Anheuser-Busch, Inc. (the Company) appeals from the district court’s1 judgment in favor of appellees Jimmy Easley, Andre Griffin, and Patricia Murphy, in an employment discrimination action brought under 42 U.S.C. § 1981 and Title VII of the Civil Rights Act of 1964. For reversal, Anheuser-Busch contends that the district court erred in (1) excluding most of the testimony of two of its witnesses; (2) finding intentional discrimination against all three appellees; (3) not considering whether appellees would have been hired absent discrimination; and (4) improperly determining the nature and extent of the relief and attorneys’ fees granted. We affirm in part and reverse and remand in part.
I. BACKGROUND.
Easley, Griffin, and Murphy, all of whom are black, filed the present civil rights action alleging that Anheuser-Busch discriminated against them in hiring because of their race. All three were unsuccessful applicants for positions as bottlers at the Company’s St. Louis plant. At issue are certain hiring practices utilized by Anheuser-Busch between 1978 and 1980.
Prior to 1979, the hiring procedure for bottlers at the St. Louis plant involved a review of written applications, one or more interviews, reference checks, and a physical examination. In 1978, Anheuser-Busch hired 427 bottlers on this basis, 157 (36.7%) of whom were black, and 270 (63.2%) of whom were white.2
Beginning in February 1979, the Company added to this procedure a written preemployment test developed by an outside consultant specifically for the bottler job. Anheuser-Busch maintains that it introduced the test in order to have a more objective means of obtaining better qualified employees.3 The Company contends that it had received complaints from supervisors about the ability of some bottlers hired under the old system, and that certain changes at the plant had underscored the need for obtaining a more competent work force. The changes included introducing more sophisticated machinery into the plant, and running the bottling operation seven days a week, three shifts a day. The change to a continuous operation created a need for additional bottlers, many of whom were hired to work on weekends and holidays to supplement the regular work force. These bottlers were called upon to perform a variety of job assignments under a number of different supervisors. According to Anheuser-Busch, the purpose of the written test was to facilitate the hiring of employees well-suited to performing the various tasks required of bottlers in these changing conditions. Nearly 1,500 applicants took the pre-employment test, roughly 46% of whom were black. Of the 342 bottlers hired under the testing program, 62 (18%) were black, and 280 (82%) were white.
Patricia Murphy submitted a written application for employment with AnheuserBusch on September 19, 1978, while the old hiring procedure was still in effect. A receptionist in the personnel office informed her that the Company was not hiring at the time, and that her application would be kept on file. She was not interviewed. On the next hiring date, October 15, 1978, the Company hired twenty-seven bottlers from a pool of at least 252 applica*255tions then on file (78 white and 174 black).4 Murphy was not among those selected, and the record contains no information about the racial composition of the group hired. No other hiring took place before the testing program began in February 1979.
Anheuser-Busch called Murphy in to be tested on May 22, 1979. She passed the test and was subsequently interviewed and given a physical examination. Although the Company determined that Murphy was fully qualified and ready for a job as a bottler, she was never hired. The reason for this, according to Anheuser-Busch, was that the demand for bottlers had slacked off by the time Murphy attained “ready” status in July of 1979. In fact, the only hiring after June 24,1979, occurred in February 1980, when the Company selected 46 (8 blacks and 38 whites) of the applicants then qualified and ready for immediate hiring.5
Jimmy Easley submitted a written application for employment as a bottler on March 28, 1979, and took the test on June 7, 1979. Andre Griffin applied on April 5, 1979, and took the test on May 18, 1979. Both men failed the test and received no further job consideration.
All three appellees filed timely charges of race discrimination against the Company with the Equal Employment Opportunity Commission (EEOC). In each instance, the EEOC found reasonable cause to believe the charge was true, and issued notice of a right to sue. This lawsuit followed.
At a three-day bench trial in March 1983, appellees advanced both disparate treatment and disparate impact theories of discrimination.6 The district court issued a lengthy memorandum opinion holding that Anheuser-Busch had unlawfully discriminated against appellees under both theories. Easley v. Anheuser-Busch, Inc., 572 F.Supp. 402 (E.D.Mo.1983). The trial judge found that Easley and Griffin were entitled to prevail on their disparate impact claims7 because the pre-employment test that eliminated them from fur*256ther job consideration had a material adverse impact on black applicants,8 and the Company failed to demonstrate that the test was a valid selection device that was justified by business necessity.9 Id. at 406, 414-15. Additionally, an alternative selection procedure (the informal pretest hiring procedure), which did not have the same adverse impact on black applicants, was available to the Company and, despite Anheuser-Busch’s allegations, had not been shown to be • unsatisfactory. Id. at 410.
