PIERCE, Circuit Judge:
Defendants-Appellants (“Defendants” or “the State”) and Intervenors-Appellants (“Intervenors”) appeal from an order and judgment of the United States District Court for the Northern District of New York, Roger J. Miner, Judge, filed October 4, 1983, 571 F.Supp. 1562, denying Defendants’ and Intervenors’ motions for summary judgment and granting Plaintiffs-Appellees’ (“Plaintiffs”) motion for summary judgment. The district judge also enjoined Defendants from making appointments to the position of Correction Captain (“Captain”) of the New York State Department of Correctional Services (“Correctional Services”) from an eligibility list that was based on certain examination scores adjusted to eliminate what Defendants perceived to be the adverse racial impact against minority candidates of a written examination administered by the New York State Civil Service Commission (“Civil Service”). The district judge agreed with Plaintiffs, who are nonminority candidates for the position of Captain, that the State’s adjustment of the minority candidates’ raw test scores discriminated against nonminority candidates in violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e-2 to 2000e-17 (1976 & Supp. V 1981) (“Title VII”).
For the reasons set forth below, we reverse the order and judgment of the district court and we remand for further proceedings consistent with this opinion.
I. BACKGROUND
This action represents the most recent chapter in the controversial history of promotional examinations administered by the Civil Service for supervisory titles in the State’s Correctional Services. The instant dispute began on January 30, 1982, when the Civil Service and the Correctional Services conducted Promotional Examination No. 37-526 for the position of Correction Captain. At the time the examination was given, no minority officers held permanent appointments as Captains in the State’s prisons. After administering the test, the Civil Service tabulated each candidate’s right and wrong answers to arrive at the candidates’ raw scores. The tabulation results indicated that nonminority candidates had passed the test at about twice the rate as minority candidates,1 as follows:
Passing Passing Test Takers Candidates Rate
Nonminority........ 243 119 49%
Minority......... 32 8 25%
In view of the “four-fifths” rule of the Equal Employment Opportunity Commission (“EEOC”) Uniform Guidelines on Employee Selection, 29 C.F.R. § 1607.4(D) (“Guidelines”),2 the State determined that the Captains’ examination had an adverse racial impact on minority candidates because the passing rate of minority candidates was approximately fifty percent lower than the passing rate of nonminority candidates. Taking into account other factors that it felt reinforced its conclusion of adverse impact,3 the State adjusted both [223] the minority and nonminority candidates’ scores by converting them to separate frequency distributions and then equating or normalizing them with the respective means. The effect was to increase to fifty percent the percentage of minorities who passed the test. From the State’s perspective, this adjustment served to correct the adverse racial impact of the test by equalizing the passing rate of minority and nonminority candidates. In practical terms, the adjustment added eight minority candidates to the eligibility list without removing any of the 119 nonminorities from the list.
By acting to eliminate the perceived adverse impact of the examination on minorities, the State sought anticipatorily to avoid litigation it assumed minority candidates would bring challenging reliance on the test to promote candidates to the position of Captain. Such litigation had resulted from the use of past promotional examinations with respect to Correction Sergeants in 1972, Kirkland v. New York State Department of Correctional Services, 374 F.Supp. 1361 (S.D.N.Y.1974), aff'd in part and rev’d in part, 520 F.2d 420 (2d Cir. 1975), cert. denied, 429 U.S. 823, 97 S.Ct. 73, 50 L.Ed.2d 84 (1976), on remand, 482 F.Supp. 1179 (S.D.N.Y.), aff'd, 628 F.2d 796 (2d Cir.1980), cert. denied, 450 U.S. 980, 101 S.Ct. 1515, 67 L.Ed.2d 815 (1981) (“Kirkland Sergeants”), and Correction Lieutenants in 1981, Kirkland v. New York State Department of Correctional Services, 552 F.Supp. 667 (S.D.N.Y.1982), aff'd, 711 F.2d 1117 (2d Cir.1983), cert. denied, — U.S. -, 104 S.Ct. 997, 79 L.Ed.2d 230 (1984) (“Kirkland Lieutenants”).
