MEMORANDUM OPINION AND ORDER I. PROCEDURAL BACKGROUND

WILLIAM T. HART, United States District Judge.

This lawsuit involves promotions within the Chicago Police Department ("CPD") and plaintiffs' allegations that White police officers were discriminated against-during the process. The promotions at issue were based on a sergeant promotional examination administered from 1985 to 1988 (the "Examination") Scoring of the Examination was adjusted based on race so that each racial group would be represented in the top 500 candidates expected to be hired in approximately the same percentage as they were in the entire candidate pool (the "Score Adjustments"). Also, some Hispanics were promoted outside the rank order of the adjusted scores on the Examination (the "Out-of-Rank Promotions"). The promotions to the rank of sergeant based on the Examination occurred on five "Promotion Dates": December 16, 1988, July 16, 1990, November 16, 1990, February 8, 1991, and September 16, 1991.See generally Petit v. City of Chicago, 31 F. Supp.2d 604, 608-10 (N.D. Ill. 1998).

When this case went to trial in January 2002, there were 82 remaining plaintiffs and the only remaining defendant was the City of Chicago. It was undisputed that defendant had adjusted the Examination scores of African-American and Hispanic candidates in a manner favorable to those candidates and that defendant also provided Out-of-Rank Promotions to Hispanics. See Jury Instr. 17 ("Defendant does not dispute that it took race and national origin into account when promoting patrol officers to sergeant."). See also Petit II, 31 F. Supp.2d at 608-10 (undisputed facts on summary judgment). Defendant contended that this intentional use of racial criteria was justified by three compelling interests: (a) remedying the effects at the sergeant rank of prior CPD discrimination in hiring and promotion (the "Discriminatory Effects Interest") (b) avoiding an adverse impact violation of Title VII of the Civil Rights Act of 1964 (the "Adverse Impact Interest") and (c) the CPD's operational need to have a diverse work force at the sergeant rank (the "Operational Need Interest"). Defendant also contended that these interests in combination (the "Combined Compelling Interest") constituted a compelling interest and that the "Selection Preferences" employed were narrowly tailored to the compelling interests. See Jury Instr. [330] 17-18.

These issues of liability were tried to the jury. The jury was to decide as to each Promotion Date, each Identified Group, and each type of Selection Preference, whether each compelling interest existed and, if so, whether the Selection Preference was narrowly tailored. See id. at 22, Verdict Form. As to most every issue, however, the jury was unable to reach a unanimous verdict and a mistrial was ultimately declared. The jury did unanimously agree and return a verdict as to the Discriminatory Effects Interest for the promotion of Hispanics on all five Promotion Dates. The jury found that interest had been proven to exist, but did not make a unanimous finding as to whether any of the Score Adjustments or Out-of-Rank Promotions of Hispanics were narrowly tailored to that interest.

At the close of the evidence, plaintiffs moved for judgment as a matter of law on the ground that a collective bargaining agreement between the CPD and FOP (the "CBA") (see Pl. Exh. 3, Def. Exh. 59) required that the CPD make promotions without regard to race. Tr. Vol. VIII 163-64 (Jan. 29, 2002). Defendant argued that the issue was waived because not previously raised during the litigation. Id. at 164-65. Plaintiffs' motion was denied. Id. at 165. At the same time, defendant moved for judgment as a matter of law based on each of its asserted compelling interests. Id. at 165-72. Ruling on defendant's motion was reserved. Id. at 172.

On February 6, 2002, a mistrial was declared and the jury discharged. The order declaring the mistrial and the filing of the verdict form were both entered on the docket on February 11, 2002. See Docket Entries 331, 333. On February 15, pursuant to Fed.R.Civ.P. 50(b), defendant moved for entry of judgment as a matter of law [335] and, on February 20, plaintiffs filed their motion for judgment as a matter of law [336]. Defendant thereafter moved to strike all aspects of plaintiffs' motion other than the argument based on the CBA, contending that the additional arguments were waived because not raised in the motion made at the close of evidence [339]. The following order was thereafter entered:

Defendant City of Chicago's motion to strike Part II of plaintiffs' motion for judgment as a matter of law is denied without prejudice. To the extent defendant successfully argues that plaintiffs have waived the opportunity for a Rule 50(b) motion, plaintiffs' motion for judgment as a matter of law will be treated as a Rule 56 motion for summary judgment. The parties will not be required to file any Local Rule 56.1 statements, but they must provide adequate citations to the record in support of their factual assertions.

Order dated Feb. 26, 2002 [340].

As was clarified in part at a subsequent hearing on defendant's motion for reconsideration, see March 13, 2002 Tr., the parties are entitled to a ruling on their Rule 50 motions for judgment as a matter of law based on the trial record. To the extent all claims can be finally resolved on defendant's motion, judgment will be entered to that effect and the case is over in the district court. To the extent plaintiffs' motion was to be granted as to all compelling interests, damages would remain to be resolved, but the issue of liability would be resolved in the district court. However, to the extent that no claims or only some of the claims can be resolved on the Rule 50 motions, the case would remain open in the district court and it would still be possible and appropriate to consider summary judgment motions. That is why the motions will be considered as summary judgment motions to the extent they raise issues not properly preserved for a Rule 50 motion or rely on evidence not presented at trial. Also, as was previously stated in court, even if the case can be resolved on one of the Rule 50 motions, the summary judgment issues will still be resolved so that there will be a complete record in the event any otherwise dispositive ruling is overturned or modified on appeal.Cf. Fed.R.Civ.P. 50(c)(1).

II. RULE 50 MOTIONS

A. Rule 50 Standards

Rule 50(a)(2) of the Federal Rules of Civil procedure provides: "Motions for judgment as a matter of law may be made at any time before submission of the case to the jury. Such a motion shall specify the judgment sought and the law and the facts on which the moving party is entitled to the judgment." If a Rule 50(a) motion made at the close of all the evidence is not granted, the movant may renew the motion within ten days after the entry of judgment. Fed. K. Civ. P. 50(b). If no verdict was returned, the court may order a new trial or direct entry of judgment as a matter of law. Id. 50(b)(2). It is clear that a timely Rule 50(b) motion on issues that have been adequately preserved may and should be considered even though a mistrial has been declared. Evanston Hospital Corp. v. Astra Pharmaceutical Products, Inc., 1992 WL 220607 *1 (N.D. Ill. Sept. 3, 1992); Bostron v. Apfel, 104 F. Supp.2d 548, 551 (D. Md. 2000), aff'd by unpublished order, 2 Fed. Appx. 235 (4th Cir.), cert. denied, 122 S.Ct. 218 (2001); Stokes v. Children's Hospital, Inc., 805 F. Supp. 79, 80 (D.D.C. 1992), aff'd by unpublished order, 36 F.3d 127 (D.C. Cir. 1994); Charles Alan Wright Arthur R. Miller, Federal Practice Procedure § 2537 at 162 (2d ed. 2002 Supp.). Also, to the extent the case cannot be resolved on the Rule 50(b) motions, it is appropriate to consider summary judgment. See Ross v. Hotel Employees Restaurant Employees International Union, 266 F.3d 236, 243 (3d Cir. 2001), cert. denied, 122 S.Ct. 1172 (2002).

Under the plain language of Rule 50(b), a Rule 50(a) motion for judgment as a matter of law must be made at the close of the evidence in order to be able to bring a posttrial Rule 50(b) motion for judgment as a matter of law. Laborers' Pension Fund v. AC Environmental, Inc., 301 F.3d 768, 775-76 (7th Cir. 2002); Mid-America Tablewares, Inc. v. Mogi Trading Co., 100 F.3d 1353, 1364 (7th Cir. 1996). Rule 50(a)(2) also requires that the motion "specify the judgment sought and the law and the facts on which the moving party is entitled to the judgment." Thus, it is also the rule that issues that were not adequately preserved in a Rule 50(a) motion made at the close of evidence may not be included in a Rule 50(b) motion. AC Environmental, 301 F.3d at 777-78; McCarty v. Pheasant Run, Inc., 826 F.2d 1554, 1555-56 (7th Cir. 1987); Rankin v. Evans, 133 F.3d 1425, 1431 (11th Cir.), cert. denied, 525 U.S. 823 (1998); Advisory Committee Notes to the 1991. Amendments to Rule 50(b). However, these waiver rules must be considered in light of Rule 50(b)'s purpose of requiring that a motion for judgment as a matter of law be made after the submission of all the evidence, but before the case is given to the jury, so as to afford the opposing party an opportunity to cure any defect in its case before the jury retires. AC Environmental, 301 F.3d at 777. The rule should not be strictly enforced when its rationale collapses under the circumstances of a particular case. Id. The Seventh Circuit has held that a failure to expressly state all grounds or expressly state a sufficient argument when the motion is presented at the close of the evidence will not result in waiver if previously presented arguments (in an earlier Rule 50(a) motion, in trial briefs, in motions in limine, on summary judgment, or otherwise) have made the moving party's position clear for the court and opposing party. Id. at 777-78;Urso v. United States, 72 F.3d 59, 61 (7th Cir. 1995). Accord Rankin, 133 F.3d at 1432-33.

As to the merits of a Rule 50 motion for judgment as a matter of law, federal law provides the applicable standard. Deimer v. Cincinnati Sub-Zero Products, Inc., 58 F.3d 341, 343 (7th Cir. 1995) (quoting Mayer v. Gary Partners Co., 29 F.3d 330, 335 (7th Cir. 1994)). The standard is whether the evidence presented, combined with all reasonable inferences permissibly drawn, is legally sufficient to support the verdict when viewed in the light most favorable to the nonmovant. Fed.R.Civ.P. 50; Mutual Service Casualty Insurance Co. v. Elizabeth State Bank, 265 F.3d 601, 612 (7th Cir. 2001); Mathur v. Board of Trustees of Southern Illinois University, 207 F.3d 938, 941 (7th Cir. 2000); Collins v. Kibort, 143 F.3d 331, 335 (7th Cir. 1998). In other words, the test is whether no rational jury could return a verdict for the nonmovant.Mathur, 207 F.3d at 941; Dadian v. Village of Wilmette, 269 F.3d 831, 837 (7th Cir. 2001); Emmel v. Coca-Cola Bottling Co. of Chicago, 95 F.3d 627, 630 (7th Cir. 1996). The court may not reweigh the evidence, resolve conflicts in the testimony against the nonmovant, or override the jury's determinations as to the credibility of witnesses. EEOC v. Board of Regents of University of Wisconsin System, 288 F.3d 296, 301 (7th Cir. 2002); Grassi v. Information Resources, Inc., 63 F.3d 596, 599 (7th Cir. 1995). The standard is essentially the same as that for a motion for summary judgment. Hall v. Gary Community School Corp., 298 F.3d 672, 675 (7th Cir. 2002); Millbrook v. IBP, Inc., 280 F.3d 1169, 1179 n. 2 (7th Cir. 2002); Klunk v. County of St. Joseph, 170 F.3d 772, 775 (7th Cir. 1999).

B. Plaintiff s' Rule 50(b) Motion

At the close of evidence, the only contention expressly raised by plaintiffs in their Rule 50(a) motion was the contention that defendant could not have had a compelling interest in providing Selection Preferences because such Preferences were prohibited by the CBA. In their Rule 50(b) motion, plaintiffs also argue that (a) defendant made an insufficient showing of past discrimination against African-Americans and Hispanics; (b) defendant failed to connect anecdotal testimony about past discrimination to its statistical evidence; (c) the testimony of defendant's statistical expert is deficient because based on assumptions that were not shown to be true; (d) the Selection Preferences were not narrowly tailored to any Discriminatory Effects Interest that was shown; (e) the Selection Preferences were not narrowly tailored to the Adverse Impact Interest because they went beyond the number necessary to avoid the regulatory 80% rule and any potential adverse impact violation could have been avoided by validating the test as job-related; and (f) the Operational Needs Interest was not shown to exist nor was there evidence of how many additional sergeants were needed to satisfy any such interest (collectively, the "Additional Issues"). Defendant contends the Additional Issues are all waived because not included in plaintiffs' Rule 50(a) motion made at the close of the evidence.

1. Plaintiffs' Additional Issues

This case was pending for almost 12 years before it went to trial. Much of that time, the parties were occupied with discovery that was conducted in consolidated proceedings with other cases involving police promotions from the same time period. However, after the consolidated proceedings were completed, the parties had substantial time and opportunity to bring substantive motions and this court issued a number of written rulings.See Petit VII, 2001 WL 1143163 at *1 (summarizing proceedings through September 2001). The parties were provided with the opportunity to move for summary judgment on more than one occasion. See Petit II, 31 F. Supp.2d 604; Petit V, 2001 WL 629306. See also Petit VII, supra (plaintiffs' motion to reinstate that essentially sought reconsideration of prior summary judgment rulings). The parties also filed various substantive motions related to collateral estoppel and a number of motions in limine that raised substantive issues. Additionally, the parties had the opportunity to address substantive issues in the final pretrial order and regarding jury instructions. Although the parties had these many opportunities to raise issues, the parties' submissions were often unfocused, particularly submissions from plaintiffs. Despite these many opportunities to raise issues as well as the opportunity to raise issues at the time the Rule 50(a) motions were presented, plaintiffs had not previously raised most of the issues contained in their Rule 50(b) motion, and certainly had not raised the more fact-oriented issues that go to the evidence that was actually presented at trial.

The only issue on which plaintiffs moved for summary judgment was the issue of narrow tailoring related to the Discriminatory Effects Interest. See Petit V, 2001 WL 629306 at *7-12. That argument was focused on certain aspects of Bernard Siskin's expert report and is different from the contentions plaintiffs attempt to raise in their Rule 50(b) motion. Also, no summary judgment issue was raised concerning a failure to provide sufficient anecdotal evidence of discrimination and a failure to prove narrow tailoring related to the Adverse Impact or Operational Needs Interests. Plaintiffs did assert on a number of occasions that the Adverse Impact and Operational Needs Interests could not constitute compelling interests, but those are purely legal questions that are distinct from the facts supporting the Interests. Plaintiffs do not assert the purely legal issues in their Rule 50(b) motion [336] nor did they make the argument during their formal objections to the jury instructions. See Tr. Vol. VIII 109-41 (Jan. 29, 2002).