The district court found in favor of all three appellees under a disparate treatment analysis.10 The court determined that each had established a prima facie case of racial discrimination under the standard set forth in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802, 93 S.Ct. 1817, 1824, 36 L.Ed.2d 668 (1973), by showing that they were black, that they applied and were qualified for the entry-level bot-tier jobs for which Anheuser-Busch was seeking and hiring applicants, and that they were rejected while the Company hired others similarly qualified. 572 F.Supp. at 413. With respect to Patricia Murphy, the trial court concluded that AnheuserBusch had failed to articulate any legitimate nondiscriminatory reason for not hiring her, and she was therefore entitled to judgment on her disparate treatment claim under Texas Department of Community Affairs v. Burdine, 450 U.S. 248, 101 S.Ct. 1084, 67 L.Ed.2d 207 (1981). The articulated reason for rejecting Easley and Griffin was their failure to pass the pre-employment test. The court found that because Anheuser-Busch instituted the test with knowledge of its adverse impact, and because the test clearly was not valid, the Company’s explanation did not provide a “legitimate defense.” 572 F.Supp. at 414. The court further concluded that statistical evidence showing gross disparities *257in the hiring of black and white bottlers,11 along with the scheduling of applicants for testing in a manner which resulted in the ineligibility of a majority of black applicants,12 compelled a finding of intentional discrimination as to all three appellees.
The district court entered judgment awarding appellees backpay together with prejudgment interest, plus enhanced attorneys’ fees and expert witness fees. In addition, the court awarded each appellee compensatory damages of $500 for emotional distress. The court also enjoined Anheuser-Busch, “its agents, consultants, contractors, and employees” from discriminating in employment on the basis of race, and from using the same or any similarly discriminatory selection procedure. Finally, the court ordered Anheuser-Busch to provide employment for each appellee with appropriate seniority rights and benefits. Anheuser-Busch appeals from the district court’s judgment and award.
II. DISCUSSION.
A. Exclusion of Testimony.
Anheuser-Busch first contends that the district court erred in precluding a substantial part of the testimony of two of the Company’s witnesses, Dr. Edmund Gaydos and Dr. Barry Seeskin, because they had not been identified as expert witnesses sufficiently in advance of trial.
On March 11, 1983, ten days prior to the March 21 trial setting,13 Anheuser-Busch identified both Dr. Gaydos and Dr. Seeskin as witnesses in a pretrial submission witness list, but did not indicate that they would testify as experts.14 At trial, Anheuser-Busch’s counsel stated that he informed plaintiffs’ attorney during two conversations in March that the witnesses were experts. However, he did not serve supplemental answers indicating the nature of their expert opinion testimony until the afternoon of March 21.
Neither Dr. Seeskin nor Dr. Gaydos was specially retained to testify at trial. Both were industrial psychologists employed by Anheuser-Busch to work as in-house experts. Because both men had been identified as witnesses, plaintiffs’ counsel did not object to their testifying to facts as employees of the Company, but did object to any expert opinion testimony. The district court sustained the objection on the ground that plaintiffs had been denied an opportunity for appropriate discovery, and ruled that both men could testify as ordinary witnesses. However, after each witness had testified briefly and the court had sustained repeated objections on the ground that the testimony was expert in nature, the trial judge concluded that the substance of the testimony offered was such that he was unable to separate Dr. Gaydos’ and Dr. Seeskin’s statements about what they did for the Company from their expertise. He therefore granted plaintiffs’ motions in limine to exclude further testimony from either witness. Anheuser-Busch contends that these evidentiary rulings constitute reversible error.
*258We disagree. A sanction imposed by a trial court for failure to make appropriate discovery is a ground for reversal only when it constitutes an abuse of discretion. See Laclede Gas Co. v. G.W. Warnecke Corp., 604 F.2d 561, 565 (8th Cir. 1979). In this instance, we cannot say that the district court abused its discretion by not allowing Dr. Gaydos and Dr. Seeskin to testify as experts. In their interrogatories, plaintiffs requested certain information about “each person whom defendant expects to call as an expert witness at trial.” Anheuser-Busch contends that the testimony of Dr. Gaydos and Dr. Seeskin should not have been classified as “expert” because it was not “acquired or developed in anticipation of litigation or for trial * * as required by Fed.R.Civ.P. 26(b)(4).15 The Company essentially conceded the expert nature of the testimony in its belated supplemental response to plaintiffs’ interrogatories concerning experts. However, even if, as in-house experts, Dr. Gaydos and Dr. Seeskin were not within the ambit of Rule 26(b)(4), they were still “experts” within the scope of plaintiffs’ discovery requests. Their identity and the facts known and the opinions they held were subject to discovery under the general discovery provisions of Rule 26(b)(1). See, e.g., In re Sinking of Barge “Ranger I”, 92 F.R.D. 486 (S.D.Tex.1981); Grinnell Corp. v. Hackett, 70 F.R.D. 326 (D.R.I.1976); Virginia Elec. & Power Co. v. Sun Shipbuilding & Dry Dock Co., 68 F.R.D. 397 (E.D. Va.1975). Anheuser-Busch possessed the information required to answer plaintiffs’ interrogatories well in advance of trial. We must agree with the district court that plaintiffs were entitled to know the substance of the expert testimony before March 21.16 “Discovery of expert opinion must not be allowed to degenerate into a game of evasion.” Voegeli v. Lewis, 568 F.2d 89, 97 (8th Cir.1977).
We believe that the trial judge erred, however, in not permitting Dr. Gay-dos or Dr. Seeskin to continue testifying as fact witnesses to their actions as employees of Anheuser-Busch. Although we recognize the difficulty of separating statements about actions taken as in-house experts from what would essentially be expert opinion testimony regarding the bottler’s test, the better course in this bench trial would have been to hear the testimony, and continue to sustain objections when appropriate. As it was, Anheuser-Busch was apparently precluded from presenting some admissible evidence.
