FAIRCHILD, Chief Judge.

This is an appeal by appellant United States of America from a decision of the district court holding that certain practices of appellees City of Chicago, et al. are not in violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq., as amended by the Equal Employment Opportunity Act of 1972. For the reasons hereinafter stated, we reverse and remand for further proceedings consistent with this opinion.

I. BACKGROUND

This action was brought by the United States to challenge certain promotion and transfer policies of the Chicago Fire Department, which allegedly discriminated against blacks and Hispanics. 1 Promotions in the Fire Department are based upon performance on written promotional examinations administered by the Chicago Civil Service Commission, efficiency ratings by supervisory personnel in the Fire Department, and seniority. After the written examinations are graded by the Civil Service Commission, the Commission publishes an eligibility list which ranks candidates on the basis of written test score (weighted 60%), efficiency rating test score (weighted 30%), and seniority (weighted 10%). Candidates with a composite score of 70 or above are placed on' an eligibility list. It is undisputed that virtually all applicants receive the maximum points awardable for seniority. Variability in composite score is therefore attributable almost entirely to the. written examinations and the efficiency rating tests.

After an eligibility list for a particular position has been published, it is used until there are no more names on the list or until a new examination has been developed. Individuals are promoted as vacancies occur according to their respective composite scores on the eligibility list, i. e., individuals with the highest composite scores are promoted first. The government contends that these promotion procedures discriminate against blacks and Hispanics and therefore violate Title VII.

Seventeen eligibility lists based on seventeen examinations have been prepared since 1960. 2 For the purposes of this appeal, however, the government is challenging eight tests administered during this period for selection of engineers, lieutenants, and captains. Of these eight tests, only three are presently posted; the examination for engineer given in 1969; the examination for lieutenant given in 1970; and the examination given for captain in 1973. Moreover, only one of these tests — the 1973 captain’s exam, was given after Title VII became applicable to the City of Chicago on March 24, 1972. Although the 1969 engineer’s exam and the 1970 lieutenant’s exam were administered before the City became subject to Title VII, current promotions, of course, would occur after the effective date of the Act.

The government has also challenged the assignment and transfer policies of the Fire Department. These policies, the government asserts, operate to confine blacks to unit locations in black neighborhoods and exclude blacks from certain specialized function units of the department.

*420 II. THE DECISION OF THE DISTRICT COURT

The district court found that appellant United States had met its burden of demonstrating that three written promotional examinations which are presently in effect had an adverse impact on blacks. 3 The district court further found, however, that appellees had demonstrated that the tests were job-related and therefore held that the tests did not violate Title VII. The efficiency ratings used by appellees were also found by the district court not to violate Title VII. Finally, the district court held that the transfer and assignment policy of appellees did not constitute a Title VII violation.

Having determined that none of the challenged practices was unlawful, the district court rejected various forms of relief requested by appellants such as a quota system for promotions, a ban on the use of any efficiency ratings until approved by appellants or the court, and involuntary assignments of certain Fire Department employees to redress the allegedly discriminatory assignment and transfer policies of appellees. The district court also refused to issue a permanent injunction prohibiting further promotions based on the results of current eligibility lists.

The district court did, however, order that appellees: (1) apply the E.E.O.C. Guidelines concerning the use of efficiency ratings; 4 (2) furnish to appellant a copy of the validation study of any written examination to be used for making promotions at least 30 days prior to its use and that any such validation study conform to relevant E.E.O.C. Guidelines; and (3) post all vacancies simultaneously in each firehouse for a period of at least 30 days before such vacancies are filled and post a transfer order in each fire station indicating relevant background information of each person transferred.

While this appeal was pending, the appellant filed a motion in this court for an injunction pending appeal enjoining appellees from making any permanent promotions based on the 1969 engineer’s exam and the 1970 lieutenant’s exam. This court concluded that appellant had established a probability of success on the merits and therefore temporarily enjoined permanent promotions based on the two examinations in question pending appeal. Temporary assignments of persons to the rank of engineer and lieutenant were permitted. Appellants did not seek to enjoin permanent promotions to captain pending appeal and therefore promotions to captain were excluded from the injunction.

On this appeal, appellant vigorously contends that the district court erred in holding that the tests were properly validated and that the assignment and transfer policies were nondiscriminatory. Appellant further urges this court to fashion appropriate remedies to correct the effects of the allegedly discriminatory promotion and assignment transfer policies. Appellees, on the other hand, argue primarily that the holding of the district court should be affirmed because there has been no showing of intentional discrimination and therefore no violation of Title VII is possible, an argument not passed upon by the district court. Appellees also argue that even if no showing of intentional discrimination is required, the district court correctly determined that no violations of Title VII had been proven.

