MEMORANDUM OPINION AND ORDER
PARSONS, District Judge.
Historically the management and work of the Fire Department of the City of Chicago has been carried on predominately by young white men, and because of its necessary paramilitary and career service nature it has been slower in changing its ethnic composition than has the city itself. Since World War II the black and Hispanic populations of Chicago have increased dramatically, until by 1980 they answered for 45% of the city’s working populations. Today minorities still constitute fewer than 25% of the city’s fire department. Race, ethnic or age based decisions in hirings, promotions or discharges in the Fire Department have spawned in recent years numerous challenges in federal court to the department’s [922]*922employment policies and practices. In 1980 the City entered into a consent decree with the United States addressing alleged violations of the Title YII of the Civil Rights Act of 1964 with respect to the denial of promotional opportunities to black and Hispanic firefighters. See No. 73 C 661 and 80 C 1590 (N.D.I11.) Due in large part to the efforts resulting from this and similar federal litigation and to the efforts of the City’s present administration, the opportunities for employment by the city for all minorities including women have been greatly enhanced. The City has affirmatively increased the percentage of minority and female firefighters within its ranks. In spite of these commendable efforts, however, there still remains a persistent disparity between the percentage of the minority representation in the ranks of the Chicago Fire Department and the percentage of minorities represented in the available workforce in Chicago. Perhaps most troubling is the continued underrepresentation of minorities in the upper ranks of the department despite the City’s commitment over eight years ago to “increase substantially the minority composition in each of the promotional ranks.” Consent Decree.
It is true that once discriminatory practices or impermissible barriers to hiring and advancement are recognized, remedial measures to ameliorate the effects of past discrimination may take some time to implement and to effect desired changes. This is especially true where the demand for increasing minority representation within the ranks must not be permitted to compromise the standards of safety and performance set by the employer in protecting the community. Yet, how long is the employer to be indulged in purging discriminatory practices and moving minorities up through the ranks so that their representation at all levels more naturally reflect the workforce in general. That is the question presented here.
The plaintiffs are black and Hispanic lieutenants of the Chicago Fire Department who challenge the eligibility rule that only those individuals who have successfully taken and passed the captain’s exam and have been promoted to captain are eligible to take the battalion chief’s exam presently scheduled to be administered on August 13, 1988. The plaintiffs have taken and passed the captain’s exam, but due to the present lack of vacancies at the captain level they have not been promoted to captain. They allege that the City in denying them the opportunity to take the battalion chief examination is foreclosing employment opportunities to them and is perpetuating the underrepresentation of minorities in the upper ranks of the department. Plaintiffs claim that the eligibility rule requiring them to first be promoted to captain adversely impacts on minorities and is without any business necessity. The Firefighters union local is joined as a defendant since it is a party to the collective bargaining agreement with the city which contains the eligibility rule challenged here. Plaintiffs seek a preliminary injunction requiring that the City permit the plaintiffs and similarly situated lieutenants to take the battalion chief examination. Because of the reasonable likelihood of plaintiffs’ success at trial based upon the evidentiary hearing held in this matter and because of the apparent irreparable harm that might occur should the present battalion chief examination be administered without the inclusion of the plaintiffs, I conclude that injunctive relief is proper and temporarily restrain the City from administering the battalion chief examination until it has provided for the inclusion of the plaintiffs and similarly situated lieutenants.
I. FACTS
The Chicago Fire Department like many municipal service organizations has a paramilitary organizational structure. Employees once hired progress up the ranks of command according to vacancies in the next higher rank and according to qualifications of the employee. The entry level position is that of firefighter. In succession, the ranks following firefighter are lieutenant, captain, battalion chief, deputy district chief, chief and commissioner.1 [923]*923The positions of deputy district chief, chief and commissioner are appointed positions. The remaining ranks, however, are filled by employees who study, take and pass a series of examinations preparatory to the next rank. The names of those employees who successfully pass the examinations are placed on an eligibility list from which names are taken to fill vacancies as they occur in that rank. When the list expires, either after a discretionary number of years or after exhaustion of all names on the list, another examination is given and a new list of eligible candidates for promotion is posted. The examinations are administered by the City’s Department of Personnel and are designed to test an employee’s familiarity with the areas of expertise expected of an employee at the next rank. An oral component of the examination is also designed to “test” an applicant’s maturity and ability to command. The administration of these examinations in the past has been fraught with difficulties. Virtually every promotional exam given from the 1960’s to the early 1980’s has been found to have had a discriminatory impact on minorities. See United States v. City of Chicago, 573 F.2d 416 (7th Cir.1978). In addition, the examinations are administered on an irregular and infrequent basis. The last battalion chief examination was given over ten years ago and the city admits it had a discriminatory impact. The result of this sporadic scheduling of examinations is that many employees are never given an opportunity to advance because they find that by the time an examination is finally administered, the list of eligible candidates is posted, and the promotions are made according to existing vacancies, the employees are due to retire because of the department’s mandatory retirement age.