Nevertheless, we are not persuaded that reversal on this basis is necessary. We have carefully reviewed the entire record before the court, including the testimony of the two witnesses, the offers of proof made in connection with their excluded testimony, Dr. Gaydos’ testimony at an arbitration proceeding concerning the validity of the test, and the arguments and statistical information set forth on appeal. Having done so, we conclude that the value of the witnesses’ testimony was directly related to their status as experts. Therefore, we cannot say that the district court’s error in excluding further fact testimony in *259this bench trial was sufficiently prejudicial to warrant remand for a new trial.
B. Disparate Treatment Claim.
Anheuser-Busch next challenges the district court’s determination that the Company intentionally discriminated against all three appellees. The Company maintains that none of the appellees is entitled to prevail under a disparate treatment theory.
A court of appeals may reverse a district court’s finding on the ultimate factual question of discriminatory intent only if it concludes that the finding is clearly erroneous. Pullman-Standard, v. Swint, 456 U.S. 273, 102 S.Ct. 1781, 72 L.Ed.2d 66 (1982). “A finding is ‘clearly erroneous’ when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.” United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 542, 92 L.Ed. 746 (1948).
1. Failure to Hire Murphy.
The district court found that AnheuserBusch intentionally discriminated against Patricia Murphy by failing to hire her on October 15, 1978, the first hiring date after she filed her application, or in 1979-80, after she had passed the test and was fully qualified for employment.
As noted above, the Company had at least 252 applications on file as of October 15, but hired only 27 bottlers on that date. The record contains no information about the racial composition of the applicants selected. We note, however, that approximately 37% of the bottlers hired in 1978 were black, a figure well in excess of the percentage of minority workers available for hiring in the St. Louis metropolitan area work force (a figure less than 23%). 17 Moreover, the evidence indicates that Murphy had not completed the hiring procedure (she had not been interviewed, had her references checked, or received a physical examination) by October 15, and thus was not ready for immediate hiring on that date. In the circumstances, the evidence is simply insufficient to support a finding that Anheuser-Busch intentionally discriminated, against Murphy because she was not among the small group of applicants hired on October 15. The district court’s finding in this regard is clearly erroneous.
Adequate support exists, however, for the finding that Anheuser-Busch intentionally discriminated against Murphy by failing to hire her under the testing program. The Company called Murphy in to be tested on May 22, 1979. Thereafter, although Company personnel determined that she was fully qualified for a job as a bottler, she was never hired. The inference from Anheuser-Busch’s evidence is that the Company did not hire her because bottler jobs were no longer available by the time she had completed the application process in July 1979. The district court found this explanation insufficient to meet the burden of articulating a legitimate nondiscriminatory reason for failing to hire Murphy, and concluded that she was therefore entitled to prevail on her claim. 572 F.Supp. at 414. See Burdine, 450 U.S. at 254-55, 101 S.Ct. at 1093-94. Even assuming, however, that the Company’s reason was adequate to rebut the presumption of discrimination raised by Murphy’s prima facie case,18 the district court also relied on statistical evidence which suggested that the Company’s proffered explanation was pretextual. As noted earlier, the trial court based its finding of intentional discrimination primarily on gross discrepan- \ *260cíes in the numbers of black and white bottlers hired in 1979-80 (see n. 11, supra), and on evidence suggesting that the Company scheduled applicants for testing in a racially discriminatory manner, with blacks as a group having a far longer wait between application and testing than white applicants (see n. 12, supra). As a result, the Company met its hiring needs before most black applicants became eligible for jobs. In the final analysis, only 18% of the bottlers hired were black, even though 34% of the applicants who passed the test were black.
To refute this evidence, appellant points out that 144 of the 204 black applicants who passed the test did so in July 1979 or later, and that of the 60 blacks who passed the test and were in the hiring pool when Anheuser-Busch did most of its hiring, 55 were selected. Appellant further submits that the reason nearly all the applicants tested in the early days of the program were white was not because of discriminatory scheduling, but because most of those initially tested were “current” 1979 applicants, many of whom were union referrals who happened to be white. Moreover, blacks who applied after the testing program began were, on the average, tested faster than comparable white applicants. With regard to the large number of blacks tested later in the program, AnheuserBusch asserts that when it discovered that blacks were failing the test at a disproportionately high rate, the Company took steps to increase the number of blacks tested in hopes of increasing the number of blacks in the applicant pool. One of the methods the Company used was to go back through 1978 applications to find more blacks for testing. In fact, AnheuserBusch maintains that it scheduled Murphy for testing only because of the Company’s affirmative efforts to test more blacks, and therefore she actually received more favorable treatment because of her race.
Anheuser-Busch fails, however, to explain why, during a period when the Company was allegedly trying to hire more black bottlers, when it knew that referrals from the union yielded mostly white applicants and it also received a substantial number of walk-in applications, and when it was concededly aware that pencil-and-paper tests generally have an adverse impact on blacks, it did not attempt to schedule more black applicants from the outset instead of later in the process when most of the hiring had been done. Although many of those initially tested were recent job applicants referred by the union, the test log reveals that from the beginning, the Company was going back in its files and selecting a substantial number of 1978 applicants to test, nearly all of whom were white.19 Furthermore, Anheuser-Busch fails to explain why, when 144 blacks passed the test after June 1979, only 8 (17%) of the 46 bottlers hired in February 1980 were black, a result hardly consistent with the Company’s assertions that it attempted to cure the adverse impact of the test.