In ¶ 8 of their Rule 50(b) motion [336], plaintiffs assert: "Without first allowing argument from plaintiffs' attorney, at the close of the evidence, before the jury got the case, the court announced its conclusion that the City had produced a strong basis in the evidence for a compelling governmental interest and a narrowly tailored plan, for all three of its alleged interests. The court made no findings to support this broad conclusion." Where a court precludes a party from making its Rule 50(b) presentation, it may be found that issues not expressly raised were not waived. See AC Environmental, 301 F.3d at 776. See also Urso, 72 F.3d at 61. Plaintiffs misstate the record. During the parties' opportunity to formally place their jury instruction objections on the record, plaintiffs objected to the instruction regarding the burden of proof. The court stated that, on the legal issue of defendant meeting the burden of production for its defenses, it would be found that defendant satisfied its burden of production as to a compelling interest and, therefore, the jury was only being instructed as to plaintiffs' burden to prove a negative. Tr. Vol. VIII 124-27 (Jan. 29, 2002). The court did not find that defendant had proved a strong basis, instead indicating only that defendant had met its burden of production. More importantly, the court expressly stated that plaintiffs would be provided with the opportunity to address these fact issues and the court would rule thereon if the issues were subsequently raised on a "directed verdict" motion.Id. at 126. When the time came for plaintiffs' presentation of their Rule 50(a) motion, they did not raise any issues regarding the evidence supporting defendant's asserted compelling interests or narrow tailoring; they only raised the CBA issue. The court was prepared to give serious consideration to an expected Rule 50(a) motion regarding all the compelling interests, though the decision may have been to reserve ruling as was done with defendant's Rule 50(a) motion. Moreover, the prior indication that defendant had met its burden of production is different from the sufficiency of the evidence issue. Even though defendant met its burden of production and plaintiffs therefore continued to bear the burden of persuasion, evidence submitted by plaintiffs still could have been such that no rational juror could have done anything but find for plaintiffs. Contrary to what had been expected, this latter issue was never raised by plaintiffs nor ruled upon by the court; all the court did during the jury instruction conference was indicate it had found that defendant met its burden of production.

This is not a situation where the Additional Issues were preserved because clearly before the court even if not expressly raised in a Rule 50(a) motion. Plaintiffs have not preserved any issue for a Rule 50(b) motion other than the CBA issue. As previously stated, though, the other issues will be considered as summary judgment motions. This is largely to the same effect. Although summary judgment permits consideration of evidence other than that actually presented at the trial, neither side has presented much additional evidence.

2. The CBA Issue

The CBA in effect from February 7, 1990 through December 31, 1991, provided in part:

Section 10.1 Equal Employment Opportunity

Section 10.2 Non-Discrimination

The Employer will continue to provide equal employment opportunity for all officers, and develop and apply equal employment practices.

The Employer shall not discriminate against officers, and employment-related decisions will be based on qualifications and predicted performance in a given position without regard to race, color, sex, religion, age (40-63), or national origin of the officer. . . . Officers shall not be transferred, assigned or reassigned for reasons prohibited by this Section 10.2.

Plaintiffs contend that, as a matter of law, there cannot possibly be a compelling interest for voluntary affirmative action when applying the Selection Preferences violated the CBA that defendant had with its employees.

Defendant contends this issue is waived because not included in the complaint nor the final pretrial order. Plaintiffs allege in their First Amended Complaint that defendant made race-conscious promotion decisions. See First Am. Compl. [45] ¶ 26. They did not have to allege facts showing that no compelling interest existed justifying the use of race-conscious decisions. And even if defendant had expressly denominated a compelling interest as an affirmative defense, which it did not (see Def. Am. Answer to First Am. Compl. [112]), the Federal Rules of Civil Procedure do not require that a plaintiff reply to affirmative defenses. See Fed.R.Civ.P. 7(a). It is true that plaintiffs did allege the existence of certain personnel ordinances and regulations of the City concerning employment examinations and equal opportunity (First Am. Compl. [45] ¶¶ 17-22), but such allegation were not essential to plaintiffs' causes of action. Plaintiffs did not waive the CBA issue by failing to include in their complaint allegations regarding the CBA.

Defendant's other contention is that plaintiffs waived the CBA issue by not including it in their final pretrial order. Rule 16(e) of the Federal Rules of Civil Procedure provides that a final pretrial order "shall control the subsequent course of the action unless modified by a subsequent order." See also Gorlikowski v. Tolbert, 52 F.3d 1439, 1443 (7th Cir. 1995). Local Rule requires that, when ordered to do so by the court, the parties must prepare a final pretrial order in compliance with the form set forth in the rules. See N.D. Ill. L.R. 16.1(a), Form 16.1.1. The form is to include an agreed statement or separate statements of contested issues of fact and law. Id. 16.1.1(2)(b). The final pretrial order also must include trial briefs which "are intended to provide full and complete disclosure of the parties' respective theories of the case. Accordingly, each trial brief shall include statements of . . . (c) the party's theory of liability or defense . . ., and (e) the party's theory of any anticipated motion for directed verdict. . . . Any theory of liability or defense that is not expressed in a party's trial brief will be deemed waived." Id. 16.1.1(9). "[T]he pretrial order is treated as superseding the pleadings and establishes the issues to be considered at trial." Gorlikowski, 52 F.3d at 1444 (quoting Erff v. MarkHon Industries, Inc., 781 F.2d 613, 617 (7th Cir. 1986)). Accord Nagy v. Riblet Products Corp., 79 F.3d 572, 575 (7th Cir. 1996). Evidence or theories not raised in the pretrial order are properly excluded at trial. Gorlikowski, 52 E.3d at 1444 n. 3. This rule also applies to raising contentions in a Rule 50 motion. See Maul v. Constan, 928 F.2d 784, 786-87 (7th Cir. 1991); Morro v. City of Birmingham, 117 F.3d 508, 515-16 n. 3 (11th Cir. 1997), cert. denied, 523 U.S. 1020 (1998). Cf. Nagy, 79 F.3d at 575 (motion for a new trial). The pretrial order can be modified at trial, but only if the "stringent standard" is satisfied that modification is necessary to "prevent manifest injustice." Fed. K. Civ. P. 16(e); Gorlikowski, 52 F.3d at 1444.

Here, plaintiffs submitted a trial brief and statements of contested issues of law and fact. None of those documents makes any mention of the CBA. The only reference to the CBA in the final pretrial order is plaintiffs' inclusion of the excerpt as a listed exhibit. The legal theory associated with the CBA, however, is not mentioned. Additionally, this issue had never been raised on summary judgment or otherwise during pretrial proceedings. Because plaintiffs did not include this issue in the final pretrial order, it was not properly preserved to be raised in a Rule 50 motion, either at the close of evidence or postverdict.

Although the CBA issue is waived for purposes of being raised in a Rule 50 motion, it is not necessarily waived for purposes of being raised on summary judgment. The pretrial order presently in place controls proceedings that occurred at trial and posttrial motions directed to that trial; it does not control subsequent proceedings following a mistrial or grant of a new trial. Johns Hopkins University v. CellPro, Inc., 152 F.3d 1342, 1357 (Fed. Cir. 1998). If defendant's Rule 50(b) motion is denied, plaintiffs would be free to raise on summary judgment any issues that have not been otherwise waived or dismissed in pretrial proceedings.

Even if the pretrial order continues to control after a mistrial, it would be appropriate to permit modification prior to a second trial. In determining whether it is appropriate to permit modification to avoid manifest injustice, the court must "weigh the possible hardships imposed on the respective parties by allowing or refusing to allow the order to be modified; the court must also balance the need for doing justice on the merits between the parties (in spite of the errors and oversights of their attorneys) against the need for maintaining orderly and efficient procedural arrangements." Gorlikowski, 52 F.3d at 1144 (quoting Matter of Delagrange, 820 F.2d 229, 232 (7th Cir. 1987)). Factors to consider include: "(1) the prejudice or surprise to the nonmoving party, (2) the ability of the party to cure the prejudice, (3) the extent of the disruption to the orderly and efficient trial of the case or other cases in the court, and (4) the bad faith and willfulness in failing to comply with the court's order." Ryan v. Illinois Department of Children Family Services, 185 F.3d 751, 763 (7th Cir. 1999).

Defendant may have been surprised when the CBA issue was first raised during plaintiffs' Rule 50(a) motion. At this point in time, however, defendant would no longer be surprised by the issue and has had adequate opportunity to brief it. Moreover, there is no contention by defendant that it would have engaged in additional discovery had it known this issue would be presented by plaintiffs. Defendant suffers no undue prejudice by permitting the issue to be raised at this time. Permitting the issue to be raised after Rule 50(b) motions are resolved and before proceeding to a new trial does not interrupt an orderly and efficient retrial of the case. There is no indication that plaintiffs intentionally delayed raising this issue to gain an advantage. To the extent the issue is meritorious and dispositive of liability, it would have been to plaintiffs' advantage to raise the issue previously and avoid some litigation costs. Additionally, to the extent the issue is meritorious, it would serve justice to permit plaintiffs to now raise this issue. Therefore, even if the final pretrial order still limits the issues that are before the court, that order would, after resolution of defendant's Rule 50(b) motion, be modified to permit plaintiffs to raise the CBA issue on summary judgment or at a second trial. This issue will be addressed below in § III (C)(1).

C. Defendant's Rule 50(b) Motion

Defendant timely moves for judgment as a matter of law on its Discriminatory Effects Interest. Its Rule 50(b) motion and supporting brief raises this Interest only; it does not raise an issue as to the other two claimed compelling interests. See Def. Motion for Entry of Judgment as a Matter of Law [335] ¶ 6; Def. Memo. of Law in Support of Motion for Judgment as a Matter of Law [346]. After the court ruled that plaintiffs' Rule 50(b) motion could, in the alternative, be treated as a summary judgment motion, defendant responded to plaintiffs' motion and additionally requested that summary judgment be granted in its favor based on all three of its claimed compelling interests. See Def. Response to Pl. Motion for Judgment as a Matter of Law [350] at 1-2 n. 2.

Summary judgment will be addressed in § III, infra. Unlike the Rule 50(b) motion, summary judgment permits consideration of evidence in addition to that presented at trial. Neither side, however, has relied on much additional evidence. The general facts and general legal principles set forth in the following two subsections apply to both sides' summary judgment motions as well.

1. General Facts

The facts taken as true for purposes of ruling on summary judgment in 1998, see Petit II, 31 F. Supp.2d at 608-10, were generally indisputably proven at trial as well. No rational jury could find other than the following facts.

Beginning in 1970, various lawsuits were filed challenging the procedures for promoting Chicago police officers as being discriminatory. The 1973 and 1978-79 sergeant examinations were found to be discriminatory and affirmative relief was granted to benefit African-Americans, Hispanics, and women, including the imposition of some quotas. In light of this past history, the City developed the Examination that is the subject of the present litigation. Claims regarding the Examination were raised in United States v. City of Chicago, No. 73 C 2080 (N.D. Ill.) ("City of Chicago"). In City of Chicago, the City moved to impose a quota that would make adjustments favoring Hispanics and women, but that motion was denied. See Petit I, 766 F. Supp. at 609. A number of the plaintiffs in the present case intervened in City of Chicago. The intervenors' claims were subsequently dismissed, some with prejudice and some without prejudice. It was previously held that this dismissal has a preclusive effect on all the individual plaintiffs in the present case. Petit I, 766 F. Supp. at 610-13. Claims dismissed with prejudice in City of Chicago were not to be relitigated in the case sub judice. Plaintiffs were not precluded from challenging the racial standardization of raw scores; they were precluded from challenging both a scoring method that resulted in scores bunched in a narrow range and the elimination of certain questions from use in the final scoring.

The Examination developed by defendant consisted of four parts for which it was determined the following weights would be applied: (a) written multiple choice (28%); (b) written short answer (29%); (c) oral examination (40%); and (d) performance evaluations (3%). The points assigned to each part added up to a total possible score of 100 points. The multiple choice examination was administered in 1985 and the short answer and oral examinations in 1988. All patrol officers who had completed their probationary period by the time the Examination began were eligible. 3416 officers took all parts of the examination, 2274 Whites (66.6%), 931 African-Americans (27.2%), and 192 Hispanics (5.6%). At the time the Examination was developed and scored, defendant anticipated promoting approximately 500 patrol officers to sergeant.

Adjustments were also made to standardize the grading of the different raters who scored the short answer and oral examination parts. This aspect of the standardization was not based on the rater's race. As to the short answer examination, one standardization was performed to adjust for raters and a separate standardization was performed to adjust for the race of the applicant. For the oral examination, though, a single standardization was performed taking into account both the rater and the race of the applicant. For this litigation, this standardization was recalculated, in separate steps for the oral examination rater standardization and the standardization based on the applicant's race. Thus, assuming no other adjustments would have been made, it can be determined what the rank order of applicants would have been had no racial standardization been performed.

Absent racial standardization, the top 500 candidates (which is the approximate number of officers the City expected to promote) would have been composed of 416 Whites (83.2%), 66 African-Americans (13.2%), and 18 Hispanics (3.6%). By comparison, the racial composition of those who completed the application process was 66.6% White, 27.2% African-American, and 5.6% Hispanic. Stated a different way, absent racial standardization, 18.3% of White applicants were in the top 500, 7.1% of African-American applicants, and 9.4% of Hispanic applicants. If only the top 458 (the number actually promoted) are considered, 16.8% of the White applicants would have been promoted, 6.4% of the African-American applicants, and 7.8% of the Hispanic applicants.

