility and broad discretionary powers of police
officers, coupled with the perception of black citizens that they
are the particular victims of police mistreatment, result in
apprehension, distrust, and hostility between the police and the
black community. As Judge Edwards wrote in 1965, "hostility
between the Negro communities in our large cities and the police
departments is the major problem in law enforcement in this
decade." Edwards, Order and Civil Liberties: A Complex Role
for the Police, 64 Mich. L. Rev. 47, 54-55 (1965) . The National
Advisory Commission on Civil Disorders found that "deep hostility
between police and ghetto communities [was] a primary cause of
the disorders . . . in practically every city that has experienced
racial disruption since the summer of 1964 . . . ," Report of the
National Advisory Commission on Civil Disorders, supra at 299,
and that discriminatory police employment and promotion practices
were among the most serious complaints in those cities, id. at'
144. Similarly, the President’s Commission on Law Enforcement
and Administration of Justice determined that there was a critical
shortage of black police officers in most large cities, and that
discrimination against blacks in the selection of police officers
110
was widespread. Task Force Report: The Police, supra at 167-71.
That Commission also found that hostility between the police and
minority communities was "as serious as any problem the police
has today." Id* at 99* As stated in the National Advisory Com
mission's report,
for police in a Negro community to be pre
dominantly white can serve as a dangerous
irritant; a feeling may develop that the
community is not being policed to maintain
civil peace but to maintain the status quo.
[In addition,] . . . contact with Negro
officers can help to avoid stereotypes and
prejudices in the minds of white officers.
Negro officers also can increase departmental
insight into ghetto problems, and provide
information necessary for early anticipation
of the tensions and grievances that can lead
to disorders. Report of the National Advisory
Commission on Civil Disorders, supra.
Hostility between the black community and the police force
continues to be a major problem in Louisville today. The fact
that black officers are grossly underrepresented on the force,
and the fact that there is a longstanding pattern of discrimina
tion against blacks in police employment practices, are widely
known in the black community. See plaintiffs' proposed findings
20-21. The.black community is also keenly aware of police harass
ment and mistreatment of black citizens. See Brown, Vol. IV,
4/28/77 at 557; Arnold, Vol. IV, 9/28/77 at 772. The Director
of Safety, Allen Bryan, acknowledged that there have been a
number of problems between the black community and the police
over the years. Bryan, Vol. Ill, 6/22/77 at 420. Mayor Sloane
also testified that the City had received many complaints about
the police from members of the black community. Sloane. Vol.
IV, 9/29/77 at 595.,
Ill
The existence of such hostility toward and distrust of the
police by the black community severely impairs the ability of the
Division of Police to perform its proper functions for the citizens
of Louisville. Its effectiveness in detecting and preventing
crime and in apprehending suspects depends to a great extent on
a network of information and communication between the police
and the community. Keil, Vol. IV, 9/29/77 at 720-21. When this
network breaks down because of hostility and distrust, the police
have great difficulty in carrying out this role. Id. As the
Chief of Police testified, the police force also provides "the
social agency contact for a large number of people in the city,
. . . [including] counseling of juveniles and adults. . . . "
Nevin, Vol. IV, 6/23/77 at 480. The Division of Police clearly
cannot provide adequate services of this kind to the black com
munity in the atmosphere of hostility and alienation which pre
sently exists. Finally, the police force and its officers serve
as visible, pervasive symbols of government in modern society.
Cf. Foley v. Connelie, supra, 55 L.Ed. 2d at 293. The black
community's negative perception of the police force can lead to
a rejection not only of "the apparatus of law enforcement," DX
85 at 75, but also of the entire governmental system. Keil,
Vol. IV, 9/29/77 at 718-19.
These effects of thp longstanding pattern of discriminatory
exclusion of blacks from the police force can be substantially
reduced by an order of this Court requiring the hiring of suf
ficient black officers to make the Division of Police representa
tive of the community it serves. Id. at 724. As Judge Edwards
112
has stated, "Perhaps nothing will do more in the long run to give
Negroes confidence in the police than the presence of black faces
on the force." Edwards, The Police on the Urban Frontier at 86
(New York 1968). The President's Commission on Law Enforcement
and Administration of Justice similarly found that, "In order
to gain the general acceptance of a community, personnel within
a police department should be representative of the community as
a whole." Id. at 167. See also, Report of the National Advisory _ 76/
Commission on Civil Disorders, suora at 315-16.