III. WHETHER A SHOWING OF INTENTIONAL DISCRIMINATION IS REQUIRED

In Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971), and Albemarle Paper Co. v. Moody, 422 U.S. 405, 95 S.Ct. 2362,45 L.Ed.2d 280 (1972), the Supreme Court articulated the governing principles for establishing a violation under *421 Title VII. These cases make clear that to establish a prima facie case of discrimination, a plaintiff need only show that facially neutral standards have a disproportionate impact on minorities. If the plaintiff can demonstrate that employment standards are discriminatory in effect, the employer must meet “the burden of showing that any given requirement [has] ... a manifest relationship to the employment in question.” Griggs, supra at 432, 91 S.Ct. at 854. If the challenged job requirements are shown to be job related by the employer, the plaintiff may then establish that other selection devices without a similar discriminatory effect would also “serve the employer’s legitimate interest. . .” Albemarle, supra at 425, 95 S.Ct. 2362, quoting McDonnell Douglas Corp. v. Green, 411 U.S. 792, 801, 93 S.Ct. 1817, 36 L.Ed.2d 668. The legal standard of Griggs and Albemarle was followed by this court in Title VII suits in United States v. City of Chicago, 549 F.2d 415 (7th Cir. 1977).

Appellees contend, however, that later decisions of the Supreme Court —General Electric Co. v. Gilbert, 429 U.S. 125, 97 S.Ct. 401, 50 L.Ed.2d 343 (1976) and Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976) — mandate that no violation of Title VII can occur absent a showing of discriminatory purpose. Appellees also assert, relying on National League of Cities v. Usery, 426 U.S. 833, 96 S.Ct. 2465, 49 L.Ed.2d 245 (1976), that the Tenth Amendment requires that Title VII be interpreted only to encompass purposeful discrimination by employers. Furthermore, appellees argue that if the constitutional basis for Title VII is the Fourteenth Amendment rather than the Commerce Clause, a finding of intentional discrimination is required. We find none of these contentions persuasive.

A. The Impact of Washington v. Davis and General Electric Co. v. Gilbert on Griggs and Albemarle

In Washington v. Davis, supra, the Supreme Court held that a showing of discriminatory purpose is necessary to establish a violation of the Fourteenth Amendment. However, the Court made clear that a showing of discriminatory purpose was not required in Title VII suits:

Under Title VII, Congress provided that when hiring and promotion practices disqualifying substantially disproportionate numbers of blacks are challenged, discriminatory purpose need not be proved ... 426 U.S. at 246-47, 96 S.Ct. at 2051.

Thus Washington v. Davis expressly refutes the contention of appellees that discriminatory purpose or intent must be demonstrated in a Title VII case.

General Electric Co. v. Gilbert is no more helpful to appellees’ argument. In Gilbert, the Court held that a plan which provided nonoccupational sickness and accident benefits to all employees but excluded disabilities arising from pregnancy did not violate Title VII. The touchstone of Gilbert was that there was no showing of discriminatory effect — no evidence was introduced to suggest that men received more benefits from the plan than did women:

As there is no proof that the package is in fact worth more to men than to women, it is impossible to find any gender-based discriminatory effect in this scheme simply because women disabled as a result of pregnancy do not receive benefits: that is to say, gender-based discrimination does not result simply because an employer’s disability benefits plan is less than all-inclusive. For all that appears, pregnancy-related disabilities constitute an additional risk, unique to women, and the failure to compensate them for this risk does not destroy the presumed parity of the benefits, accruing to men and women alike, which results from the facially evenhanded inclusion of risks. 429 U.S. at 138-140, 97 S.Ct. at 409-10.

Moreover, the Court expressly reaffirmed the holding of Griggs:

our cases recognize that a prima facie violation of Title VII can be established in some circumstances upon proof that the effect of an otherwise facially neutral plan or classification is to discriminate against members of one *422 class or another. See Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 2051, 48 L.Ed.2d 597 (1976). For example in the context of a challenge, under the provisions of § 703(a)(2), to a facially neutral employment test, this Court held that a prima facie case of discrimination would be established if, even absent proof of intent, the consequences of the test were ‘invidiously to discriminate on the basis of racial or other impermissible classification.’ Griggs v. Duke Power Co., 401 U.S. 424, 431, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971).

