Nash v. Consolidated City of Jacksonville

895 F. Supp. 1536, 1995 U.S. Dist. LEXIS 16397, 1995 WL 500293

• Court: District Court, M.D. Florida
• Decided: August 4, 1995
• Docket: 92-903-Civ-J-20
• Status: Published

Opinion

MEMORANDUM OPINION

SCHLESINGER, District Judge.

Trial in this cause was held before the undersigned. Based on the testimony and evidence received during trial, and the applicable legal standards, the Court makes the following findings of fact and conclusions of law, as required by Rule 52(a) of the Federal Rules of Civil Procedure.

Plaintiff, Winston F. Nash, an African-American male, is employed by Defendant, the Consolidated City of Jacksonville (“City”) in the Department of Public Safety, Fire Department (“Fire Department”). The City is a municipal corporation. Defendant Local 122, International Association of Firefighter (“Local 122”), an association, was allowed to intervene in this action on October 26, 1993. See Order (Doc. No. 24) dated October 26, 1993.

This is a race discrimination case brought pursuant to Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000(e), et seq., and 42 U.S.C. §§ 1981 and 1983. See Pretrial Stipulation (Doc. No. 28) at 1. Plaintiff alleges that the denial of his promotion to the position of captain constituted unlawful race discrimination since Plaintiff was qualified to be promoted to captain and all such promotions went to white males. This case is based upon two theories of recovery, a disparate impact theory and a disparate treatment theory.

Plaintiff has been employed with the City in the Fire Department since November 17, 1972. Plaintiff is a certified firefighter and a certified emergency medical technician in the State of Florida. Plaintiff is currently a fire combat lieutenant (“lieutenant”). Plaintiff also has a masters degree in management supervision from Central Michigan University, a bachelor’s degree in business administration from the University of North Florida, an associate’s arts degree in fire science from Florida Community College (“FCC”) and an associate’s arts degree in business administration from FCC.

Plaintiff applied for promotion to fire combat captain (“captain”), the rank immediately above that of lieutenant. Plaintiff met all of the requirements for eligibility and on October 26, 1990, Plaintiff took the fire combat captain promotional examination (“1990 exam”). This was a written examination.

If an applicant passed the 1990 exam by scoring 70 or higher, seniority credits and veteran preference points were added to the test score to come up with a final score. Plaintiff is not challenging the seniority points or veteran points added to the test score. 1

Alter the final score was calculated, the candidates were placed on an eligibility list in which the candidates were ranked according to their final scores. When an opening occurred, the top ranked candidate on the list, i.e., the candidate with the highest final score, received the promotion. This is known as the rule of one, or alternatively, as rank-ordering. The eligibility list remained open for two years, from 1990 to 1992.

The rule of one is used throughout every department in the City for promotions. However, the rule of three is used when an elected official needs to fill a position immediately below him or her. The rule of three allows some discretion in hiring, but the person hired must be ranked in the top three.

After a person was promoted, there was a six month probationary period in which the recent promotee was evaluated in the new job. Also, those items which were determined not to be testable by written exam were looked at during the six month probationary period. Failure to perform satisfactorily during the probationary period would constitute cause to revert the promotee back to the previous position. Nobody was reverted back to lieutenant after being promoted to captain. The only demotion anyone could recall was back in the 1970’s.

Plaintiff and Larry Norris were the only two African-Americans out of the ninety-four applicants to take the 1990 exam. Ninety one of the applicants passed the exam. The three applicants who failed the 1990 exam were white males. Neither of the two African-American applicants were selected for promotion. The thirty applicants that were promoted from the eligibility list were all white.

Plaintiff is challenging both the 1990 exam and the rule of one.

Plaintiff’s raw score on the 1990 exam was 82.653. See Defendants’ Ex. No. 19. In terms of raw score, this ranked Plaintiff eighty-sixth. Out of the 94 people who took the test, 85 of them scored higher than Plaintiff.

Norris received a raw score of 94.898, ranking him twenty-third.

Plaintiff then received 9.389 seniority points, out of a possible 10, bringing his total score to 92.042. Mr. Norris, on the other hand, only received 6.729 seniority points, bringing his total score to 101.627.

Question number one was the only question out of the 98 allowed questions on the 1990 exam (two questions were thrown out) which both African-American candidates scored incorrectly. However, question number one had the highest rate of incorrect answers with 68% of the examinees getting it wrong.

On the Promotional Eligibility List (Defendants’ Ex. No. 20), which rank-ordered the candidates based on their final scores, Norris ranked 44th and Plaintiff ranked 84th. Since there were only thirty promotions before the Promotional Eligibility List expired, neither Plaintiff nor Norris were promoted to captain.

Plaintiff and Norris were not treated differently than any other person who participated in the promotional process.

The Fire Department is a paramilitary organization, that is, there is a rank structure in which superior officers give orders to subordinates. These orders cannot be disobeyed. The following are the ranks in the Fire Department, from lowest to highest: firefighter (entry level); engineer; lieutenant; captain; district chief; and battalion chief (senior district chief). Then there are the politically appointed positions.

There are three shifts at a fire station. Each shift is led by an officer. A captain is on one shift and the other two shifts have a lieutenant as the lead officer.

A normal full assignment at a fire scene is two engine companies and one ladder company. The Fire Department tries to arrange it so that at least one captain will be present for normal full assignment. To attempt to ensure this, the Fire Department tries to have the rotation so that when everyone shows up at a fire there is at least one captain on the scene

A captain is in charge of the station house. A captain makes the decisions on logistical matters. Even when the captain has the day off, the captain is ultimately responsible for what transpires at the station and acts through his subordinate officers, the lieutenants.

The captain supervises not only his own shift but also the other two shifts. A captain meets with one lieutenant when the captain is relieved in the morning, and with the other lieutenant when the captain relieves him or her the following morning, and they discuss things that have happened and things that need to be done, and the captain makes the necessary decisions. A captain will also leave messages, i.e., orders, when he wants certain things done. A captain does not have to physically be there for his orders to be carried out. He, of course, does not supervise the other shifts at fire scenes because he is not present.

The captain sets the policy at the station house within the rules and regulations, is responsible for the scheduling at the station house, decides where things go, e.g., hoses, first aid boxes, etc., takes responsibility for harmony in the station house, and is responsible for ordering and inventorying supplies. The captain is the one ultimately responsible for the apparatus, i.e., the fire vehicle. This includes maintenance of the vehicle. Only a captain can do the paperwork to permanently transfer someone. Also, a captain initially authorizes leave and captains are first in line in the grievance process.

As for disciplinary matters, the general rule in the Fire Department is that discipline is handled at the lowest level, if possible. Thus, the lieutenant will handle the discipline on his shift if possible. If it is of a magnitude where the lieutenant cannot handle the problem, the captain will become involved. If it still needs to be referred further up the line, the captain will refer it to the district chief.

Generally, whichever officer gets to a fire scene first, that is, the officer with the first company to arrive, is in charge, and that officer is the Incident Commander, and remains as such until relieved by a senior officer. The senior officer has to tell the Incident Commander he or she is relieving him or her and broadcast it over the radio so that everybody knows a change of command has occurred.

If a lieutenant arrives first, the lieutenant will be the incident commander. When the captain subsequently arrives, the captain is' in charge if he or she relieves the lieutenant, otherwise the lieutenant remains in charge as the incident commander. There was conflicting testimony as to whether the captain must take over if the Incident Commander is a lieutenant. It appears to be in the captain’s discretion. In the downtown area, the general practice is for the first to arrive to remain as the incident commander until the chief arrives. This way, time is not wasted changing command.

In rare circumstances, an engineer may even be the Incident Commander. If an officer has the day off and is not replaced by a roving officer, then the engineer on the engine may be in charge for that shift at lieutenant’s pay. If that engineer is the first to arrive at a fire scene, then the engineer becomes the Incident Commander and remains as such until relieved. The engineer will be riding as a lieutenant on that day since the lieutenant is off.

The difference between a captain and a lieutenant is a matter of degree, with the primary difference being custodial duties.