The defendants in this case have acknowledged that the police
force should be representative of the racial composition of the
population of the City of Louisiville. For example, George Burton,
the Director of Safety in 1970-1971, testified that there was an
imbalance between the proportion of black officers on the force
and the proportion of blacks in the City population. Burton,
Vol. Ill, 6/22/77 at 392, 399. He recognized that an "organization
... so close to the people should pretty well represent a cross-
section number-wise of the city." Id. at 339. Later, in applying
for federal funds for a minority recruitment program in 1975, the
defendants again stated that they were "aware of the low percentage
76/ Patrick V. Murphy, a former head of the police departments in
Detroit, Washington, D.C., and New York City, testified before the
Commission that "one of the serious problems facing the police in
the Nation today is the lack of adequate representation of Negroes
in police departments. I think the police have not recruited
enough Negroes in the past and are not recruiting enough of them
today. I think we would be less than honest if we didn't admit
that Negroes have been kept out of police departments in the
past for reasons of race discrimination." Report of the National
Advisory Commission on Civil Disorders, supra at 316.
113
of black officers as compared to the notable percentage
of black citizens within the city," and they recognized that it
was "desirable that a police department be representative of the
community which it serves. . . . " DX 76, "Index — First Quarter,
1975," at "Project Narrative"; Priebe, Vol. I, 9/26/77 at 156-58.
Thus, the defendants, like the courts, have long been aware
that
all citizens profit when the city achieves
a racially integrated police force of
qualified individuals who are knowledgable
of the diverse porblems of different ethnic
groups and who are not prey to destructive
hostility from minorities who feel excluded
from full participation in city government
life. Clearly, the general harmony of the
community is enhanced by the city's obtain
ing a police force representative of its
population. Officers for Justice v. Civil
Service Commission, supra, 371 F. Supp. at
1330-31.
As the Second and Fifth Circuits have stated in .upholding ratio
hiring remedies like that sought by plaintiffs here,
perhaps the most cruical consideration in
our view is that this is not a private
employer and not simply an exercise in pro
viding minorities with equal opportunity
employment. This is a police department
and the visibility of the Black patrolman
in the community is a decided advantage
for all segments of the public at a time
when racial divisiveness is plaguing law
enforcement. Bridgeport Guardians, Inc, v.
Bridgeport Civil Service Commission, supra,
482 F.2d at 1341; NAACP v. Allen, supra,
493 F .2d at 621.
See also, Arnold v. 3allara, supra, 390 F. Supp. at 736.
Because of the defendants' long history of racial discrimina
tion and the resulting gross underrepresentation of blacks on the
Louisville police force, an order which merely required the
defendants to engage in nondiscriminatory recruitment and hiring
114 -
practices from this date forward would delay for decades the
achievement of a reasonable representation of black officers on
the force. Based on assumptions that 25% of the future applicants
77/
will be black, that as a result of nondiscriminatory hiring
practices 25% of the officers hired in the future would therefore
22/be black, that both the increase in the size of the force and
the numbers of new officers hired annually will continue at the
23/same rates as occurred from 1964 to 1975, and that no current
or future black officers will leave the force during the period
fiQ/in question, it would take more than 34 years for enough black
officers to be hired to bring black representation on the force
01/up to the level of 25%. Even under the affirmative one-to-one
77/ From November 1976 to January 1977, 23.5% of the applicants
were black. PX 35.
73/ .Cf. Teamsters, supra, 431 U.S. at 340 n.20.
79/ See plaintiffs' proposed findings 6, 8, 31.
80/ This is an extremely conservative assumption which leads to
an underestimate of the number of years it will take to achieve
a level of 25% black representation on the force.
31/ This estimate was derived from the following formula:
% of blacks on force =
# of blacks on force in X years x }_qq =
total # of officers in X years
(# of blacks + (%.of blacksx (total officersx (X years)
on 1/1/75)______ hired)________ hired/year)______________ .̂oo
(size of force + (rate of annual growth)x(X years) x
on 1/1/75)
„ . . _ .. 53 + (.25) (38.8) (X)Solving for X years: 25% = (24 1) "(x) — X 100
X = 34.15 years.