See also the separate concurring opinions of Justice Stewart and Justice Blackmun both adhering to Griggs and stating that plaintiffs in Gilbert had failed to prove a discriminatory effect. 429 U.S. at 146, 97 S.Ct. 401, 408.

Cases subsequent to Washington and Gilbert have continued to adhere to Griggs. See, e. g., Nashville Gas Co. v. Satty, 434 U.S. 136, 138, 98 S.Ct. 347, 54 L.Ed.2d 356 (Dec. 6, 1977); Dothard v. Rawlinson, 433 U.S. 321, 97 S.Ct. 2720, 2726, 53 L.Ed.2d 786 (1977). We therefore find no merit to appellees’ contention that subsequent cases have eroded the vitality of Griggs. 5

B. The Impact of the Tenth and Fourteenth Amendments on Title VII

[2] In the 1972 Amendments to Title VII, Congress, acting under § 5 of the Fourteenth Amendment, expanded the coverage of Title VII to include state and local governments. Fitzpatrick v. Bitzer, 427 U.S. 445, 447, 96 S.Ct. 2666, 49 L.Ed.2d 614 (1975). “There is no dispute that in enacting the 1972 Amendments to Title VII to extend coverage to the states as employers, Congress exercised its power under § 5 of the Fourteenth Amendment.” Id. at 453 n. 9, 96 S.Ct. at 2670.

Section 5 of the Fourteenth Amendment provides:

The Congress shall have power to enforce, by appropriate legislation, the provisions of this article.

In enacting the 1972 Amendments, Congress intended to extend to government employees “the same benefits and protections in equal employment as the employees in the private sector of the economy.” S.Rep.No. 92-415, 92nd Cong., 2nd Sess., p. 9. Cf. Chandler v. Roudebush, 425 U.S. 840, 841, 96 S.Ct. 1949, 48 L.Ed.2d 416 (1976). Moreover, Congress was fully aware of the legal standards articulated in Griggs and consciously desired to apply these standards to government employees. S.Rep.No. 92-415 at 5, 10, 14. Thus the question before us is whether the 1972 Amendments with their authorization of the Griggs rationale is “appropriate legislation” within the meaning of § 5 of the Fourteenth Amendment to enforce that Amendment’s prohibition against discrimination. Appellees contend that since the 1972 Amendments of Title VII are based on the Fourteenth Amendment, the legal standard in proving a claim of discrimination under the Fourteenth Amendment — a showing of discriminatory purpose — must be incorporated into Title VII. 6 We disagree.

The Supreme Court has considered an analogous claim to that of appellees in Kat *423 zenbach v. Morgan, 384 U.S. 641, 86 S.Ct. 1717, 16 L.Ed.2d ,828 (1966). In Lassiter v. Northampton Election Bd., 360 U.S. 45, 79 S.Ct. 985, 3 L.Ed.2d 1072 (1959), the Court had sustained a North Carolina English literacy test for voting as not necessarily violative of the Fourteenth Amendment. Subsequent to Lassiter, Congress enacted the Voting Rights Act of 1965, § 4(e) of which provides that persons who have satisfied certain requirements not here pertinent cannot be denied the right to vote. The issue in Katzenbach v. Morgan was whether § 4(e) was constitutional.

Under appellees’ argument in this case, § 4(e) would have been unconstitutional since it prohibited certain state regulations of voting which had previously not been found unconstitutional. However, the Court specifically rejected this approach, focusing instead on the power given to Congress under § 5 of the Fourteenth Amendment:

Thus our task in this case is not to determine whether the New York English literacy requirement as applied to deny the right to Vote to a person who successfully completed the sixth grade in a Puerto Rican school violates the Equal Protection Clause. Accordingly, our decision in Lassiter v. Northampton Election Bd., 360 U.S. 45, [79 S.Ct. 985, 3 L.Ed.2d 1072] sustaining the North Carolina English literacy requirement as not in all circumstances prohibited by the first sections of the Fourteenth and Fifteenth Amendments, is inapposite.

Lassiter did not present the question before us here: Without regard to whether the judiciary would find that the Equal Protection Clause itself nullifies New York’s English literacy requirement as so applied, could Congress prohibit the enforcement of the state law by legislating under § 5 of the Fourteenth Amendment? In answering this question, our task is limited to determining whether such legislation is, as required by § 5, appropriate legislation to enforce the Equal Protection Clause. 384 U.S. at 649-50, 86 S.Ct. at 1722.

