BOYCE F. MARTIN, Jr., Circuit Judge.

Deniece Scales, a female, brought this action under Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e, et seq., and the Equal Pay Act, 29 U.S.C. § 206(d), against defendant, J.C. Bradford and Company. Scales claimed that Bradford violated Title VII by denying her promotions because of her gender and by retaliating against her for filing a gender discrimination charge with the Equal Employment Opportunity Commission. Scales also claimed that Bradford violated the Equal Pay Act, alleging that she was paid lower wages than similarly situated male employees.

Pursuant to 42 U.S.C. § 2000e-5(f)(3), the district court referred this matter to a United States magistrate to conduct all necessary proceedings. During the process of discovery, a dispute arose as to the proper scope of discovery. Scales requested, inter alia, the production of Bradford's "master payroll record" for the years of 1981 to 1986. Bradford objected to this request as overly broad and unduly burdensome. The magistrate ordered Bradford to comply with this production request with the exception of partners-owners who were deemed to be irrelevant. Both parties filed objections to the magistrate's ruling. The district court reversed the magistrate's ruling in part, limiting Scales' access of payroll information to nonpartner employees who worked in the same department as Scales for the time period from June 1, 1983, to December 31, 1986.

A full trial was conducted before the magistrate on August 1-5, 1988. On January 6, 1989, the magistrate issued his report recommending that judgment be entered for Scales on her claims of discriminatory promotion practices and retaliation in violation of Title VII, but that judgment be entered for Bradford on plaintiff's claim under the Equal Pay Act. Both parties filed objections to the report and sought review by the district court. The district court adopted the magistrate's report except as to Scales' Title VII claim under a disparate impact theory. The district court ruled that Scales had failed to establish a disparate impact claim under Title VII and overruled the magistrate's report with respect to this finding. The district court reduced Scales' recovery from $48,820 to $21,634 to reflect its decision concerning her Title VII claim.

Scales subsequently filed an application for attorney's fees and costs pursuant to 42 U.S.C. § 2000e-5(c). Scales requested a "lodestar" amount of $83,057.75, with an upward adjustment of between thirty-three percent and seventy-five percent. Bradford timely filed a memorandum in opposition to Scales' application for fees. The district court reduced the hourly rate charged by Scales' attorneys and reduced the lodestar amount due to plaintiff's limited success. The district court gave plaintiff an enhancement of five percent to account for various factors. The total fee award was $18,773.83.

Scales has appealed from the decisions of the district court concerning a number of issues. First, Scales contends that the district court erred in limiting her discovery to requests to Bradford employees working in her department. Scales further contends that the district court erred in both affirming the magistrate's report with respect to her disparate treatment claim and by reversing the magistrate's report concerning her disparate impact claim. Finally, Scales contends that the district court abused its discretion by denying her prejudgment interest on her award and by reducing her requested attorney fees.

Scales was hired by Bradford as a secretary on August 11, 1980. Scales, who is a high-school graduate, had prior clerical and secretarial skills, but no prior brokerage experience. In October, 1981, she was promoted to bookkeeper in the municipal bond department. Scales consulted with EEOC officials on May 29, 1984, concerning the hiring of Steve Scott. However, she wrote a letter dated June 11, 1984, in which she stated that "steps have been taken to remedy the situation to my current satisfaction." On June 12, 1985, Scales filed the EEOC charge that gave rise to this action. On September 11, 1985, plaintiff was promoted to the position of broker service representative, but did not receive a raise immediately. Scales did not perform broker representative duties on a full-time basis until December 1, 1985. Scales continued to work as a broker representative until September, 1986, when she went on maternity leave. On November 12, 1986, Scales notified Bradford she was moving to Memphis, Tennessee, because her husband had been transferred there. She submitted her resignation to become effective on November 30, 1986.

In Bradford's municipal bond department, there are three categories of nonpartner positions: municipal institutional sales representative, broker services representative, and broker sales representative bookkeeper. There were no written qualifications for the broker representative and municipal sales positions, and no written evaluations or formal application process for promotions.