I. DISPARATE IMPACT

Disparate impact cases challenge employment practices that are facially neutral in their treatment of different groups, but in fact fall more harshly on one group than another. See Int’l Bhd. of Teamsters v. United States, 431 U.S. 324, 335 n. 15, 97 S.Ct. 1843, 1854 n. 15, 52 L.Ed.2d 396 (1977). “[G]ood intent or absence of discriminatory intent does not redeem employment procedures or testing mechanisms that operate as ‘built in headwinds’ for minority groups and are unrelated to measuring job capability....” Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971). Title VII “proscribes not only overt discrimination but also practices that are fair in form, but discriminatory in operation.” Id. at 432, 91 S.Ct. at 854. A disparate impact theory is much different from a disparate treatment theory in that it does not require a showing of animus on the employer’s part. See Powers v. Alabama Dep’t of Educ., 854 F.2d 1285, 1291 (11th Cir.1988), cert. denied, 490 U.S. 1107, 109 S.Ct. 3158, 104 L.Ed.2d 1021 (1989) (citing Dothard v. Rawlinson, 433 U.S. 321, 328, 97 S.Ct. 2720, 2726, 53 L.Ed.2d 786 (1977); Griggs, 401 U.S. at 430-32, 91 S.Ct. at 853-54). That is, proof of intent is not required. See Mitchell v. Jefferson County Bd. of Educ., 936 F.2d 539, 546 (11th Cir.1991).

In a disparate impact case, the plaintiff has the initial burden of establishing a prima facie case by showing that the employment practices complained of “select applicants for hire or promotion in a racial pattern significantly different from that of the pool of applicants.” Albemarle Paper Co. v. Moody, 422 U.S. 405, 425, 95 S.Ct. 2362, 2375, 45 L.Ed.2d 280 (1975). The burden then shifts to the employer to demonstrate that the tests are “job related.” Id. The plaintiff then has the opportunity to show that “other tests or selection devices, without a similarly undesirable racial effect, would also serve the employer’s legitimate interest in ‘efficient and trustworthy workmanship,’ ” which would be “evidence that the employer was using its tests merely as a ‘pretext’ for discrimination.” Id. (citation omitted).

Unlike in disparate treatment cases, the similarities of sections 1981 and 1983 with Title VII do not exist when liability under Title VII is premised on the disparate impact of a facially neutral employment practice. In such cases, a Title VII violation can be established without a showing of discriminatory motive. However, liability cannot be imposed under section 1981 without proof of intentional discrimination. See General Bldg. Contractors Ass’n, Inc. v. Pennsylvania, 458 U.S. 375, 383 n. 8, 389, 102 S.Ct. 3141, 3146 n. 8, 73 L.Ed.2d 835 (1982); Stallworth v. Shuler, 777 F.2d 1431, 1433 (11th Cir.1985); Lincoln v. Bd. of Regents of Univ. Sys. of Georgia, 697 F.2d 928, 935 n. 6 (11th Cir.1983), cert. denied, 464 U.S. 826, 104 S.Ct. 97, 78 L.Ed.2d 102 (1983); Williams v. Dekalb County, 582 F.2d 2 (5th Cir.1978). 2 Similarly, section 1983 also requires discriminatory intent. See Ramirez v. Sloss, 615 F.2d 163, 168 n. 8 (5th Cir.1980); see also Foster v. Wyrick, 823 F.2d 218, 222 (8th Cir.1987) (“a Title VII disparate impact claim may not be asserted in a § 1983 action.”).

Even though proof of disparate impact alone is insufficient to establish a section 1981 or section 1983 violation, see Progressive Officers Club, Inc. v. Metro. Dade County, 1990 WL 270786 at * 4 n. 1 (S.D.Fla.1990) (citing Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976)), disparate impact is a form of evidence with which Plaintiff can demonstrate disparate treatment. See Howard v. B.P. Oil Co. Inc., 32 F.3d 520, 526 (11th Cir.1994). Evidence of a disproportionate impact is relevant to the question of intent since an invidious discriminatory purpose may often be inferred from the totality of the relevant facts.

A PRIMA FACIE CASE

Disparate impact eases typically focus on statistical disparities and on the various explanations for those disparities. To make out a prima facie case under a disparate impact theory of employment discrimination, the complaining party must demonstrate that the defendant employed a facially neutral employment practice that had a significant discriminatory effect. See Stephen v. PGA Sheraton Resort, Ltd., 873 F.2d 276 (11th Cir.1989). “[T]he plaintiffs burden in establishing a prima facie case goes beyond the need to show that there are statistical disparities in the employer’s work force.” Watson v. Ft. Worth Bank & Trust, 487 U.S. 977, 994, 108 S.Ct. 2777, 2788, 101 L.Ed.2d 827 (1988). “[T]he plaintiff must offer statistical evidence of a kind and degree sufficient to show that the practice in question has caused the exclusion of applicants for jobs or promotions because of their membership in a protected group.... [Statistical disparities must be sufficiently substantial that they raise such an inference of causation.” Id. at 994-95, 108 S.Ct. at 2789.

“In the end, a defendant may not be held liable on a claim of disparate impact on the basis of less evidence than is required to prove intentional discrimination.” Armstrong v. Flowers Hosp., Inc., 33 F.3d 1308, 1314 (11th Cir.1994) (citing Watson, 487 U.S. at 987, 108 S.Ct. at 2785). Plaintiff must produce competent evidence which shows that the failure to promote falls disproportionately on African-American employees.

“The plaintiff must first show an imbalance in the workforce through statistics. The proper comparison is ‘between the racial composition of the qualified persons in the labor market and the persons holding at-issue jobs.” Pelli v. Stone Savannah River Pulp and Paper Corp., 878 F.Supp. 1559, 1566 (S.D.Ga.1995) (quoting Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 650, 109 S.Ct. 2115, 2121, 104 L.Ed.2d 733 (1989)). Merely showing an imbalance in the workforce is not enough. See id. at 1567 (citing MacPherson v. Univ. of Montevallo, 922 F.2d 766, 771 (11th Cir.1991)). Plaintiff must establish a “ ‘meaningful statistical comparison’ between those employees eligible for promotion and those actually promoted.” Alford v. City of Montgomery, Alabama, 879 F.Supp. 1143, 1149 (M.D.Ala.1995). When an organization, such as the Fire Department, promotes from within, the relevant labor pool is the actual group of employees from which the organization fills these jobs. See Griffin v. Carlin, 755 F.2d 1516, 1526 (11th Cir.1985).

There is no set mathematical threshold that must be met in order to show significant disparate impact. See Richardson v. Lamar County Bd. of Educ., 729 F.Supp. 806, 817 (M.D.Ala.1989), aff'd, 935 F.2d 1240 (11th Cir.1991) (citing Moore v. Southwestern Bell Tel. Co., 593 F.2d 607, 608 (5th Cir.1979)). Various formulas can be used to measure the degree of impact in a specific case.

The Equal Employment Opportunity Commission (“EEOC”) generally regards a selection rate that is less than %, or 80%, of the rate for the group with the highest rate as an indication of significant adverse impact. See 29 C.F.R. § 1607.4D. “Other courts have stated that, if the difference between the expected value and the observed number is greater than two or three standard deviations, then adverse impact has been shown.” Richardson, 729 F.Supp. at 816 (citing Hazelwood School Dist. v. United States, 433 U.S. 299, 308-11 & nn. 14 & 17, 97 S.Ct. 2736, 2741-43 & nn. 14 & 17, 53 L.Ed.2d 768 (1977); Castaneda v. Partida, 430 U.S. 482, 496 n. 17, 97 S.Ct. 1272, 1281 n. 17, 51 L.Ed.2d 498 (1977)). One commentator has proposed, in cases involving employment tests, using the “test for difference between independent proportions.” Id. (citing Shoben, Differential Pass-Fail Rates in Employment Testing: Statistical Proof Under Title VII, 91 Harv.L.Rev. 793, 797-800 (1981)).

Plaintiff has argued that the EEOC’s % rule was violated because 0% of the African-Americans were promoted and approximately 33% of the white candidates were promoted. Thus, the selection rate, i.e., promotion rate, for African-Americans was less than 80% of the selection rate for whites.

However, if one more African-American was promoted, the African-American promotion rate would have been 50%, bringing it above the white promotion rate. Additionally, the white promotion rate, approximately 33%, would be less than 80% of the African-American promotion rate. Thus, the numbers are too unpredictable to say there is adverse impact. That is, the numbers are not statistically significant.