There exists no other requirement for movement to the next rank except for the successful completion of the examination and, of course, an existing vacancy. The department requires no minimum amount of time to be served in each position before an employee can advance to the next level. Although, as a practical matter, on the average, an employee usually serves a number of years as a lieutenant or captain before advancing to the next position.
All positions except firefighter are supervisory positions with concomitant administrative responsibilities. The duties and responsibilities of employees in the various levels above firefighter differ primarily in the degree of supervision expected of the employee. Each level has supervisory responsibility over those employees of the ranks below it and in that respect each rank has somewhat increased supervisory responsibility over the ranks below. But the nature of the supervisory responsibility does not differ especially in calls to alarms. The superior officer first arriving at the incident whether it is a lieutenant or battalion chief has full supervisory responsibility in directing the suppression of the fire until a superior officer to him arrives relieving him of this duty. In other respects, the differences between the ranks is purely administrative. For example, a captain is assigned some housekeeping and inventory responsibilities not expected of a lieutenant.
After years of discriminatory practices resulting in underrepresentation of minorities in the department, in 1980 as part of the Consent Decree with the United States, the City agreed to increase minority representation at all ranks. They agreed as to promotions to lieutenants that for every three white firemen promoted, one black or Hispanic fireman would be promoted. Thus they were ensuring, at least at the lieutenant level, a commitment to an eventual 25% minority representation in that rank. Additionally and at this same time, the City entered into a collective bargaining agreement with the firefighters’ union. This agreement provided that the City would take affirmative action to include black and Hispanic employees in all categories in all ranks in order to “reach as quickly as possible a level as close to 45% as is reasonably achievable.” Collective Bargaining Agreement, Appendix G.
[924]*924This plan floundered, however, until 1983 when under the city’s new administration newly appointed Fire Commissioner Louis Galante committed the department to honoring the requirements of court decrees and of the collective bargaining agreement’s affirmative action provisions. Since 1983 Commissioner Galante and the City have endeavored to meet these affirmative action goals by offering promotional examinations and affirmatively increasing the minority representation in the ranks. At present, although substantial increase of minorities has occurred at the lower ranks of firefighter, there remains substantial underrepresentation at the higher ranks.2 According to Commissioner Galante’s own estimates it will require at least two cycle’s of examinations over the course of at least six to eight years before the percentage of minorities in the upper ranks approach the percentage of minorities available in the workforce. This percentage in Chicago is approximately 45%. But according to the statistics presented by the parties under the current program and considering the history of affirmative action already begun, it still would take until the year 2013 before the 45% figure among battalion chiefs would be reached.
II. DISCUSSION
The plaintiffs allege that the eligibility rule that an employee must take and pass the captain’s examination and be promoted to captain before taking the battalion chief examination discriminatorily impacts upon minority lieutenants and results in the perpetuation of past discrimination. The disparate impact theory prohibits employers from applying policies of job requirements which, although neutral on their face, disproportionately disqualify members of a protected class from employment opportunities. Proof of discriminatory motive is not required. Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158. If a plaintiff can prove that a job requirement or policy impacts disproportionately on a protected group, and the requirement or policy is not related to job performance, it violates Title VII. Connecticut v. Teal, 457 U.S. 440, 102 S.Ct. 2525, 73 L.Ed.2d 130 (1982).
Once the plaintiff proves a disparity in treatment between protected and unprotected groups because of a facially neutral standard, the defendant may rebut it by showing that the plaintiff’s proof is flawed or by showing that the challenged standards causing the discriminatory pattern are job related. Washington v. Davis, 426 U.S. 229, 246-47, 96 S.Ct. 2040, 2050-51, 48 L.Ed.2d 597 (1976); Shidaker v. Carlin, 782 F.2d 746, 750 (7th Cir.) vacated, 480 U.S. 616, 107 S.Ct. 1442, 94 L.Ed.2d 615 (1987) opinion reinstated on remand sub. nom. Shidaker v. Tisch, 833 F.2d 627, 631 (7th Cir.1987). Generally, the plaintiff’s evidence is statistical in nature. Only in very limited instances may a court find discriminatory impact from non-statistical evidence. Dothard v. Rawlinson, 433 U.S. 321, 330, 97 S.Ct. 2720, 2727, 53 L.Ed.2d 786 (1977); Carpenter v. Board of Regents, 728 F.2d 911, 914 (7th Cir.1984). Thus, the defendant’s rebuttal to the plaintiff’s evidence requires a showing of statistical evidence of his own that is “more refined, accurate and valid.” Movement for Opportunity and Equality v. General Motors, 622 F.2d 1235, 1245 (7th Cir.1980). This he accomplishes by performing a statistical analysis upon the same database or by challenging the database upon which the plaintiff relies. See Id. at 1245-46.