In determining whether § 4(e) is “appropriate legislation” as required by § 5, the Court stated that § 5 granted to Congress “the same broad powers expressed in the Necessary and Proper Clause,” Id. at 650, 86 S.Ct. at 1723, and that therefore:

Correctly viewed, § 5 is a positive grant of legislative power authorizing Congress to exercise its discretion in determining whether and what legislation is needed to secure the guarantees of the Fourteenth Amendment. Id. at 651, 86 S.Ct. at 1723.

Under this discretionary power given to Congress, the test of whether a statute to enforce the Equal Protection Clause is “appropriate legislation” is whether the statute is “ ‘plainly adapted to that end’ and whether it is not prohibited by but is consistent with ‘the letter and spirit of the constitution.’ ” Id. at 651, 86 S.Ct. at 1724 (quoting from Chief Justice Marshall’s opinion in McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 421, 4 L.Ed. 579). After articulating these guiding principles, the Court upheld the constitutionality of § 4(e). See also, Oregon v. Mitchell, 400 U.S. 112, 91 S.Ct. 260, 27 L.Ed.2d 272 (1970).

The present case is governed by these same principles. It is undisputed that the 1972 Amendments to Title VII are an enactment to enforce the anti-discrimination prohibitions of the Equal Protection Clause and are “plainly adapted to that end.” Indeed, the whole purpose of the 1972 Amendments was to give public employees the same protections against discrimination as those enjoyed by employees in the private sector. Cf. Chandler v. Roudebush, 425 U.S. 840, 96 S.Ct. 1949, 48 L.Ed.2d 416 (1976). It was well within congressional authority to weigh the competing policy considerations and determine that public employees required the safeguards against discrimination given to private employees by the Griggs standard. Cf. Katzenbach v. Morgan, supra at 653, 86 S.Ct. 1717. Thus, since the 1972 Amendments are clearly rationally related to and consistent with “the letter and spirit” of the Fourteenth Amendment, and the means chosen were not in themselves unconstitutional, we *424 conclude that Congress could constitutionally incorporate the Griggs test into the 1972 Amendments.

Appellees also argue that as a matter of federalism under the Tenth Amendment, it is necessary for a plaintiff bringing a Title VII action against a governmental unit to prove intentional discrimination. This argument, however, is inapposite as Congress, as discussed above, relied on § 5 of the Fourteenth Amendment rather than the Commerce Clause as a basis for the 1972 Amendments to Title VII. Moreover, our holding that Congress could constitutionally incorporate the Griggs standard in the 1972 Amendments under § 5 makes it unnecessary to decide whether Congress could achieve the same result under the Commerce Clause. Cf. Fitzpatrick v. Bitzer, supra.

IV. WHETHER THE PROMOTIONAL PROCESS EMPLOYED BY APPELLEES VIOLATES TITLE VII

Having determined that no showing of intentional discrimination is required, it is necessary to consider whether the promotional exams and the efficiency ratings— the two variable components of the promotional process — are violative of Title VII under the discriminatory impact standard of Griggs and Albemarle.

A. The Number of Promotional Exams At Issue

Plaintiff-appellant presented evidence in the district court relating to the pass rates on seventeen promotional exams given from 1960-1973. Of these, appellant contends only the eight tests during this period for engineer, lieutenant, and captain are relevant for this appeal.

Title VII did not become applicable to the City of Chicago until 1972. Of the eight tests challenged on appeal, only the 1973 captain’s examination was given after this date. Of the seven tests given before the effective date of the statute, five are no longer posted. Since these five tests were given before the effective date of the statute and eligibility lists based on these tests are no longer in use, these five tests cannot be subject of a Title VII violation.

There remain for consideration the 1969 engineer’s exam and the 1970 lieutenant’s exam. Both of these tests were given before the effective date of the statute but are still posted. Thus promotions based on these allegedly discriminatory tests will occur after the effective date of the statute.

In International Brotherhood of Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977), the Supreme Court held that a seniority system adopted before the effective date of Title VII was not unlawful, absent an intent to discriminate, even if it locked in the effects of prior discrimination. The Court’s holding was premised on an interpretation of § 703(h) of Title VII which provides in relevant part:

(h) Notwithstanding any other provision of this subchapter, it shall not be an unlawful employment practice for an employer to apply different standards of compensation, or different terms, conditions, or privileges of employment pursuant to a bona fide seniority or merit system . . . provided that such differences are not the result of an intention to discriminate because of race .