The Supreme Court has indicated that statistical evidence can be challenged on various grounds, including “small or incomplete data sets and inadequate statistical technique.” Watson, 487 U.S. at 996-97, 108 S.Ct. at 2790. In the instant case, the data set is too small to reach any conclusions since only two African-Americans took the 1990 exam. See, e.g., Martinez v. U.S. Sugar Corp., 880 F.Supp. 773, 781 (M.D.Fla.1995) (six individuals receiving discipline is an incomplete data set); Sims v. Montgomery County Comm’n, 873 F.Supp. 585, 602 (M.D.Ala.1994) (“a sample size may be too small to support a determination of whether there is adverse impact.”) (citing Question 21 of the Questions and Answers to Clarify and Provide a Common Interpretation of the Uniform Guidelines, 44 Fed.Reg. 11,999 (March 2, 1979)). A statistical pool of only two African-American test takers is too shallow for the Court to dive into with a disparate impact analysis.

However, while statistics can play a significant role in employment discrimination cases, “statistics are inherently slippery in nature.” Williams v. Mead Coated Bd., Inc., 836 F.Supp. 1552, 1565 (M.D.Ala.1993), aff'd, 41 F.3d 668 (11th Cir.1994). A plaintiff may also use nonstatistical evidence to establish a prima facie case of disparate impact. See Bunch v. Bullard, 795 F.2d 384, 395 (5th Cir.1986); Page v. U.S. Indus., Inc., 726 F.2d 1038, 1053 (5th Cir.1984). This nonstatistical evidence may include a pattern of past racial discrimination.

Disparate impact as a theory of liability is a means of dealing with the residues of past discrimination. See Armstrong, 33 F.3d at 1317 (a pregnancy discrimination case); Troupe v. May Dept. Stores, 20 F.3d 734, 738 (7th Cir.1994); Finnegan v. Trans World Airlines, Inc., 967 F.2d 1161, 1164 (7th Cir.1992). “The concept of disparate impact was developed for the purpose of identifying situations where, through inertia or insensitivity, companies were following policies that gratuitously — needlessly—although not necessarily deliberately, excluded black or female workers from equal employment opportunities.” Finnegan, 967 F.2d at 1164.

Therefore, besides statistics, the Court should also consider that the pool of African-American applicants eligible for promotion to captain might have been constricted by the City’s past discrimination in hiring and in promoting African-Americans. See U.S. v. Paradise, 480 U.S. 149, 168, 107 S.Ct. 1053, 1065, 94 L.Ed.2d 203 (1987); Sims, 873 F.Supp. at 602; Stuart v. Roache, 951 F.2d 446, 452 (1st Cir.1991), cert. denied, 504 U.S. 913, 112 S.Ct. 1948, 118 L.Ed.2d 553 (1992). In other words, as the Fifth Circuit explained, “the limited size of the population in question [does not preclude] [the Court] from recognizing a decisive pattern emerging from a history of experience.” Bunch, 795 F.2d at 395 (quoting Rivera v. City of Wichita Falls, 665 F.2d 531, 536 n. 7 (5th Cir.1982)).

The City has a history of discriminating against African-Americans in the Fire Department. An example of this is with Plaintiff himself. Plaintiff applied for a promotion to lieutenant, his current rank, in 1976, 1978 and 1981. The City required, as it does now, a score of 70 or higher on the written promotional examination in order to be considered for promotion. The candidates were then rank-ordered based on their scores, as in this case. Nash scored an 84 on the test in 1976 and was placed on the promotional eligibility list. The list expired before he was promoted. In 1978 and 1981, Nash failed the promotional examination. See Nash v. Consolidated City of Jacksonville, Duval County, Fla., 837 F.2d 1534, 1535 (11th Cir.1988), vacated by, 490 U.S. 1103, 109 S.Ct. 3151, 104 L.Ed.2d 1015 (1989), reinstated by, 905 F.2d 355 (11th Cir.1990), cert. denied, 498 U.S. 1089, 111 S.Ct. 967, 112 L.Ed.2d 1054 (1991). Nash then sued the City for discrimination. In the previous Nash case, three out of 20 African-Americans passed while 57 out of 100 whites passed. This violation of the % rule, in conjunction with additional expert statistical evidence, resulted in Nash making out a prima facie case of discriminatory impact. Nash eventually prevailed in his suit against the City and the Court ordered that he be promoted to lieutenant. There were no African-American lieutenants until Nash was promoted.

The results of the 1988 lieutenant promotional eligibility list, see Plaintiffs Ex. No. 66, are as follows:

20 people were promoted to lieutenant— one African-American and 19 whites. There were 116 people on the eligibility list — sixteen black and one hundred whites. Thus 17.24% of the people on the eligibility list were promoted — 6.25% of the blacks and 19% of the whites. This violates the % rule because the African-American promotional rate is only 32.89% (6.25/19) of the white promotional rate.

There are currently 56 fire combat captains, none of which are from African-American ancestry. There are 17 fire combat district chiefs, none of which are African-Amer ican. There are 125 fire combat lieutenants, 11 of them are African-Americans. However, in 1990, when Plaintiff sat for the 1990 exam, there were only two African-American lieutenants. 3 In 1991, 24.35% of the City was comprised of African-American citizens. Out of 251 engineers, thirty-eight are African-American.

Even though the Court finds that Nash cannot make out a prima facie case on statistics alone, he can meet his burden because of past discrimination in the Fire Department. After all, the reason only two African-Americans took the 1990 exam was because there were so few African-Americans eligible to take that promotional exam because of past discrimination. “Therefore, any method of measuring adverse impact ... based on the number of applicants would ... inaccurately reflect[ ] the history of discrimination.” Sims, 873 F.Supp. at 602. The Court finds that “a lower selection rate continued over a period of time, so as to constitute a pattern,” id, at 602, and that the City’s past discrimination has resulted in “a departmental hierarchy dominated exclusively by non-minorities.” Paradise, 480 U.S. at 168, 107 S.Ct. at 1065. See, e.g., Ensley Branch, N.A.A.C.P. v. Seibels, 31 F.3d 1548, 1564 (11th Cir.1994), (the “Board’s discrimination against blacks seeking entry-level police and firefighter jobs made it almost inevitable that the effects of discrimination had worked their way up to taint the City and Board promotional positions, too.”); Sims, 873 F.Supp. 585 (the court found that for the 15 years between 1973 and 1988 the racial pattern was self-evident — no African-Americans were promoted or transferred to any supervisory position in the enforcement division).

The Court finds that Plaintiff has made out a prima facie case of disparate impact. 4

B. CITY’S BURDEN

Since Plaintiff has made out a prima facie case, the burden now shifts to Defendants to show “business necessity.” The question arises, though, as to what burden Defendants have, a burden of production or a burden of persuasion. This confusion arises because the captain’s promotional eligibility list in question remained open from 1990-1992, and the law changed in 1991. However, since the Court finds that Defendants have met the higher burden imposed by the Civil Rights Act of 1991, this issue need not be decided.

Prior to the Supreme Court’s 1989 decision in Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 109 S.Ct. 2115, 104 L.Ed.2d 733 (1989), defendants “bore the burden of proof and risk of nonpersuasion” to show that the challenged business practice was a business necessity. Fitzpatrick v. City of Atlanta, 2 F.3d 1112, 1117 n. 5 (11th Cir.1993) (citing Dothard, 433 U.S. at 329, 97 S.Ct. at 2726-27; Griggs, 401 U.S. at 432, 91 S.Ct. at 854).

However, the Supreme Court changed the law with Wards Cove, holding that the defendant bore only the burden of coming forward with an alleged business-related justification for the challenged practice which the plaintiff would then have to disprove in order to prevail. See Wards Cove, 490 U.S. at 659, 109 S.Ct. at 2126. The Supreme Court in Wards Cove held that the defendant’s burden on the “business necessity” defense is only one of production — with the burden of persuasion remaining at all times with the plaintiff. See id. at 659, 109 S.Ct. at 2126. The Court also broadened the scope of the business necessity defense by holding that practices causing a disparate impact were permis sible, even if they could not be shown to be absolutely necessary, so long as they “served, in a significant way, the legitimate employment goals of the employer.” Fitzpatrick, 2 F.3d at 1117-18 n. 5. “These changes were statutorily reversed by the Civil Rights Act of 1991, Pub.L. No. 102-166, § 105(a), 105 Stat. 1071, 1074-75 (codified at 42 U.S.C. §§ 2000e-2(k)(1)(A)).” Id.