These three positions have separate and distinct responsibilities. A municipal sales representative is a commissioned municipal bond salesperson who sells bonds to potential investors. Bradford requires a municipal sales representative to have a Series 7 Registration from the National Association of Securities Dealers. On the other hand, broker representatives do not deal directly with the public; they evaluate bonds for Bradford's municipal sales representatives and brokers and, with supervision, purchase bonds for Bradford's bond inventory.

Scales claims that four hirings or promotions by Bradford resulted in a delay of her promotion to the broker representative position, and she claims this delay was based upon her gender. The individuals hired or promoted were Doug Eley, Steve Scott, Galen Foster and John Murfee.

One occasion on which Scales alleges that she was denied a promotion involved Bradford's hiring of Galen Foster in April, 1985. According to the partner in charge of this department, Foster had been in the trucking business and did not know much about bonds. Foster was trained on the municipal bond desk for a period of four months as a municipal sales representative.

Scales claims that the district court abused its discretion by limiting her discovery of Bradford's payroll records. During the course of discovery, Scales requested, inter alia, that Bradford produce its "master payroll record" for all its employees for the years 1981-1986. Bradford has approximately 1,000 partners and employees in 51 offices located in 13 states. Bradford objected to this request. The district court limited Scales' discovery request to employees of Bradford's Nashville municipal bond department for the time period from June 1, 1983, to December 31, 1986. In reviewing the district court's decision to limit discovery, this court will intervene only if it was an abuse of discretion resulting in substantial prejudice. Ghandi v. Police Dept. of City of Detroit, 747 F.2d 338, 354 (6th Cir. 1984), cert. denied sub nom. Ghandi v. Fayed, 484 U.S. 1042, 108 S.Ct. 774, 98 L.Ed.2d 861 (1988); Humble v. Mountain State Construction Co., 441 F.2d 816, 818-819 (6th Cir. 1971).

Scales contends that the district court's ruling limiting her discovery made it impossible for her to prove her Title VII gender discrimination claim. It is well-settled that information concerning an employer's general employment practices is relevant even to a Title VII individual disparate treatment claim. See McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973); EEOC v. Roadway Express Inc., 750 F.2d 40 (6th Cir. 1984) (holding that general employment practices are relevant to EEOC investigation of individual disparate treatment claim). A plaintiff who must shoulder the burden of proving that the reasons given for her denial of promotions are pretextual should not normally be denied the information necessary to establish that claim. See, e.g., Burns v. Thiokol Chemical Corp., 483 F.2d 300 (5th Cir. 1973). However, this desire to allow broad discovery is not without limits and the trial court is given wide discretion in balancing the needs and rights of both plaintiff and defendant. See Trevino v. Celanese Corp., 701 F.2d 397 (5th Cir. 1983). The focus of plaintiff's complaint was on the treatment she received from Bradford while working in its municipal bond department in Nashville. In her complaint, Scales alleged that on three separate occasions Bradford hired a man to fill a position in which she had expressed an interest. Where, as here, the employment decisions were made locally, discovery may be properly limited to the employing unit. See Earley v. Champion International Corp., 907 F.2d 1077 (11th Cir. 1990).

Furthermore, the district court found that plaintiff's husband, who was a former employee of Bradford and was now employed at a competitor of Bradford, had attempted to hire personnel away from Bradford. The court therefore concluded that discovery on a company-wide basis could cause Bradford substantial injury. Taking these factors into consideration, we cannot say that the district court abused its discretion in limiting the scope of discovery. See Marshall v. Westinghouse Electric Corp., 576 F.2d 588, 592 (5th Cir. 1978) (affirming denial of plaintiff's motion to compel information about all company employees where discrimination against only one employee alleged in complaint).

Scales next contends that the district court erred in not finding that Bradford's proffered nondiscriminatory reason for hiring Galen Foster was a mere pretext for gender discrimination. To prevail on a Title VII gender discrimination claim under disparate treatment theory, the plaintiff has the burden of proving a prima facie case of discrimination by a preponderance of the evidence. Texas Department of Community Affairs v. Burdine, 450 U.S. 248, 252-53, 101 S.Ct. 1089, 1093-94, 67 L.Ed.2d 207 (1981) (citing McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973)). Once a plaintiff has established a prima facie case, the defendant may rebut this by showing a legitimate nondiscriminatory reason for its actions. Burdine, 450 U.S. at 252-53, 101 S.Ct. at 1093-94. If the defendant can show a legitimate nondiscriminatory reason for its promotion and hiring decisions, the plaintiff must then show that the defendant's proffered reason is just a pretext for discrimination. Id.