The State did not succeed in staving off litigation; this time it was initiated by the nonminority candidates, who brought the instant action contending, inter aliaf that the State’s adjustment of candidates’ raw scores involved “reverse discrimination” in violation of Title VII. In its answer to the complaint, the State responded that it acted in good faith and in compliance with applicable law by voluntarily adjusting the raw scores to eliminate the examination’s adverse impact. A group of minority candidates moved to intervene as defendants to assert that the written test had a facially discriminatory impact against them, that the use of separate frequency distributions to eliminate such adverse impact was proper, and that the remedial action taken by the State was the minimum necessary given the past pattern of discrimination that had been the subject of the Kirkland Sergeants and the Kirkland Lieutenants actions. The district judge granted the application to intervene by an order dated February 16, 1983.
In June, 1983, all parties moved for summary judgment. After hearing oral argument and reviewing the papers submitted by the parties in support of their respective positions,5 Judge Miner denied Defendants’ [224] and Intervenors’ motions, granted Plaintiffs’ motion for summary judgment, and enjoined Defendants from making appointments to the position of Captain based on the eligibility list.
The district judge based his conclusion on three principal grounds. He held that the factors considered by the State — the distribution of scores from the written test, the litigation history of prior Correctional Services examinations for supervisory titles, the absence of empirical data indicating that minority and nonminority candidates would not perform equally well as Captains, and the availability of some data regarding the job performance of minority candidates in the two supervisory titles directly below Captain — 6 did not establish a prima facie case of adverse racial impact vis-a-vis minority candidates. Moreover, he held that even assuming, arguendo, that the State had shown in defense of its actions a prima facie case of adverse racial impact, it still had acted in violation of Title VII because it had not proved that the inference of racial discrimination arising from a showing of adverse impact could not be rebutted by proof that the differences in test performance were job-related. In effect, the district judge held that before taking any voluntary action to eliminate the adverse racial impact of its selection process, the State, as an employer, had to meet the following two burdens: (1) make out a prima facie ease of discrimination (in this context, adverse impact), and (2) prove that such prima facie case was not rebut-table. Finally, the district judge concluded that “[e]ven assuming the propriety of the need for defendants’ actions in remedying the alleged discriminatory impact of [the] promotional procedures, the method by which defendants chose to effect their remedy was itself fundamentally flawed.”
Upon review, we hold that the district court erred with respect to each of these grounds in granting summary judgment in favor of Plaintiffs. Consequently, we reverse and remand.
II. DISCUSSION
A. Prima Facie Case of Adverse Impact
Plaintiffs argue that the district court held that the raw scores on the examination did not establish a prima facie case of adverse impact because, in Plaintiffs’ words, the “statistical sample of Black and Hispanic candidates was simply too small to serve as a basis for a finding that the adverse impact was due to race or national origin.” This contention, however, mischaracterizes the holding of the district judge, who expressly stated that he was “not prepared to hold that a pool consisting of thirty-two minorities and 240 or so non-minorities, is, as a matter of law, an inadequate number from which to draw statistical inferences.” Instead, relying on a report submitted by Dr. Kavanagh, Plaintiffs’ statistical expert, the district judge held that the admitted differences in score distributions resulted from differences in the prior experiences of several of the minority candidates. The record does suggest that of the candidates who took the examination, at least eleven of them qualified because of their experience as Correctional Supervising Officers with the Office of Drug Abuse Services (“ODAS”). Of those eleven, six were minority candidates. The district court, relying on the affidavit of Plaintiffs’ expert, reasoned that the difference in test scores resulted not from an adverse racial impact of the test, but from this difference in employment experience between minority and nonminority candidates.