Section 105 of the Civil Rights Act of 1991 reversed Wards Cove by adding a new amendment, codified as 42 U.S.C. § 2000e-2(k)(1), restoring pre-Wards Cove law in which, inter alia, “business necessity” is an affirmative defense on which the defendant bears the burden of persuasion and “business necessity” really means “necessity.” Now, after a prima facie ease of disparate impact is put forth, the employer must prove that the identified practice bears a “manifest relationship” to the employment in question. See New York City Transit Auth. v. Beazer, 440 U.S. 568, 587 n. 31, 99 S.Ct. 1355, 1366 n. 31, 59 L.Ed.2d 587 (1979); Stephen, 873 F.2d at 279 (quoting Connecticut v. Teal, 457 U.S. 440, 446, 102 S.Ct. 2525, 2530, 73 L.Ed.2d 130 (1982) (citing Griggs, 401 U.S. at 430-32, 91 S.Ct. at 853-54)). The employer now bears the burden of proving that the challenged practice is “job-related.” Powers, 854 F.2d at 1292 n. 11; see Albemarle Paper Co., 422 U.S. at 425, 95 S.Ct. at 2375; Craig v. Alabama State Univ., 804 F.2d 682, 689 (11th Cir.1986); Crawford v. Western Elec. Co., Inc., 745 F.2d 1373, 1384 (11th Cir.1984).

Courts look at whether the challenged procedure measures “important skills, abilities and knowledges that are necessary for the successful performance of the job.” E.E.O.C. v. Atlas Paper Box Co., 868 F.2d 1487, 1490 (6th Cir.), cert. denied, 493 U.S. 814, 110 S.Ct. 63, 107 L.Ed.2d 30 (1989). Under the stricter standard imposed by the Civil Rights Act of 1991, “the business purpose must be sufficiently compelling to override any racial impact; the challenged practice must effectively carry out the business purpose it is alleged to serve; and there must be available no acceptable alternative policies or practices which would better accomplish the business purpose advanced or accomplish it equally well with less differential racial impact.” Craig, 804 F.2d at 689 (citing Pettway v. American Cast Iron Pipe Co., 494 F.2d 211, 245 (5th Cir.1974)).

The business necessity doctrine is very narrow. See Parson v. Kaiser Aluminum & Chem. Corp., 575 F.2d 1374, 1389 (5th Cir.1978). An employer does not meet its burden of establishing the job relatedness of a test by merely showing a rational basis for the test. See Fisher v. Procter & Gamble Mfg. Co., 613 F.2d 527, 544 (5th Cir.1980), cert. denied, 449 U.S. 1115, 101 S.Ct. 929, 66 L.Ed.2d 845 (1981). It must be validated by a demonstration that the examination is “appropriate for the selection of qualified applicants for the job in question.” Washington v. Davis, 426 U.S. 229, 247, 96 S.Ct. 2040, 2051, 48 L.Ed.2d 597 (1976).

1. Content Validity

Under the Uniform Guidelines, there are three ways to test the validity of a selection device: criterion validation, content validation and construct validation. For the 1990 exam, the City chose content validation. To show the content validity of the 1990 exam, the City needs to prove that the know-ledges, skills and abilities covered in the 1990 exam were the knowledges, skills and abilities required for the job.

The City hired Robert Anderson, an industrial psychologist, to ensure that the examination development process conformed to the Uniform Guidelines and resulted in a valid examination. Robert B. Anderson worked for the City from 1988 to 1992. His title was Industrial Psychologist. Dr. Anderson provided the technical supervision for the development of the 1990 exam. However, the City cannot escape liability just because its test was developed by Dr. Anderson. See, e.g., James v. Stockham Valves & Fittings Co., 559 F.2d 310 (5th Cir.1977) (employer could be held liable for test developed by consultant).

Dr. Anderson performed a job-analysis for the 1990 exam.

First, Dr. Anderson personally visited fire stations to observe how they operated. He also interviewed captains and district chiefs (who supervise captains) and reviewed the class specification for captain with them. Delaney Williams and Vicki Demiranda, also made a few information gathering visits to fire stations for Dr. Anderson. Next, the subject matter experts (i.e., captains and district chiefs) came to Anderson’s offices in City Hall. The visits to the fire stations, in conjunction with the interviews of the subject matter experts at City Hall, enabled Anderson to determine the job duties of a captain and the knowledges necessary to perform those duties.

Next, Anderson drafted a list of task statements describing the various work behaviors of captains. Anderson then examined, with the subject matter experts, the task statements by systematically examining each of the work behaviors. This was the “task editing” stage.

Next, “task screening” was performed, where those work behaviors “critical” to the performance of a captain’s duties were determined. A work behavior was considered critical if it was something a captain would be counseled or disciplined for if he or she did not perform it adequately.

The next step in the job analysis was to determine if a newly promoted captain should be expected to perform certain tasks on entry into the position, that is, is the task something a newly appointed captain could perform with nothing other then a brief orientation. If it was a task a captain was expected to be able to perform only after learning on the job, then it was eliminated.

At this point, Anderson had a pool of critical tasks. The non-critical tasks were eliminated. A committee of subject matter experts was then appointed to rank order these critical tasks. The committee consisted of five to seven people. After the committee ranked the critical tasks, weights were then given to each critical task, based on their relative importance to one another.

After the critical tasks were weighted, the next stage in the job analysis was to go through a somewhat similar process for the knowledges, skills and abilities (“KSAs”) considered to be essential and necessary for the position of captain. The KSAs were then linked up with the specific tasks. This was done by the committee of subject matter experts. It was determined which KSAs were necessary to perform a particular task. The committee then weighted the KSAs given to a particular task in order of importance. This was done for each critical task.

At this point, Dr. Anderson had the relative importance of both the critical tasks and the KSAs for each of those tasks. To get a test plan, Dr. Anderson determined the relative importance of the KSAs for the entire test by multiplying each KSA weight for the task with the task weight. The goal was to develop a test plan that gave a representative sample of the KSAs necessary to perform the functions of captain.

After the KSA weights were worked out, a preliminary reading list allowing one to find the knowledge needed was developed. This would eventually be made public so that all of the test takers could study for the examination.

Next, questions for the test were developed. Teams of subject matter experts were organized to write the test questions for the different topics. Approximately six teams, with two people on a team, were used. The subject matter experts wrote the exam since Anderson believed that they were the best qualified to determine the relevancy, i.e., job relatedness, of the questions. As questions were developed, Dr. Anderson would review them with the various teams and discuss the questions and he would point out and fix technical problems. The process involved continuous revisions of questions and the reading list. It was always ensured that the knowledge being tested could be found in the reading list.

When the test questions were narrowed, Dr. Anderson had the subject matter experts link the questions back to the KSAs to ensure the job-relatedness of the questions.

After Anderson selected the questions for the test, the reading list was finalized and published. The exam was announced around the same time.

The development of the 1990 exam took approximately three months.

The makeup of the 1990 exam was as follows: there were 22 questions on the abili ty to read, understand and apply Fire Division Rules and Regulations, Civil Service and Personnel Rules, and collective bargaining agreements; 14 questions on the knowledge of fire fighting equipment such as hoses, ladders and ventilation equipment; 15 questions on the knowledge of fire fighting vehicles and apparatus, tankers, ladders and engines; 12 questions on the knowledge of techniques and procedures for finding, containing and extinguishing fires; 7 questions on the knowledge of techniques and procedures of overhaul and salvage operations; 13 questions on the knowledge of basic life support systems such as CPR and oxygen inhalation; 10 questions on the knowledge to apply fire combat standard operating procedures; and 7 questions on the knowledge of building construction and fire protection equipment and systems.

After the test was given, and before it was scored, there was a protest period of ten days for test takers to review the answer key so that they could complain about any questions they thought unfair, or those they thought had multiple answers, or those questions they thought were not job related. A protest panel, made up of one personnel examiner and four subject matter experts, two of which were selected by the candidates, reviewed the protests. On the 1990 exam, the exam in question, protests were filed. Two questions were excluded and on two other questions two answers for each were acceptable. This shows that the protest period was not just a sham but validly used for the purpose intended.

Throughout his testimony, Dr. Shapiro, Plaintiffs expert, stressed how the lieutenant and captain examinations are similar. However, the Court does not believe that Dr. Shapiro ever actually compared the two by looking at the questions and reading them. On cross-examination, Dr. Shapiro admitted that he did not compare the captain’s and lieutenant’s exams by looking at the wording of the questions. He just did a statistical analysis and looked at the material required to pass the test.