We are mindful that statistics must, of course, be used carefully. “[T]hey come in an infinite variety and * * * their usefulness depends on all of the surrounding facts and circumstances.” International Brotherhood of Teamsters v. United States, 431 U.S. at 340, 97 S.Ct. at 1856. Nevertheless, with respect to individual claims of discrimination, statistics may be of probative value on the issue of intent. See, e.g., Craik v. Minn. State Univ. Bd., 731 F.2d 465, 471 (8th Cir.1984); Harper v. Trans World Airlines, Inc., 525 F.2d 409, 412 (8th Cir.1975); King v. Yellow Freight System, Inc., 523 F.2d 879, 882 (8th Cir. 1975). It is a form of circumstantial evidence from which an inference of discriminatory animus may be drawn. Having carefully examined the record as a whole, we conclude that the evidence is sufficient to support a finding that Anheuser-Busch’s *261reason for not hiring Murphy was pretextual.
Murphy was tested approximately 265 days after she filed her application, a period substantially longer than the average wait for blacks or whites. The Company asserts that it tested her as part of its efforts to remedy the adverse impact of the test. Yet even though she passed the test on May 22, the Company fails to explain why she was not processed in time to be considered for one of the 121 bottler positions filled in June.20 Nor does the Company give any satisfactory reason for not including Murphy among those hired in February 1980, beyond noting that many persons applied for only a few jobs. In the circumstances, we cannot say that the district court clearly erred in finding that Anheuser-Busch intentionally discriminated against Patricia Murphy by not hiring her under the testing program.
2. Jimmy Easley and Andre Griffin.
Easley’s wait from the date of application to the date of testing was approximately 70 days. Griffin waited only 45 days. Since both had a waiting period less than the average time for white applicants, the argument of discriminatory test scheduling cannot be said to apply to either of them. Moreover, both men failed the test and, like all applicants who failed — black or white21 — they received no further consideration. Although the court later determined that the test was not a valid selection procedure, for purposes of Easley’s and Griffin’s disparate treatment claim, failure on the test provided a legitimate, nondiscriminatory reason for not hiring them. Thus, the evidence principally relied on by the district court in finding improper motive in Anheuser-Busch’s hiring practices has no application to these appellees.22 We conclude that the district court’s finding of intentional discrimination was clearly erroneous with respect to Easley and Griffin.
C. Failure to Determine Whether Appellees Would Have Been Hired Absent Discrimination.
Anheuser-Busch next contends that, after finding that the Company unlawfully discriminated against appellees, the district court erred in simply assuming that all three would have been hired absent discrimination and were therefore entitled to backpay. The Company argues that, because the number of qualified applicants far exceeded the number of job openings, the trial court should have determined whether Easley, Griffin, and Murphy would have been among those who received the jobs that should have gone to black applicants. Awarding backpay without making this finding, the Company asserts, subjects Anheuser-Busch to potential liability at the hands of all 617 rejected black applicants, clearly an unfair result. Anheuser-Busch submits that the court should apply the analysis set forth in Hameed v. International Ass’n. of Bridge, Structural and Ornamental Iron Workers, Local 396, 637 F.2d 506 (8th Cir.1980).
In Hameed, a class action involving a disparate impact claim, the court determined the number of class members who were “actual discriminatees” by estimating how many positions were denied black applicants by reason of the employer’s discriminatory policies. The court made this estimate by applying the percentage of applicants who were black to the total number of persons hired. The difference between the figure derived and the number *262of blacks actually hired constituted the number of actual discriminatees entitled to receive backpay. Id. at 520-21. For example, in the present case, 46% (679 of 1,477) of the applicants tested were black. Forty-six percent of 342 (the number of job openings for bottlers during the period in question) is 157. The difference between 157 and the number of blacks actually hired (62) is 95. Thus, under the Hameed formula, 95 additional positions should have gone to black applicants, and the total backpay awarded would be limited in light of that fact.
Unlike Hameed, this lawsuit is not a class action. Nevertheless, we agree that it is appropriate to apply similar reasoning in the circumstances of the present case in order to avoid unfairly penalizing Anheuser-Busch. The next question under Ha-meed, then, is which of the rejected black applicants would most likely have been selected for the 95 positions and should therefore be awarded backpay?