The highest possible score was 100. On this 100-point scale, a difference of 2 or 3 points could make a substantial difference in the rank order. For example, using the racially unstandardized scores, the person ranked 630 had an 81.90 score and the person ranked 430 had an 83.59 score. The lowest-scoring Hispanic applicant among the racially standardized top 500 had a racially usstandardized score of 80.95 (rank 774) and a racially standardized score of 83.43 (rank 417). The lowest-scoring African-American applicant among the racially standardized top 500 had a racially un standardized score of 80.70 (rank 854) and a racially standardized score of 82.82 (rank 497). During this litigation, plaintiffs contended that the differences are made to appear artificially small by arbitrarily selecting a 100-point scale and by the standard deviation used for the rater standardization. As plaintiffs have pointed out, the raw score differences may be substantially larger (at least viewed on an absolute scale). Plaintiffs, however, have been precluded (based on prior litigation) from directly challenging the scaling and standard deviation methodology the defendant chose to employ. Additionally, defendant presented undisputed evidence at trial supporting that the test had a standard measure of error of more than three points. That indicates that differences in scores of a few points are not significant. Moreover, plaintiffs presented no evidence from which a rational jury could find that differences of a few points on the test would be a significant predictor of job performance.

The City chose to standardize scores by race. In doing this, the City assumed that, by group, Whites, African-Americans, and Hispanics all would have scored equally on a non-biased test. Therefore, the scores for African-Americans and Hispanics were adjusted so that each group would be represented in the top 500 candidates in essentially the same proportion as their representation among applicants. Following racial standardization, the top 500 applicants consisted of 332 Whites (66.4%), 138 African-Americans (27.6%), and 30 Hispanics (6.0%). Stated another way, after racial standardization, 14.60% (332/2274) of White applicants were ranked in the top 500, 14.82% (138/931) of African-American applicants, and 15.63% (30/192) of Hispanic applicants. The raciallyunstandardized scores were 82.98 for the 332nd White applicant, 80.70 for the 138th African-American applicant, and 80.95 for the 30th Hispanic applicant.

Between December 1988 and September 1991, 458 patrol officers were promoted to sergeant. 402 of the promotions were based on the rank order of the 1988 eligibility list derived from the racially standardized 1985-88 examination. The other 56 promotions were Out-of-Rank Promotions. Of the 458 officers promoted, 298 (65.07%) were White, 119 (25.98%) were African-American, and 41 (8.95%) were Hispanic. The 1988 eligibility list was canceled on October 9, 1992. A new sergeant promotional examination was administered in 1993.

2. General Legal Principles

Because this case concerns promotional decisions of a municipal entity that were influenced by race, strict scrutiny applies to defendant's use of the Selection Preferences. Majeske, 218 F.3d at 819. To survive strict scrutiny, the Selection Preferences must have promoted a compelling interest. Reynolds, 296 F.3d at 526; Majeske, 218 F.3d at 820. There must be a strong basis in the evidence supporting the existence of a compelling interest. Builders Association of Greater Chicago v. County of Cook, 256 F.3d 642, 645 (7th Cir. 2001); Clark v. Putnam County, 293 F.3d 1261, 1273 (11th Cir. 2002); Engineering Contractors Association of South Fla. Inc. v. Metropolitan Dade County, 122 F.3d 895, 906-07 (11th Cir. 1997), cert. denied, 523 U.S. 1004 (1998); Contractors Association of Eastern Pa., Inc. v. City of Philadelphia, 91 F.3d 586, 597 (3d Cir. 1996), cert. denied, 519 U.S. 1113 (1997); Concrete Works of Colorado, Inc. v. City County of Denver, 36 F.3d 1513, 1521-22 (10th Cir. 1994), cert. denied, 514 U.S. 1004 (1995).

To survive strict scrutiny, the Selection Preferences must also have been narrowly tailored. Reynolds, 296 F.3d at 526; Chicago Firefighters Local 2 v. City of Chicago, 249 F.3d 649, 657 (7th Cir.), cert. denied, 122 S.Ct. 465 (2001); Majeske, 218 F.2d at 820. "To determine whether an affirmative action plan is narrowly tailored, the test we use is whether the racially preferenced measure is 'a plausible lower-bound estimate of a shortfall in minority representation' that is caused by past discrimination." Majeske, 218 F.3d at 823 (quoting McNamara v. City of Chicago, 138 F.3d 1219, 1224 (7th Cir.), cert. denied, 525 U.S. 981 (1998)). Aside from the numbers, general factors to consider include "the necessity for the relief and the efficacy of alternative remedies; the flexibility and duration of the relief, including the availability of waiver provisions; the relationship of the numerical goals to the relevant labor market; and the impact of the relief on the rights of third parties." United States v. Paradise, 480 U.S. 149, 171 (1987) (plurality opinion) (cited in Majeske, 218 F.3d at 824).

Here, the asserted compelling interests are not limited to remedying past discrimination. Defendant also raises the Operational Need Interest and the Adverse Impact Interest. While the basic measuring standard is still "a plausible lower-bound estimate of a shortfall in minority representation," it must be measured by the particular compelling interest being considered. The above-quoted passage from Majeske, 218 F.3d at 823, concerned a discriminatory effects interest and therefore was measured by the minority shortfall caused by past discrimination. For the Operational Need Interest, the measuring standard involves the shortfall of minority sergeants necessary to serve the CPD's operational needs and for the Adverse Impact Interest the measuring standard involves the shortfall of minority promotions necessary to avoid an adverse impact violation. Thus, for purposes of narrow tailoring, each particular Interest is measured against a different goal. As the discussions below will indicate, this means the estimated lower-bound number of minority promotions needed to serve a particular Interest must be particularized to that Interest, thereby producing a different minimum estimated quantity of promotions that would serve each particular Interest. Because narrow tailoring for each Interest is measured against a different estimated quantity, it would not be inconsistent to hold that (a) the undisputed facts only support that narrow tailoring is satisfied for one Interest; (b) the undisputed facts only support that narrow tailoring cannot be satisfied as to a second Interest; and (c) a genuine factual dispute remains as to whether narrow tailoring is satisfied for the third Interest.

Defendant had the burden of producing evidence of a compelling interest and narrow tailoring. With the burden of production met, the burden of persuasion remained with plaintiffs to show a lack of a compelling interest and/or a failure to narrowly tailor. Majeske, 218 F.3d at 820;Dade County, 122 F.3d at 917; Philadelphia, 91 F.3d at 957.

Consistent with these legal principles, the jury was instructed as to plaintiffs' burdens regarding compelling interests and narrow tailoring.See Jury Instr. [330] 22-26, 34-36. The burden of production of defendant was a legal question for the court and therefore no instruction was given as to defendant's burden. See Petit X, 2002 WL 171978 at *1 n. 2. Initially, this court ruled that post-implementation or post-enactment evidence could not be the support for the compelling interest showing. "As to each particular compelling interest, the City must show that, during the pertinent time period, it had a strong basis for finding the particular compelling interest to be present. It cannot rely on a post-1991 factual basis as constituting the necessary strong basis in the evidence." Petit IX, 2002 WL 10481 at *4, On the other hand, it was held that post-implementation evidence was admissible to show narrow tailoring. Id. On reconsideration prior to the beginning of the trial,see Order dated Jan. 14, 2002 [319], it was held that post-implementation evidence was also admissible as evidence of a compelling interest and that it was only necessary to have some evidence of a compelling interest at the time the Selection Preference was being implemented. See Jury Instr. [330] 23, 25-26; Adarand Constructors, Inc. v. Slater, 228 F.3d 1147, 1166 n. 11 (10th Cir. 2000), cert. dismissed, 534 U.S. 103 (2001); Dade County, 122 F.3d at 911; Contractors Association of Eastern Pa., Inc. v. City of Philadelphia, 6 F.3d 990, 1003-06 (3d Cir. 1993), cert. denied, 519 U.S. 1113 (1997); Coral Construction Co. v. King County, 941 F.2d 910, 920-21 (9th Cir. 1991),cert. denied, 502 U.S. 1033 (1992). See, e.g., Reynolds, 296 F.3d at 529-30 (referring to expert testimony in support of compelling interest that would not have existed at the time the promotions at issue were implemented); Majeske, 218 F.3d at 820-21 (same). But see In re City of Memphis, 293 F.3d 345, 350-51 (6th Cir. 2002); Rothe Development Corp. v. United States Department of Defense, 262 F.3d 1306, 1324-28 (Fed. Cir. 2001).

The jury was instructed as follows and the law stated therein will be applied in today's ruling.

In order to find that, on a particular Promotion Date, a compelling interest existed to apply a particular Selection Preference for one of the Identified Groups, each of the following propositions must be found:

(1) The particular compelling interest motivated defendant to apply a Selection Preference(s) for the particular Identified Group;

(2) At the time, some evidence supported that the particular compelling interest existed as to that Identified Group; and

(3) A strong basis supports that the particular compelling interest existed as to that Identified Group.

In making this determination, you must keep in mind that the burden of proof is on plaintiffs. Therefore, you must find that a compelling interest existed as to a particular Promotion Date and Identified Group unless plaintiffs prove by a preponderance of the evidence that one or more of the above propositions is not true.

* * *

As used in these instructions, "some evidence" means specific evidence known and considered by defendant (through its agents and employees) on or before the Promotion Date and which provided support for the existence of the compelling interest.

As used in these instructions, a "strong basis" means sufficient evidence upon which reasonable persons, making decisions about serious and important matters, would rely to infer that a particular fact is likely true.

A strong basis does not require proof that the underlying facts actually are true, only such evidence from which a reasonable person could infer the particular facts are likely true.

In determining whether a strong basis existed, you are to consider all the evidence that is before you regardless of whether it was actually known by or considered by defendant as of the Promotion Date.

* * *

In making the narrowly tailored determination you must keep in mind that the burden is on plaintiffs to prove by a preponderance of the evidence that a particular Selection Preference was not narrowly tailored to the particular compelling interest, Promotion Date, and Identified Group.

* * *

"Narrowly tailored" means that the Selection Preference was implemented in a manner that resulted in the promotion of a lower-bound estimate of the number of Identified Group members necessary to be promoted in order to serve the compelling interest which you found existed. In other words, that it discriminates against Whites as little as possible consistent with effective remediation.

Aside from the number of Identified Group members promoted through the application of the Selection Preference, general factors you may consider include the necessity for the relief and the efficacy of possible alternatives; the flexibility and duration of the relief; the relationship of the numerical goals to the relevant labor market; and the impact of the relief on the rights of third parties.

In determining whether the Selection Preference was narrowly tailored to the particular compelling interest, Promotion Date, and Identified Group under consideration, you are to consider all the relevant evidence in the case, including facts and evidence that were or were not known to defendant (through its agents and employees) as of the particular Promotion Date. The question to resolve is whether the Selection Preference was narrowly tailored as implemented.

Jury Instr. [330] 23, 25-26, 35-36.

3. Discriminatory Effects Interest

A combination of persuasive statistical data and anecdotal evidence of past discrimination is adequate to establish a compelling interest to cure the effects of past discrimination. Majeske, 218 F.3d at 822; McNamara, 138 F.3d at 1223-24; Adarand, 228 F.3d at 1166; Wessmann v. Gittens, 160 F.3d 790, 802-03 (1st Cir. 1998). Since defendant met its burden of production on this issue, the burden was on plaintiffs to show by a preponderance of the evidence that the Discriminatory Effects Interest did not exist. As to the promotion of Hispanics, the jury found that the Discriminatory Effects Interest existed as of each Promotion Date. Since plaintiffs did not preserve any challenge based on the sufficiency of the evidence supporting that finding, see § 11(B)(1), supra, it must be taken as established that a Discriminatory Effects Interest existed regarding the promotion of Hispanics. For defendant to be entitled to judgment as a matter of law on any of those claims, however, it must also be held that, as to particular Selection Preferences and Promotion Dates, a rational jury could only find the particular Selection Preference for Hispanics was narrowly tailored.

As to the promotions of African-Americans, it must also be considered whether a rational jury could only find that the Discriminatory Effects Interest existed. Plaintiffs do not raise any issue as to defendant being motivated by a perceived Discriminatory Effects Interest or there being some contemporary evidence of the Discriminatory Effects Interest. Plaintiffs' contention is that the strong basis requirement is not satisfied.

As to the Discriminatory Effects Interest, the jury was instructed as follows.

An interest in remedying the effects of prior discrimination can constitute a compelling interest. For the Discriminatory Effects Interest to constitute a compelling interest, the prior discrimination by Chicago Police Department supervisors must have continued to have had a significant and pertinent then-present effect. In the present case, that means sufficient evidence of a prior discrimination against the Identified Group that significantly affected the number of Identified Group sergeants as of the Promotion Date under consideration.

Jury Instr. [330] 27.

Pretrial rulings in this case established that plaintiffs were estopped from denying that African-American and Hispanic police officers and applicants had suffered a significant degree of certain employment discrimination prior to 1989. See Petit IX, 2002 WL 10481 at *7; Petit III, 1999 WL 66539 at *5, reconsideration denied, Petit IV, 1999 WL 261706. The established facts were stated as follows for the jury.

Based on prior court findings, the following facts are established.

1. In the decades prior to 1989, African-American patrol officers were subjected to intentional, unfavorable treatment in assignments (for example, segregated beats, restricted duties, and unfair efficiencies).

2. In the decades prior to 1989, African-Americans and Hispanics were subjected to intentional, unfavorable treatment in hiring for patrol officer positions (for example, in the use of medical and entrance qualifications).

3. In the decades prior to 1989, supervisory police officials acted in ways that were hostile to and tolerated acts of hostility toward African-American and Hispanic patrol officers.

The exact extent of the discrimination and the effect, if any, that it had on the number of African-American and Hispanic sergeants in the Chicago Police Department as of the years at issue in the present case is for you to decide.

Jury Instr. [330] 19.