Dr. Shapiro testified that the captain and lieutenant promotional exams are virtually indistinguishable with regard to the materials covered. Yet, the 1990 exam required reading eleven chapters from the Fire Chiefs Handbook, whereas the 1991 lieutenant’s promotional exam only required a reading of Chapter 7 of the Fire Chiefs Handbook. Dr. Shapiro stated that he did not know whether another reading source on the lieutenant’s exam covered the same material in the Fire Chiefs Handbook. On the 1990 exam, 49 questions were from the Fire Chiefs Handi-book. The 1991 lieutenant’s exam had zero questions from the Fire Chiefs Handbook.

The Fire Chiefs Handbook is a more knowledgeable, comprehensive book than IFSTA (First Responder and Essentials of Firefighting), which is used on the lieutenant’s exam. The Fire Chiefs Handbook and IFSTA clearly parallel each other, but the Fire Chiefs Handbook is in more detail. IFSTA is a basic fireperson’s handbook, and the Fire Chiefs Handbook is more advanced study material.

Both Dr. Barrett and Dr. Greenwood reviewed Dr. Anderson’s approach in preparing the 1990 exam, examined all the materials in the City’s personnel file concerning the process and concluded that the 1990 exam was content-valid. Dr. Barrett went back and read each item on the test and looked at the sources. It was evident to him that the questions tapped relevant knowledge and that the 1990 exam was content valid.

The Court finds that a thorough job analysis was accomplished and critical work behaviors were identified. From these work behaviors, important knowledges were gleaned which were tied back to sources which contained that knowledge. From these sources, questions were written to tap the knowledge base which were all tied back again into the work behaviors. This was done by people who were subject matter experts (fire captains and district chiefs) and Dr. Anderson personally reviewed each and every question to be sure there was no bias of any sort and to ensure that each item was relevant to the job.

Dr. Shapiro testified that there was no relationship between the job analysis and the test questions. The Court disagrees. The Court finds that a job task analysis in con formity with the Uniform Guidelines was undertaken. After personally reviewing the questions on the 1990 exam, 5 and after listening to all of the testimony in this case, the Court is convinced that the 1990 exam was linked to the knowledges one needs to be a captain. The City has met its burden of persuasion that the 1990 test was job-related and had a manifest relationship to the position of captain.

2. Reliability

According to Dr. Shapiro, the 1990 exam is not reliable and therefore it cannot be valid. Reliability is “the extent to which the exam would produce consistent results if applicants repeatedly took it or similar tests.” Guardians Ass’n of New York City Police Dep’t, Inc. v. Civil Service Comm’n of City of New York, 630 F.2d 79, 101 (2d Cir.1980), cert. denied, 452 U.S. 940, 101 S.Ct. 3083, 69 L.Ed.2d 954 (1981).

Dr. Shapiro did an “item analysis” of the test, which he ran on a computer program called “Reliability,” which is in the computer package known as “SPSS,” which is short for “Statistical Package for the Social Sciences.” See Plaintiffs Exhibit No. 1. It is a computer analysis designed to analyze the relationships among the test items and the relationships between the individual items and the test as a whole. “Reliability” does not deal with the validity of the test, that is, whether the test relates to the job in question. The program that Dr. Shapiro ran simply deals with the internal consistency of the test performance.

According to Dr. Shapiro, the test is not valid because there are too many negative correlations between questions. But the 1990 exam was a comprehensive test, where one question may pertain to fire hoses and another with calculating overtime pay. Common sense indicates that they would not relate, or correlate, with each other. Additionally, Dr. Greenwood testified that the program Dr. Shapiro ran was “totally inappropriate” for the 1990 exam because SSPS Reliability should only be done on a homogeneous test, that is, a test that measures the same traits throughout. However, the 1990 captain promotional exam was a heteroge-nous test, i.e., it tested different knowledges.

Dr. Shapiro also questions how it can be that people who know more about one thing know less about another when the whole test is for the same occupation. This, however, is flawed logic. Everybody has strengths and weaknesses.

The “alpha value” on Dr. Shapiro’s computer run statistical analysis is 0.8922. However, Dr. Shapiro says this alpha value is meaningless because the computer program stated, “warning ... statistics based on inverse matrix for scale alpha are meaningless....”

Dr. Greenwood ran a different reliability test than the one Dr. Shapiro used. Dr. Greenwood ran the Kuder-Riehardson Formula 20 test. This test actually looks at the top numbers of people and the correct answers they indicated and it does the same for the lower portion of people. Dr. Greenwood looked at the top 27% of the test scorers and how many questions they actually answered correctly and the lower 27% of the test scorers and the questions they answered correctly and compared the two to come up with a reliability coefficient. The reliability coefficient for the 1990 exam was 0.87. A reliability coefficient can range between zero and one. Anything above 0.70 is considered to be reliable. 0.87 indicates that the 1990 exam was highly internally reliable. Dr. Barrett also said that it was a reliable test.

The Court finds that the 1990 exam was a highly reliable test.

Supervisory skills were not tested on the 1990 exam. The 1990 exam was a knowledge based test. Supervision is a construct, that is, it is presumed to exist because of its impact or effect, but cannot be directly observed. However, work behaviors that comprise supervision were defined, it was determined which ones were critical, and the knowledge necessary to perform those work behaviors were assessed.

During Dr. Shapiro’s rebuttal testimony, he said the 1990 exam was not content valid because the 1990 exam omitted questions on supervision. The Court finds Dr. Shapiro’s opinion questionable. The Uniform Guidelines state that “a content strategy is not appropriate for demonstrating the validity of selection procedures which purport to measure traits or constructs, such as intelligence aptitude, personality, commonsense, judgment, leadership, and spatial ability.” 29 C.F.R. § 1607.14C(1). Additionally, in the previous Nash case, the City admitted that “supervisory skills ... cannot be tested by written examination.” Nash, 837 F.2d at 1535.

The Court also questions Dr. Shapiro’s opinion because, as will be discussed shortly, he used the wrong numbers when discussing the results of the 1990 exam. Additionally, Dr. Shapiro argued that a system based on performance appraisals should be used. But he said, “Now, exactly how you want to do it, you know, it’s not my job to tell them how to revise their system.” When asked if he knew whether such a system would reduce adverse impact, Dr. Shapiro stated that if you use the system correctly it would be valid and if the valid exam ends up having adverse impact, so be it.

S. Rank-Ordering

a. Use of tests

Even though the Court finds that the 1990 exam is content-valid, it may be that the 1990 exam was accurate enough to determine which applicants were qualified, but not accurate enough to rank fairly those who passed. See Seibels, 31 F.3d at 1574 n. 12 (citation omitted). The Uniform Guidelines recognize that “a test which may be valid as a pass/fail test, may not be valid as a ranking test....” Brunet v. City of Columbus, 642 F.Supp. 1214, 1248 (S.D.Ohio 1986), appeal dismissed, 826 F.2d 1062 (6th Cir.1987), cert. denied, 485 U.S. 1034, 108 S.Ct. 1593, 99 L.Ed.2d 908 (1988). See 29 C.F.R. § 1607.5(G).

“[CJontent validity is not an all or nothing matter; it comes in degrees. A test may have enough validity for making gross distinctions between those qualified and unqualified for a job, yet may be totally inadequate to yield passing grades that show positive correlation with job performance.” Guardians, 630 F.2d at 100; see also N.A.A.C.P., Ensley Branch, v. Seibels, 14 Empl.Prac.Dec. (CCH) ¶ 11,504, at 6804, 1977 WL 806 (N.D.Ala.1977) (noting that many tests may be accurate enough “to determine which individuals will exceed a given minimum standard of performance” but unable “to predict individuals’ relative standing”), aff'd in part and rev’d in part on other grounds, 616 F.2d 812 (5th Cir.), cert. denied, 449 U.S. 1061, 101 S.Ct. 783, 784, 66 L.Ed.2d 603 (1980)). “[U]se of an exam to rank applicants, when the exam cannot predict applicants’ relative merits, offers nothing but a false sense of assurance based on a misplaced belief that some criterion — no matter how arbitrary — is better than none.” Seibels, 31 F.3d at 1574 n. 12 (citing Guardians 630 F.2d at 104).