In class actions, which are usually bifurcated into a liability phase and a remedial phase, the employer carries the burden of demonstrating at the remedial phase that an individual applicant would not have been hired even absent unlawful discrimination. See id. at 520 (quoting Teamsters, 431 U.S. at 362, 97 S.Ct. at 1868). This court has applied similar reasoning in an individual discrimination action, noting that, after a job applicant has proved unlawful discrimination, the employer is entitled to prove that the “job applicant would not have been hired anyway in order to limit the job applicant’s relief.” King v. Trans World Airlines, Inc., 738 F.2d 255, 257-58 (8th Cir.1984). Accord Day v. Mathews, 530 F.2d 1083 (D.C.Cir.1976) (employer who was a “proven discriminator” could rebut plaintiff’s claims to relief in an individual employment discrimination action by establishing that plaintiff would not have been selected even absent discrimination). See also Nanty v. Barrows Co., 660 F.2d 1327, 1333 (9th Cir. 1981). Anheuser-Busch has, however, failed to make such a showing with respect to the present appellees.23 The Company’s bare assertion that the odds were against any rejected applicant being selected cannot suffice. Anheuser-Busch did not demonstrate at trial that any of the appellees failed to satisfy valid job qualifications. Nor did the Company provide the district court with information from which the court could have determined the likelihood of appellees being selected over other rejected black applicants. Indeed, the record indicates that Anheuser-Busch purged the only information which might have allowed the court to compare the applicants’ relative qualifications. Thus, while we agree in principle that it may be unfair to award full backpay to all rejected black applicants, based on the evidence before the district court, the trial judge did not err in assuming that Murphy, Easley, and Griffin would have been hired. They are therefore entitled to backpay.24
*263D. Remedies.
1. Backpay for Murphy.
The district court computed backpay for Murphy based on its determination that, absent discrimination, Anheuser-Busch would have hired her on October 15, 1978. Because we conclude that Anheuser-Busch did not intentionally discriminate against Murphy by failing to hire her on that date, the district court should, on remand, recompute the amount of backpay and interest to which she is entitled. Murphy could not have been tested before February 23, 1979, the first date on which the Company administered the test. The first hiring thereafter occurred on March 4, 1979. Her damages should therefore be measured from that date.
2. Award of Compensatory Damages.
The district court awarded $500 compensatory damages for mental and emotional distress to each of the appellees. Compensatory damages are available under section 1981, but not under Title VII. Muldrew v. Anheuser-Busch, Inc., 728 F.2d 989, 992 (8th Cir.1984). To prevail on a claim under section 1981, a claimant must prove discriminatory intent, as that section reaches only purposeful discrimination. General Building Contractors Ass’n. v. Pennsylvania, 458 U.S. 375, 388-89, 102 S.Ct. 3141, 3149, 73 L.Ed.2d 835 (1982). Because we reverse the finding of discriminatory intent with respect to appellees Easley and Griffin, they are not entitled to receive compensatory damages. We affirm the award of compensatory damages for Patricia Murphy.
3. Injunctive Relief.
The district court’s judgment included the following injunctive language:
IT ' IS HEREBY FURTHER ORDERED AND ADJUDGED that defendant, Anheuser-Busch, Inc., its agents, consultants, contractors, and employees be and the same are permanently enjoined from engaging in employment discrimination on the basis of race, and more particularly, from instituting or using the pre-employment testing program which is the subject of this action, or any similar test, program, or method of hiring which results in racially discriminatory hiring practices, and from instituting or using any such test program or method of hiring which is invalid under the Uniform Guidelines on Employee Selection Procedures promulgated by the Equal Employment Opportunity Commission of the United States, 29 C.F.R. § 1607.1, et seq. (1982).
572 F.Supp. at 417-18. Anheuser-Busch argues that this language is vague and overbroad. Read literally, the injunction applies to all forms of race discrimination against every employee or applicant for every job at every Anheuser-Busch plant. It envelops not only Anheuser-Busch, but its contractors and consultants, whether engaged in work for Anheuser-Busch or not. As such, it goes far beyond the employment practices challenged here.
“[OJnce discrimination has been established in a Title VII action, the issuance of an injunction rests in the sound discretion of the district court * * Taylor v. Teletype Corp., 648 F.2d 1129, 1135 (8th Cir.), cert. denied, 454 U.S. 969, 102 S.Ct. 515, 70 L.Ed.2d 386 (1981). However, while injunctive relief is appropriate in this case, and while courts may look beyond the individual plaintiffs’ situation in fashioning relief, we must agree that the language of the present injunction is far too broad. On remand, we direct the district court to recast the injunction to apply only to the hiring practices for bottlers at the St. Louis plant that are at issue in this lawsuit.
In addition, the district court ordered Anheuser-Busch to provide all three appellees with employment as bottlers “at currently prevailing wages and with such seniority rights and accompanying benefits as would be appropriate to an employee having already accrued work experience commensurate with the award of back pay made to each plaintiff * * *.” 572 F.Supp. at 418. The court determined that Jimmy Easley and Andre Griffin should have been *264hired on April 15, 1979, and computed their backpay awards from that date. However, the district court also noted that bottlers hired on April 15 were laid off in 1980 and have not worked for Anheuser-Busch since that time. Id. at 411. In the circumstances of this case, we believe an award of backpay to these two appellees is sufficient to serve the “make whole” purpose of Title VII. See Albemarle Paper Co. v. Moody, 422 U.S. 405, 417-18, 95 S.Ct. 2362, 2371-72, 45 L.Ed.2d 280 (1975) (the purpose of Title VII is to make persons whole for injuries suffered because of unlawful employment discrimination). We therefore reverse the district court’s order insofar as it requires Anheuser-Busch to provide immediate employment, to Easley and Griffin. If, however, Anheuser-Busch recalls bottlers hired on April 15, 1979, the Company should consider appellees as having seniority rights commensurate with their backpay awards.
We have determined that Patricia Murphy should have been hired on March 4, 1979. The record does not reflect whether bottlers hired on that date are still working for Anheuser-Busch. On remand, the district court should consider whether bottlers employed on March 4 have also been laid off, and whether, in light of that determination, appropriate relief for Murphy should include immediate employment with full seniority rights and benefits.