Amicus Local No. 2, International Association of Firefighters, AFL/CIO, argues, relying on the language of § 703(h) which exempts “a bona fide seniority or merit system ” (emphasis added), that the two tests given before the effective date of Title VII are “bona fide merit systems” and are therefore exempt from Title VII under the Teamsters rationale. Neither the parties nor the district court addressed this argument. In the typical case, our normal procedure would be to decide this question ourselves, possibly with the assistance of additional briefing by the parties. However, since it will be necessary to remand this case to the district court for other reasons, we leave the question of the applicability of § 703(h) as interpreted by Teamsters to the 1969 engineer’s exam and the 1970 lieutenant’s exam for consideration in the district court. We emphasize that we intimate no views on this question.

*425 B. Further Findings Regarding The Promotional Exams Which Must Be Made On Remand By The District Court

The district court in this case concluded that a prima facie case of discrimination had been shown because the three examinations presently posted had a disproportionate impact on minorities. The district court further concluded, however, that defendants had satisfied their burden of proof by demonstrating that the tests were job related. Defendants-appellees contend, relying on our decision in United States v. City of Chicago, 549 F.2d 415 (7th Cir. 1977), that the findings of the district court should be upheld unless clearly erroneous. While we agree with this statement in principle, we cannot apply the clearly erroneous standard because the district court made inadequate findings of fact on the key issues involved in this appeal.

1. The Prima Facie Case

The district court resolved the question of whether a prima facie case had been established in one sentence:

In the first place, defendants have in effect admitted that plaintiff made out a prima facie case of discrimination on three promotional exams (Engineer, Lieutenant and Captain) by merely showing the disparate marks received by whites and blacks (defendants’ proposed Finding of Fact 37(c)).

We have examined the record and are unable to locate the proposed finding referred to or any other admission by defendants that a prima facie case had been established. On appeal, defendants dispute whether a disproportionate impact had been shown. Because of the complexity of the evidence on this point, findings of fact will be necessary to resolve the issue.

2. The Job-Relatedness Of The 1973 Captain’s Examination

Assuming that a prima facie case has been shown, defendants-appellees attempt to justify the use of the 1973 captain’s exam by arguing that it was “content” valid. For a test to be content valid, its content must closely approximate tasks to be performed on the job by the applicant. Washington v. Davis, supra 426 U.S. at 247, n. 13, 96 S.Ct. 2040. Constructing a content valid examination requires a thorough task analysis of the job to be performed. Once this task analysis is performed, an examination can be devised which is carefully tailored to reflect relevant job functions. The E.E.O.C. Guidelines, 29 C.F.R. § 1607.5(a), state that an examination can be content valid if it consists “of suitable samples of the essential knowledge, skills or behaviors composing the job in question.” Similarly, the Standards for Educational and Psychological Tests of the American Psychological Association, which the reader is referred to in the E.E.O.C. Guidelines state at p. 29:

An employer cannot justify an employment test on grounds of content validity if he cannot demonstrate that the content universe includes all, or nearly all, important parts of the job. (Emphasis added.)

In holding that the 1973 captain’s exam involved in the present case was content valid, the district court stated:

. plaintiff’s expert acknowledged that the description for the Captain’s position was thoroughly done. She did not attempt to rebut the “content” validation for this test, the only one given after defendant became subject to Title VII.

The district court appears to have confused the requirement of a job analysis with the further requirement that an examination, to be content valid, must closely approximate tasks to be performed on the job. Thus the district court made no findings-as to whether the 1973 captain’s exam tested “all or nearly all important parts of the job” of captain as required by the APA Standards. 7 What is required on remand is *426 for the district court to ascertain from the job analysis the various functions of the job of captain and then analyze the captain’s exam to determine whether these various functions are tested in proportion to their importance. It is not enough, therefore, that the various functions of a captain are tested — there must be a correlation between the importance of a job function as determined by the job analysis and the weight given to this function on the examination. Without such findings comparing the job analysis to the test itself, no decisions can be made on whether the captain’s examination is content valid.

3. The Job-Relatedness Of The 1969 Engineer’s Exam And The 1970 Lieutenant’s Exam

No argument has been made by defendants-appellees that the 1969 engineer’s exam and the 1970 lieutenant’s exam are content valid. Instead, appellees have attempted to demonstrate that these examinations are “criterion” valid. A criterion validity study is one in which test scores are shown to be correlated with identifiable criteria which indicate successful job performance. Washington v. Davis, supra at 247, n. 13, 96 S.Ct. 2040. Appellees relied upon two criterion related validity studies. First, appellees correlated the test results with pre-existing efficiency ratings of candidate’s performance in the positions they held before they took the promotional exam. 8 Second, a correlation was made between the candidate’s efficiency rating and ratings given in drill tests.