“Ranking is a valid, job-related selection technique only where the test scores vary directly with job performance.” Brunet v. City of Columbus, 1 F.3d 390, 410 (6th Cir.1993), ce rt. denied, — U.S. -, 114 S.Ct. 1190, 127 L.Ed.2d 540 (1994) (quoting Williams v. Vukovich, 720 F.2d 909, 924 (6th Cir.1983)); see Guardians, 630 F.2d at 100). Thus, “if a test is to be used to rank-order applicants, it must be more than merely content valid.” Brunet, 642 F.Supp. at 1248. The Guidelines provide that rank-ordering should be used only if it can be shown that “a higher score ... is likely to result in better job performance.” 29 C.F.R. § 1607.14C(9).

In Guardians, the Second Circuit stated that “[t]he relationship between higher scores and better job performance may permissibly rest on an inference, [rather than a criterion-related study], but where ... the test scores reveal a disparate racial impact, and that disparity is greater at high passing scores than at low passing scores, the appropriateness of inferring that higher scores closely correlate with better job performance must be closely scrutinized.” Guardians, 630 F.2d at 100. The Second Circuit went on to state that “[i]f a test is content valid, it may be reasonable to infer that the test scores make some useful gross distinctions between candidates. Candidates with high scores may well be expected to perform the job better than candidates with low scores. And it may even be that within some range of scores, some incremental improvements in scores show some positive correlation with improvements in job performance.” Id.

“Even without resorting to a criterion-related study, the test maker still has several ways to increase the justification for rank-ordering sufficiently to use it. First, he can conduct a job analysis and construct the test with a high degree of adherence to Guideline requirements. [E]ven content-validity for rank-ordering does not require literal compliance with every aspect of the Guidelines. But there must be a substantial demonstration of job relatedness and representativeness to show a sound basis for making rank-ordering hiring decisions. Second, the test-maker can achieve an adequate degree of reliability by careful design of the exam so that the questions will achieve a satisfactory degree of consistent results.” Id. at 104.

The breakdown of the raw scores on the 1990 exam is as follows:

Number of people Raw Score

1 100.000

1 98.980

2 97.959

9 96.939

9 95.918

13 94.898

7 93.878

6 92.857

7 91.837

8 90.816

6 89.796

4 88.776

5 87.755

4 86.735

1 85.714

2 84.694

2 82.653

1 79.592

1 78.571

1 76.531

1 73.469

1 64.289

1 62.245

1 51.020

Dr. Shapiro contends that the test scores of the 1990 exam do not predict who will be the best captain. He does not believe that a knowledge based test is the best predictor. According to Dr. Shapiro, the best predictor would be a person’s performance as a lieutenant. Dr. Shapiro contends that the test was constructed in such a way that if nonfire-fighters read and studied the reading materials there is no reason to suspect that the nonfirepersons would do any worse than the firepersons, and clearly, the firepersons are more qualified to be captains than the non-firepersons.

Dr. Shapiro testified that the 1990 exam was a test in which virtually everybody got the vast majority of items correct. Shapiro believes that ranking people who get such high scores is absurd because they all can perform the job. But Nash was not one of those people who received a high score. His raw score was 82.653, yet the average of the test was 90.656. Even if it is true that those attaining “high scores” cannot be distinguished from one another, Plaintiff scored significantly less than the average. Since the average of the exam was high, it makes Plaintiffs performance on the exam seem worse since he scored approximately 8 points below the average.

Dr. Shapiro testified that only three people received a score of less than 90. Thus, Dr. Shapiro concluded that the 1990 exam had no variability to it. Dr. Shapiro’s conclusion, however, is based on his erroneous view of the facts. In fact, thirty-one people received a grade of less than 90 on the 1990 exam. In an attempt to rehabilitate himself, after it was pointed out that he was in error, he said that perhaps it was 89 people instead of 91. Dr. Shapiro admitted that he placed “great emphasis” on his mistaken fact that 91 out of 94 received a raw score of greater than 90. Thus, the Court seriously questions Dr. Shapiro’s opinion.

Dr. Shapiro then said there is no difference between what everybody made on the test except for the bottom three, four of five. So in essence, Dr. Shapiro is saying that there is no difference between a 78.571 and a 100.00. Then, when asked how low do you have to go until it makes a difference, Dr. Shapiro said 80. Dr. Shapiro reached his conclusion that there is no difference between scores until you get to the bottom three, four or five scores out of thin air, giving no statistical proof or explanation to back up this claim. Dr. Barrett, whom the Court finds to be a more credible witness than Dr. Shapiro, testified that there is a substantial difference between a score of 82 and a 94, that is, there is a tremendous difference between the score Plaintiff attained and the score Norris attained.

b. Rule of one

Dr. Greenwood stated that the rule of one is fair and it has been scientifically proven that this particular type of strategy, rank-ordering when using a content valid approach with a highly reliable exam, serves as a tremendous predictor of success on the job. Dr. Barrett also stated that knowledge based tests are the best predictors of job performance and that there are empirical studies that show that the more job knowledge one has, the higher probability of success on the job. This is particularly true in the firefighting business which has become very technical and, thus, fire personnel are required to know a tremendous amount of information. Using a written examination is the best way to find out what one knows of this technical information. 6

Dr. Barrett testified that there is a justification for rank ordering because the highest probability of success is based upon those who have the highest test score. Because of the 1990 exam’s content validity and high reliability, Dr. Greenwood felt that it served as a good predictor of success for the job. Thus, Dr. Greenwood felt that those that scored higher were more likely to be better captains than those who scored lower.

In Guardians, the Second Circuit found the defects in the job analysis and the test construction to be substantial enough to preclude an inference that passing scores will correlate with job performance closely enough to justify rank-ordered selections. See Guardians, 680 F.2d at 101. This Court, however, does not find the identification of pertinent knowledges, the demonstration of their relationship to the job tasks, and the process of developing the questions to have been flawed for the 1990 exam. Additionally, Plaintiff scored an 82.653 on the 1990 exam, whereas Mr. Norris, the other African-American male that took this test, scored a 94.898. The Court cannot conclude that there is a greater disparity at high passing scores than at low passing scores. Thus, the inference that rank-ordering is proper remains.

In Police Officers for Equal Rights v. City of Columbus, 916 F.2d 1092, 1102-1103 (6th Cir.1990), the Sixth Circuit upheld the lower court’s finding that the test was “sufficiently reliable” to justify rank-order scoring. In that case, composite reliability was 0.95, the standard error of measurement was 1.3, there was a spread of more than 40 points among 71 test takers, the highest score was 89.66 and the passing score was 70. The top fifteen scores ranged from 78.6 to 89.66.

The raw scores on the 1990 exam ranged from a 51.020 to 100. The top 22 scores ranged from 95.918 to 100. The top 55 scores ranged from 91.837 to 100. The raw scores of the thirty persons promoted ranged from 94.898 to 100. The total scores, raw scores plus seniority points, of those promoted ranged from 104.263 to 108.313, with the lowest amount of seniority points given to someone promoted being 6.729 and the most being ten. The total scores ranged from 83.469 to 108.313. The 1990 fire combat captain examination had a reliability coefficient of 0.87. See Plaintiffs Ex. No. 62. Thus, the test was highly reliable. The average score was 90.656 and the standard deviation was 7.575. 7

The Court finds that rank-ordering was proper since the 1990 exam was highly reliable and the City demonstrated the substantial job-relatedness of the exam.

Additionally, the probationary period is significant because those people promoted were evaluated in their new jobs for six months and the failure to perform satisfactorily would be cause for a reversion back to the previous position. Dr. Shapiro argues that because nobody is ever demoted after the probationary period, that the probationary period is meaningless. The Court, however, sees it the other way — the fact that nobody was reverted back is some indication that the written test works by giving the Fire Department qualified captains. There was also testimony that the use of a written exam coupled with the rule of one produces good officers.

The Court finds that the selection process utilized by the City, the 1990 exam coupled with the rule of one, was a business necessity because it was an objective way for the City to obtain top-notch captains.

C. ALTERNATIVE PRACTICES

Since the Court finds that the 1990 exam was sufficiently job-related, the burden shifts to Plaintiff to demonstrate that other practices or criteria have less discriminatory effect but also suitably serve the City’s needs. See Watson, 487 U.S. at 977, 108 S.Ct. at 2777 (Blackmun, J., concurring in part and concurring in judgment). Such a showing would be “evidence that the employer was using its tests merely as a ‘pretext’ for discrimination.” Id. at 977, 108 S.Ct. at 2777.