The district court upheld the magistrate's report which found that the hiring of Foster for the municipal sales position was not a pretext for gender discrimination. The magistrate found that Scales had established a prima facie case under the disparate treatment theory. However, the magistrate found that Bradford had shown a legitimate business reason for its decision concerning Foster. The magistrate found, and the district court agreed, that the hiring of Foster was not a pretext for gender discrimination because Foster was more qualified for the municipal sales position than was Scales. Specifically, the magistrate found that Foster possessed sales experience and that Scales did not. The magistrate also found that Scales had never expressed any interest in this particular position and was really interested in obtaining the position of broker representative. We review these findings under a "clearly erroneous" standard. Johnson v. United States, 600 F.2d 1218, 1222 (6th Cir. 1979). Under this standard, a finding of fact should not be set aside unless the court "is left with the definite and firm conviction that a mistake has been committed." Id. (quoting United States v. United States Gypsum Co., 333 U.S. 364, 68 S.Ct. 525, 92 L.Ed. 746 (1958)). Plaintiff has failed to demonstrate how these findings are clearly erroneous.

Scales next argues that the district court erred in finding that she had failed to establish a prima facie case of gender discrimination based upon disparate impact theory. Specifically, plaintiff objects to the determination by the district court that plaintiff's statistical evidence was insufficient to prove a prima facie case of disparate impact discrimination. This determination of the district court is subject to the clearly-erroneous standard on appellate review. Bazemore v. Friday, 478 U.S. 385, 398, 106 S.Ct. 3000, 3007, 92 L.Ed.2d 315 (1986). Under a theory of disparate impact discrimination, a plaintiff may establish a prima facie case of discrimination by showing the existence of an employment practice which, although neutral on its face, has the effect of disproportionately affecting persons in a legally protected group. Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971). This theory is available to challenge promotion practices based upon subjective criteria. Watson v. Fort Worth Bank Trust Co., 487 U.S. 977, 108 S.Ct. 2777, 101 L.Ed.2d 827 (1988). In order to prevail on a disparate impact discrimination claim, the plaintiff must establish a prima facie case by first identifying the specific employment practice that is challenged. Id. at 994, 108 S.Ct. at 2788. Next, the plaintiff must show an adverse effect caused by the employment practice by offering "statistical evidence of a kind or degree sufficient to show that the practice in question has caused the exclusion of applicants for jobs or promotion because of their membership in a protected group." Id.

Once a plaintiff establishes a prima facie case of disparate impact by statistical evidence, the defendant must articulate a legitimate, non-discriminatory reason for the employment practice. At that point, the burden shifts back to the plaintiff to show either that the employer's reason is a pretext for discrimination, or that there exists an alternative employment practice which would achieve the same business ends with less discriminatory impact. Id. at 998, 108 S.Ct. at 2790.

To support her claim of gender discrimination, plaintiff introduced statistical studies done by a Dr. Hendrix, which concluded that it took females longer than males to be promoted at Bradford. These studies consisted of taking the average length of time it took both males and females to be promoted at Bradford to the positions of vice-president and broker representative. These averages revealed that it took males an average of 5.83 years to be promoted to vice-president whereas it took females an average of 12.33 years. To be promoted to the position of broker representative, it took males on average 2.38 years whereas it took females 4.75 years.

In his report and recommendation, the magistrate accepted these statistical studies and found that Scales had established a prima facie case of disparate impact discrimination. The magistrate noted that Bradford's statistical expert did not rebut or criticize these statistical studies. The magistrate also noted other evidence in the record to corroborate these statistics, such as: (1) Bradford did not advertise for their openings, (2) all of those selected for the promotions at issue were males who were personal friends or business associates of Bradford partners or supervisors, and (3) subjective criteria were used to select individuals to fill these vacancies.