These collateral estoppel facts do not establish the precise extent of the past discrimination nor the impact it had on the number of CPD sergeants as of 1988 through 1991. See Petit III, 1999 WL 66539 at *5;Petit IV, 1999 WL 261706 at *1-2. In combination with sufficient statistical evidence, however, it would be sufficient to establish a Discriminatory Effects Interest as to the promotion of African-Americans. Anecdotal evidence would not necessarily have been required as well, but was nevertheless presented. A number of current and former African-American police officers testified regarding their experiences. Plaintiffs contend their testimony as to discriminatory hiring practices is impeached by the fact that all four were hired and were promoted to high ranks: lieutenant, deputy superintendent, or superintendent. It is not surprising, however, that the witnesses that defendant found to testify were long-time officers of high rank; persons who would be more likely to have an established relationship with current City officials. That these witnesses succeeded in being promoted does not significantly undermine anecdotal testimony as to their treatment and the treatment of other African-Americans. While plaintiffs' cross-examination raised questions as to some of the specific anecdotes — questions that must be resolved in plaintiffs' favor on defendant's Rule 50(b) motion — there is substantial unrebutted, admissible evidence of discrimination against African-Americans. Viewed alone — or even more so when considered along with the collateral estoppel facts — this evidence conclusively establishes substantial discrimination that could have led to a substantial decrease in the number of African-American sergeants as of the Promotion Dates of 1988 through 1991. The key question is whether the statistical evidence is sufficient — when considered with the other evidence — to conclusively establish a substantial diminution of African-American sergeants caused by discrimination.

The statistical evidence consisted of Siskin's testimony and documentary evidence establishing the statistics that underlay Siskin's analysis. The key aspect of Siskin's analysis was his projection as to the number of African-American and Hispanic sergeants that there would have been from 1988 through 1991 if not for prior CPD discrimination against African-Americans and Hispanics. Siskin compared his projected figures with the actual number of African-American and Hispanic sergeants.

The Chicago Police Department only promotes from within. To become a sergeant, one must first be a patrol officer. For purposes of his analysis, Siskin had to determine how many minorities and women would have been hired as patrol officers if not for discrimination in hiring. Siskin assumed that hiring discrimination had been eliminated by 1977. For 1977 to 1990, he then compared the percentages for actual hires to the percentages of those minorities in the 20-34 age group of Chicago's population. For this time period, he found that African-Americans among the new hires represented .95 of their proportion of the Chicago population and Hispanic new hires represented .676 of their proportion of the Chicago population. Using these ratios and population data for each year, Siskin estimated how many African-Americans and Hispanics would have been hired as patrol officers from 1950 to 1974 if not for discrimination. Taking into account the percentage of actual patrol officer hires from each past year who actually were sergeants during the 1987-91 time period, Siskin estimated the number of African-American and Hispanic sergeants there would have been during the 1987-91 time period if not for discrimination in hiring and promotion.

According to Siskin's calculations, the actual number of African-American sergeants as compared to the expected number (that is, the estimated number if not for discrimination) were as follows for each year: 1987-269/338; 1988-294/367; 1989-282/354; 1990-287/365; 1991-282/371. For Hispanics, the yearly figures are as follows: 1987-44/55; 1988-62/66; 1989-61/65; 1990-66/70; 1991-67/74. Thus, according to Siskin's analysis, both before and after the promotions involving Selection Preferences, there were less African-American and Hispanic sergeants than there would have been if there had been no discrimination in hiring or promotions.

At trial, plaintiffs did not present an alternative analysis of the data, seek to recalculate Siskin's projections, or otherwise present their own projections. Instead, plaintiffs' present contention is that Siskin's analysis should be disregarded and accorded no weight because it is based on assumptions that are unsupported by the evidence. The two assumptions that plaintiffs challenge as unsupported are that (a) discriminatory hiring of patrol officers did not occur between 1977 and 1990 and (b) hiring requirements did not change during that time period.

The assumption of a period of nondiscriminatory hiring (at least as against Whites) is essential to Siskin's analysis. If African-Americans and/or Hispanics were receiving favorable treatment in hiring during this time period, hiring statistics from that time period cannot be used (as a yardstick) to determine the nondiscriminatory hiring ratios that were necessary for projecting nondiscriminatory hiring for prior years. If plaintiffs presented evidence from which a rational jury could find by a preponderance of the evidence that discrimination against Whites was occurring during this time period, then defendant could not be entitled to judgment as a matter of law.

Evidence at trial (as well as judicial notice of court orders) indisputably show that, beginning with use of the 1971 hiring examination, the CPD imposed certain racial quotas for hiring patrol officers. However, the evidence also indisputably shows that those quotas ended in 1976. No formal quotas for hiring existed from 1977 forward.

Plaintiffs rely on documents that they submitted in evidence, but for which no or little related testimony was presented at trial. See Pl. Exh. 36-39. Exhibits 36 and 37, which concern a 1976 eligibility list used for patrol officer hires from October 1976 through March 1982, are not self-explanatory. The data that plaintiffs seek to derive from these exhibits is not information that can be considered to have been before the jury. An affidavit of Robert Joyce that defendant provides in response to plaintiffs' motion, however, does confirm the data plaintiffs seek to derive from these two exhibits. See Joyce Aff. ¶¶ 5-11 ([351] Exh. 4). Therefore, the information could be considered on summary judgment. But even assuming the data as to hiring from the 1976 list was before the jury, it is not a basis for finding that discriminatory hiring existed. The data plaintiffs rely upon is that White males represented 50.2% of the qualified candidates, but only 48.8% of those actually hired. Plaintiffs contend this shows discriminatory hiring.

First, as this court has previously held, the group to focus on is Whites of both genders, not just White males. See Petit IX, 2002 WL 10481 at *9 But a breakdown between Whites (including Asians and others) and African-Americans/Hispanics also shows a discrepancy similar to that for White males alone. Whites represented 64.2% (700/1104) of the qualified candidates, but only 59.8% (904/1511) of those actually hired. Still, the data is incomplete because the percent qualified data includes "well qualified" and "qualified" candidates, but does not include "conditionally qualified" candidates, at least 407 of which were hired (1511 minus 1104). But even assuming Whites also represented 64.2% of the conditionally qualified candidates, discrimination cannot be inferred. As plaintiffs repeatedly argue themselves, discrimination cannot be inferred from statistics alone. See Chavez v. Illinois State Police, 251 F.3d 612, 640 (7th Cir. 2001); Rummery v. Illinois Bell Telephone Co., 250 F.3d 553, 559 (7th Cir. 2001); Radue v. Kimberly-Clark Corp., 219 F.3d 612, 617-18 (7th Cir. 2000); Adreani v. First Colonial Bankshares Corp., 154 F.3d 389, 400 (7th Cir. 1998). Plaintiffs do not point to evidence that, combined with the statistical evidence, would be sufficient to draw an inference of discrimination. Moreover, as defendant points out, there are other possible explanations for the discrepancy between eligibility rates and actual hiring. For example, it may be that Whites have a wider range of employment opportunities and therefore follow up on their eligibility at a somewhat lower rate than African-Americans and Hispanics. See Middleton v. City of Flint, Mich., 92 F.3d 396, 408-09 (6th Cir. 1996), cert. denied, 520 U.S. 1196 (1997). There is no evidence that the differences between eligibility rates and hiring rates resulted from CPD officials intentionally offering positions to qualified African-Americans and Hispanics at a greater rate than they offered positions to Whites.

The evidence regarding the 1982 and 1986 eligibility lists for patrol officer hiring is different. Exhibits 38 and 39, as well as the trial testimony of Joyce (Jan. 16, 2002 Tr. at 174-79), support that certain racial percentages were used at two steps of the hiring process for both eligibility lists. The 1982 list was used from June 1982 until August 1986. The 1986 list was used from September 1986 until May 1988. For both lists, a written examination was given and scored. There is no evidence that race or gender was considered in scoring the written examination for either list. The test was scored on a pass/fail basis. Only those who passed were eligible to go to the Assessment Center for further evaluation. However, Plaintiffs' Exhibits 38 (1982 list) and 398 (1986 list), as well as Joyce's testimony (as to both lists), support that race and gender were used in determining who would be sent to the Assessment Center. Although only those with passing scores were sent to the Assessment Center and they were sent in score order, candidates were sent in approximately the same ratio as their group was represented in the applicant pool. The categories were White male, minority (African-American and Hispanic) male, and female. In determining the 1982 list, each group sent to the Assessment Center was 30% White male, 42% minority male, and 28% female. In determining the 1986 list, each group sent to the Assessment Center was 31% White male, 37% minority male, and 32% female. A final score was then determined without regard to race or gender. However, candidates were then selected from the qualified lists in approximately the same ratio as their group had been represented in the eligible pool for the 1982 list and the original applicant pool for the 1986 list.

As to the 1982 list, the racial breakdown of those who took the written test was 51.0% African-American, 9.8% Hispanic, and 39.2% White. Pl. Exh. 38 at 1. Broken down by the racial/gender groups that were the categories used in the contemporary litigation, candidates taking the written test were 42.2% minority male, 29.8% White male, and 28.0% female. Id. Those passing the written examination and thus eligible for the Assessment Center were: 33.3% African-American, 8.2% Hispanic, and 58.5% White; or 28.6% minority male, 43.8% White male, and 27.6% female.Id. at 2. Those eligible following the Assessment Center ratings were 35.4% minority male, 34.0% White male, and 30.6% female. Id. at 3, Plaintiffs also point to evidence, Pl. Exh. 50 at 2, that the hired African-American with the lowest score had a score of 44.02 while the hired White male with the lowest score had a score of 46.13. From Pl. Exh 36, it can be computed that the breakdown of those actually hired from the 1982 list was 34.6% minority males, 34.2% White males, and 31.2% female.

As to the 1986 list, the breakdown of those who applied to take the written tests was 39.4% minority male, 23.4% White male, 32.0% female, and 5.2% were missing data. Pl. Exh. 39B. However, only two-thirds of those who applied actually took the written tests. The breakdown of those who took the written tests was 37.3% minority male, 31.1% White male, and 31.6% female. White males had a higher pass rate on the written test (85.4%) than females (67.2%) and minority males (56.9%). Thus, the breakdown for those candidates who passed the written test was 30.7% minority male, 38.5% White male, and 30.8% female. Those persons who were sent on to the Assessment Center were at the same approximate ratio as those who took the written test. The breakdown of those who passed following the Assessment Center's scoring and therefore were eligible for hire was 31.8% minority male, 37.0% White male, and 31.2% female. As with the written test, the pass rate for the Assessment Center scoring was highest for White males (88.0%), as compared to females (82.7%) and minority males (72.7%). From Pl. Exh 36, it can be computed that the breakdown of those actually hired from the 1986 list was 34.3% minority males, 33.8% White males, and 31.9% female.

The undisputed evidence is that, from June 1982 through May 1988, the City hired new police officers in a manner designed to hire Identified Group males and females in approximately the same percentages as they applied for the position. Because White males tended to score higher on the written tests and assessments used to score candidates, to the extent the tests were valid and were to be used, such a policy was favorable to Identified Group males as compared to White males. Whether it was favorable to Identified Group members overall as compared to Whites overall is not actually shown by the data because most of the data as to women is not broken down by race. A reasonable jury, however, could (but would not be required to) infer that, overall, the procedures employed favored Identified Group members. The data, however, is insufficient to determine the extent to which African-Americans and Hispanics were favored, that is, the data provides no sufficient basis for determining how many additional Identified Group members were hired because of the policy of forwarding and selecting in the specified percentages. On the other hand, a reasonable jury could also find that the data does not show Identified Group members were favored. There was no conclusive evidence presented at the Petit trial that the written tests and Assessment Center scoring was a valid and accurate predictor of who could perform well as a patrol officer, nor was evidence presented that the testing and assessments were not racially biased. Applying the percentages may have simply corrected for biases in the evaluation procedure and thus could still have resulted in nondiscriminatory hiring. Additionally, as defendant points out, White males were still being hired at a greater rate than their representation in the applicant pool. Whether this was because of (a) White males being, on average, better qualified to be patrol officers, (b) discrimination favoring White males, or (c) some other reason cannot be conclusively determined from the evidence presented at trial. On defendant's Rule 50(b) motion, though, the inferences favorable to plaintiffs must be drawn and genuine factual disputes must be resolved in plaintiffs' favor. A reasonable jury could find that patrol officer hiring from June 1982 through May 1988 was discriminatory in favor of African-Americans and Hispanics. No determination, however, could be made as to the actual effect in terms of an estimated number of additional Identified Group members hired because of the hiring policies in place.

Even taking as true that patrol officer hiring from June 1982 through May 1988 favored African-Americans and Hispanics, defendant still contends that a rational jury would still have to find that the Discriminatory Effects Interest was proven as to African-Americans. It contends that Siskin's analysis as to shortfalls at the sergeant rank cannot be undermined absent specific numbers as to the hiring effect and expert testimony as to how that would affect Siskin's analysis. As the jury was instructed at trial, though, "[t]he fact that [an expert] has given an opinion does not mean that you are required to accept it. Give the testimony whatever weight you think it deserves, considering the reasons given for the opinion, the witness's qualifications, and all the other evidence in the case." Jury Instr. [330] 10. Accord 7th Cir. Jury Instr. Crim. 3.07 (1999); Burroughs v. FEP Operating Partners, L.P., 28 F.3d 543, 550 (5th Cir. 1994). Where assumptions underlying an expert's analysis are undermined, it is well within the factfinder's power to reject the analysis. See In re TMI Litigation, 193 F.3d 613, 675-76 (3d Cir. 1999), amended, 199 F.3d 158 (3d Cir.), cert. denied, 530 U.S. 1225 (2000); Cardiac Pacemakers, Inc. v. St. Jude Medical, Inc., 2002 WL 1801525 *74.75 (S.D. Ind. July 5, 2002); Community Association for Restoration of Environment v. Henry Bosma Dairy, 2001 WE 1704240 *11 (E.D. Wash. Feb. 27, 2001), aff'd, 305 F.3d 943, 2002 WE 31051547 (9th Cir. Sept. 16, 2002). Statistical evidence presented by an expert is substantially undermined if the underlying data is not trustworthy because based on assumptions that do not match the evidence in the case. See McCleskey v. Kemp, 481 U.S. 279, 288 n. 6 (1987).

