920 F.2d 932
54 Fair Empl.Prac.Cas. 1560
Unpublished Disposition
NOTICE: Sixth Circuit Rule 24(c) states that citation of unpublished dispositions is disfavored except for establishing res judicata, estoppel, or the law of the case and requires service of copies of cited unpublished dispositions of the Sixth Circuit.
BLACK LAW ENFORCEMENT OFFICERS ASSOCIATION, Edward Irvine,
Harold Craig, William W. Ellison, Donnie W. Whitworth,
Dennis Johnson, Jr., Leonard Mitchell, Jr., O'Dell Daniels,
and Douglas Prade, Plaintiffs-Appellants,
v.
CITY OF AKRON, Fraternal Order of Police, Joseph P. Wheeler,
Sidney Foster, Virgil Collins, and Phillip Barnes,
Defendants-Appellees.
No. 89-3743.
United States Court of Appeals, Sixth Circuit.
Dec. 11, 1990.
Before KENNEDY and MILBURN, Circuit Judges, and ENGEL, Senior Circuit Judge.
PER CURIAM:
In 1984 and 1985 two separate law suits were filed against the City of Akron, the Fraternal Order of Police, Joseph P. Wheeler, President of the Civil Service Commission, Sidney Foster and Virgil Collins, members of the Civil Service Commission, and Phillip Barnes, Chief of Police of the City of Akron, in their official and individual capacities (appellees or the city), alleging discrimination in promotions within the police department. On October 4, 1985, the two cases were consolidated and the appellants, the Black Law Enforcement Officers Association, Edward Irvine, Harold Craig, William W. Ellison, Donnie W. Whitworth, Dennis Johnson, Jr., Leonard Mitchell, Jr., O'Dell Daniels, and Douglas Prade, black officers in the City of Akron Police Department (appellants), proceeded with their claims of discrimination. The appellants later moved for, and were granted, class certification. This appeal arises from the District Court's judgment for the city. For the reasons set forth below, we AFFIRM.
The appellants' original complaint alleged violations of Title VII of the 1964 Civil Rights Act, 42 U.S.C. Sec. 2000 et seq. (Title VII), as well as violations of 42 U.S.C. Secs. 1981 and 1983. In May 1985, the appellants moved for an injunction to prevent the police department from administering a sergeants examination to be given in June. The District Court allowed the exam, but restrained appellees from making any promotions based on it. One hundred sixty officers took the exam, eighteen of whom were black. After the exam was given the District Court conducted a hearing and found that the performance appraisal component of the promotional process was shown to be substantially disparate in its impact on black officers, and the city was enjoined from making any permanent promotions to the rank of sergeant. Although the city had intended to promote only fourteen officers, under the court's ruling, if the city chose to make any temporary promotions, it must promote the top twenty-two candidates from a court ordered ranking which excluded the performance appraisal component. Under the new ranking, sixty-five whites and seven blacks passed the exam. The city made the twenty-two temporary promotions, which included two blacks.1 The city appealed the preliminary injunction, and this Court affirmed. Much of the factual background developed prior to trial appears in the decision on the earlier appeal. Black Law Enforcement Officers Ass'n v. City of Akron, 824 F.2d 475 (6th Cir.1987) (Akron I ).
The police department in Akron is divided into three subdivisions, uniform, investigative, and services. Each of the subdivisions is further divided into units and bureaus. The uniform division includes both the traffic and patrol bureaus. The investigative division includes the detective bureau, and the services division includes many of the administrative functions. The Akron Police Department has an authorized strength of 526 persons, although average actual strength is approximately 488 officers. Ninety-nine of those positions are supervisory.
The supervisory positions within the police department are sergeant, lieutenant, captain, deputy chief and police chief. Promotional exams are given periodically to determine eligibility for promotion to those positions. On appeal, the appellants focus on the sergeants exam because the number of officers taking the lieutenants and captains exams are too small to analyze. In 1975, the sergeants exam consisted of a job knowledge component, which involved a written test aimed at testing the knowledge, skill and ability needed by sergeants to perform specific tasks and duties. Once a candidate passed the written test, seniority and service ratings (reports completed for each officer by supervisory personnel twice a year), were taken into consideration for the rankings. From 1975 to 1985, the promotional exams also included an additional component called the test battery, a written portion which attempted to measure reasoning ability, memory, leadership, and lack of impulsiveness. In 1985, the city added a performance appraisal component, which consisted of an evaluation of candidates by their superiors. The performance appraisal component was abandoned by the city after it was found susceptible to adverse impact in the District Court's decision on the preliminary injunction motion.