4. Attorneys’ Fees.
The district court awarded plaintiffs $24,-648.23 in attorneys’ fees. It reached that figure by determining that $75 was a reasonable hourly rate of compensation, and multiplying that amount by the number of hours reasonably expended to reach a total of $18,532.50 (the “lodestar” amount). The court then enhanced the lodestar figure by 33%, based upon
the skills of plaintiffs’ attorneys and the acknowledged expertise of counsel, required for the successful prosecution of this case, the results obtained on behalf of plaintiffs, the clear vindication of the public interest contemplated with respect to actions of this nature, the contingent character of the compensation available, and the thorough and diligent efforts made in plaintiffs’ interest * * *.
Id. at 417, 95 S.Ct. at 2371.
Anheuser-Busch contends that the district court erred in enhancing the fee award. We agree. As prevailing parties in this civil rights action, appellees are entitled to recover reasonable attorneys’ fees. 42 U.S.C. § 1988; see Craik v. Minn. State Univ. Bd., 738 F.2d 348, 349 (8th Cir.1984). See also Hameed, 637 F.2d at 522 (prevailing party in a Title VII action is entitled to an award of reasonable attorneys’ fees). “[I]n some cases of exceptional success an enhanced award may be justified.” Hensley v. Eckerhart, 461 U.S. 424, 435, 103 S.Ct. 1933, 1940, 76 L.Ed.2d 40 (1983). Appellees’ success in this action, especially in light of our amended findings, was not so exceptional as to warrant an enhancement of the fee award.
Moreover, the reasons offered by the district court for enhancing the award involve factors normally considered in determining the basic fee.25 They are not *265sufficient in the circumstances of this case to provide an independent basis for enhancing the lodestar amount. See generally Blum v. Stenson, — U.S.-, 104 S.Ct. 1541, 1548-50, 79 L.Ed.2d 891 (1984). We conclude that enhancement was not necessary to provide fair and reasonable compensation to plaintiffs’ counsel. Accordingly, we reverse the district court’s award of an enhanced fee. Furthermore, the district court may wish to consider on remand whether the lodestar amount should be reduced in light of our amended findings.
In sum, we affirm the district court’s finding that Anheuser-Busch intentionally discriminated against Patricia Murphy, but reverse the finding of intentional discrimination against Jimmy Easley and Andre Griffin. We affirm the district court’s finding of discrimination against Easley and Griffin under a disparate impact analysis. Easley and Griffin are not entitled to receive compensatory damages, nor is Anheuser-Busch required to provide immediate employment for them. We affirm the backpay awards for all three appellees. However, on remand, the district court should recompute Murphy’s award of back-pay to date from March 4, 1979. In addition, we direct the district court to modify the injunctive decree to apply only to the hiring practices at issue in this lawsuit. The district court should also consider whether Anheuser-Busch should be required to employ Murphy immediately as a bottler, and whether, in light of our amended findings, the award of attorneys’ fees should be reduced. Although we leave the question of the basic fee award to the discretion of the district court, we reverse the award of the enhanced fee.
Accordingly, the judgment of the district court is affirmed in part and reversed in part, and remanded for further proceedings consistent with this opinion.
. The Honorable William L. Hungate, United States District Judge for the Eastern District of Missouri.
. The record contains no information about the total number of applicants for bottler positions in 1978, or the racial composition of the overall applicant pool.
. The company used the pre-employment test package as a selection device for bottlers from February 1979 through Feburary 1980. Scores on four separate parts of the paper-and-pencil test were used in the selection process: Production Arithmetic, Production Problems, Code Interpretation, and Oral Instructions. In order to pass the test and receive further consideration for employment, an applicant had to achieve either (1) a designated cutoff score on any three parts of the test, or (2) the cutoff score on two of the parts, and a score within 80% of the cutoff score on the remaining two parts.
. These figures and most of the statistics presented at trial were derived from analysis of a "testing log” maintained by Anheuser-Busch. The log, which provides the only available data about the Company’s processing of bottler applicants in 1979 and 1980, contains information about each applicant’s date of application, race, sex, date tested, test scores, and hiring status. The district court found that the Company had purged other data relating to applicants despite the pendancy of appellees’ EEOC charges and this lawsuit. Statistics thus reflect only those 1978 and 1979 applicants selected for testing, with no clear indication of how representative this group is of the total applicant pool.
. The district court found that the Company made these selections from a pool of 60 white and 61 black applicants as reflected in Defendant's Exhibit A. Anheuser-Busch claims that Exhibit A reflects the ready list of applicants as of the date of trial in 1983. The Company has not included Exhibit A as part of the record on appeal.
. The Supreme Court discussed the distinctions between disparate treatment and disparate impact theories of discrimination in International Brotherhood of Teamsters v. United States, 431 U.S. 324, 335-36 n. 15, 97 S.Ct. 1843, 1854-55 n. 15, 52 L.Ed.2d 396 (1977). Disparate treatment occurs where ”[t]he employer simply treats some people less favorably than others because of their race * * Id. Proof of discriminatory intent is critical. In contrast, disparate impact "involve[s] employment practices that are facially neutral in their treatment of different groups but that in fact fall more harshly on one group than another and cannot be justified by business necessity.” Id. Proof of discriminatory motive is not required. Both theories may be applied to a particular set of facts. See also Jones v. International Paper Co., 720 F.2d 496, 499 (8th Cir.1983).