Here, the patrol officer hiring rate for African-Americans and Hispanics in a nondiscriminatory procedure was central to Siskin's projections as to how many African-American and Hispanic sergeants there would have been from 1987 through 1991 if not for discrimination. In determining that rate, Siskin used hiring data from 1977 through 1990, which he assumed represented nondiscriminatory hiring. A reasonable jury could have found that data from half of those 14 years did not actually represent nondiscriminatory hiring, but instead represented hiring policies that favored the Identified Groups. Perhaps the numerical effect on additional Identified Group hirings was minimal so that Siskin's projections were still close, though it must be kept in mind that the projected shortfalls of Hispanic sergeants for the years 1988, 1989, and 1990 were only four. or perhaps the analysis could have been recomputed leaving out the seven years from 1982 through 1988, but that would likely require a statistical expert to redo the analysis. On the evidence presented at trial, a reasonable jury could have rejected Siskin's analysis as being based on an assumption that was inconsistent with the evidence. Without Siskin's analysis, there is an insufficient basis for connecting the past discrimination to a continuing effect on the number of African-American sergeants from 1988 through 1991. Therefore, defendant cannot be entitled to judgment as a matter of law on its Discriminatory Effects Interest regarding Selection Preferences for African-Americans.

The issue is different as to the Discriminatory Effects Interest and the Selection Preferences for Hispanics. The jury found such an Interest existed and plaintiffs failed to properly preserve a challenge to that verdict. Therefore, it must be taken as true that the Discriminatory Effects Interest as to Hispanics was shown. However, defendant still would not be entitled to judgment as a matter of law unless a reasonable jury could only conclude that the Selection Preferences for Hispanics were narrowly tailored to that Interest. As previously discussed, a reasonable jury could find that a key assumption of Siskin's analysis was undermined. With a key assumption undermined, Siskin's projections as to the number of Hispanic sergeants there would have been absent discrimination would not represent a plausible lower-bound estimate of the shortfall. Since a reasonable jury could find that the narrow tailoring requirement is not satisfied, defendant would not be entitled to judgment as a matter of law as to the Discriminatory Effects Interest and Selection Preferences for Hispanics.

Defendant's Rule 50(b) motion will be denied.

III. SUMMARY JUDGMENT MOTIONS

A. Summary Judgment Standard

On a motion for summary judgment, the entire record is considered with all reasonable inferences drawn in favor of the nonmovant and all factual disputes resolved in favor of the nonmovant. Lesch v. Crown Cork Seal Co., 282 F.3d 467, 471 (7th Cir. 2002); Hilt-Dyson v. City of Chicago, 282 F.3d 456, 462 (7th Cir. 2002); Schneiker v. Fortis Insurance Co., 200 F.3d 1055, 1057 (7th Cir. 2000). The burden of establishing a lack of any genuine issue of material fact rests on the movant. Outlaw v. Newkirk, 259 F.3d 833, 837 (7th Cir. 2001); Wollin v. Gondert, 192 F.3d 616, 621-22 (7th Cir. 1999). The nonmovant, however, must make a showing sufficient to establish any essential element for which he, she, or it will bear the burden of proof at trial. Celotex Corp. v. Catrett, 477 U.S. 317, 322 (1986); Billings v. Madison Metropolitan School District, 259 F.3d 807, 812 (7th Cir. 2001). The movant need not provide affidavits or deposition testimony showing the nonexistence of such essential elements. Celotex, 477 U.S. at 324. Also, it is not sufficient to show evidence of purportedly disputed facts if those facts are not plausible in light of the entire record. See NLFC, Inc. v. Devcom Mid-America, Inc., 45 F.3d 231, 236 (7th Cir.), cert. denied, 515 U.S. 1104 (1995); Covalt v. Carey Canada, Inc., 950 F.2d 481, 485 (7th Cir. 1991);Collins v. Associated Pathologists, Ltd., 844 F.2d 473, 476-77 (7th Cir.), cert. denied, 488 U.S. 852 (1988). As the Seventh Circuit has summarized:

The party moving for summary judgment carries the initial burden of production to identify "those portions of the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, which it believes demonstrate the absence of a genuine issue of material fact." Logan v. Commercial Union Ins. Co., 96 F.3d 971, 978 (7th Cir. 1996) (citing Celotex Corp. v. Catrett, 477 U.S. 317, 323, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986) (citation and internal quotation omitted)). The moving party may discharge this burden by "'showing' — that is, pointing out to the district court — that there is an absence of evidence to support the nonmoving party's case." Celotex, 477 U.S. at 325, 106 S.Ct. 2548. Once the moving party satisfies this burden, the nonmovant must "set forth specific facts showing that there is a genuine issue for trial." Fed.R.Civ.P. 56(e). "The nonmovant must do more, however, than demonstrate some factual disagreement between the parties; the issue must be 'material.'" Logan, 96 F.3d at 978. "Irrelevant or unnecessary facts do not preclude summary judgment even when they are in dispute." Id. (citation omitted). In determining whether the nonmovant has identified a "material" issue of fact for trial, we are guided by the applicable substantive law; "[o]nly disputes that could affect the outcome of the suit under governing law will properly preclude the entry of summary judgment." McGinn v. Burlington Northern R.R. Co., 102 F.3d 295, 298 (7th Cir. 1996) (citation omitted). Furthermore, a factual dispute is "genuine" for summary judgment purposes only when there is "sufficient evidence favoring the nonmoving party for a jury to return a verdict for that party." Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). Hence, a "metaphysical doubt" regarding the existence of a genuine fact issue is not enough to stave of f summary judgment, and "the nonmovant fails to demonstrate a genuine issue for trial 'where the record taken as a whole could not lead a rational trier of fact to find for the non-moving party . . . .'" Logan, 96 F.3d at 978 (quoting Matsushita Elec. Indus. Co., Ltd. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986)).

Defendant's motion for summary judgment will be addressed first because it need succeed on only one compelling interest to be entitled to a judgment dismissing all remaining claims. Plaintiffs must be entitled to summary judgment as to all three compelling interests in order to be entitled to a finding of liability. Plaintiffs' CBA issue is the only contention that goes to all three compelling interests.

B. Defendant's Summary Judgment Motion

1. Discriminatory Effects Interest

Defendant has provided some additional evidentiary documents that were not presented at trial, including affidavits of Joyce. None of the additional evidence, however, alters the analysis contained in the ruling on defendant's Rule 50(b) motion. See § II (C)(3), supra. Defendant is not entitled to summary judgment as to the Discriminatory Effects Interest.

2. Adverse Impact Interest

The Adverse Impact Interest is discussed in more detail regarding plaintiffs' summary judgment motion. See § III (C)(3), infra. Evidence at trial showed that African-Americans and Hispanics were promoted at rates well above that necessary to avoid violating the 80% rule. Therefore, a factfinder would not be required to find that the Score Adjustments were narrowly tailored to the Adverse Impact Interest. Defendant is not entitled to summary judgment based on the Adverse Impact Interest.

3. Operational Need Interest

As to the Operational Need Interest, the jury was instructed:

An interest in diversifying law enforcement personnel so as to have an effective law enforcement organization may constitute a compelling interest. For the Operational Need Interest to be compelling, for the Promotion Date, there must be sufficient evidence that (a) an important aspect of the effectiveness of the Chicago Police Department's law enforcement mission was significantly diminished by an underrepresentation of the Identified Group at the sergeant rank and (b) increasing the representation of the Identified Group at the sergeant rank was necessary to the accomplishment of an important law enforcement objective.

Jury Instr. [330] 33.

As to narrow tailoring of the Operational Need Interest, the jury was further instructed:

In connection with narrow tailoring and the Operational Need Interest, a factor you should consider is whether the number of Identified Group members promoted to sergeant on a particular Promotion Date was a plausible lower-bound estimate of the number of members of that Identified Group that needed to be promoted to sergeant in order to sufficiently diversify the sergeants rank so as to accomplish an important law enforcement objective.

Consistent with the instructions given at the trial, the Seventh Circuit subsequently held that a police department's operational need for diversity in its workforce can be a compelling interest for engaging in affirmative action promotions. Reynolds, 296 F.3d at 529-30. TheReynolds case involved CPD promotions to lieutenant that occurred in 1990 and 1991. See id. at 525. "Many courts . . ., including our own have at least left open a small window for forms of discrimination that are supported by compelling public safety concerns, such as affirmative action in the staffing of police departments and correctional institutions. Especially in a period of heightened public concern with the dangers posed by international terrorism, effective police work must be reckoned a national priority that justifies some sacrifice of competing interests." Id. at 530. In Reynolds, it was held that the City conclusively proved a two-fold operational need to have more Hispanics at the lieutenant rank. One was for internal operation in the supervision of employees, including setting a tone and sensitizing non-Hispanic officers. The second was an external operational need to serve as ambassadors to the community. Id. at 529-30. "The imperative need for this discrimination had to be proved and not merely conjectured. It would not have done for the City merely to have presented plausible argumentation or to have appealed merely to common sense. It had to substantiate its position with evidence." Id. at 530.

In support of its Operational Need Interest in the present case, defendant points to the testimony of two experts, testimony of high-ranking CPD officials, and census data as to the City's African-American and Hispanic population.

Samuel Walker, Ph.D., testified at trial. He is a professor in the Department of Criminal Justice at the University of Nebraska, where he has taught since 1947. He has published a number of books and articles on policing, including on civil rights, race, and ethnicity in the criminal justice system. This has included research as to the employment of African-Americans and Hispanics as police officers in urban areas. Related to the present case, Walker conducted a survey as to the perceptions and attitudes of Chicago residents regarding minority employment in supervisory and command positions in the CPD.

Walker testified that there was a consensus of opinion among experts in his field that policing is a human service activity where "issues of communication, issues of trust, issues of understanding are absolutely crucial to the effectiveness of a police department." Tr. Vol. VII 20 (Jan. 28, 2002). He further testified that accreditation standards for police departments include that it reflect the diversity of the community it polices. Id. at 21-22. He testified that he did not know of a single study supporting that minority representation is unimportant, id. at 22, and that the consensus opinion in his field was that discriminatory employment practices and lack of minority officers is a problem that will impede the effectiveness of a police force, id. at 20. He also testified that, "in a typical police department, sergeants operating — field sergeants in patrol area and actually in the criminal investigation area will be directly involved in many specific calls and contacts between citizens. . . . [A] good sergeant will back up officers in any incident where there is potentially a problem, they often will give advice instruction to officers, they will sometimes take command if it is a particularly difficult or problem situation." Id. at 23.

As to his 1996 survey of Chicago residents, the conclusions were "that the employment of minority police officers in command position matters to Chicago residents. The people we surveyed said that it is important, they said they noticed, they said that increased minority officers helps to improve the police department. Not a single one of our respondents said that the Chicago Police Department had actually got worse because of more minority employment." Id. at 19. See also id. at 23-36. He also testified that "quite a number" of respondents, both African-American and White, believed there were not enough minority officers in the command ranks.Id. at 33.

Defendant also presented the testimony of Tom Potter, former Chief of Police for the City of Portland and a police consultant since 1993. He testified that diversity in the supervisory ranks of a police department is necessary to being a more effective law enforcement organization. Id. at 70.

. . . And the reason I do is that . . . I think the sergeant's job is not only the most significant promotion a person gets in the organization, I think it's one of the most important jobs in an organization. It's where the rubber meets the road, it's where the people, the sergeants go out there with the officers, they cover them, they mentor them, they make decisions out there in the field, and that's where most people have contact with the police, both minorities and Whites, and that that's where you can make a good impression or a bad impression, and that I found consistently that in minority communities, it worked to our advantage that we could have minority command and supervisors out in the field covering minority officers and White officers, but having greater entry into the communities, their ethnic communities that they represented.

Potter testified that minority sergeants increase the effectiveness of a police organization internally, by decreasing overt racism within a department, id. at 66-67, and by serving as mentors and role models for minority patrol officers, id. at 80-81. He also testified that minority sergeants increase the effectiveness of a police department through community interaction because they are often the first supervisors on a scene and recognized as such, and may have a better ability or be perceived as relating better to a particular community. Id. at 78-82. He testified that building such trust is particularly important for community policing strategies, which the CPD began implementing in the mid-1980's. Id. at 68-69. Additionally, he testified that diversity in a police force is also particularly important in cities like Chicago where there are neighborhoods with high concentrations of minorities. Id. at 75-76.

Defendant also presented the testimony of current and former CPD supervisory officials. Terry Hilliard is the current Superintendent and began as a patrol officer in 1968. Id. at 120. Gerald Cooper was a CPD patrol officer and sergeant from 1961 until 1978 and employed at the lieutenant rank as an attorney in the CPD legal department from 1985 until his retirement in 1995. Id. at 95-100. LeRoy Martin was Superintendent under three mayors from October 1987 until his retirement in April 1992. Tr. Vol. VIII at 16-17 (Jan. 29, 2002). He began as a CPD patrol officer in 1955, became a sergeant in 1965, a lieutenant in 1975, a captain in 1981, and, subsequently, a commander. Id. at 18-21. Joseph DeLopez had been a deputy superintendent for approximately two years at the time he testified. Tr. Vol. V at 115 (Jan. 23, 2002). He began as a CPD patrol officer in 1971. Id. at 117. All four testified regarding the need for minority sergeants.

Hilliard testified that sergeants have to supervise patrol officers, including training, disciplining, mentoring, and supporting; patrol officers interact with their sergeants more than any other CPD management official. Vol. VII at 132. See also id. at 183-84 (Cooper). Sergeants also have substantial interaction with the community, especially under community policing. Id. at 133. All four testified race and ethnicity played a role in police-community interactions and that the presence of minorities at the sergeant rank was important. Vol. V at 126-30, 137-41; Vol. VII at 134-35, 140-45, 182-85; Vol. VIII at 31-33. Martin and Cooper both testified about a need for minority sergeants, particularly Hispanic sergeants, at the time of the promotions at issue in the present case and that this was considered in making the promotions at issue. Vol. VII at 105-07, 182-86; Vol VIII at 24-27, 31-33, 37.