Those who receive a passing grade on the promotional examination are ranked in order of their scores. Those promoted are taken only from the list of those who passed, and are generally promoted in the order of their rank. The list of those who have passed remains the list of officers eligible for promotion until the next examination is given, at which time all of the officers who still wish to be considered for promotion must take the test again.
After holding a bench trial, the District Court issued a memorandum opinion, finding that the appellants had failed to establish a prima facie case. The District Court made permanent the temporary promotions that had been made in compliance with its preliminary injunction. The District Court also held that the appellants had abandoned their claims under sections 1981 and 1983. Midway through the trial the appellants had moved to amend their complaint to include a claim that the appellees violated the consent decree entered in an earlier case, Arnold v. Ballard, 390 F.Supp. 723 (N.D.Ohio 1975), in which the district court held that the City of Akron Police Department discriminated in its hiring practices. The District Court below denied the appellants' motion to amend their complaint to include allegations that the city failed to adhere to the consent decree. This appeal followed.
In order to prevail in a disparate impact case, a plaintiff must first establish a prima facie case that specific employment practices have a disparate impact on blacks. If that is established, then the burden of production shifts to the defendant to show that those employment practices are a business necessity. Wards Cove Packing Co. v. Atonio, 109 S.Ct. 2115, 2124-26 (1989). The District Court held that the appellants did not establish a prima facie case of impermissible disparate impact. In reaching that conclusion, the District Court made several factual determinations that can only be overturned by this Court if they are clearly erroneous. See United States v. United States Gypsum Co., 333 U.S. 364 (1948). Therefore, where the District Court has employed the correct legal analysis, this Court may only reverse where "the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed." Id. at 395.
In making its finding, the District Court credited the appellees' experts, Frank J. Landy and Judith Stoikov, and discounted the appellants' expert, Douglas Schultz. Schultz testified that the sergeants exam adversely impacted on blacks, although he acknowledged that statistics are not as helpful with small samples such as that involved here. He used two methods of showing the disparate impact, the " 4/5 Rule" and effect size analysis. Schultz testified that the " 4/5 Rule,"2 showed no adverse impact when the pass rates of black and white officers are compared. However, Schultz asserted that in this instance, the better sample to use is the group that might have been promoted had the District Court not entered its preliminary injunction. According to Schultz, the adverse impact ratio is less than fifty percent at each number below twenty-two. When twenty-two are promoted, the ratio is seventy-nine percent.
Next, Schultz analyzed the situation in terms of effect size, which he described as a measure of the difference between the means of different groups. Schultz calculated the mean scores for blacks and whites and divided by the standard deviation. According to Schultz, the effect size revealed a disparate impact on each component of the exam. The mean score on the 1985 sergeants exam was 71.10 for white officers and 63.84 for black officers, with a significant difference of five percent.
Both of appellees' experts concluded that use of the " 4/5 Rule" in this case is inappropriate. Stoikov, however, used the " 4/5 Rule" analysis for the pass rate on all of the exams from 1970 to 1985, and concluded that there was no adverse impact. Stoikov also used the multiple regression and chi-square models and found that there was no significant statistical difference in the number of blacks and the number of whites who passed the exams or in the number actually promoted.3 In addition to reaching the same conclusion that Stoikov did by using the chi-square analysis, defense expert Landy testified that Schultz's effect size is not an accepted way of analyzing statistics.4
The District Court accepted Landy's and Stoikov's analyses and rejected Schultz's. It is within the realm of the District Court to make determinations of credibility, not only of lay but also of expert witnesses. Appellants argue, however, that the District Court erred in relying on the "bottom line" of the number of officers who actually passed the examination. The appellants correctly point out, however, that only a small portion of those who pass the exam were actually promoted. The number promoted is fortuitous in that it is affected by outside circumstances. Although the appellants may not have proven disparate impact in their analysis of the final numbers, they could still have shown discrimination in the particular parts of the examination. Reliance, therefore, solely on the "bottom line," is impermissible. See Connecticut v. Teal, 457 U.S. 440 (1982). This case is an example of the need for an analysis that goes beyond the "bottom line." Where the points scored on an exam are important, not only in terms of passing, but in terms of ranking, if the appellants can show that some components or questions had a disparate impact on blacks, then they may have established a prima facie case. The District Court recognized the need to consider individual components of the examination in its treatment of the performance appraisal component at the preliminary injunction stage.