The district court recognized that the two criterion validity studies are not in accord with the E.E.O.C. Guidelines which require a correlation between test scores and “important elements of work behavior which comprise or are relevant to the job or jobs for which candidates are being evaluated.” 29 C.F.R. § 1607.4(c) (emphasis added). See also, Albemarle, supra 422 U.S. at 431, 95 S.Ct. 2362. In this case, the efficiency ratings measured performance in a lower level job rather than the job for which the test was being administered. For a criterion validity study of this type to be consistent with the E.E.O.C. Guidelines, it would have been necessary to correlate test scores with post-promotion efficiency ratings rather than performance ratings in lower level jobs. As we stated in our order granting in part appellant’s motion for an injunction pending appeal, the only way a correlation between pre-promotion efficiency ratings and test scores could comply with the E.E.O.C. Guidelines would be if the employer could establish that requirements of the lower level job and the job for which the test is being administered are identical. However, no such showing has been made in this case.

Similarly, the correlation between drill test scores and the efficiency ratings as a basis for showing criterion validity do not comply with the E.E.O.C. Guidelines. Since there was no job analysis nor any findings by the district court to establish that drill test scores measure relevant aspects of job performance in the job tested for, 9 data based on drill tests fails to predict performance in promotional ranks. 10

*427 Despite this non-compliance with the E.E. O.C. Guidelines, the district court concluded that the criterion related studies were adequate to satisfy the defendants’ burden apparently because “[t]he tests themselves . appear on their face to be job-related.” The APA Standards, however, specifically reject face validity as a method for validation of tests. APA Standards, p. 29. Moreover, the district court did not give adequate weight to the fact that the criterion validity studies failed to comply with the E.E.O.C. Guidelines. The Supreme Court has stated that the Guidelines are entitled to considerable deference as the administrative interpretation of Title VII by the agency whose task it is to enforce it. Albemarle, supra at 430-31, 95 S.Ct. 2362; Griggs, supra, 401 U.S. at 433-34, 91 S.Ct. 849. This court has stated that “[c]ompliance with these Guidelines is generally required absent some showing that a cogent reason exists for non-compliance.” United States v. City of Chicago, supra at 430. In our order granting in part appellant’s injunction pending appeal, we emphasized that while compliance with the E.E.O.C. Guidelines was not the only possible validation method, “the City’s burden is much heavier if it has not used such a method.”

On remand, therefore, the district court should make a detailed step-by-step analysis to determine whether the appellees have met their heavy burden of establishing job-relatedness in the absence of compliance with the E.E.O.C. Guidelines. Appellees should be required to make a strong showing that the two criterion validity studies are predictive of successful performance in the jobs being tested for.

4. The Efficiency Ratings

Efficiency rating scores are weighted 30% in the determination of a candidate’s composite score. 11 The district court stated that any disparity in the efficiency rating scores and those of minorities does not trigger Griggs analysis and therefore held that no violation of Title VII was shown:

Plaintiff relies solely upon statistical tabulations which show that blacks generally received lower efficiency ratings than whites, but evidence is lacking as to why this occurs. Griggs does not shift the burden to defendants to show that efficiency ratings, as distinguished from written tests which are not job related, must be validated merely because blacks are rated lower than whites. The five efficiency components on their face are job related, and we know of no way to eliminate a subjective element from the ratings. ... In short, we do not believe that Griggs or Albemarle Paper Co. v. Moody, 422 U.S. 405, 95 S.Ct. 2362, 45 L.Ed.2d 280 (1975) require defendants to go forward with proof of nondiscrimination on the issue of allegedly discriminatory efficiency ratings in this case.