. To establish a prima facie case on a disparate impact claim, plaintiffs must show that a facially neutral employment practice has a significantly adverse impact on a protected group. Once that showing is made, the burden shifts to the employer to demonstrate that the practice has a manifest relationship to the employment in question and is justified by business necessity. If the employer meets this burden, the plaintiffs may then show that other practices, which lack a similarly discriminatory effect, would satisfy the employer’s legitimate interests. Such a showing would be evidence that the employer was using the practice as a mere pretext for discrimination. See, e.g., Connecticut v. Teal, 457 U.S. 440, 446-47, 102 S.Ct. 2525, 2530-31, 73 L.Ed.2d 130 (1982); 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, 430-32, 91 S.Ct. 849, 853-54, 28 L.Ed.2d 158 (1971).
. Of the nearly 1,500 applicants tested (679 black and 798 white), only 30% of the blacks passed the test, while 50% of the white applicants passed. As the district court noted, these results indicate that the test had a material adverse impact under the "four-fifths” or “80%” rule established in the EEOC's Uniform Guidelines on Employee Selection Procedures, because the black pass rate was less than four-fifths of the white pass rate. 572 F.Supp. at 406. Moreover,
[u]nder the chi-square test of statistical significance, the probability that the difference in pass rates was the result of chance rather than race is less than one in 1,000. Applying the standard deviation method, comparing the expected pass rate for blacks (if race were not a factor in passing the test) to the actual or observed pass rate for blacks, the difference is 6.64 standard deviations.
Id. ■ The Supreme Court has noted that a difference greater than two or three standard deviations would appear to be statistically significant. Castaneda v. Partida, 430 U.S. 482, 496-97 n. 17, 97 S.Ct. 1272, 1281 n. 17, 51 L.Ed.2d 498 (1977). See also Hameed v. International Ass'n. of Bridge, Structural and Ornamental Iron Workers, Local 396, 637 F.2d 506, 513-14 n. 8 (8th Cir.1980).
. The trial court found, inter alia, that Anheuser-Busch failed to demonstrate that its outside consultant performed a thorough job analysis in connection with the development or validation of the test, or that the test program clearly approximated tasks performed on the job. Nor did the Company produce adequate evidence regarding the validity or reliability of the test. 572 F.Supp. at 409.
. To establish a prima facie disparate treatment case, plaintiffs must prove that they applied for an available position for which they were qualified, but were "rejected under circumstances which give rise to an inference of unlawful discrimination." Texas Dept. of Community Affairs v. Burdine, 450 U.S. 248, 253, 101 S.Ct. 1089, 1093, 67 L.Ed.2d 207 (1981). See also McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802, 93 S.Ct. 1817, 1824, 36 L.Ed.2d 668 (1973). Establishment of a prima facie case creates a "legally mandatory, rebuttable presumption" of discrimination. Burdine, 450 U.S. at 254 n. 7, 101 S.Ct. at 1094 n. 7. The burden of production then shifts to the employer to articulate a legitimate, nondiscriminatory reason for the employees’ rejection. ”[I]f the employer is silent in the face of the presumption, the court must enter judgment for the plaintiff because no issue of fact remains in the case.” Id. at 254, 101 S.Ct. at 1094. However, if the employer carries the burden of production, the presumption of illegal discrimination drops from the case, and the plaintiffs must show that the employer’s proffered reasons were merely a pretext. Plaintiffs retain the burden of persuasion throughout, and must convince the trier of fact by a preponderance of the evidence that they were victims of intentional discrimination. Id. at 253-54, 101 S.Ct. at 1093-94; Craik v. Minn. State Univ. Bd., 731 F.2d 465, 469 (8th Cir.1984).
. For example, of the more than 600 applicants who passed the test, roughly 34% were black, yet only 18% of the bottlers hired under the testing program were black. The overall selection rate for blacks who passed the test was 30.4%, a figure substantially lower than the selection rate for whites (70.2%). Even in February 1980, despite the Company’s claim that it actively sought to increase the number of qualified black applicants, and despite the fact that 144 blacks passed the test after June 1979, only 8 (17%) of the 46 bottlers hired were black.
. The court relied on evidence indicating that, on the average, black applicants submitted their applications earlier than whites, but had to wait longer to be tested (214.2 days from the date they filed their applications, as opposed to 101.6 days for white applicants). In addition, blacks were generally tested later in the year than whites, and as a result, were not in a position to be hired when most of the Company’s hiring was done. With the exception of 46 bottlers hired in February 1980, Anheuser-Busch completed its hiring by June 1979, yet only 41.6% of the blacks were tested before June, as compared with 90.1% of the white applicants.
. Because of a continuance, the trial actually began on March 23, 1983.
. Plaintiffs-appellees had requested identification of Anheuser-Busch's expert witnesses in interrogatories filed August 16, 1982.