However, the District Court in fact went beyond the bottom line, although it stated that it was unnecessary to do so. The District Court analyzed the components of the sergeants exam, and the factual determinations it made in doing so must be analyzed under the clearly erroneous standard. In order to establish a prima facie case, the appellants must show that a particular component of the examination accounts for the disparate impact. "As a general matter, a plaintiff must demonstrate that it is the application of a specific or particular employment practice that has created the disparate impact under attack." Wards Cove, 109 S.Ct. at 2124. If the appellants are unable to particularize their claim in that manner, they have not met their burden, and have not established a prima facie case.5
In their attempt to establish that a particular component impacted disparately on blacks, appellants point to their claim that blacks in the Akron Police Department had been deprived of certain assignments and had been assigned almost exclusively to cases involving victims or suspects. They claim that their limited assignments caused them to score lower on the exam. The appellants presented no statistical evidence of this, since the police department does not keep records of assignments other than the general assignments to subdivisions. The appellants did, however, present anecdotal evidence from several black police officers who testified to their limited assignments within the subdivisions. The District Court, however, held that the appellants had not established that black officers were assigned to different positions within the subdivisions. The District Court did credit the testimony of nine police officers who testified that they were assigned almost exclusively to cases involving black suspects or victims. The District Court held, however, that the appellants had not established that this pattern of assignment had caused them to score poorly on the examination. The appellants do not point to any example of how this assignment pattern is connected to any particular questions. Because it is not clearly erroneous, we AFFIRM the District Court's finding.
The appellants next claim that the District Court erred in holding that they had abandoned their claims under sections 1981 and 1983. The District Court determined that the appellants had abandoned those claims because they did not address them in their post-trial brief. In addition, they had abandoned their legal claims for relief, and at the time of trial sought only equitable relief. The appellants do not point to any evidence they presented at trial to support their claims under sections 1981 and 1983. Therefore, it was not improper for the District Court to determine that the appellants had abandoned those claims.
Finally, the appellants claim that the District Court erred in refusing to allow the appellants to amend their complaint to include allegations that the appellees have failed to comply with the consent decree entered in an earlier case, Arnold, 390 F.Supp. at 723. The District Court denied appellants' motion, finding that amending the complaint to include that part of Arnold that dealt with promotions would be redundant, and that including the other issues in Arnold would be prejudicial to the appellees. The District Court noted that the appellants offered no explanation for their failure to include the allegations of noncompliance with Arnold in their original complaint although the appellants had expressed knowledge of that decision in earlier proceedings.
"The grant or denial of a motion for leave to amend is within the sound discretion of the District Court and will be reversed only for an abuse of discretion." Troxel Mfg. Co. v. Schwinn Bicycle Co., 489 F.2d 968, 970 (6th Cir.1973) (citing Zenith Radio Corp. v. Hazeltine Research, Inc., 401 U.S. 321, 330 (1971)), cert. denied, 416 U.S. 939 (1974). The District Court did not abuse its discretion in denying the appellants' motion to amend. The District Court clearly stated its reasons for denying the motion, and those reasons were proper considerations. See id.
The District Court's judgment is AFFIRMED.
Under the new ranking, the seven blacks who passed ranked: 20, 22, 47, 63, 64, 65 and 73
The "4/5 Rule" holds that there is adverse impact when the selection rate for one group is less than four-fifths (80 percent) of the selection rate for the other group
The multiple regression model analyzes a situation in terms of different factors that may have affected the outcome. The chi-square test measures deviation from expected behavior. Although some deviation is expected to occur by chance, too much deviation must be accounted to something other than chance
Schultz testified that use of the chi-square test showed no adverse impact even when he changed the passing score to exclude all blacks
Even if Schultz's analysis is accepted the appellants must go beyond that and show disparate impact in a particular element of the exam. Therefore, this Court need not determine whether the District Court erred in concluding that the important numbers in the analysis are only the pass and promotion rates. Even if the set preferred by appellants is used, the appellants must still go beyond that initial indicia of disparate impact