Of course, the defendant may accept the disparity shown by the plaintiff, but seek to justify it as the product of a job related requirement of business necessity. New [925]*925York Transit Authority v. Beazer, 440 U.S. 568, 587 n. 32, 99 S.Ct. 1355, 1367, n. 32, 59 L.Ed.2d 587 (1979); United States v. Town of Cicero, Illinois, 786 F.2d 331 (7th Cir.1986). In the event the defendant selects this tactic, the plaintiff is entitled to present evidence to show that the proffered nondiscriminatory reasons are pretextual or that alternatives exist with a lesser discriminatory impact. Teal, 457 U.S. at 447, 102 S.Ct. at 2530-31.
Plaintiffs’ prima facie showing here consists of statistics demonstrating that the captain eligibility rule has an adverse impact on minorities in two ways. First, plaintiffs’ statistics show that the success rate of minorities is less than 80% of the success rate for whites. When minority captains and lieutenants are compared to the white captains and lieutenants eligible to take the battalion chief examination it is clear that the minority candidates have a lower success rate. It is generally recognized that the failure of a sex, race or ethnic group to have a success rate which is at least 80% of the rate of the most successful group is considered evidence of adverse impact. See Watson v. Fort Worth Bank and Trust, — U.S.-, 108 S.Ct. 2777, 2789 n. 3, 101 L.Ed.2d 827 (1988). See also EEOC’s Uniform Guidelines on Employee Selection Procedures, 29 C.F.R. 1607.4(4).
Second plaintiffs’ evidence shows, in general, significant underrepresentation of minorities in upper level positions, significantly less than the percentage of minorities in the lower level positions. Plaintiffs’ statistics specifically show an underrepresentation of minorities at the rank of battalion chief versus the rank of captain and a significant underrepresentation of minorities at the rank of captain versus lieutenant. In general, minorities occupy fewer than 10% of the supervisory positions in comparison to nearly 28% of the lower level positions. These comparisons are all statistically significant and are strongly suggestive of discrimination absent some explanation. Courts have long recognized that nondiscriminatory employment practices will generally result in a workforce representative of the ethnic and racial composition of those available workers. See Teamsters v. United States, 431 U.S. 324, 340 n. 20, 97 S.Ct. 1843, 1856-57 n. 20, 52 L.Ed. 2d 396 (1976). The existence of disparities in percentages of minorities in the upper versus the lower level positions is potent evidence of discrimination where the employer promotes from within and obviously is selecting more non-minorities for promotion than would normally be expected given the composition of the workforce. See Shidaker, 833 F.2d at 631.
In response the city challenges the plaintiffs’ proof and offers a legitimate non-discriminatory reason: the captain’s eligibility rule is job related. The City first disputes that the plaintiffs’ selection of the combined pool of captain and lieutenant as the relevant labor pool for comparison to battalion chief was proper since the captain eligibility rule by definition requires an applicant to be a captain. According to the City, lieutenants are therefore not minimally qualified and should not be considered for statistical purposes as part of the pool. While the City’s stated qualifications for promotion to battalion chief are entitled to presumptive validity and should be considered when determining proper labor pools, such qualifications must be carefully scrutinized especially where as here the purported qualification is the challenged criterion resulting in adverse impact. The analysis therefore cannot be neatly dissected into a tripartite discussion of prima facie case, rebuttal, and pretext. Simply, if the plaintiffs succeed also in showing the asserted job-relatedness of the criterion is not worthy of credence, then any suggestion that the labor pool for comparison should only include captains must also fail. As fully discussed below, I find that testimony and evidence presented during the hearing strongly supports plaintiffs’ claim that the captain eligibility rule can not be justified by business necessity. Therefore, captains and lieutenants are equally eligible to take the exam and the City’s protestations that the lieutenants should not be included for [926]*926comparison must be dismissed. I find the plaintiffs’ selection of labor pools for comparison appropriate under the circumstances.
