TERENCE T. EVANS, Circuit Judge.

As anyone who was awake during the fall of 2004 knows, the label of "flip-flopper" got considerable play during the presidential campaign. The term lives again in this appeal, which involves yet another challenge to the promotion procedures of the Chicago Police Department. The plaintiffs are minority police sergeants who claim their failure to be promoted after taking the 1998 lieutenant examination violated Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e to e-17. During the course of litigation, the plaintiffs elected to renounce one theory of recovery in order to pursue another. But later, when this fresh approach appeared doomed, they tried to resuscitate their original claim. A federal district court did not allow it and entered summary judgment in favor of the City of Chicago.

First, some background. The promotional process at issue had three parts: a written qualifying test, an assessment exercise, and a merit selection process. See Allen v. City of Chicago, 351 F.3d 306 (7th Cir. 2003). Applicants were required to pass the written qualifying test to be considered for promotion under the other two factors. The City then promoted 70 percent of applicants from a rank-order list of candidates based on their scores on the assessment exercise, which required written responses to questions relating to officer supervision, subordinate evaluation, review of reports, and situational judgment. See id. at 309. The remaining 30 percent were promoted on merit. The written qualifying test had a disparate impact on African-Americans, while the rank-order assessment exercise had a disparate impact on African-American and Hispanic officers. Merit promotions had no disparate impact.

In their original complaint, filed in November of 1998, the plaintiffs alleged that the City's employment test (the written qualifying test and the rank-order assessment exercise) unlawfully discriminated against them on the basis of race. The district court certified two subclasses of officers adversely affected by the 1998 promotion process. Subclass A plaintiffs are minority officers who failed the written qualifying test and are thus ineligible for promotions based on assessment or merit. Subclass B plaintiffs are those who passed the written qualifying test but did not score high enough on assessment to earn a promotion.

In July of 2000, the plaintiffs requested leave to file a "conditional amended complaint" and a stay of discovery pending resolution of the petition for a writ of certiorari filed in Bryant v. City of Chicago, 200 F.3d 1092 (7th Cir. 2000). In Bryant, minority sergeants challenged the 1994 lieutenant promotion process and its use of test results to make rank-order promotions. We concluded that the use of test scores for rank-order promotions was content-valid and thus not violative of Title VII. 200 F.3d at 1100. In their request for a stay, our plaintiffs conceded that the written qualifying test and assessment exercise of the 1998 exam were valid under Bryant. They also stated that if Bryant survived, they would "limit their case to whether making more than 30% of the promotions through merit selection would be an `equally valid, less discriminatory alternative'" under Title VII. The district court granted their request and stayed the proceedings. The Supreme Court denied the petition for a writ of certiorari in Bryant in October of 2000.

A month later, the plaintiffs filed an amended complaint; this time alleging that merit promotions are an equally valid, less discriminatory alternative to the rank-order promotions and that the City violated Title VII by limiting them to 30 percent of all promotions. When discovery resumed, the plaintiffs, responding to the City's request for admissions under Fed.R.Civ.P. 36(a), conceded the validity of the written qualifying test and the assessment exercise under Title VII. By the time discovery closed in August of 2002, the plaintiffs had failed to designate an expert to substantiate their claims.

The City moved for summary judgment in November of 2002, based on the plaintiffs' failure to secure an expert, as well as the decision of a federal district court in Allen v. City of Chicago, 2002 WL 31176003 (N.D.Ill. Sept.30, 2002). In that case, the court rejected a similar challenge to the 1998 lieutenant examination, concluding that the plaintiffs failed to demonstrate that increasing the percentage of merit-based promotions from 30 percent was an equally valid, less discriminatory alternative. The plaintiffs in our case responded by moving to stay summary judgment proceedings pending disposition of the appeal in Allen. The court granted their request and stayed the proceedings for a second time. In December of 2003, we affirmed in Allen, concluding that the plaintiffs failed to set forth evidence establishing that increased merit-based promotions are an equally valid, less discriminatory alternative. See Allen v. City of Chicago, 351 F.3d 306, 315 (7th Cir. 2003).

The plaintiffs then tried to flip-flop, asking under Fed.R.Civ.P. 36(b) to withdraw their earlier admissions that the written qualifying test and assessment exercise are kosher under Title VII. The plaintiffs acknowledged that " Allen is fatal to any `equally valid, less discriminatory alternative' argument in this case" and argued that they should be allowed to resurrect their claims relating to test validity because, in their view, Allen altered the City's burden of proof for justifying a discriminatory selection method. The district court denied the request and then entered summary judgment in favor of the City.

On appeal, the plaintiffs contend that the district court erred by not allowing them to withdraw their admissions. Admissions, in some ways, are like sworn testimony. Once one is made, there is no need to revisit the point. Under Rule 36(b), matters admitted are "conclusively established." In our case, the plaintiffs admitted, among other things, that the written qualifying test and assessment exercise of the 1998 exam "do not violate or contribute to any violation of Title VII." Now, they wish to take back these admissions. A sort of flip-flop, if you will.

