MESKILL, Circuit Judge:

This is an appeal from a judgment entered in the United States District Court for the District of Connecticut, Daly, J., dismissing appellants' action instituted under § 703(a) of Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e-2 (1976), for failure to make out a prima facie case of race discrimination. This appeal raises the question whether a plaintiff in a Title VII action may make a prima facie showing of discriminatory impact based upon the disparate results produced by one portion of an employee selection process, where the overall results of the selection process concededly reveal no such imbalance. For the reasons stated below, we hold that where a plaintiff establishes that a component of a selection process produced disparate results and constituted a pass-fail barrier beyond which the complaining candidates were not permitted to proceed, a prima facie case of disparate impact is established, notwithstanding that the entire selection procedure did not yield disparate results.

BACKGROUND

The plaintiffs, Winnie Teal, Rose Walker, Edith Latney, and Gracie Clack, are black American citizens employed by the State of Connecticut, Department of Income Maintenance. All four plaintiffs were promoted provisionally to the position of Welfare Eligibility Supervisor and served in that capacity for periods in excess of two years. Several of the plaintiffs' superiors testified at trial that the plaintiffs possessed the qualifications for permanent positions as supervisors and praised their performance on the job.

To attain permanent status as supervisors, plaintiffs had to participate in a selection process which requires, as the first step, a passing score on a written examination. Those candidates who pass the written examination form an eligibility pool from which the appointing authority selects persons to fill the permanent positions. In making the final determinations the appointing authority considers the past work performance of the candidates, recommendations of the candidates' supervisors, and to a lesser degree, the candidates' seniority. Additionally, in this final step of the process, the defendants employ an affirmative action program to insure a large representation of minority candidates on the supervisory level. However, only if a candidate passes the written examination and enters the eligibility pool will he be exposed to the rest of the selection process and thus benefit from the consideration of these other factors.

The written test was administered on December 2, 1978 to 329 candidates. The mean score on the examination was 70.4 percent. However, because the black candidates had a mean score 6.7 percentage points lower than the white candidates, a disparity that the defendants admitted was statistically significant, and because a cutoff score of 70 would have resulted in a disproportionately large number of black candidates failing the exam, the defendants set the passing score at 65. With the passing score set at 65, the following results were obtained: Group Number Passing Score (%)

More than a year after this action was instituted, and approximately one month prior to trial, the defendants made their first promotions from the eligibility list generated by the written exam. Forty-six persons were promoted to permanent supervisory positions, 11 of whom were black and 35 of whom were white. Of the 48 identified black candidates who entered the selection process, 11, or approximately 23 percent, were promoted; of the 259 identified white candidates who applied, 35, or about 13.5 percent, were promoted. Thus, the actual promotion rate of blacks (23 percent) was approximately 170 percent that of the actual promotion rate (13.5 percent) of identified whites.

The district court initially determined that the plaintiffs' discrimination claims should be evaluated under the disparate impact analysis formulated by the Supreme Court in Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971). Judge Daly evaluated the results of the selection process under the four-fifths rule of the Uniform Guidelines of Employee Selection Procedures prepared by the Equal Employment Opportunity Commission, 29 C.F.R. § 1607.4(D) (1979). The four-fifths rule provides that "[a] selection rate for any race, sex, or ethnic group which is less than four-fifths (4/5) (or eighty percent) of the rate for the group with the highest rate will generally be regarded . . . as evidence of adverse impact." While noting that the results of the written test did not satisfy the four-fifths rule, Judge Daly concluded that the results of the entire selection process should be used to determine whether the plaintiffs had made out a prima facie case of race discrimination under the disparate impact formula. Since the results of the entire selection procedure actually were more favorable to the black candidates than to the white applicants, Judge Daly dismissed the plaintiffs' action for failure to prove a prima facie case. Thus, the district court found it unnecessary to reach the issue whether the written examination was, indeed, job-related.

DISCUSSION

The plaintiffs contend on this appeal that the district court erred in using the results of the overall selection process in determining that the plaintiffs failed to make a prima facie showing of disparate impact. Alternatively, the plaintiffs argue that the proof adduced at trial was sufficient to establish a prima facie case under the McDonnell Douglas disparate treatment formula. The defendants, on the other hand, maintain that the district court was correct in its decision to dismiss the plaintiffs' complaint and, in any event, contend that the written examination given to the plaintiffs was job-related.

Although the disparate impact and treatment analyses are not necessarily mutually exclusive, see, e. g., Grant v. Bethlehem Steel Corp., 635 F.2d 1007 (2d Cir. 1980) (prima facie case established under both analyses), we believe that the district court was correct in evaluating this case under the disparate impact formula. Contrary to the assertion of the plaintiffs, their qualifications for the permanent supervisory positions were in dispute. Indeed, the resolution of that issue turned directly upon the validity of the written examination. But employers certainly cannot be compelled to come forward and prove the job-relatedness of an eligibility examination upon a plaintiff-employee's bald assertion that he or she is qualified for the position in question. Indeed, since the burden of proof is upon the plaintiff to prove his or her qualifications for the job to establish a prima facie case of disparate treatment, see McDonnell Douglas Corp. v. Green, supra, evaluating a case such as the one at bar under that analysis might require placing the burden upon the plaintiffs to prove that the written examination was not job-related. We believe that the resolution of the latter issue is best undertaken under the disparate impact analysis that was formulated by the Supreme Court precisely to accommodate cases such as this.