Defendant presented evidence of the increase in the percentage of African-American and Hispanic residents of the City, from 22.9% African-American and 3.1% Hispanic in 1960 to 38.6% African-American and 19.6% Hispanic in 1990. Def. Exh. 34B. See also Def. Exh. 36A, attachments 9 (at 41), 29. Defendant also presented evidence as to which neighborhoods had high concentrations of African-Americans or Hispanics as of each census from 1960 to 1990. Def. Exh. 36A, attachments 18-21, 22-24. As of December 31, 1987, the CPD had 269 (22.6%) African-American sergeants and 44 (3.7%) Hispanic sergeants. As of December 31, 1991, the numbers were 287 (23.8%) African-American sergeants and 66 (5.5%) Hispanic sergeants. Def. Exh. 28A-B. Martin, who was Superintendent in 1988, testified that, considering days off and the need to staff three shifts seven days a week, 44 Hispanic sergeants allowed for, at most, eight Hispanic sergeants to be working in the entire city on a particular shift. That was not enough to provide a diverse workforce for the then four predominantly Hispanic police districts, let alone maintain diversity in other districts and at other assignments for sergeants. Vol. VIII at 31-33.

Plaintiffs point to other evidence that they contend creates at least a factual dispute as to whether the Operational Need Interest was proven to be compelling. Plaintiffs point to testimony supporting that police officers can be effective with persons of other races. However, no witness testified that only African-American officers could be effective with African-Americans or only Hispanic officers can be effective with Hispanics. The testimony was that having a sufficient level of minority sergeants changed the perceptions of both minority citizens and the minority patrol officers who were supervised by sergeants; it sensitized all officers to cultural issues; and it increased the possibility of having an officer available who would better be able to handle the problem. No one testified that only African-Americans should work in African-American neighborhoods, only Hispanics in Hispanic neighborhoods, and only Whites in White neighborhoods. No such finding is necessary to support a compelling interest in a need for diversity. See Wittmer v. Peters, 87 F.3d 916, 920 (7th Cir. 1996) (that belonging to the same race as an inmate is not a requirement for effectiveness as a guard did not cast doubt on a prison's need for an African-American male lieutenant). Additionally, the fact that there are differences within the African-American community and different nationalities or other differences within the Hispanic community, does not undermine the testimony as to a need for diversity. Neither does it matter, as plaintiffs assert, whether particular Hispanic sergeants have the ability to speak Spanish. See Reynolds, 296 F.3d at 530.

Plaintiffs contend that Potter's testimony may be disregarded because he did not rely on an actual study and was unfamiliar with Chicago. Potter's credentials were established and plaintiffs did not present any expert of their own who disputed Potter's or Walker's testimony as to a police department's need for diversity. Potter' s minimal assumptions as to Chicago, that it had neighborhoods with high concentrations of minorities and that it was using community policing techniques, were both supported by uncontroverted evidence. No basis in the evidence existed for rejecting Potter's and Walker's testimony as to a police department's internal and external needs for diversity.

Plaintiffs now question the validity of Walker's survey based on the number of nonresponses. Walker, however, testified the survey had sufficient respondents. Vol. VII at 27-29. See also Def. Exh. 31 at 12-13. Plaintiffs did not question Walker about this nor did they present their own expert to raise questions as to the survey's validity or reliability. Walker's testimony as to the adequacy of the survey's methodology must be accepted since uncontested. Moreover, even if the survey were to be completely disregarded, it is not essential to finding a compelling interest.

Apparently taking Walker's survey as valid, plaintiffs also contend that it supports that diversity at the sergeant level is not important to the public because only 18.5% of those surveyed had contact with a command officer. Def. Exh. 31 at 20, App. II at 14. Plaintiff, however, ignores that 9.1% had said they had contact within the previous year.Id., Vol. VII at 42-43. Walker testified that national surveys had shown that 20% of people have contact with any police officer during a year.Id. at 41-42. Thus, evidence supports that a substantial percentage of those who actually have contact with the police have contact with a command officer.

Plaintiffs also point to the facts that a majority of those surveyed who had contact with a command officer had contact with a White officer (56.7%), Def. Exh. 31, App. II at 14, and that, of all those who had contact, 80.5% were satisfied with the contact, id. at 15. First, no breakdown is provided as to whether satisfaction rates differed based on the races of the command officers, the races of the respondents, and a combination thereof. In any event, even if citizens are generally satisfied with the conduct of command officers, that does not mean that the presence of minority sergeants cannot contribute substantially to an improvement in police effectiveness.

Defendant presented substantial evidence establishing the Operational Need Interest, both as to internal supervision needs and external community contact needs. Plaintiffs did not present any evidence raising any significant issue as to defendant's Operational Need Interest. A reasonable trier of fact could only find that the Operational Need Interest existed from 1987 through 1991. Cf. Reynolds, 296 F.3d at 529-30. There is no genuine factual dispute as to this issue.

However, it still must be considered whether there is no genuine factual dispute as to the Selection Preferences being narrowly tailored to the Operational Need Interest. Martin's testimony that 44 Hispanic sergeants was not adequate (for diversity purposes) to cover districts with high concentrations of Hispanics as well as other districts and assignments is uncontested. After implementing the Selection Preferences at issue in the present case, there were 66 Hispanic sergeants, a 50% increase. That would allow for approximately 12 Hispanic sergeants per shift. The only reasonable inference from Martin's uncontested testimony would be that amount is still not enough to serve all of the CPD's operational needs regarding Hispanic sergeants. Therefore, the Selection Preferences did not result in an increase of Hispanic sergeants that exceeded the plausible lower-bound estimate of the shortfall that needed to be overcome in order to serve defendant's compelling Operational Need Interest.

As to the years and promotions at issue in this case, the uncontested facts support both that (a) there was a compelling Operational Need Interest for increasing the number of Hispanic sergeants and (b) the Selection Preferences for Hispanic sergeants that were imposed were narrowly tailored.

The evidence as to African-American sergeants is different. Unlike Martin's testimony as to Hispanic sergeants, there is no evidence specifically addressing the number of African-American sergeants and spreading them out among the various police districts. While Walker and Potter testified about a need for diversity, they did not testify as to a particular number, percentage., or ratio that would be necessary to satisfy that need. Cooper testified that, during the pertinent time period, the feedback he received from community members was that the CPD did not have an adequate representation of African-American sergeants.See Vol. VII at 105-07, 184. He also testified that he believed more African-American sergeants were needed to serve the CPD's operational needs. Id. at 182-86. Martin also testified that more African-American sergeants were needed to serve the CPD's operational needs. Id. at 37. Plaintiffs did not present any expert or CPD official who testified that there were already a sufficient number of African-American sergeants to serve the CPD's operational needs nor do they presently provide additional, nontrial evidence to that effect. The cross-examination of Cooper and Martin also raised no genuine dispute as to their testimony of a need for more African-American sergeants. Therefore, on summary judgment, it must be taken as true that the CPD's operational needs required more African-American sergeants.

Neither Cooper nor Martin specified exactly how many more African-American sergeants were needed to serve this need. Regarding internal operational needs, though, Martin referred to a need to have a sergeant rank that, in terms of race, reasonably represented the patrol ranks. Id. at 27. However, as to the percentage of African-Americans at the patrol officer and sergeant ranks, parity had essentially been achieved prior to the promotions at issue, see Def. Exh. 22; Petit V, 2001 WL 629306 at *11 n. 17, and therefore, as to internal operational need, parity alone could not be a justification for a Selection Preference favoring the promotion of African-Americans to sergeant nor a lower-bound estimate of the promotions necessary to satisfy that need.

Defendant points to the African-American population of the City as compared to the percentage of African-Americans at the sergeant rank. However, no evidence before the court establishes a particular ratio between minority resident population in the City and minorities in the sergeant rank that is minimally necessary to serve external operational needs.

As of December 31, 1987, the CPD had 269 African-American sergeants, which was 22.6% of sergeants. Def. Exh. 28A. As of December 31, 1991, that is after the promotions at issue had all been made, there were 287 African-American sergeants, which was 23.8% of sergeants. Def. Exh. 28B. There is uncontroverted evidence that the 269 African-American sergeants as of the end of 1987 were insufficient to serve the CPD's operational needs. By the end of the promotions at issue, the absolute number of African-American sergeants had only increased by 18. Because the overall number of sergeants had also increased (from 1191 to 1208, See Def. Exh. 28A-B), the difference in the percentage of African-American sergeants was 1.2%. Stated another way, had the total number of sergeants stayed at 1191, there would have had to have been 14 more Afridan-American sergeants for 23.8%. Viewed as a percentage of White sergeants, 14 is 1.6% of the 876 White sergeants employed as of the end of 1987 and 18 is 2.1% of the 855 White sergeants employed as of the end of 1991. See Def. Exh. 28A-B. The increases in the number of African-American sergeants (and corresponding decreases in the number of White sergeants) were so minimal that a reasonable finder of fact could only find that the promotions of African-American sergeants was well within a lower-bound estimate of the number necessary to overcome the uncontroverted shortfall in the number of African-American sergeants needed to serve the CPD's operational needs. Compare Reynolds, 296 F.3d at 530-31.

In addition to keeping the promotions to a plausible lower-bound estimate of that necessary to serve the CPD's external operational needs, the promotions were otherwise narrowly tailored. The additional identified group members that were promoted had all qualified under the Examination and had scored within two points of the lowest scoring White officer who was promoted, which was within the Examination's standard measurement of error. Cf. Boston Police Superior Officers Federation v. City of Boston, 147 F.3d 13, 24 (1st Cir. 1998). The Selection Preferences were used for only a three-year period and there is no evidence of an efficacious alterative that would have had less of an impact on White candidates. The effect on many or most of the remaining plaintiffs was a delayed promotion to sergeant, not a complete exclusion from that position, nor any loss of an existing position. Cf. Majeske, 218 F.3d at 824; Cotter, 192 F. Supp.2d at 354. Additionally, the decrease in the percentage of White sergeants was not large, from 73.7% on December 31, 1987 to 70.7% as of December 31, 1991. See Def. Exh. 28A-B. The only reasonable conclusion based on the undisputed evidence is that narrow tailoring was satisfied.

Defendant satisfied its burden of production as to narrow tailoring, thereby placing the burden of persuasion on plaintiffs. Cf. Petit V, 2001 WL 629306 at *10. Therefore, to defeat summary judgment as to this issue, plaintiffs must present sufficient evidence from which it could be found that narrow tailoring was not satisfied. Plaintiffs did not directly dispute, or point to sufficient contrary evidence, that narrow tailoring was not satisfied for the promotion of African-Americans. Thus, plaintiffs do not show the existence of a genuine factual dispute meriting another trial on this issue. The Selection Preferences for African-American sergeants were narrowly tailored to defendant's Operational Need Interest.

Resolving all genuine factual disputes and drawing all reasonable inferences in plaintiffs' favor, it can only be found and held that, from 1988 through 1991, a compelling Operational Need Interest existed for promoting African-Americans and Hispanics to sergeant and the Selection Preferences employed were narrowly tailored. Defendant is therefore entitled to summary judgment dismissing all remaining claims of plaintiffs. As was previously indicated, see § I, supra, all aspects of plaintiffs' summary judgment will still be addressed.

C. Plaintiffs' Summary Judgment Motion

1. The CBA

Section 10.1 Equal Employment Opportunity

Section 10.2 Non-Discrimination

The Employer will continue to provide equal employment opportunity for all officers, and develop and apply equal employment practices.

The Employer shall not discriminate against officers, and employment-related decisions will be based on qualifications and predicted performance in a given position without regard to race, color, sex, religion, age (40-63), or national origin of the officer. . . . Officers shall not be transferred, assigned or reassigned for reasons prohibited by this Section 10.2.

As is discussed in § II(B)(2), supra, there is no evidence as to the CBA provisions that may have been in effect when the Examination was being prepared and scored and also when the first set of promotions were made in December 1988. Therefore, plaintiffs' contention as to the OBA could only apply to the promotions made on the last four Promotion Dates.

Defendant contends the cited clauses are inapplicable to the promotions at issue because sergeants were not members of the bargaining unit and promotions to positions outside the bargaining unit are not covered by the CBA. Defendant cites to a 1998 arbitration decision to that effect see Sergeant Promotion Examination Grievance 129-97-015 (Arbitrator Cox May 11, 1998) ([351] Exh. 1), and contends this "court" is bound to follow the arbitration decision. The arbitration, however, concerned the 1995 labor agreement that was in effect in 1997. Even if the pertinent contract language is the same as that in the CBA, a construction of the 1995 Agreement would not necessarily be a binding construction of the 1990 CBA. More importantly, unlike the cases cited by defendant (see United Steelworkers of America v. Enterprise Wheel Car Corp., 363 U.S. 593, 599 (1960); Rose v. Bridgeport Brass Co., 487 F.2d 804, 811 (7th Cir. 1973)), which found parties, not the court, bound by prior arbitration decisions, defendant does not contend that any of the remaining plaintiffs were parties to the arbitration through their representative, the FOP. Even if not binding on any plaintiffs, the Arbitration Decision contains persuasive reasoning.

Notwithstand language in § 23.3 of the 1995 Agreement requiring that seniority be considered for promotions, the Arbitrator followed other arbitration decisions holding that a collective bargaining agreement generally does not apply to promotions outside the bargaining unit covered by the agreement. Arb. Dec. at 4 n. 2. The Arbitrator described this principle as being well settled in other jurisdictions and followed in Illinois. See id. at 6 (citing Village of Franklin Park v. Illinois State Labor Relations Board, 265 Ill. App.3d 997, 638 N.E.2d 1144, 1148-49 (1st Dist.), appeal denied, 157 Ill.2d 498, 642 N.E.2d 1277 (1994)). See also Oney v. Kansas City Southern Railway Co., 3 F. Supp.2d 729, 731 (S.D. Tex. 1997). This reasoning applies to the language contained in the CBA.