This reasoning does not comport with adverse impact analysis. The case law clearly provides that a prima facie case is established by a showing that an examination has an adverse racial impact on minority candidates. Thereafter, legitimate, job-related explanations for differences in score distributions become relevant to rebut the prima facie showing of adverse [225] racial impact. See, e.g., Albemarle Paper Co. v. Moody, 422 U.S. 405, 425, 95 S.Ct. 2362, 2375, 45 L.Ed.2d 280 (1975); Griggs v. Duke Power Co., 401 U.S. 424, 431-36, 91 S.Ct. 849, 853-56, 28 L.Ed.2d 158 (1971); Kirkland Lieutenants, 711 F.2d at 1132; Guardians Association of New York City Police Department, Inc. v. Civil Service Commission, 630 F.2d 79, 86-87 (2d Cir. 1980), cert. denied, 452 U.S. 940, 101 S.Ct. 3083, 69 L.Ed.2d 954 (1981); Vulcan Society of New York City Fire Department, Inc. v. Civil Service Commission, 490 F.2d 387, 391-92 (2d Cir.1973). Contrary to this approach, Judge Miner held that there had been no prima facie showing of adverse racial impact because the ODAS experience of some of the minority candidates explained the differences in score distributions. In other words, he turned to job-related explanations, which might be appropriate to rebut a prima facie showing of adverse impact, in order to determine that the prima facie case had not been made in the first place. This court previously rejected such an approach, urged by nonminority intervenors in the Kirkland Lieutenants case, by noting that “[although lack of experience may be relevant to the question of a test’s job validity, it does not affect the question whether a prima facie case has been properly established.” Kirkland Lieutenants, 711 F.2d at 1132. Even if prior experience were relevant to the prima facie showing of adverse impact, which it is not, we note, as the court in Kirkland Lieutenants stated, that the difference “in responsibility between Office of Drug Abuse [Services] officers and officers working at minimum and medium security [Correctional Services] facilities has been held to be negligible.” Id. (citing Stokes v. New York State Department of Correctional Services, 569 F.Supp. 918 (S.D.N.Y.1982)).
Having put aside the district court’s erroneous interpretation of the prima facie ease, we turn to Defendants’ assertion that the test scores, and the application of the EEOC’s “four-fifths” rule, established a prima facie showing of adverse impact. It is well-settled in the case law that “[a] prima facie violation of [Title VII] may be established by statistical evidence showing that an employment practice has the effect of denying the members of one race equal access to employment opportunities.” New York Transit Authority v. Beazer, 440 U.S. 568, 584, 99 S.Ct. 1355, 1365, 59 L.Ed.2d 587 (1979); see also Albemarle v. Moody, 422 U.S. at 425, 95 S.Ct. at 2375 (prima facie case of adverse impact established by proof “that the tests in question select applicants for hire or promotion in a racial pattern significantly different from that of the pool of applicants”); Kirkland Lieutenants, 711 F.2d at 1130 (upholding district court’s finding of a “prima facie case of employment discrimination through a statistical demonstration of disproportionate racial impact”); Guardians v. Civil Service, 630 F.2d at 88 (“statistics showing a significantly disparate racial impact have consistently been held to create a presumption of Title VII discrimination”); Vulcan Society v. Civil Service, 490 F.2d at 392-93 (upholding district court’s finding of adverse impact by comparing test passing rates of minority and nonminority candidates). Similarly, “[w]hile courts are not bound by the EEOC Guidelines, the Supreme Court has declared that the guidelines should be shown ‘great deference.’ ” Teal v. Connecticut, 645 F.2d 133, 137 n. 6 (2d Cir.1981) (quoting Griggs, 401 U.S. at 433-34, 91 S.Ct. at 854-55, and citing Albemarle v. Moody, 422 U.S. at 431, 95 S.Ct. at 2378), aff'd, 457 U.S. 440, 102 S.Ct. 2525, 73 L.Ed.2d 130 (1982); see also Guardians, 630 F.2d 88 (“By any reasonable measure, including ... the four-fifths rule of the EEOC Guidelines, Exam No. 8155 had a disparate racial impact.”).
Herein, the differences between the score distributions of minority and nonminority candidates were sufficient to establish a prima facie showing of adverse impact. Twenty-five percent of the minority candidates passed the test whereas nonminority candidates passed at a rate of forty-nine percent. Thus, the passing rate of minority candidates was approximately fifty percent lower than the passing rate of [226] the nonminority candidates. In other words, about five minority candidates passed the Correction Captain examination for every ten nonminority candidates who passed it.