“[J]ust suggesting an alternative practice is insufficient to meet the plaintiffs burden: the alternative practice ‘must be equally effective as [the employer’s chosen practice] in achieving [its] legitimate employment goals.’ ” MacPherson, 922 F.2d at 771. “The burdens imposed on the employer— including cost — by the alternative practices are ‘relevant in determining whether they would be equally effective as the challenged practice in serving the employer’s legitimate business goals.’” Id.

One of the possible alternative procedures suggested was to look at a person’s performance as a lieutenant to determine if that person would be a good captain, that is, use performance appraisals. However, no evidence was produced that a performance appraisal system was valid or that it would have less adverse impact.

From past experience, Dr. Barrett testified that even though performance appraisals sound like a good idea, they do not work because there is too much subjectivity from too many people. For instance, in the Fire Department, 56 captains would be rating the two lieutenants under them. Also, since some captains and lieutenants are friends, favoritism would factor into the appraisals. Dr. Barrett testified that the reliability of performance appraisals is very low.

Previous cases have found that performance appraisals are not adequate since “[promotion systems utilizing subjective evaluations by all white supervisors provide a ready mechanism for discrimination.” Fisher, 613 F.2d at 546 (citing Hamilton v. General Motors Corp., 606 F.2d 576 (5th Cir.1979), ce rt. denied, 447 U.S. 907, 100 S.Ct. 2990, 64 L.Ed.2d 856 (1980)); see also Powers, 854 F.2d at 1295 (experts found that black employees were more likely to receive low evaluation scores from their supervisors than white employees).

Nash also argued that rather than using a rank-order system based on the scores of a written exam, points should be given for education. But Dr. Barrett testified that this suggestion is ironic because Dr. Barrett’s first case in 1973 involved African-American firefighters arguing that they did not want high school education as a requirement to be a firefighter. Additionally, Dr. Shapiro, Nash’s expert, testified that because of his educational background, if he studied the material for the 1990 exam he would get a 100, yet he would not make a good fireper-son. Thus, Nash’s own expert testified that education does not correlate to being a good fireperson. Furthermore, if education is used as a selector of who should be promoted to captain, there is going to be adverse impact in terms of what is known with regard to the general statistics in this country at this time. Also, no evidence was introduced that education is just as valid a predictor than what was used.

Another suggested alternative selection device was to use work samples, that is, to set up a fire scene and to see how the candidate performs. This, however, is very expensive. Dr. Barrett envisions Universal Studio coming in and building various fire scenes. Additionally, subjectivity could factor in because somebody has to be doing the evaluating of the person in the simulation. No evidence was produced that this system would be as valid as what was used nor was any evidence produced that it would have less adverse impact.

Plaintiff also suggested the use of a pass/ fail exam and then whoever is Personnel Director should decide what system should be used to pick those to be promoted. There is no evidence that this system would be equally valid or have less adverse impact. When asked how the Personnel Director would determine who gets promoted from those who pass if a pass/fail system is used, Nash replied that it was not his job to figure this out and that he is not paid to come up with the promotional system.

Dr. Shapiro also stated that it was not his job to tell them how to revise the system, and that he does not envy the Court in having to make such a difficult decision. However, it is not for the Court to come up with an alternative practice. Title VII does not vest the federal court with power to sit as a personnel director or review board for every employment decision. Rather, Title VII provides a remedy for employment discrimination.

Anything other than a written exam coupled with the rule of one would open the process to favoritism, politics and tokenism, just what the City is trying to avoid by using the rule of one. One of the benefits of the system used by the City is that it removed politics and favoritism from the process. The rule of one is more protective of minorities by not allowing any racial discrimination into the process, since the rule of one is color blind, than if an alternative method was used such as evaluations, which are subjective.

Additionally, without the rule of one combined with written tests, people would not seriously study, causing the Fire Department to be less knowledgeable as a whole and resulting in a Fire Department that lacks professionalism, since they would know that other factors besides test scores are in the process, such as whether your buddy could get you the promotion. Studying is a way of training — a way of learning the technical side of fighting fires. Extensive knowledge is required for the position of captain, and studying for an exam helps one attain this knowledge. Because of the tests, people learn more about their jobs and how to properly perform the job. Testing gives the personnel more incentive to keep on top of their knowledge than any other promotional system.

Also, eliminating the rule of one would devastate the morale of many people in the Fire Department. The rule of one keeps the morale of the Fire Department relatively high because the firepersons know it keeps out prejudice, politics and favoritism. Morale leads directly to safety on the fire ground since firepersons need to be able to work as a team.

Plaintiff has not shown an alternative selection device which has equal validity and less adverse impact and does not have a burden upon the City in terms of greatly increased costs. Accordingly, the Court finds that Plaintiff cannot prevail under a disparate impact theory.

II. DISPARATE TREATMENT

In McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973), and Texas Dep’t of Community Affairs v. Burdine, 450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981), the Supreme Court established the basic allocation of burdens and order of presentation of proof to be used in eases alleging discriminatory treatment under Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e, et seq., which the Court will discuss momentarily.

The Title VII claim merges with the section 1981 and section 1983 claims under the disparate treatment theory. Section 1981 and Title VII are not coextensive. When section 1981 is used as a basis for relief parallel to that provided by Title VII, the elements of a section 1981 claim are identical to the elements of a Title VII claim. See Parr v. Woodmen of the World Life Ins. Co., 791 F.2d 888, 892 (11th Cir.1986); Walters v. City of Atlanta, 803 F.2d 1135, 1143 (11th Cir.1986) (citing Whiting v. Jackson State University, 616 F.2d 116, 121 (5th Cir.1980)); Lincoln v. Bd. of Regents of Univ. System, 697 F.2d 928, 935 n. 6 (11th Cir.), cert. denied, 464 U.S. 826, 104 S.Ct. 97, 78 L.Ed.2d 102 (1983); see also Chaline v. KCOH, Inc., 693 F.2d 477, 479 (5th Cir.1982) (“The elements of a case of racial discrimination as articulated in McDonnell Douglas and Burdine apply to cases brought under section 1981 as well.”) (citation omitted). Similarly, “when section 1983 is used as a parallel remedy for violation of § 703 of Title VII [42 U.S.C. § 2000e-2], the elements of the two causes of action are the same.” Cross v. State of Alabama, State Dept. of Mental Health and Mental Retardation, 49 F.3d 1490, 1508 (11th Cir.1995) (citing Hardin v. Stynchcomb, 691 F.2d 1364, 1369 n. 16 (11th Cir.1982)) (citing Whiting, 616 F.2d at 123). Thus, the Court’s references to Title VII under the discussion of the disparate treatment theory will encompass Plaintiffs section 1981 and section 1983 claims as well.

Section 703(a)(1) of Title VII of the Civil Rights Act of 1964 provides in relevant part:

It shall be an unlawful employment practice for an employer-(l) to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s race, color, religion, sex, or national origin ...

42 U.S.C. § 2000e-2(a).

Intentional discrimination can be established by either direct or circumstantial evidence. Green v. School Bd. of Hillsborough County, Fla., 25 F.3d 974, 978 (11th Cir.1994). Where the evidence is circumstantial, there is a three-step framework for the analysis of intentional discrimination.

In any Title VII employment discrimination case, the plaintiff must first make out a prima facie case. This initial burden, which is “not onerous,” Burdine, 450 U.S. at 253, 101 S.Ct. at 1093-94, is discharged by showing that (1) he is a member of a protected group; (2) he is qualified for the position in question; (3) he suffered an adverse employment decision; and (4) a person outside the protected class was awarded the position in question. See McDonnell Douglas, 411 U.S. at 802, 93 S.Ct. at 1824; Jones v. Firestone Tire and Rubber Co., Inc, 977 F.2d 527, 537-38 (11th Cir.1992), cert. denied, - U.S. -, 113 S.Ct. 2932, 124 L.Ed.2d 682 (1993). Then, when the defendant produces a legitimate, non-discriminatory reason for the adverse employment decision, the burden of production shifts backs to the plaintiff to establish that this reason is pretextual. Burdine, 450 U.S. at 253, 101 S.Ct. at 1093-94.