115
hiring remedy requested by plaintiffs, it would take more than
9 years for the proportion of black officers on the force to
82J
reach 25%.
Many courts have recognized that it is necessary to hire
minorities at rates above their percentages in the applicant
population or relevant labor market for a sufficient period of
time to correct the effects of past discrimination. See, e.g.,
Arnold v. Ballard, supra, 390 F. Supp. at 736, aff'd, 12 FEP Cases
1613 (6th Cir.), vacated and remanded on other grounds, 16 FEP
Cases 396 (6th Cir. 1976); Stamps v. Detroit Edison Co., 365 F.
Supp. at 122, aff'd in pertinent part sub nom. EEOC v. Detroit
Edison Co., 515 F.2d 301, 317 (6th Cir. 1975), vacated and re
manded on other grounds, 431 U.S. 951 (1977); Davis v. County
of Los Angeles, supra, 566 F.2d at 1343-44; United States v.
City of Chicago, supra, 549 F.2d at 436-47; Crockett v. Green.
supra, 534 F.2d at 718; Boston Chapter, NAACP, Inc, v. Beecher,
supra, 504 F.2d at 1026-28; Erie Human Relations Commission v.
Tullio, supra, 493 F.2d at 374-75; NAACP v. Allen, supra, 493
F .2d at 617, 620-22; Bridgeport Guardians, Inc, v. Bridgeport
Civil Service Commission, supra, 482 F.2d at 1339-40; Carter v.
Gallagher, supra, ,452 F.2d at 330-31; League of United Latin
American Citizens v. City of Santa Ana, supra, 410 F. Supp. at 911.
82 / This estimate was derived from the same formula:
2 5% = 53 + (.50) (38.8) (X)
714 + (24.1) (X) x 100
X = 9.38 years.
116
Where, as in Louisville, discriminatory employment practices have
produced an underrepresentation of blacks on the police force
which is so extreme that it will take many years to correct even
under a one-to-one remedial hiring order, the Court may properly
impose such hiring relief at this time and require that it remain
83/
in effect until further order of the Court. See, e.g., Arnold
v. Ballard, supra, 390 F. Supp. at 736; United States v. City of
Chicago, supra, 549 F.2d at 436.
83/ Plaintiffs submit that, because of the length of time which
will be necessary to achieve any reasonable level of black repre
sentation on the force, the Court should immediately institute
accelerated hiring procedures even ,if the Court believes that
further proceedings are necessary to determine the ultimate per
centage goal of such hiring procedures. See n.71, supra.
117
IV. PLAINTIFFS ARE ENTITLED TO AN INTERIM
AWARD OF REASONABLE ATTORNEYS' FEES.
Both section 706(k) of Title VII, 42 U.S.C. § 2000e-5(k),
and the Civil Rights Attorneys' Fees Awards Act of 1976, Pub.
L. No. 94-559, 90 Stat. 2641, codified in 42 U.S.C. § 1988, pro
vide that "the court, in its discretion, may allow the pre
vailing party . . . a reasonable attorney's fee as part of the
costs." The Supreme Court has held that,except in special cir
cumstances, prevailing plaintiffs in Title VII cases ordinarily
are entitled to recover counsel fees under § 706(k). Christiansburg
Garment Co. v. EEOC, 434 U.S. 412, 420-22 (1978). The same
standards apply to awards to prevailing plaintiffs in § 1981
and § 1983 employment discrimination cases under the 1976 Act.
See S. Rep. No. 94-1011, 94th Cong., 2d Sess. 4 (1976). There is
no Eleventh Amendment bar to an award of counsel fees against
state or local government defendants under either Title VII,
Fitzpatrick v. Bitzer, 427 U.S. 445 (1976) , or the 1976 Act,
Hutto v. Finney, 46 U.S.L.W. 4817 (June 23, 1973).
The Supreme •Court has recognized that in civil rights cases
" [a] district court must have discretion to award fees and costs
incident to the final disposition of interim matters." Bradley
v. Richmond School Board, 416 U.S. 696, 723 (1974). Such awards
may appropriately be made upon 'fthe entry of any order that de
termines substantial rights of the parties." Id. at 722 n. 28.