In short, the test data established a prima facie case of adverse impact against minority candidates as a group. We cannot conclude, however, that the test data, by itself, establishes adverse impact as to Hispanic candidates, who were included within the minority grouping, since the record indicates that fifty percent — two of the four — of the Hispanic candidates who took the test passed it. It remains an open question, to be explored fully on remand, whether as to the Hispanic candidates there exists, on the basis of other factors cited by the State, “a sufficiently serious claim of discrimination to serve as a predicate” for the State’s voluntary remedial actions.7 Kirkland Lieutenants, 711 F.2d at 1130.
B. Burden to Rebut
The district court also held, and Plaintiffs argue on appeal, that even if Defendants had made a prima facie showing of adverse racial impact, before adopting remedial measures they still had to demonstrate that such a case was not rebuttable by proof that the score distribution differences were not job-related. In other words, the district judge ruled that before taking steps to rectify a perceived violation of Title VII, the State, as an employer, had to meet two burdens — first establish a prima facie case of discrimination and then prove that such prima facie case was not rebuttable through job-related explanations. We hold that, in the context of this case, the imposition of the latter burden on a party seeking to comply voluntarily with Title VII is contrary to the case law and the statute’s underlying policy.
As this court noted recently, “[i]t is settled that voluntary compliance is a preferred means of achieving Title VII’s goal of eliminating employment discrimination.” Kirkland Lieutenants, 711 F.2d at 1128; see also Carson v. American Brands, Inc., 450 U.S. 79, 88 n, 14, 101 S.Ct. 993, 998-99 n. 14, 67 L.Ed.2d 59 (1981) (“[in] enacting Title VII, Congress expressed a strong preference for encouraging voluntary settlement of employment discrimination claims”); Alexander v. Gardner-Denver Co., 415 U.S. 36, 44, 94 S.Ct. 1011, 1017, 39 L.Ed.2d 147 (1974) (“[cooperation and voluntary compliance were selected as the preferred means for achieving this goal”). By requiring proof of not only a prima facie case of adverse impact, but also of the inability to rebut such a case, the district court in effect posited a rule, contrary to the stated policy of voluntary compliance, permitting an employer, here the State, to take remedial actions only where there could be a judicial determination of racial discrimination.
This court expressly rejected such an approach when it was urged in the context of a voluntary settlement by nonminority employees who intervened to oppose the settlement agreed to by the parties in the Kirkland Lieutenants case. 711 F.2d at 1129-30. Herein, Plaintiffs’ situation is analogous to the position of the intervenors in Kirkland Lieutenants, where minority employees challenged the State’s use of an examination to promote candidates to the position of Correction Lieutenants. Id. at 1121. In Kirkland Lieutenants, the non-minority intervenors argued that “before any race-conscious relief can be granted to plaintiff class, there must be a judicial determination that Exam 36-808 and its resulting eligibility list are not job-related and are therefore racially discriminatory, i.e., a mere statistical showing of disproportionate impact does not amount to a proper basis for settlement.” Id. at 1129. [227] This contention is essentially the same one raised by Plaintiffs herein and adopted by the district court; that is, this contention amounts to a claim that the State must prove the inability to rebut the prima facie case by showing that the differences in score distributions were not job-related. In Kirkland Lieutenants, this court rejected this argument on the ground that it “would turn Title VII law on its head since ... job-relatedness is never presumed ____” Id.; see Guardians v. Civil Service, 630 F.2d at 86-87; Vulcan Society v. Civil Service, 490 F.2d at 391-92. Moreover, “if intervenors’ [in the instant case, Plaintiffs’] position were adopted, no Title VII testing case could be settled by agreement until a judicial determination on the test’s job validity was made. Such a result would seriously undermine Title VII’s preference for voluntary compliance ____” Kirkland Lieutenants, 711 F.2d at 1130 (citations omitted). Consequently, as Judge Lumbard wrote in Kirkland Lieutenants in the context of voluntary settlements, “a prima facie case of employment discrimination through a statistical demonstration of disproportionate racial impact constitutes a sufficiently serious claim of discrimination to serve as a predicate for a voluntary compromise containing race-conscious remedies.” Id.