Plaintiffs cite § 23.3 of the CBA which provides:

"Seniority shall be considered in the promotion of officers covered by this Agreement. In considering officers for promotion, seniority shall in competitive testing, be utilized as a tiebreaker." Plaintiffs contend this shows the CBA applies to all promotions. There are, however, promotions within the bargaining unit, e.g., promotions to detective. Moreover, § 4 of the CBA provides that the City retains the right to "operate and manage its affairs in each and every respect," specifically including promoting officers, unless "expressly" provided otherwise in the CBA. There is no express provision applying the non-discrimination clause to promotions outside the bargaining unit. Therefore, the general rule prevails that a labor contract does not apply to positions outside the bargaining unit. See Oney, 3 F. Supp.2d at 731-32.

It could not have been a breach of § 10.1 or § 10.2 of the CBA for the CPD to consider race in promotions to the nonbargaining unit position of sergeant. Since the underlying contention is without merit, it is unnecessary to consider whether such a breach would otherwise preclude the possibility of an overriding compelling interest.

Plaintiffs' contentions as to the CBA are no basis for granting them summary judgment nor do they involve any genuine factual disputes that preclude granting summary judgment to defendant based on defendant's Operational Need Interest.

2. Discriminatory Effects Interest

Plaintiffs contend that Siskin's analysis must be rejected because it relies on the incorrect assumption that the CPD's qualifications for hiring did not change during the years under consideration in his analysis. See Tr. Vol. VI at 155-56 (Jan. 24, 2002). Plaintiffs point to evidence that a high school education requirement was imposed in 1975, the citizenship requirement was changed to lawful resident in 1975, a psychological test was added in 1981, and a drug screen was instituted in 1985. Plaintiffs, however, point to no evidence showing that any of these changes resulted in more African-Americans or Hispanics being hired than would have been the case without the changes. As Siskin testified, failing to take the changes into account may have understated the shortfall or the effects may have been a wash. Id. at 156-58. Plaintiffs do not present conclusive evidence that Siskin's analysis overstates the shortfall of African-Americans or Hispanics because the analysis fails to consider changes in qualification criteria. Cf. Petit V, 2001 WL 629306 at *10.

Even accepting Siskin's analysis as true, plaintiffs contend narrow tailoring is not satisfied. One of plaintiffs' contentions is that African-American males were overrepresented. This contention was previously addressed and rejected on summary judgment. See id. at *11. Plaintiffs also contend that narrow tailoring was not satisfied because African-Americans and Hispanics were no longer underrepresented at the sergeant rank in that their representation at that rank was within reasonable parity of their representation at the patrol officer rank. That argument was also rejected on summary judgment. Parity within ranks is not the proper measure of a discriminatory effects shortfall. See id.

On plaintiffs' summary judgment motion, resolving all genuine factual disputes and drawing all reasonable inferences in defendant's favor, it must be taken as true that the Discriminatory Effects Interest exists for both African-Americans and Hispanics. Siskin's analysis and his projected shortfalls also must be taken as true. Taking the shortfalls as true, the Selection Preferences were narrowly tailored. See id. at *9-42. Plaintiffs would not be entitled to summary judgment as to the Discriminatory Effects Interest.

3. Adverse Impact Interest

Prior to trial, it was held that the Adverse Impact Interest was a possible compelling interest. See Petit IX, 2001 WL 10481 at *3; Petit II, 31 F. Supp.2d at 61.3-14. See also Cotter, 193 F. Supp.2d at 347-51;Boston Police, 147 F.3d at 21. This Interest is only claimed to justify the Score Adjustments; it is not contended to be a justification for the Out-of-Rank Promotions of Hispanics.

As to the Adverse Impact Interest, the jury was instructed:

Title VII is a federal statute that prohibits discrimination in employment based on a person's race or national origin. In addition to prohibiting intentional discrimination, Title VII prohibits certain employment practices that have an adverse impact on a racial or ethnic group, such as African-Americans or Hispanics. Defendant's claimed Adverse Impact Interest is based on its contention that it had a compelling interest in avoiding a violation of the adverse impact provisions of Title VII.

For the Adverse Impact Interest to be compelling, there must be sufficient evidence of (a) a substantial likelihood that not applying Score Adjustments would result in an adverse impact violation of Title VII and (b) no reasonable alternative to Score Adjustments, such as using a different examination or selection method or validating the 1985-88 Examination within a reasonable period of time before it was necessary to make promotions to the rank of sergeant.

Jury Instr. [330] 30.

The jury was also instructed as to certain Title VII terms that were used at trial including, employment practice, selection rate, adverse impact, and validating a promotional examination. See id. at 31-32. As to adverse impact, the jury was instructed: "An employment practice generally has an 'adverse impact' when the selection rate of one racial or ethnic group is less than four-fifths (4/5) or 80% of the selection rate of the racial or ethnic group with the highest selection rate." Id. at 31. See 29 C.F.R. § 1607.4 (D); Fickling v. New York State Department of Civil Service, 909 F. Supp. 185, 188-89 (S.D.N.Y. 1995).See also Boston Police, 147 F.3d at 21 (80% rule may be a benchmark against which to gauge efforts to remedy past discrimination).

As to narrow tailoring of the Adverse Impact Interest, the jury was instructed:

In connection with narrow tailoring and the Adverse Impact Interest, a factor you should consider is whether the number of Identified Group members promoted to sergeant on a particular Promotion Date was a plausible lower-bound estimate of the number of members of that Identified Group that needed to be promoted to sergeant in order to avoid an adverse impact violation as of that particular Promotion Date.

Plaintiffs contend they are entitled to summary judgment because, even assuming a compelling interest has been adequately proven, the evidence before the court cannot support that the Selection Preferences were narrowly tailored to the Adverse Impact Interest. One of plaintiffs' contentions is that defendant did not attempt the alternative of avoiding an adverse impact violation by validating the Examination. However, testimony supports that there was a need for sergeants and the CPD did not want to delay hiring while attempting to create a validated examination. Compare Billish v. City of Chicago, 989 F.2d 890, 894-95 (7th Cir.), cert. denied, 510 U.S. 908 (1993) (no evidence of urgent need to promote firefighter captains that might justify racially adjusting test results). On plaintiffs' summary judgment motion, this testimony must be taken as true.

Plaintiffs' other ground for contending narrow tailoring cannot be satisfied is that the promotions went well beyond avoiding a violation of the 80% Rule. Defendant could have avoided an adverse impact violation by adjusting the scoring so that, out of the top 500 persons they expected to hire, African-Americans and Hispanics were included at approximately 80% of the rate that Whites were included. Instead, the City attempted to adjust the scores so that each racial group would have essentially the same selection rate.

At the time defendant was racially standardizing the Examination, it expected to promote 500 patrol officers to sergeant. Assuming 500 patrol officers were to be promoted, the distribution would have to be as follows to promote African-Americans and Hispanics at approximately 80% of the White selection rate.Race Candidates Promoted Rate White Rate 192 24 12.50

Looking to the patrol officers actually promoted, the numbers are even further beyond avoiding falling below the 80% benchmark. The 458 actually promoted were as follows:

Selection Race Candidates Promoted Rate White 2274 298 13.10% Afric. 931 119 12.78 Hisp. 192 41 21.35 All 3397 458 13.48%

80% of the 13.10% White selection rate is 10.48%. African-Americans and Hispanics were promoted well above that rate. A 10.48% selection rate for African-Americans would be 98 promotions and for Hispanics it would be 20. As the promotions actually occurred, African-Americans and Whites were selected at less than 80% of the Hispanic selection rate (17.08%).

Defendant does not contend that the Adverse Impact Interest justified the Out-of-Rank promotions of Hispanics. Even if those candidates and promotions are excluded from consideration, the 402 promoted were as follows:

SelectionRace Candidates Promoted Rate White 2252 276 12.26% Afric. 921 109 11.83 Hisp. 168 17 10.12 All 3341 402 12.03%

80% of the White selection rate is 9.80%. A 9.80% selection rate for African Americans would result in 90 promotions and a 9.80% selection rate for Hispanics would result in 16 promotions (16.46 to be precise). Although, Hispanics hired directly from the racially standardized list were at approximately 80% of the White selection rate, it cannot be ignored that, at the same time, an additional 24 Hispanics were being promoted Out-of-Rank, some of whom would have otherwise been promoted based on the racially standardized list. Cf. Billish, 989 F.2d at 895 (combination of racially standardizing examination results and out-of-rank promotions not justified).

If there had been no Out-of-Rank Promotions and, instead, the first 458 candidates on the racially standardized list had been promoted, the distribution would have been as follows:

Selection 80% ofRace Candidates Promoted Rate Rate White 2271 304 13.39% 10.71% Afric. 928 124 13.36 10.66 Hisp. 192 30 15.63 12.50 All 3391 458 13.51%

If the racially standardized list had been strictly followed for 458 promotions, there would not have been a violation of the 80% benchmark. However, more African-Americans and Hispanics would have been hired than would be necessary to avoid a violation of the 80% benchmark. A 10.71% selection rate for African-Americans would be 99 promotions and for Hispanics would be 21 promotions.

Under the above analyses, it is clear that African-Americans and Hispanics were promoted at rates significantly above the 80% benchmark used in adverse impact analysis and defendant does not contend otherwise. Defendant concedes that undisputed evidence shows that the Score Adjustments resulted in the promotion of African-Americans and Hispanics at a rate significantly above that necessary to avoid an adverse impact violation. Defendant, though, contends this is appropriate because, when there is a violation, it is appropriate to provide full relief. That contention relies on two unfounded assumptions. First, it assumes that a violation is being remedied. That is not true of the Adverse Impact Interest. The purpose behind this Interest is to avoid a benchmark violation, not remedy a violation that has already occurred. It only takes an 80% selection rate to avoid violating the benchmark and narrow tailoring required that the CPD aim for a plausible lower-bound estimate of that necessary to avoid the violation. But even assuming the CPD was permitted to aim for the so-called full remedy, the second unfounded assumption is that an appropriate remedy is that each racial group have the same selection rate. There is, however, no evidence that each racial group would have been selected at the same rate had the CPD been able to develop a valid promotional examination. See Billish, 989 F.2d at 896 (quoting United States v. City of Chicago, 870 F.2d at 1256, 1261 (7th Cir. 1989)) ("There is no iron law of human behavior that every racial or ethnic group will perform equally well on nonbiased examinations in all fields of human endeavor."); Middleton, 92 F.3d at 408 (same). It cannot be assumed that what the City did was simply provide a full remedy for the statutory violation it otherwise expected to commit.

Defendant's other contention is that it was not feasible to adjust scores in a manner designed to satisfy the 80% Rule and that any such "partial standardization" would be a "half-hearted remedy" and not "principled." Def. Response [350] at 24 n. 23. But just as scores can be given a standard distribution to produce the same results for different tests and different racial groups, it should be feasible to adjust them so that groups have success in different ratios. See, e.g., Billish, 989 F.2d at 895 (test results adjusted so that minority selection rate was 84% of White selection rate). Moreover, it is no less principled to adjust to an arbitrary 80% selection rate than it is to adjust to an arbitrary 100% selection rate. If an adjustment to the 80% rate is unprincipled, then the Adverse Impact Interest that is primarily measured by that rate cannot be a principled compelling interest.

Regardless of whether adjusting to an 80% rate can be feasible or principled, the law is clear that affirmative action imposed based on a compelling interest must be narrowly tailored to that interest, which means limiting any preferences to a plausible lower-bound estimate of what is needed to serve the compelling interest. If avoiding an adverse impact violation constitutes a compelling interest, it must be narrowly tailored to avoiding a violation of the 80% Rule. See Cotter, 193 F. Supp.2d at 352. In this case, defendant did not narrowly tailor its remedy to that objective. Therefore, plaintiffs would be entitled to summary judgment as to the Adverse Impact Interest.

4. Operational Need Interest

Defendant is entitled to summary judgment based on the Operational Need Interest. Therefore, plaintiffs cannot be entitled to summary judgment on this Interest.

IV. CONCLUSION

Defendant is entitled to summary judgment in its favor based on the Operational Need Interest. Therefore, judgment will be entered dismissing the remaining claims in this case. Alternatively, if this decision were to be overturned, the case would be subject to being retried, but limited to defendant raising the Operational Need Interest and Discriminatory Effects Interest. Also, at any retrial, the jury would have to be instructed that a Discriminatory Effects Interest as to Hispanics is already established.

IT IS THEREFORE ORDERED that the cause of action of plaintiff Stanley Surdej (deceased) is dismissed without prejudice. Defendant's motion for entry of judgment as a matter of law [335] is denied. Defendant's motion for summary judgment is granted. Plaintiffs' motion for judgment. as a matter of law (treated in part as a motion for summary judgment) [336] is denied. The Clerk of the Court is directed to enter judgment in favor of defendants and against plaintiffs denying plaintiffs' causes of action with prejudice except that the following are dismissed without prejudice (a) plaintiff Stanley Surdej's cause of action, (b) plaintiff Lodge 7 of the Fraternal Order of Police's cause of action, and (c) prayers for injunctive relief.

MEMORANDUM OPINION AND ORDER

WILLIAM T. HART, United States District Judge.

Following a mistrial, the. remaining parties in this action filed Rule 50(b) motions for judgment as a matter of law and also moved for summary judgment. The Rule 50(b) motions were denied and defendant's motion for summary judgment was granted. See Petit v. City of Chicago, 2002 WL 31409589 (N.D. Ill. Oct. 24, 2002) ("Petit XI"). On October 24, 2002, a final judgment was entered in favor of all defendants. See Docket Entry [365].