A court, in its discretion, may permit a party to rescind admissions when doing so better serves the presentation of the merits of the case and the party who benefits from the admissions (usually by relying on them) is not prejudiced. Fed.R.Civ.P. 36(b). We review a court's refusal to allow the withdrawal of admissions only for an abuse of discretion. United States v. Kasuboski, 834 F.2d 1345, 1350 n. 7 (7th Cir. 1987); Perez v. Miami-Dade County, 297 F.3d 1255, 1265 (11th Cir. 2002).

Here, there was no abuse of discretion. In the first instance, the plaintiffs cannot, in the words of Rule 36(b), demonstrate that permitting them to withdraw their admissions "subserves the merits." Their original, abandoned claims are likely foreclosed by Bryant, whether they admit it or not. Still, they contend the district court abused its discretion because there was a change in our analysis of disparate impact. To prevail on such a cause of action, the plaintiffs must establish that the City's promotion process caused a disparate impact on minority sergeants. If this is shown, the burden shifts to the City to demonstrate that the promotion process is "job related" and "consistent with business necessity." 42 U.S.C. § 2000e-2(k)(1)(A); Allen, 351 F.3d at 311. The City can show that its process is "job related" by any one of three tests: criterion-related, content validity, or construct validity. See Uniform Guidelines on Employee Selection Procedures, 29 C.F.R. § 1607.5(B); Bryant, 200 F.3d at 1094. If the City makes this showing, the plaintiffs can still prevail by showing that an alternative practice exists and the City refused to adopt it. 42 U.S.C. § 2000e-2(k)(1)(A)(ii) and (C); Allen, 351 F.3d at 311-12.

The plaintiffs claim that our interpretation of this test has changed. Specifically, they contend that before Bryant, an employer could demonstrate the validity of rank-order selection only by using empirical evidence of an association between performance on the exams and on the job. In their view, Bryant abandoned this requirement, leaving them no choice but to admit the validity of the 1998 test. But then later in Allen, so the argument goes, the requirement of statistical evidence demonstrating a correlation between test and job performance was reinstated.

The plaintiffs are misreading these cases. As we stated earlier, employers can establish job-relatedness by one of three methods, including "content validity," which entails showing that "the test measures the job or adequately reflects the skills or knowledge required by the job." Gillespie v. Wisconsin, 771 F.2d 1035, 1040 n. 3 (7th Cir. 1985) (internal quotation omitted). A typing test for secretaries exemplifies this kind of approach. Billish v. City of Chicago, 989 F.2d 890, 895 (7th Cir. 1993) (en banc). This method does not require empirical evidence, but instead "should consist of data showing that the content of the selection procedure is representative of important aspects of performance on the job." 29 C.F.R. § 1607.5(B). In contrast, the "criterion related" approach evaluates whether a test is adequately correlated with future job performance and is constructed to measure traits thought to be relevant to future job performance. An IQ test is a typical criterion-related method. See Billish, 989 F.2d at 896. Unlike content validity, this method requires "empirical data demonstrating that the selection procedure is predictive of or significantly correlated with important elements of job performance." 29 C.F.R. § 1607.5(B).

Contrary to the plaintiffs' assertion, our pre- Bryant cases authorized proceeding under the content-validity approach to establish job-relatedness, which as we have said does not require employers to come forth with statistical evidence. See Billish, 989 F.2d at 896; Gillespie, 771 F.2d at 1044. Bryant was no departure from this. There, we concluded that the City demonstrated that the test scoring system used in the 1994 lieutenant exam was content-valid and that the use of these scores to make promotions in rank order was job-related. 200 F.3d at 1099-1100. Finally, Allen had no bearing on the employer's burden of establishing job-relatedness, as the plaintiffs in that case conceded the validity of the 1998 exam. 351 F.3d at 312. Instead, Allen concerned step three of the burden-shifting analysis, i.e., whether the plaintiffs established an equally valid, less discriminatory alternative to merit promotions. Id. at 315. Thus, our analysis of disparate impact has not shifted, and permitting the plaintiffs to withdraw their admissions would be pointless.

Moreover, merits notwithstanding, the City met its burden of establishing prejudice. This lawsuit has been around since 1998. The plaintiffs forced the City to litigate the issue of test validity, only to abandon the theory with a vow not to litigate it. Discovery was then completed on the plaintiffs' new theory that merit promotion was an equally valid, less discriminatory alternative. Once it became apparent that the plaintiffs could not win under this theory, they tried to go back where they started. Throughout this litigation, the City has been subjected to long delays and voluminous discovery. Enough is enough.

AFFIRMED.