We think the district court applied the wrong legal standard in evaluating the efficiency rating scores. Face validity does not justify the use of efficiency ratings any more than it justifies the promotional exams. Moreover, we do not agree with the district court that the Griggs standard is inapplicable to efficiency ratings because they are not written tests. In United States v. City of Chicago, supra at 427, 429-430, for example, we affirmed a finding that a Police Department’s background investigation disqualified a significantly greater percentage of minority applicants than white applicants and was not demonstrably related to job performance. 12

*428 On remand, the district court should follow this approach in analyzing the legality of the efficiency ratings. The district court should first make specific findings on whether whites received higher efficiency rating scores than minorities. 13 If such a disparity exists, the district court should make additional findings on whether defendants have met their burden of establishing job relatedness. This determination should involve more than face validity; an inquiry into whether the efficiency ratings accurately predict performance in the job being tested for will be required. 14

V. WHETHER THE ASSIGNMENT AND TRANSFER POLICIES OF AP-PELLEES VIOLATED TITLE VII

The second major issue in this appeal is the assignment and transfer policies of defendants-appellees. The district court found that: since 1966 assignments and transfers have been based primarily on seniority within each rank; 15 that transfers are granted to blacks when a vacancy is available; that the absence of blacks from certain specialized units was attributable to the failure of qualified blacks to apply 16 and to factors other than race; that blacks had been transferred to other specialized units; and that the fact that predominantly black stations are located in predominantly black areas results partly from discriminatory policies before 1966 (before defendants became subject to Title VII) and partly from personal preferences of blacks. The district court concluded, therefore, that “[t]he weight of the evidence does not show an existing pattern of discriminatory assignments or failure to transfer, either generally or to specialized units.” The court nevertheless ordered defendants to post all vacancies simultaneously in each firehouse for a period of at least 30 days before such vacancies are filled and to post transfer orders indicating background information about each person transferred. The court rejected a request by plaintiff-appellant to order involuntary transfers to remedy the effects of prior discrimination.

We have reviewed the record on this aspect of the case and cannot say the findings of the district court are clearly erroneous. Thus we affirm the district court’s finding that the assignment and transfer policies of defendants did not violate Title VII and denial of relief requested by plaintiffs on this appeal. There has been no appeal from the requirement that vacancies and transfer orders be posted and this aspect of the judgment of the district court will remain in effect.

VI. THE TREATMENT OF HISPANICS

The district court held that plaintiff had failed to establish that Hispanics were victims of discrimination. Indeed, the small number of Hispanic candidates makes any such conclusion based on test results virtually impossible. In United States v. City of Chicago, supra, we held that Hispanics could be included in a remedy designed to correct discrimination in promotional practices even though it could not be shown that Hispanics considered by themselves were victims of discrimination. Id. *429 at 429, 436. Thus, if the district court determines on remand that current use of any of the promotional exams violate Title VII, Hispanics should be included in any remedy.

VII. INTERIM RELIEF

In our injunction pending appeal, we enjoined permanent promotions based on the 1969 engineer’s and the 1970 lieutenant’s examinations but allowed temporary promotions to these positions. Promotions based on the 1973 captain’s examination were not included in the injunction.

Clearly the creation of new eligibility lists based on carefully validated tests will obviate for the future most of the problems presented in this case. We are not aware whether there has been progress towards creating such tests and lists while this appeal has been pending. Our evaluation of the case leads us to the conclusion that unless such lists have been created there should be some form of interim relief.

We now modify our injunction pending appeal. Permanent promotions based on the 1969 engineer’s, 1970 lieutenant’s, and 1973 captain’s examinations are prohibited unless such promotions are first offered to minority candidates on the respective lists in the order of their rank thereon. After the minority candidates on any eligibility list have been promoted, further vacancies may be filled by promoting whites on the list according to their rank.

This injunction as modified shall remain in effect until our mandate is issued, and therefore until judgment on the merits, except that the district court is given leave to modify it (or vacate it) in the light of developing circumstances.

Insofar as the judgment appealed from requires posting of vacancies and transfer orders, and denies any further relief with respect to the assignment and transfer policies of defendants-appellees, it is affirmed. In all other respects, it is reversed and the cause is remanded to the district court for further proceedings consistent with this opinion. Circuit Rule 18 shall apply on remand. The district court may, in its discretion, hear additional evidence.

Each party shall bear its own costs of appeal.

REVERSED AND REMANDED.

Notes

. The complaint as originally filed also challenged recruitment and hiring policies of appellees. This aspect of the case has been settled by consent decree.

. The seventeen eligibility lists which have been prepared are: engineer in 1965 and 1969; lieutenant in 1962, 1966 and 1970; captain in 1964, 1968 and 1973; battalion fire chief in 1960, 1964 and 1969; division fire marshal in 1962 and 1968; second deputy chief fire marshal in 1964; first deputy chief fire marshal in 1973; and chief fire marshal in 1964.