The City next objects to the time period which plaintiffs select for comparison of the results to determine underrepresentation. The plaintiffs’ statistics focused on the composition of the ranks at present. The City argues that this is misleading since as a result of the recent promotion (July 1, 1988) to captain rank for a number of minorities it is not surprising that until promotions from the captain rank are made to the battalion chief rank a substantial disparity exists between the minority representation of these two ranks. This the City claims can reasonably explain the disparity and is not discrimination resulting from the captain eligibility rule. However, what the City concedes is more powerful than what it posits. The City acknowledges that the existing disparity and underrepresentation between the battalion chief rank and the rank that feeds into it is a direct result of the prior history of discrimination in the Chicago Fire Department. The City also acknowledges that while it currently seeks as part of its remedial action pursuant to the 1980 Consent Decree to make battalion chiefs more representative of the minority population at the captain level, its efforts will still at best only produce a battalion chief force of less than 10% minority. Thus, while the City’s promised future remedial actions may dissipate to some degree the strength of plaintiffs’ showing, it nonetheless fails to address the admitted severe underrepresentation which continues years after the signing of the consent decree and by the City’s own admission will continue for a number of years into the future. Although this court does not seek to discourage the City from its continued good faith efforts to comply with the 1980 Consent Decree, Title VII does not contemplate that a finding of existing impediments resulting in discrimination can be excused or salved by promises to do better in the future. Existing practices having an adverse, discriminatory impact on minorities must be treated for what they are: unconscionable violations of this nation’s Civil Rights Act. Accordingly, I find the City’s tender of statistics that future promotions will relieve the existing disparity to be encouraging, but nonetheless insufficient to defeat the plaintiffs’ showing of an existing disparity.
The City also challenges the plaintiffs’ statistical test and results in determining whether the success rate of minorities was at least 80% of the success rate of non-minorities in the relevant ranks. The City argues that results obtained do not sufficiently support the discriminatory inference plaintiffs request the court to draw. I recognize that “statistics come in an infinite variety and their usefulness depends on all of the surrounding facts and circumstances.” Teamsters v. United States, 431 U.S. 324, 340, 97 S.Ct. 1843, 1856-57, 52 L.Ed.2d 396 (1977). I find, however, that upon consideration of plaintiffs’ statistics that they are the product of approved statistical method and technique in these particular types of cases. There is some suggestion by defendants that the plaintiffs should have refined the results by use of multiple regression analysis. See, e.g., Campbell, Regression Analysis in Title VII Cases, 36 Stan.L.Rev. 1299 (1984). If as defendants suggest such analyses were critical it confounds the court why defendants did not undertake such task themselves. Where the defendant discerns “fallacies or deficiencies in the data offered by plaintiffs, he is free to adduce countervailing evidence of his own.” Dothard v. Rawlinson, 433 U.S. 321, 331, 97 S.Ct. 2720, 2727-28, 53 L.Ed.2d 786 (1977). The defendants did not convincingly do so here.
Finally, the City offers its alternate response — the captain eligibility rule is job related. The City contends that to strike down the captain eligibility rule would have the effect of throwing unqualified employees into impossible situations and exposing the community to considerable risk. This would result presumably because lieutenants promoted to battalion chief would be given supervisory responsibility far beyond their capabilities and because by mucking with an ordered hierarchy of command, the [927]*927court would plant seeds of disaffection and dissension in those who have progressed orderly through the ranks. I find these proffered justifications for the retention of the rule under the present circumstances to be insubstantial.
Despite the City’s characterization of a lieutenant’s responsibilities as dramatically different from the duties of a captain the evidence showed substantial similarity in the positions. The only difference between the functions each performs is that the captain performs some administrative duties not usually assigned to a lieutenant. However, there was testimony that when captains are not assigned to a shift or firehouse the lieutenants frequently are required to assume those traditional administrative duties of a captain. In any event, the fact that a lieutenant may not have immediate familiarity with the filling out of forms is for all practical purposes not likely to cause great and unavoidable risk to public safety should as a lieutenant he be promoted without having had this type of experience.
Presumably, the business necessity claim upon which the City most relies is that the time spent in a supervisory position as a captain matures one as an individual and makes a better battalion chief. I have no doubt that in general this might be the case. But the City’s own checkered adherence to this requirement reveals the relative unimportance to which the City assigns it. In 1978, twelve white lieutenants were permitted to take a battalion chiefs examination without first serving as captains. The Commissioner has appointed several lieutenants to deputy district chiefs, positions which have supervisory authority over battalion chiefs, despite the fact that these lieutenants never served as captains. The recent promotions to captain in July of this year produced individuals who will now be eligible for taking the battalion chief examination and for promotion to battalion chief with as little as six months experience as captain. Indeed, the City has stated that even one day’s experience as Captain is sufficient for promotion to battalion chief. I find the relative lack of experience of these candidates who have made battalion chief or will soon be eligible for promotion to battalion chief undermines the City’s claim that one must first serve a period of time as captain before taking the battalion examination. Even if the City could demonstrate that the cumulative supervisory experience that one gains as a captain is an essential criterion for promotion to battalion chief, then those lieutenants who take the exam could be required to serve a specified stint as captain before promotion to battalion chief.