Rule 59(e)'S ten-day time period for filing a motion to alter or amend expired on November 7, 2002. On November 4, 2002, plaintiffs timely filed a motion to alter and set aside judgment [368] which they presented on November 6, 2002. On the same dates, plaintiffs also filed and presented a motion to extend the time to file an amended motion to alter and set aside judgment [367]. The motion for extension of time was denied, see Minute Order dated November 6, 2002 [369], because this court lacks authority to extend the time period for filing a Rule 59(e) motion. Fed.R.Civ.P. 6(b); Talano v. Northwestern Medical Faculty Foundation, Inc., 273 F.3d 757, 761 (7th Cir. 2001). Later on November 6, 2002, plaintiffs filed their amended motion to alter and set aside judgment (366] ("Pl. Am. Motion to Alter"). Because the amended motion was filed within the time for bringing a Rule 59(e) motion, it will be considered. Additionally, because the amended motion includes all the contentions contained in the initial motion, only the amended motion will be discussed and the, initial motion to alter and set aside judgment will be denied without prejudice,

This lawsuit has been pending for more than 12 years. The parties have had substantial, opportunity for discovery and the briefing of issues.See Petit XI, 2002 WL 31409589 at *5; Petit v. City of Chicago, 2001 WL. 1143163 *1 (N.D. Ill. Sept. 27, 2001) ("Petit VII"). The parties were also provided with substantial opportunity and time to brief the Rule 50(b) and summary judgment motions. See Docket Entries [335]-[363]. A Rule 59(e) motion is not the appropriate mechanism for presenting new issues or facts that could have been raised during the pendency of the original proceeding, but were not. Bordelon v. Chicago School Reform Board of Trustees, 233 F.3d 524, 529 (7th Cir. 2000); Divane v. Krull Electric Co., 194 F.3d 845, 850 (7th Cir. 1999); United States v. 47 West 644 Route 38, Maple Park, Ill., 190 F.3d 781, 783 (7th Cir. 1999), cert. denied, 529 U.S. 1005 (2000); Moro v. Shell Oil Co., 91 F.3d 872, 876 (7th Cir. 1996); Caisse Nationale de Credit Agricole v. CBI Industries, Inc., 90 F.3d 1264, 1269-70 (7th Cir. 1996); Publishers Resource, Inc. v. Walker-Davis Publications, Inc., 762 F.2d 557, 561 (7th Cir. 1985) (quoting Keene Corp. v. International Fidelity Insurance Co., 561 F. Supp. 656, 665-66 (N.D. Ill. 1982), aff'd, 736 F.2d 388 (7th Cir. 1984)).

In the present motion, plaintiffs first dispute the holding that the CBA's non-discrimination provision does not apply to promotions to sergeant. See Petit XI, 2002 WL 31409589 at *27-29 (§ III(C)(1)). Plaintiffs again rely on CBA § 23.3's provision that "Seniority shall be considered in the promotion of officers covered by this Agreement" as showing all provisions of the CBA apply to all promotions. Plaintiffs contend the court mistakenly assumed that provision could be referring to promotions to detective. See Petit XI, 2002 WL 31409589 at *28. Plaintiffs point to Robert Joyce's testimony at the Majeske trial, which they contend shows that detective is an assignment, not a promotion. The testimony, though, is that a detective is a "police officer assigned as a detective" and that it is an "appointive position," not a "permanent career service appointment" as with a sergeant. Joyce did not testify that a detective appointment, unlike a sergeant appointment, is not a promotion. To the contrary, opinions in both the Majeske case and consolidated discovery proceedings repeatedly refer to detective as a promotion. See e.g., Majeske v. City of Chicago, 218 F.3d 816, 818 (7th Cir. 2000), cert. denied, 531 U.S. 1079 (2001); Majeske v. Fraternal Order of Police, Local Lodge No. 7, 94 F.3d 307, 308 (7th Cir. 1996);Majeske v. City of Chicago, 29 F. Supp.2d 872, 873 (N.D. Ill. 1998);Maleske v. City of Chicago, 1998 WL 312016 *1 (N.D.Ill. June 4, 1998); In re Chicago Police Officer Promotions, 1995 WL 103296 *1 (N.D. Ill. March 2, 1995); In re Chicago Police Officer Promotions, 1993 WL 322834 *1 (N.D. Ill. Aug. 23, 1993). The Illinois Appellate Court has referred to the Chicago detective position as a promotion covered by the collective bargaining agreement in effect in 1995. See McArdle v. Rodriguez, 277 Ill. App.3d 365, 659 N.E.2d 1356, 1360 (1st Dist. 1995) ("Unlike lieutenants, detectives were not members of the career service, and merit promotions for detectives were specifically authorized by a Police Department general order and the detectives' collective bargaining agreement between the City and the police union."). The Arbitration Decision ([351] Exh. 1, at 6), also refers to detective as a promotion. Plaintiffs do not point to evidence or law adequately raising an issue that appointment as a detective is not a promotion.

But even if the detective position is not a promotion, there is no evidence that there are not other possible promotions to which § 23.3 is referring. Moreover, there is still no express language in the CBA providing that the CBA in general or the non-discrimination provision in particular applies to promotions to the non-bargaining unit position of sergeant. For the reasons stated in Petit XI, 2002 WL 31409589 at *28, the lack of an express provision requires that the non-discrimination provision of the CBA be construed as not applying to sergeant promotions. The holding contained in § III(C)(1) of Petit XI will not be modified.

Plaintiffs object to the alternative holding in Petit XI that, in the event of any retrial, it would stand as established that a Discriminatory Effects Interest exists as to the promotion of Hispanics. Id. at *13, 19, 29, 33. Plaintiffs, however, do not make any argument against the holding that they failed to adequately preserve any Rule 50 challenge to that jury determination. See id. at *5-6 (§ II(B)(1)), 13. The alternative holding will stand.

Plaintiffs contend that, in rejecting their summary judgment motion as to the Discriminatory Effects Interest, see id. at *29-30, an incorrect burden of proof was applied. The prior holdings as to the applicable burden of proofs and their application in a jury trial will not be modified. See id. at *5 n. 11, 11-12; Petit v. City of Chicago, 219 F. Supp.2d 917 (N.D. Ill. 2002) ("Petit X"). The expert testimony of Bernard Siskin was the principal evidentiary basis for holding that a factual dispute exists as to the Discriminatory Effects Interest. Plaintiffs object to the consideration of that testimony. Plaintiffs' objections to permitting Siskin to be substituted for defendant's previously retained expert were previously denied and plaintiffs were provided with Siskin's report years before the trial was held and were provided with more than adequate opportunity to retain their own expert and/or analyze Siskin's opinions. See Petit v. City of Chicago, 1999 WL 66539 *1-2 (N.D. Ill. Feb. 8, 1999) ("Petit III"); Petit v. City of Chicago, 2002 WL 10481 *6 (N.D. Ill. Jan. 3, 2002) ("Petit IX"). See also Petit v. City of Chicago, 2001 WL 629306 *9-11 (N.D. Ill. May 25, 2002) ("Petit V"). Plaintiffs' present contention that they had inadequate opportunity for the discovery necessary to contest Siskin's testimony is without any merit.

Summary judgment in the City's favor was based on its Operational Need Interest. See Petit XI, 2002 WL 31409589 at *20-27(§ III(B)(3)). As to the promotion of Hispanics, it was based on both internal and external operational needs. See id. at *25-26. As to the promotion of African-Americans, summary judgment was based on external operational needs, there being a genuine factual dispute as to the promotions being narrowly tailored to internal operational needs for more African-American sergeants. Id. at *26-27. Plaintiffs raise various objections to the Operational Need Interest holdings.

First, plaintiffs contend defendant did not move for summary judgment based on this ground. To the contrary, defendant did move for summary judgment based on all three of the asserted compelling interests. See Petit XI, 2002 WL 31409589 at *8 (citing Def. Response to Pl. Motion for Judgment as a Matter of Law [350] at 1-2 n. 2).

Next, plaintiffs rely on. . . deposition testimony of LeRoy Martin, Pl. Am. Motion to Alter [366], Exh. E at 63-64, that they contend states he did not believe additional Hispanics and African-Americans were needed to. serve the CPD's operational needs. Since this evidence was not provided or relied upon in response to defendant's summary judgment motion, it is inappropriately relied upon to support the Rule 59(e) motion. Moreover, plaintiffs had the opportunity to cross examine Martin at trial, but did not attempt to use this deposition testimony to impeach his trial testimony. Further, even if the deposition testimony were to be considered, it does not support plaintiffs' present contention. It is unclear what Martin stated before and after the limited deposition excerpt provided by plaintiffs. As to what is provided, Martin said the greatest need for more African-Americans was among detectives, but he did not state that this was the only part of the force that needed more African-Americans, just the division where they were "mainly" needed. There is no deposition testimony that there was no need for additional African-American or Hispanic sergeants.

Plaintiffs dispute Petit XI's factual statements as to the existence of an internal operational need. Plaintiffs' present argument mainly goes to the internal operational need for more African-American sergeants. That, however, was not a basis for granting summary judgment in favor of defendant. In any event, Petit XI adequately states the basis for holding there were no genuine factual disputes as to the existence of internal operational needs, including narrow tailoring in regard to the promotion of Hispanics.

As to external operational needs, plaintiffs contend that Martin's testimony is misdescribed. See Petit XI, 2002 WL 31409589 at *65 (citing Martin testimony, Tr. Vol. VIII 37) Contrary to plaintiffs' contention, Martin's testimony refers to all the sergeant promotions that occurred in 1988, 1990, and 1991, see Def. Exh. 40, not just the Out-of-Rank Promotions of Hispanics and women.

Q [defendant's counsel]. Now, at the time you, as Superintendent of Police, made the promotions listed there (Def. Exh. 40] to sergeant in 1988, 1990 and 1991, did you believe it was necessary to make those promotions off the 1988 list and make the out-of-rank promotions to Hispanics and women to remedy past discrimination and to meet the department's operatignal needs?

A. [Martin] Yes.

Tr. Vol. VIII 37 (emphasis added).

Also, plaintiffs now question the reliability of Cooper's testimony as to the need for African-American sergeants because, during the promotion period at issue, he was in a legal department desk job, not assigned to the field. At most, that contention goes to the weight of Cooper's testimony. As stated in Petit XI, 2002 WL 31409589 at *26, though, there was no evidence counter to Cooper's and Martin's testimony and plaintiffs still do not point to any such evidence. A reasonable finder of fact could only find that there existed a need for additional African-American sergeants to serve the CPD's external operational needs.

Plaintiffs' present contentions as to the Walker survey do not matter since the survey was not essential evidence and the discussion of contact with commanders concerns a point raised by plaintiffs, not evidence relied upon to support the operational need holding. See id. at *25.

Plaintiffs contend that testimony about a need for minorities to serve in community policing programs is misplaced because the CPD did not begin community policing until 1995, after the promotions at issue in this case. In response to defendant's summary judgment motion, though, plaintiffs did not contend that expert witness Tom Potter's testimony regarding the use of community policing in Chicago was inaccurate. See id. at *24. The actual testimony is that the CAPS community policing program started in late 1994 and went citywide in 1995. Tr. Vol. VII 161. Even if reliance on this testimony is not waived for having failed to cite it in response to defendant's motion for summary judgment, it would not affect the outcome. Although community policing was cited as involving greater community interaction on the part of sergeants, testimony also supported that there is substantial interaction between sergeants and the community even without community policing and that having minority sergeants improves a police department's performance in such interaction. See Petit XI, 2002 WL 31409589 at *22-24. Again, there is no contrary evidence supporting a genuine factual dispute on this issue. The date community policing began would not change the holding regarding external operational need.

Plaintiffs' contentions that certain evidence should not be considered because not adequately disclosed in discovery have all either already been rejected or have been waived for failure to adequately object at the time the evidence was presented at trial.

Plaintiffs dispute the statement in Petit XI, 2002 WL 31409589 at *27, that "[t]he effect on many or most of the remaining plaintiffs was a delayed promotion to sergeant, not a complete exclusion from that position, nor any loss of an existing position." They point to the fact that 31 of 81 plaintiffs were promoted to sergeant based on the Examination. Thirty-one is still "many" of the plaintiffs being promoted. Moreover, 31 represents those who were promoted by 1991. An additional number of plaintiffs were promoted to sergeant based on subsequent promotional examinations.

Plaintiffs point to one of defendant's proposed facts contained in the final pretrial order. That proposed fact states in part: "The remaining 33 out-of-rank promotions to females are not at issue in this litigation." Final Pretrial Order,

Sched. B, Def. Stmt. ¶ 167. That statement is correct, the gender discrimination claims contained in the complaint were no longer pending at the time the case went to trial; § 1983 claims of race and national origin discrimination were the only pending claims. See Petit XI, 2002 WL 31409589 at *1 n. 6; Petit IX, 2002 WL 10481 at *9. It is unclear how plaintiffs jump from the fact that 33 women received out-of-rank order promotions to the conclusion that 33 additional women are now entitled to be promoted based on operational need. They may be contending that 33 women should have been promoted based on operational need and that the 33 chosen should have been those women who scored the highest on the Examination without adjustments for racial standardization. Whatever the basis is for the logical leap, the contention is without merit for a number of reasons. One, gender claims were no longer pending by the time this case went to trial. Two, the City would not have been obliged to use the unstandardized Examination in determining which women to promote to satisfy its operational needs. Three, plaintiffs did not previously contend that operational need justified the promotion of women and continue to contend that race and gender cannot be an operational need. Four, even., if operational need would justify more promotions of women, a municipality is not obliged to engage in affirmative action just because a compelling interest exists; narrowly tailored affirmative action is permitted by a compelling interest, not required by it. Coalition for Economic Equity v. Wilson, 122 F.3d 692, 709 (9th Cir.), cert. denied, 522 U.S. 963 (1997); L. Tarango Trucking v. County of Contra Costa, 181 F. Supp.2d 1017, 1037 (N.D. Cal. 2001).

As to issues raised in the Rule 59(e) motion that are not expressly addressed herein, the analysis contained in Petit XI speaks for itself.

IT IS THEREFORE ORDERED that plaintiffs' motion to alter and set aside judgment [368] is denied without prejudice. Plaintiffs' amended motion to alter and set aside judgment [366] is denied.