Turning to the disparate impact claim, we conclude that the district court erred in ruling that the results of the written examination alone were insufficient to support a prima facie case of disparate impact in violation of Title VII. While we noted in Kirkland v. New York State Department of Correctional Services, 520 F.2d 420, 425 (2d Cir. 1975), cert. denied, 429 U.S. 823, 97 S.Ct. 73, 50 L.Ed.2d 84 (1976), that proof concerning the disparate impact of certain sub-tests was of "little relevance . . . on the issue of whether or not the examination as a whole had an unconstitutional discriminatory impact," there, the passing grade was dependent upon the cumulative results of the sub-tests. Thus, all of the candidates in Kirkland were exposed to the challenged selection device in its entirety, and, therefore, were able to benefit from the components of the exam that apparently offset the allegedly discriminatory sub-test. In other words, all of the candidates in Kirkland were subjected to the complete selection process, which, when viewed as a whole, did not produce unlawful disparate results. Likewise, in Smith v. Troyan, 520 F.2d 492, 497-98 (6th Cir. 1975), cert. denied, 426 U.S. 934, 96 S.Ct. 2646, 49 L.Ed.2d 385 (1976), a disparate impact case in which the Sixth Circuit rejected a plaintiff's attempt to fragment a selection process in an effort to challenge a written "General Classification Test," the passing score was cumulative.

Viewing the overall results of a selection process ordinarily is a prudent course to pursue, since it places the burden upon the defendant to demonstrate job-relatedness only in situations in which courts can determine with some degree of certainty that a selection device has discriminatorily denied an employment opportunity to members of a protected class. Where all of the candidates participate in the entire selection process, and the overall results reveal no significant disparity of impact, scrutinizing individual questions or individual sub-tests would, indeed, "conflict[ ] with the dictates of common sense." Kirkland v. New York State Department of Correctional Services, 374 F.Supp. 1361, 1370 (S.D.N.Y. 1974), aff'd, 520 F.2d 420 (2d Cir. 1975). Where, however, an identifiable pass-fail barrier denies an employment opportunity to a disproportionately large number of minorities and prevents them from proceeding to the next step in the selection process, a different result must obtain. Otherwise we would be adopting the position that, regardless of the language of the statute, Congress intended Title VII to protect faceless groups rather than individuals. Title VII cannot be interpreted to permit an obvious denial of employment opportunities to candidates that is solely attributable to non-job-related characteristics they possess that are incident to their race.

In Brown v. New Haven Civil Service Board, 474 F.Supp. 1256 (D.Conn. 1979), the case upon which the defendants principally rely, Judge Newman, then sitting as a district court judge, was confronted by the same issue presented in this case in a similar factual setting. In Brown, the plaintiffs alleged that of the 99 blacks who had taken a written employment examination, only 30 passed, whereas 120 of the 194 whites passed. Thus, as to that part of the selection process, the passing rate for blacks was less than half the passing rate of whites. The results of the entire selection process, however, revealed no significant disparity between the actual hiring rates of blacks and whites. Judge Newman held that the plaintiffs could not dissect the overall selection process and challenge the disparate results of one of its components where the final results of the selection procedure did not produce disparate results in actual hiring. Judge Newman advanced two principal bases for this holding. First, he suggested that to

examine each component of an entire application process [would] launch[ ] a court on a course that has no boundaries and no clear end. . . .

[E]ven if only one question has a disproportionate impact on a minority group, an individual who answers that question incorrectly and then fails by one point to accumulate the minimum required score for hiring can argue that he has been the victim of discrimination in the hiring process. The prospect of a court's examining subtests, sub-subtests, and even individual questions within test segments argues in favor of examining only the end result of an entire hiring process.

Id. at 1262. Second, Judge Newman contended that Title VII should not be interpreted to prohibit employers from adopting voluntary affirmative action programs:

Specifically, Title VII should not be construed to prohibit a municipality's using a hiring process that results in a percentage of minority policemen approximating their percentage of the local population, instead of relying on the expectation that a validated job-related testing procedure will produce an equivalent result, yet with the risk that it might lead to substantially less minority hiring.

Id. at 1263.

With respect to the first contention that courts should not be burdened with having to evaluate the allegedly disparate impact that an isolated sub-test or question might have had, we think that Judge Newman's concern in that regard is justifiable only in connection with cases that might involve a challenge to a disparate "subscore" that was subsumed in a non-disparate, cumulative, overall score. While it is difficult to conceive of how the dry scores of a multi-component selection process could discriminate in part, but not in the aggregate, without the influence of some affirmative action effort designed to achieve the nondiscriminatory overall result, in such a case the multi-component selection process would by necessity have to include some internal mechanism that offsets the disparate impact caused by the discriminatory component. Since all of the candidates in such a process would have been afforded the opportunity to receive the benefit of the offsetting mechanism, however, the overall results of the process should be deemed a fair barometer of the fairness of the process.