In employment discrimination cases, interim counsel fee awards
have been made under both Title VII and the 1976 Act. See James
v . Stockham Valves & Fittings Co., 559 F.2d 310, 358-59 and n. 63
118
(5th Cir. 1977), cert, denied, 434 U.S. 1034 (1978); Johnson v.
Ryder Truck Lines, Inc., 15 CCS E.P.D.l 7969 (W.D.N.C. 1977);
Commonwealth of Pennsylvania v. O'Neill, 14 CCH E.P.D. 11 7699
(E.D. Pa. 1977); Chandler v. Roudebush, 13 CCH E.P.D. 1[ 11,574
(C.D. Cal. 1977); Lewis v. Philip Morris, Tnc,, 13 CCH E.P.D.
11 11,350 (E.D. Va. 1976); Patterson v. American Tobacco Co.,
9 CCH E.P.D. 1[ 10,039 (E.D. Va. 1975), As the Fifth Circuit
said in James v. Stockham Valves, supra,
There is a danger that litigants will
be discouraged from bringing such suits
because of the risks of protracted liti
gation and the extended financial drain
represented by such a risk. An award of
interim attorneys' fees will prevent ex
treme cash-flow problems for plaintiffs
and their attorneys. 559 F .2d at 358-59,
Moreover, in enacting the Civil Rights Attorneys' Fees
Awards Act of 1976, both the. Senate and the House specifically
approved interim awards. The Senate Committee report stated that,
In appropriate circumstances, counsel fees
under S. 2278 may be awarded pendente lite.
See Bradley v. School Board of the City of
Richmond, 416 U.S. 696 (1974). Such awards
are especially appropriate where a party
has prevailed on an important matter in
the course of litigation, even when he
ultimately does not prevail on all issues.
See Bradley, supra; Mills v. Electric
Auto—Lite Co., 396 U.S. 375 (1970). S, Rep,
94-1011, 94th Cong., 2d Sess. 5 (1976).
Similarly,the House committee report stated that
the word "prevailing" is not intended to
require the entry of a final order before
fees may be recovered. . . . Such awards
pendente lite are particularly important
in protracted litigation, where it is^
difficult to predicate with any certainty
the date upon which a final order will be
119
entered. . .H. R. Rep. No. 94-1558, 94th Cong.,
2d Sess. 8 (1976), (emphasis in original).
The Court should make an interim award of attorneys' fees
to counsel for the plaintiffs in this case. This action has
been before the Court for more than four years. Many issues,
including the claims of discrimination in training, assignment,
promotion and discipline, have yet to be tried. Even after the
Court decides the issues as to defendants’* liability for dis
crimination in recruitment and hiring and as to the appropriate
affirmative hiring relief, many issues will remain to be decided
as to other classwide injunctive relief as well as back pay,
retroactive seniority, and other forms of relief for individual
victims of discrimination in recruitment and hiring. See
Teamsters, supra, 431 U.S. at 347-48, 357-71. Assuming that
the Court now decides that defendants have engaged in unlawful
discrimination in recruitment and hiring and enters an affirmative
hiring order, the plaintiffs will have prevailed on matters deter
mining substantial rights of the parties, Bradley, supra, 416
at 722 n.28, but this lengthy and costly litigation will be far
i
from over. In these circumstances, interim 'counsel fees
should be awarded at this, time, the amount of such award to be
deducted from the sum finally awarded.for attorneys' fees for
the full course of the litigation. ■ If an interim award is not
made, the defendants "may be tempted to seek victory through
an economic war of attrition against the plaintiffs." James
v. Stockham Valves, supra, 559 F.2d at 359.
120
CONCLUSION
For the reasons stated above, plaintiffs urge the Court to
grant the relief requested.
Respectfully submitted,
''Q. --------
WILLIAM H. ALLISON, JR.
PAUL SOREFF
3208 West Broadway
Louisville, Kentucky 40211
JUANITA LOGAN CHRISTIAN
Suite 240, Hart Block Building
730 West Main Street
Louisville, Kentucky 40202
FREDERIC J. COWDEN
1300 West Broadway
Louisville, Kentucky 40203
JACK GREENBERG
KRISTINE S . KNAPLUND
PATRICK O. PATTERSON
10 Columbus Circle
Suite 2030
New York, New York 10019 .