. The district court held that there was insufficient evidence to support an allegation that either the promotional exam or the assignment and transfer policies had an adverse impact on Hispanics.

. The district court also ordered that any efficiency ratings utilized in the future be made in ink.

. In Nashville Gas Co. v. Satty, supra, 434 U.S. 142, 98 S.Ct. 347, the Court while recognizing that Griggs held that a violation of § 703(a)(2) of Title VII can be established by proof of discriminatory effect, reserved the question of whether it is necessary to prove intent to establish a prima facie violation of § 703(a)(1). The Court stated that § 703(a)(1) was the proper section of Title VII under which to analyze questions of sick leave or disability payments. Id. We need not pass on this question for the promotional exams and the assignment and transfer policies are the types of practice analyzed under the discriminatory impact test in Griggs and Albemarle rather than in the nature of sick leave or disability payments.

. Two district courts are split on this question. Compare Scott v. City of Anniston, Alabama, 430 F.Supp. 508 (N.D.Ala.1977) (holding that discriminatory purpose must be shown under Title VII as applied to government employers) with Harrington v. Vandalia-Butler Bd. of Educ., 418 F.Supp. 603 (S.D.Ohio 1976) (discriminatory intent standard of Washington v. Davis inapplicable to suits brought against state and local governments). Scott v. City of Anniston, Alabama, supra, is presently on appeal in the Fifth Circuit.

. Thus in Firefighters Institute For Racial Equality v. United States, 549 F.2d 506 (8th Cir. 1977), the Eighth Circuit held that a promotional examination which did not measure supervisory skills was not content valid even though the test did measure other necessary skills of the promotional job. See also, Kirkland v. New York Dept. of Correctional Services, 374 F.Supp. 1361, 1378 (S.D.N.Y.1974), aff’d in relevant part, 520 F.2d 420 (2d Cir. 1975); Vulcan *426 Society of New York City Fire Department, Inc. v. Civil Service Commission, 360 F.Supp. 1265, 1274 (S.D.N.Y.1973), aff’d 490 F.2d 387 (2d Cir. 1973).

. The efficiency ratings used as a validation device were also used as an independent selection device constituting 30% of a candidate’s composite score.

. Indeed, the district court recognized that drill tests “do not clearly distinguish between the individual’s performance and that of the unit under his direction.” Moreover, even if the performance of the drill test leader could be effectively isolated, drill test scores would still be an inadequate validation device absent a showing that the drill tests measured the major responsibilities of the promotional positions.

. Defendants contend on appeal that the correlation between efficiency ratings and drill tests demonstrates that the promotional exams are construct valid. A construct valid study is one in which examinations are structured to measure the degree to which job applicants have identifiable characteristics that have been determined to be important in job performance. Washington v. Davis, supra, 426 U.S. at 247, n. 13, 96 S.Ct. 2040. This argument does not help defendants, however, because there has been *427 no attempt to determine from a job analysis what tracts are necessary for job performance. E.E.O.C. Guidelines, 29 C.F.R. § 1607.5(a). Thus the drill tests fail to demonstrate either criterion validity or construct validity because they fail to predict performance in promotional ranks.

. Efficiency ratings consist of a supervisory evaluation of five factors: quality and quantity of work, dependability, personal relationships, and attendance.

. In United States v. City of Chicago, 385 F.Supp. 543, 560-61 (N.D.Ill.1974), the court held that efficiency ratings used by the Police Department of defendants which were also used as 30% of a candidate’s composite score and as a validation device for promotional exams were not job related. The court stated *428 that “[n]o evidence was offered of any attempt to validate the efficiency ratings as predictors of job success. . . .” The court later found, however, that the efficiency ratings were not discriminatory because of absence of a clear showing of disparate impact.

. Appellees assert that any disparity between the efficiency rating scores received by whites and those received by blacks was too slight to establish a prima facie case of discrimination. In the absence of specific findings by the district court on whether the efficiency ratings had a disproportionate impact, however, we leave this question to the district court.

. Since supervisory ratings are by their very nature subjective, the district court should make every effort to insure that they are free from bias and otherwise free “from factors which would unfairly depress the scores of minority groups.” E.E.O.C. Guidelines, 27 C.F.R. § 1607.5(6)(4).

. Before 1966, appellees had followed a policy of segregation with respect to assignments and transfers within the Fire Department.

. The district court stated that there was only one example of an alleged failure to transfer a qualified black to a specialized unit and this occurred in 1960.