Where a selection process contains a discriminatory pass-fail barrier, however, we see little justification for judicial restraint. In the case of the pass-fail barrier, both the discriminatory component and the affected individuals are readily identifiable. Courts are under an obligation to entertain the claims of individuals who have been victimized by a discriminatory pass-fail barrier, notwithstanding any belated "corrective" action taken by the employer. In such cases only those candidates who do not bear the incidences of their racial group would surmount the discriminatory barrier and benefit from the corrective, affirmative action measure.

Nor are we persuaded that construing Title VII to invalidate a selection process that includes an affirmative action component to insure non-disparate overall results will frustrate the general objective of achieving racial balance in the workforce. While Title VII may permit voluntary affirmative action in some instances, see United Steelworkers of America v. Weber, 443 U.S. 193, 200, 99 S.Ct. 2721, 2726, 61 L.Ed.2d 480 (1979), the statute was never intended to be interpreted as an affirmative action device. Indeed, § 703(j) of Title VII expressly provides that Title VII shall not be "interpreted to require any employer . . . to grant preferential treatment to any individual or to any group because of the race, color, religion, sex, or national origin of such individual or group on account of an imbalance which may exist" in that respect in the employer's workforce. 42 U.S.C. § 2000e-2(j) (1976) (emphasis added). But more specifically, an employer's adoption of affirmative action policies has never been held by the courts to relieve the employer of liability under Title VII to persons who were injured by the employer's previous discriminatory conduct. In International Brotherhood of Teamsters v. United States, 431 U.S. 324, 341-42, 97 S.Ct. 1843, 1857-1858, 52 L.Ed.2d 396 (1977), the Supreme Court stated:

The company's later changes in its hiring and promotion policies could be of little comfort to the victims of the earlier post-Act discrimination, and could not erase its previous illegal conduct or its obligation to afford relief to those who suffered because of it.

Id. (emphasis added). See also Furnco Construction Corp. v. Waters, 438 U.S. 567, 584, 98 S.Ct. 2943, 2953, 57 L.Ed.2d 957 (1978) (opinion of Marshall, J.). Indeed, "[a] racially balanced work force cannot immunize an employer from liability for specific acts of discrimination." Id. at 579, 98 S.Ct. at 2950. In the case at bar, an employee selection device produced a readily discernible disparate impact upon the black candidates. The affirmative action effort taken by the defendants at the end of the process was "of little comfort" to the candidates who were not permitted to proceed beyond the allegedly discriminatory pass-fail barrier.

Title VII was designed to protect the rights of individuals. "It is clear beyond cavil that the obligation imposed by Title VII is to provide an equal opportunity for each applicant regardless of race . . . ." Furnco Construction Corp. v. Waters, supra, 438 U.S. at 579, 98 S.Ct. at 2951. "[A] primary objective of Title VII is prophylactic: to achieve equal employment opportunity and to remove the barriers that have operated to favor white male employees over other employees. . . . An equally important purpose of the Act is `to make persons whole for injuries suffered on account of unlawful employment discrimination.'" International Brotherhood of Teamsters v. United States, supra, 431 U.S. at 364, 97 S.Ct. at 1869, 52 L.Ed.2d 396; (quoting Albermarle Paper Co. v. Moody, 422 U.S. 405, 418, 95 S.Ct. 2362, 2372, 45 L.Ed. 280 (1975)). Permitting the overall selection process result in this case to validate what may be the denial of an employment opportunity to a disproportionate number of black candidates is an expedient that transgresses the purpose of Title VII. Where an employment practice forecloses a disproportionate number of persons in a protected group from a job opportunity, it matters very little to the victimized individuals that their group as a whole is well represented in the group of hirees. "[G]ood intent or absence of discriminatory intent does not redeem employment procedures or testing mechanisms that operate as `built-in headwinds' for minority groups and are unrelated to measuring job capability." Griggs v. Duke Power Co., supra, 401 U.S. at 432, 91 S.Ct. at 854.

Since Judge Daly determined that the plaintiffs failed to establish a prima facie case of disparate impact under Title VII, he evidently thought it unnecessary to reach the question of the job-relatedness of the written examination, even though this issue had been fully tried. Of course, we make no comment on the resolution of that issue. On remand, the district court should evaluate the job-relatedness of the written examination in accordance with the EEOC Guidelines deferred to by the Supreme Court in Albermarle Paper Co. v. Moody, supra, 422 U.S. at 430-31, 95 S.Ct. at 2377-2378. See 29 C.F.R. pt. 1607 (1979). This panel will retain jurisdiction of any appeal from the district court's decision on that issue.

The judgment of the district court is reversed and this action is remanded for further proceedings consistent with this opinion.

* * * * * *

(h) Notwithstanding any other provision of this subchapter, it shall not be an unlawful employment practice . . . for an employer to give and to act upon the results of any professionally developed ability test provided that such test, its administration or action upon the results is not designed, intended or used to discriminate because of race, color, religion, sex or national origin.