Attorneys for Plaintiffs
Dated: September 13, 1978
. New York, New York
121
UNITED STATES DISTRICT COURT
WESTERN DISTRICT OF KENTUCKY
AT LOUISVILLE
LOUISVILLE BLACK POLICE OFFICERS
ORGANIZATION, INC., et al., :
Plaintiffs, :
CIVIL ACTION NO.
: C 74-106 L (A)vs .
CITY OF LOUISVILLE, et al.,
Defendants.
AFFIDAVIT OF JOSHUA TANKSL
STATE OF NEW YORK )
) ss. :
COUNTY OF NEW YORK )
I, JOSHUA TANKEL, first being duly sworn, hereby state as
follows:
1. I am employed by the N.A.A.C.P. Legal Defense and
Educational Fund, 10 Columbus Circle, New York, New York 10019,
as a research analyst/statistician.
2. I reviewed Plaintiffs' Exhibit #11, eligibility lists;
Plaintiffs' Exhibit #71, log books lrlO; and'Plaintiffs' Exhibit
#72, examination results lists, relating to this case.
3. There are many parts of these data that are illegible,
such as some applicants' names, the positions applied for, the
test scores, and racial identifications.
4. On information and belief, the log books theoretically
list all applicants for positions of employment with the City of
Louisville and identify which position the person applied for.
Additionally, beginning with log book #6, starting on July 10,
1973, there is a racial identification of the applicants.
5. On information and belief, the examination results lists
(Plaintiffs' Exhibit #72) ("ERLs") are supposed to contain the
name of each applicant who took the written examination for a
position of employment with the City of Louisville, to identify
the position for which position he or she applied, and to report
his or her test scores.
6. Many of the names and positions applied for in both the
log books and the ERLs are illegible. Additionally, many of the
applicants' test scores and racial identifications are either
illegible or non-existent.
7. In order to determine from these data the impact upon
black applicants of a selection procedure, such as a written test,
it is necessary to cross-reference the test taker's racial identi
fication contained in the log book with the corresponding appli
cant's name in the ERL where the score is reported.
8. In attempting to correlate the applicants listed in the
log books with the applicants in the ERLs, I discovered that some
applicants who were listed in the log books as having successfully,
completed the application process for police patrolman jobs were
not on the ERLs; and that some people who were listed on the ERLs
as having taken the written examination for those jobs were nowhere
to be found in the log books.
2
9. Some examples of the lack of correspondence between the
ERLs and the log books are:
a. William Payton (ERL 4/2/74) not in Applicants' log;
b. James Seng (ERL 2/5/74) not in Applicants' log;
c. Roger Savage (ERL 2/18/74) not in Applicants' log;
d. Dwight Williams (ERL 2/18/74) not in Applicants' log;
e. Gary D. Wall (Applicants’ log 7/12/73) not in ERL;
f. Edward T. Livers (Applicants' log 7/12/73) not in ERL.
Thera are many more such examples too numerous to detail contained
in the log books and ERLs.
10. Because of the numerous instances of illegibility, the
lack of test scores and racial identifications for many of the
applicants, and the lack of internal consistency of the data as
demonstrated by the fact that the applicants listed in the ERLs
do not match the applicants listed in the log books, I determined
that the data provided by the defendants and contained in plain
tiffs' Exhibits 11, 71, and 72 are so unreliable, inaccurate, and
incomplete that they cannot be used to make an adequate determina
tion of the impact of the written test or other selection procedures
upon black applicants. For these same reasons I was unable to
verify the data contained in Defendants' Exhibit #62.
JOSHUA TANKSL/
Subscribed and sworn to before
1978
Notary Public
CERTIFICATE OF SERVICE
I hereby certify that copies of the foregoing Plaintiffs'
Post-Trial Brief were served by depositing same in the
United States mail, postage prepaid, this 13th day of
September, 1978, addressed as follows:
Sallie S. Haynes, Esq.
Laurence J. Zielke, Esq.
City Law Department
200 City Hall
Louisville, Kentucky 40202
Richard Frockt, Esq.
1417 Citizens Plaza
Louisville, Kentucky 40202
Patrick 0. Patterson