. Rule 26(b)(4) provides in part:
Discovery of facts known and opinions held by experts, otherwise discoverable under the provisions of subdivision (b)(1) of this rule and acquired or developed in anticipation of litigation or for trial, may be obtained only as follows:
(A)(i) A party may through interrogatories require any other party to identify each person whom the other party expects to call as an expert witness at trial, to state the subject matter on which the expert is expected to testify, and to state the substance of the facts and opinions to which the expert is expected to testify and a summary of the grounds for each opinion.
. Dr. Gaydos’ testimony and proposed testimony focused mainly on the validity of-the test, his actions and recommendations regarding the adverse impact of the test, and alternative selection methods. Dr. Seeskin sought in his testimony and proposed testimony to present an alternative statistical analysis of the data in the test log to counter the analysis presented by plaintiffs’ expert. The district court’s exclusion of expert witness testimony in this case does not operate as a bar to Anheuser-Busch presenting such evidence in any other lawsuit challenging the hiring practices in question.
. Data showing the actual percentage of blacks and whites in the applicant pool from which the Company selected the 37% would, of course, be very relevant in a statistical analysis. See Hazel-wood School Dist. v. United States, 433 U.S. 299, 308 n. 13, 97 S.Ct. 2736, 2742 n. 13, 53 L.Ed.2d 768 (1977). It appears, however, that AnheuserBusch purged this data for 1978.
. Appellant suggests that, although the company did not hire Murphy, she was never rejected, and her name remained in the Company’s "ready” file. We agree with the district court that the failure to hire Murphy was, in the circumstances, tantamount to rejecting her for purposes of establishing a prima facie case. 572 F.Supp. at 406.
. An examination of the chronological test log indicates that, of the first 100 test takers listed, 47 were white applicants with current 1979 applications (most of whom were apparently referred by the union); 48 were white applicants who applied between January and August of 1978; 5 were black applicants who applied between May and September of 1978.
. The Company did not interview Murphy until June or give her a physical examination until July.
. Some evidence exists on the record that the Company processed 72 applicants (37 white and 35 black) beyond the test hurdle even though they did not pass the test. Apparently 27 of the whites in that category were ultimately hired and 2 of the blacks.
. The district court also relied on its finding that Anheuser-Busch knew in advance the bottlers’ test would have an adverse impact on blacks but implemented it anyway. 572 F.Supp. at 410. The evidence supports a finding only that the Company knew that written tests generally have an adverse impact on minorities.
. In King, the court indicated that the appropriate burden of proof to be placed on the employer once liability has been established is that of “clear and convincing evidence." 738 F.2d at 257. But cf. Craik v. Minn. State Univ. Bd., 731 F.2d at 470 n. 8 (suggesting that the appropriate standard of proof in such civil litigation is that of a preponderance of the evidence). In the present case, Anheuser-Busch has failed to carry its burden under either standard.
. In Hameed, because of the complexity and uncertainty of determining which applicants were entitled to backpay, this court concluded that a classwide backpay remedy was appropriate. We directed the district court to compute the total backpay due based on the number of “actual discriminatees” and to distribute the award in an equitable manner among class members. Here, because no class is before the court, it is impossible to determine what share of the award should be allocated to each potential claimant. However, based on the record before this court, we are not aware that Anheuser-Busch has incurred any other backpay obligations to claimants because of the hiring practices at issue here. The Company has therefore failed to demonstrate prejudice from the trial court’s assumption that Easley, Griffin, and Murphy would have been among those applicants entitled to be hired. Thus, while it may be appropriate, under the principles enunciated in Hameed, to put a cap on the Company's total exposure to backpay claims by rejected black applicants, we need not consider that issue here.
. This court has adopted the guidelines for determining attorneys’ fees set forth in Johnson v. Georgia Highway Express, Inc., 488 F.2d 714 (5th Cir. 1974). See Allen v. Amalgamated Transit Union, Local 788, 554 F.2d 876, 884 (8th Cir.), cert. denied, 434 U.S. 891, 98 S.Ct. 266, 54 L.Ed.2d 176 (1977).
The 12 factors are: (1) the time and labor required; (2) the novelty and difficulty of the questions; (3) the skill requisite to perform the legal service properly; (4) the preclusion of employment by the attorney due to acceptance of the case; (5) the customary fee; (6) .whether the fee. is fixed or contingent; (7) time limitations imposed by the client or the circumstances; (8) the amount involved and the results obtained; (9) the experience, reputation, and ability of the attorneys; (10) the “undesirability" of the case; (11) the nature and length of the professional relationship with the client; and (12) awards in similar cases.
488 F.2d at 717-19.
The district court indicated that it had considered “all appropriate [Johnson ] factors” in determining the basic fee award. 572 F.Supp. at 416. Anheuser-Busch complains that the district court used the Johnson factors first to de*265termine a reasonable hourly rate and then again to determine that an enhancement of 33% was warranted. A trial court may consider the Johnson factors in its initial calculation of hours reasonably expended at a reasonable hourly rate, and it may also consider such factors in determining whether the lodestar amount should be adjusted upward or downward.
However, the court should not engage in “double counting" by considering essentially the same factors twice to arrive at a total award. See Blum v. Stenson, — U.S. -, 104 S.Ct. 1541, 1548-50, 79 L.Ed.2d 891 (1984); Hensley v. Eckerhart, 461 U.S. at 434 & n. 9, 103 S.Ct. at 1940 & n. 9.