Plaintiffs attempt to distinguish Kirkland Lieutenants on the ground that the compliance plan therein was adopted on the basis of “arms’ length” settlement negotiations. This distinction, however, is unpersuasive. It implies that the interests of nonminority employees were somehow more protected in Kirkland Lieutenants, where the State adopted the remedial plan under negotiation pressure from minority candidates, than they were in the instant case, where the State adopted the plan without such direct pressure.
Moreover, Plaintiffs’ purported distinction would create an anomalous situation. It would require an employer, in this case the State, to issue a presumptively discriminatory eligibility list, wait to be sued by minority candidates, and only then seek a settlement pursuant to Kirkland Lieutenants. Such an approach would serve no purpose other than to impede the process of voluntary compliance with Title VII and cause the proliferation of litigation in all such cases, thereby generating litigation costs and favoring litigious over nonlitigious employees.
The “rebuttal burden” urged by Plaintiffs and adopted by the district court also is contrary to the Supreme Court’s approach in United Steelworkers of America v. Weber, 443 U.S. 193, 99 S.Ct. 2721, 61 L.Ed.2d 480 (1979).8 Without requiring a judicial determination of racial discrimination (in fact, the defendant therein denied that it had engaged in past discrimination), the majority in Weber upheld, as permissible under Title VII, an affirmative action plan “designed to eliminate conspicuous racial imbalance in traditionally segregated job categories.” Id. at 209, 99 S.Ct. at 2730 (footnote omitted). In doing so, the Court rejected the contention raised by nonminority plaintiffs that a judicial determination of racial discrimination was a necessary predicate to the adoption of race-conscious remedial measures. The broadness of the majority’s view is suggested by Justice Blackmun’s concurrence, which in urging a narrower ground nevertheless noted that “[t]he Court ... declines to consider the narrow ‘arguable violation’ approach and adheres instead to an interpretation of Title VII that permits affirmative action by an employer whenever the job category in [228] question is ‘traditionally segregated.’ ” Id. at 212, 99 S.Ct. at 2731.9
In short, we hold that consistent with Kirkland Lieutenants, 711 F.2d at 1130, and Weber, 443 U.S. at 209, 99 S.Ct. at 2730, a showing of a prima facie case of employment discrimination through a statistical demonstration of disproportionate racial impact constitutes a sufficiently serious claim of discrimination10 to serve as a predicate for employer-initiated, voluntary race-conscious remedies.11
C. Adjustment Methodology
The district judge also held that “[e]ven assuming the propriety of the need for defendants’ actions in remedying the alleged discriminatory impact of [the] promotional procedures, the method by which defendants chose to effect their remedy was itself fundamentally flawed.” Despite the possible statistical weaknesses in the State’s approach,12 the relevant question under Weber is whether in practical terms the plan “unnecessarily trammeled] the interests” of the nonminority employees. 443 U.S. at 208, 99 S.Ct. at 2730. Herein, as in Weber, the adjustment plan did not displace any nonminority candidates from the eligibility list (it merely added eight minority candidates to the list), did not place an absolute bar to the advancement of nonminority candidates, and was only temporary in nature. Moreover, the plan in the instant case was tailored narrowly to eliminate the adverse impact of the test and did not aim at maintaining a racial balance in futuro. Yet, we agree with the Court of Appeals for the District of Columbia that “Weber [does not] support[] the proposition that no purported affirmative action plan is ever unlawful unless it requires discharge, permanently bars advancement, or maintains racial balance ____” Parker v. Baltimore and Ohio Railroad Co., 652 F.2d 1012, 1014 (D.C.Cir. [229]*2291981). We conclude that herein the record is insufficient to determine whether the State’s adjustment plan trammeled the interests of the nonminority candidates. Therefore, we remand to the district court for a full exploration of this disputed issue. See id. (concluding that the record did not contain sufficient information on the effect of affirmative action plan on nonminority employees, and holding that therefore “a crucial fact remained disputed, and ... summary judgment was premature”).
III. CONCLUSION
For the foregoing reasons, we reverse the judgment and order of the district court and we remand for further proceedings consistent herewith.