The City also contends that the captain eligibility rule is required because only by taking a captain’s examination is a lieutenant instructed and tested on that material and information contained within that test. Of course, this reasoning fails against the named plaintiffs who allegedly have already taken the captain’s examination and passed it but who were never promoted to captain. Beyond this, there is evidence that in the past the examinations for both captain and lieutenant tested on identical information. The court cannot say that the plaintiffs may not have already been instructed on the same information now being presented in the captains examination.
I conclude that the City has failed to offer any persuasive reasons why the captain eligibility rule is job related. I have no doubt that, as the City warns, this finding may result in morale problems in the ranks. But a court cannot bend to whether there might follow some resistance to an order striking down employment practices viola-tive of Title VII. As a society we must be nudged sometimes to discover that we advance only when we all advance. Firefighters within the rank disenchanted by today’s decision will eventually discover that the impetus behind the plaintiffs’ request and this court’s order is to secure as soon as possible a better working environment within the Chicago Fire Department that enures to the benefit of all.
I find therefore that the plaintiffs have demonstrated at this preliminary injunction stage that they will likely prevail on the merits at trial on the issue of whether the captain eligibility rule violates Title [928]*928VII. In determining whether plaintiffs are entitled to the issuance of a preliminary injunction, I must also consider whether they will suffer irreparable harm if I do not issue injunctive relief, the degree of harm the defendants may suffer if I grant in-junctive relief, and whether the injunctive relief will harm the public interest. See Curtis v. Thompson, 840 F.2d 1291, 1296 (7th Cir.1988). I find that the plaintiffs have satisfied these requirements.
Should the battalion chief examination occur as scheduled without the inclusion of the present plaintiffs, the plaintiffs may suffer irreparable injury in that the promotions to battalion chief would then be filled from the list of present applicants unless another test is immediately issued to plaintiffs as well and selections from the successful examinees of the two groups made at the same time. The City has offered no indication that this would occur and given the infrequent and irregular scheduling of the battalion chief exam the plaintiffs may be prejudiced in waiting years before again having this opportunity. See Viscuso v. City of St. Louis, 552 F.Supp. 15 (E.D.Mo.1982).
The only harm the City may suffer in requiring the plaintiffs to be included is the additional preparation and training costs associated with the additional applicants. On the evidence presented to me I find these costs to be insubstantial. As to the City’s claim that it will suffer harm in promoting inexperienced lieutenants to battalion chiefs, I find such a claim insupportable in light of its past action and further in light of the scope of this order. Plaintiffs do not seek and this court is not ordering that plaintiffs be promoted. Simply, the city must make the battalion chief examination available for lieutenants such as the plaintiffs interested in taking it. In this spirit, I also conclude that the relief the plaintiffs seek is in the public interest.
I conclude as plaintiffs suggest that a race conscious remedy is appropriate here where the goal is to eliminate as soon as possible the effect of the department’s long term discrimination of minorities from the upper ranks. I believe that the only expeditious way in which the City can maintain its commitment to the Consent Decree and its affirmative action obligations is to permit minority lieutenants the opportunity to take the battalion chief examination. This broad remedial action is permitted here and I believe that after a full trial such relief would be required. See United States v. Paradise, 480 U.S. 149, 107 S.Ct. 1053, 94 L.Ed.2d 203 (1987).
Accordingly, I conclude that plaintiffs and the class they represent, to wit: all black and Hispanic lieutenants in the Chicago Fire Department who passed the captain’s examination but have not yet been promoted to captain, are entitled to preliminary injunctive relief in postponing the battalion chief written examination presently scheduled for August 13, 1988 until such time as the plaintiffs and similarly situated individuals may be adequately prepared and arrangements made for their inclusion. In no event should such postponement of the written examination be delayed more than sixty days. The City is directed to report back to this court within ten days on the necessary steps it will take to administer at the earliest feasible date its battalion chief examination in light of this ruling to the applicants who have already signed up for it as well as to the plaintiffs and the members of the class who desire to take the examination. In so ruling, the court has limited its immediate relief to the limited scope of the plaintiffs’ request for a preliminary injunction and does not preclude broader remedial action that might be appropriate under the circumstances.