This is merely a division of intermediate evidentiary burdens. Although this allocation of burdens and creation of a presumption by the establishment of a prima facie case “is intended progressively to sharpen the inquiry into the elusive question of intentional discrimination,” id. at 255 n. 8, 101 S.Ct. at 1094 n. 8, it must be stated emphatically that the ultimate burden of persuading the trier of fact that the defendant intentionally discriminated “remains at all times with the plaintiff.” Id. at 253, 101 S.Ct. at 1093; Equal Employment Opportunity Commission v. Reichhold Chemicals, Inc., 988 F.2d 1564, 1570 (11th Cir.1993). This point recently was reiterated by the Supreme Court in St. Mary’s Honor Center v. Hicks, — U.S. -, 113 S.Ct. 2742, 125 L.Ed.2d 407 (1993), where the Court stated that even if the factfinder disbelieves a de fendant’s proffered reasons, this will only permit it to infer the ultimate fact of intentional discrimination. Id. at -, 113 S.Ct. at 2749. “Nothing in the law,” however, permits the substitution of the required finding that the defendant’s action was the product of illegal discrimination with “the much different (and much lesser) finding that the employer’s explanation of its action was not believable.” Id. at -, 113 S.Ct. at 2751. Thus, notwithstanding the establishment of “certain modes and orders of proof,” id., the ultimate burden of persuasion remains with the plaintiff.

In a Title VII lawsuit, a defendant need not establish that the individual hired or promoted instead of the plaintiff was more qualified than the plaintiff. Burdine, 450 U.S. at 258, 101 S.Ct. at 1096.

Importantly, Title VII does not vest federal courts with the power to sit as review boards for every personnel decision. The Act does provide a remedy for employment discrimination — and the Court will entertain such complaints with its full attention — but Title VII does not even require the employer to have good cause for its decisions. Nix v. WLCY Radio/Rahall Communications, 738 F.2d 1181, 1187. (11th Cir.1984). Rather, as the court of appeals has stated, “[t]he employer may fire an employee for a good reason, a bad reason, a reason based on erroneous facts, or for no reason at all, as long as its action is not for a discriminatory reason.” Id. at 1187 (citing Megill v. Board of Regents, 541 F.2d 1073, 1077 (5th Cir.1976)).

Plaintiff has established a prima facie case. Plaintiff, an African-American male, is a member of a protected group. Having passed the 1990 exam, Plaintiff has shown that he is qualified for the position in question. Plaintiff was not promoted to captain and thus he suffered an adverse employment decision, and white males, persons outside the protected class, were awarded the position in question.

The City has produced a legitimate, non-diseriminatory reason for the adverse employment decision. The City had in place a fair and objective promotional process that produced well qualified captains. The City’s promotional process, known as the “rule of one,” keeps subjective biases out of the promotional process. No longer is the “good ol’ boy network” in place. In order to get promoted, one must perform well on the promotional examination and pray that his or her ranking is high enough. All persons are treated the same under this procedure, whites, blacks, hispanics, men, women, etc.

Additionally, there is no evidence of intentional discrimination with regard to the 1990 exam. The City hired Dr. Robert Anderson to act as the lead person in the development of the process. The City appears to have done everything in its power to give a neutral examination.

Plaintiff argues that the City’s failure to implement Title XI, Chapt. 400, Section 400.212, a City ordinance, which states that a pass/fail scoring method should be used for selection, is evidence of discriminatory intent. The Court disagrees. There is no evidence that the “rule of one” was used to intentionally discriminate against blacks in the selection of captains in the 1990 promotional process. The evidence presented to the Court was that the City historically used rank-ordering for promotions, and did not just recently change from pass/fail to rank-ordering in order to discriminate against African-Americans.

Chief Clark has been with the Fire Department since 1960, and as far as he knows, the City, and the Fire Department in particular, have always promoted through the rule of one using a written test.

In 1990, Clarence Rease was Chief of Recruiting and Examining for the Personnel Department of the City. Rease was chief of the division responsible for making up and administering exams. At the time of trial, Rease was employed by the City for approximately 22.5 years and he testified that during that time the City almost always used the rule of one. Rease testified that the rule of one was used, rather than a pass/fail system, because the City was nervous about the political fallout of doing otherwise. This goes against Plaintiffs claim that the City used the rule of one to intentionally discriminate against minorities. Rease can only remem ber one time in which the City did not use the rule of one, but cannot specifically remember when this was or for what position. As far as Rease recalls, the rule of one has always been used on promotional exams. During Rease’s four years as Chief Examiner, from July, 1987 to June, 1991, about 400 promotional exams were given, and the rule of one was used for all of them. According to Rease, the Fire Department, Sheriffs Office, Jacksonville Electric Authority, Park and Recreation Service and Public Works all use the rule of one.

Additionally, there is the issue of whether Chapt. 400.212 was superseded by state law in 1982. The Court need not decide this. It is clear that the City’s nonuse of a pass/fail system for promotions was not intentional discrimination. Walter Zabomiak, Personnel Director for the City from 1987 to 1991, testified that the General Counsel told him that the rule of one applied to promotions. There is no evidence of racial animus.

The Court finds that the Plaintiff has not proven a case of disparate treatment

Accordingly, it is ORDERED AND ADJUDGED that the Clerk shall enter Judgment in favor of Defendants and against Plaintiff, and tax costs accordingly.

DONE AND ORDERED.

1 . Seniority systems are given special protection under Title VII. See 42 U.S.C. § 2000e-2(h). "Employees may not challenge the effects of a seniority system under the disparate impact theory.” Beavers v. American Cast Iron Pipe Co., 975 F.2d 792, 798 (11th Cir.1992). "Absent a discriminatory purpose, the operation of a seniority system cannot be an unlawful employment practice even if the system has some discriminatory consequences.” Trans World Airlines, Inc. v. Hardison, 432 U.S. 63, 82, 97 S.Ct. 2264, 2276, 53 L.Ed.2d 113 (1977).

2 . All cases decided by the former Fifth Circuit Court of Appeals prior to the close of business on September 30, 1981, are binding on the Eleventh Circuit and on all district courts within the Eleventh Circuit. See Bonner v. City of Prichard, 661 F.2d 1206, 1207 (11th Cir.1981) (en banc).

3 . Nine African-American engineers were promoted to lieutenant off of the 1992 lieutenant eligibility list.

4 . Plaintiff urged the Court to look at the adverse impact of the 1981 lieutenant’s promotional examination as being a similarly situated situation in order to prove disparate impact. 29 C.F.R. § 1607.4(D) provides in pertinent part that "[wjhere the user's evidence concerning the impact of a selection procedure indicates adverse impact but is based upon numbers which are too small to be reliable, evidence concerning the impact of the procedure over a longer period of time and/or evidence concerning the impact which the selection procedure had when used in the same manner in similar circumstances elsewhere may be considered in determining adverse impact.” The Court, however, finds that the 1981 lieutenant's exam and the 1990 exam are not similarly situated. They tested for different positions, and were clearly different tests. This issue, however, does not need to be addressed since the Court found that Plaintiff made out a prima facie case of disparate impact.

5 . The Court, in order to avoid the error that occurred in first Nash case, see Nash, 837 F.2d at 1536, 1538, thoroughly reviewed each and every question on the 1990 exam.

6 . As to the physical aspect of firefighting, the employees of the Fire Department are given a physical agility test and a medical physical.

7 . There was testimony that the standard deviation for the 1990 exam was 7.575, but no one explained its significance. ‘‘[Sjtatistical analysis, based on the normal distribution curve, shows that there is a 68% probability that successive scores would fall within a range of one standard deviation from an actual score and a 95% probability (generally a satisfactory confidence level) that successive scores would fall within a range of two standard deviations from the actual score. It is also possible to estimate, again for any test, how many raw score points above and below the applicant’s actual score are within the range of one or more standard deviations.” Guardians, 630 F.2d at 102 (emphasis added). "Thus, to have 95% confidence that an applicant's grade has statistical reliability, grades within two standard deviations should theoretically be treated as equivalent to his grade, for in fact there is a 95% likelihood that each applicant at each grade would score within such a range on successive takings of equivalent tests.” id. at 102-03. This reasoning, however, does not apply to the 1990 exam because, according to Dr. Greenwood, the test results did not come out to a bell shaped curve — it was not a normal distribution — rather, it was skewed in that there were many scoring in the upper limit.