The district court reversed this finding by the magistrate as clearly erroneous. The district court held that Scales had failed to establish a prima facie case of disparate impact discrimination, concluding her statistical studies were unreliable because they failed to account for any factors other than gender which may have influenced promotion times.

The Supreme Court has rejected rigid mathematical formulas in analyzing statistics purporting to show disparate impact. Watson, 487 U.S. at 994-95, 108 S.Ct. at 2788-89. "A plaintiff in a Title VII suit need not prove discrimination with scientific certainty; rather, his or her burden is to prove discrimination by a preponderance of the evidence." Bazemore, 478 U.S. at 400, 106 S.Ct. at 3009 (citing Texas Dept. of Community Affairs v. Burdine, 450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981)). Bazemore further instructs that, "the `entire evidence' in the record must be considered in determining whether a claim for discrimination has been proven." Id. 478 U.S. at 404, 106 S.Ct. at 3010.

Based upon our reading of the record, we find that Scales did introduce sufficient evidence to establish a prima facie case of disparate impact discrimination. We agree with the district court that the statistical study regarding the average length of time required to be promoted to Vice-President at Bradford is irrelevant for the purposes of this case. However, we do find that the statistical study concerning the average length of time it took to be promoted to broker representative, coupled with the other evidence in the record, was sufficient to establish a prima facie case. The fact that this statistical study did not include the employee's education or prior work experience does not make it irrelevant. Bazemore, 478 U.S. at 400 n. 10, 106 S.Ct. at 3009 n. 10. Throughout this litigation, Bradford has claimed that it bases its promotion decisions on highly subjective criteria. It cannot now claim that plaintiff's statistics are unreliable for not including factors that it never claimed to be important in the decision-making process. Bradford was free at all times to rebut these statistics with evidence of what factors it considered in making its promotion decisions. "`If the employer discerns fallacies or deficiencies in the data offered by the plaintiff, he is free to adduce countervailing evidence of his own.'" Watson, 487 U.S. at 996, 108 S.Ct. at 2789 (quoting Dothard v. Rawlinson, 433 U.S. 321, 331, 97 S.Ct. 2720, 2727, 53 L.Ed.2d 786 (1977)). Bradford has failed to do so in this case. We, therefore, reverse the decision of the district court with respect to Scale's disparate impact claim.

Scales next contends that the district court erred in denying her pre-judgment interest on her damage award. The award or denial of pre-judgment interest is within the sound discretion of the trial judge. EEOC v. Wooster Brush Co. Employees Ass'n., 727 F.2d 566, 578-79 (6th Cir. 1984). Because Scales did not request any prejudgment interest in her damage calculation, we find no abuse of discretion in this case.

Finally, Scales argues that the district court erred in not awarding attorney fees in the requested amount. Specifically, Scales' attorneys object to the district court reduction of their hourly rates and to the reduction of the lodestar amount by seventy-five percent. We review the district court's decision with respect to the amount of attorney fees to which a prevailing party is entitled for an abuse of discretion. Hensley v. Eckerhart, 461 U.S. 424, 440, 103 S.Ct. 1933, 1943, 76 L.Ed.2d 40 (1983); Allen v. Allied Plant Maintenance Co. of Tennessee, 881 F.2d 291 (6th Cir. 1989). This deference, "is appropriate in view of the district court's superior understanding of the litigation and the desirability of avoiding frequent appellate review of what essentially are factual matters." Hensley, 461 U.S. at 437, 103 S.Ct. at 1941.

In her application for attorney's fees, plaintiff's attorney, Ms. Behm, requested compensation at a rate of $140 per hour. An associate in Ms. Behm's firm who also worked on the case, Ms. Margaret Mazzone, requested compensation at $120 per hour. In support of these billing rates, plaintiff's attorney submitted three affidavits of Nashville civil rights attorneys, all of whom attested to the fact that $140 is the prevailing rate for this type of work in the Nashville community. The district court rejected these rates and instead utilized the hourly rates provided earlier by plaintiff's attorneys during the course of discovery. Based upon this information, the district court determined Ms. Behm's usual hourly rate to be $100 per hour and Ms. Mazzone's to be $75 per hour.

In arguing against the reduction of their hourly rates, plaintiff contends that the district court erred in not considering the prevailing rate in the community in determining their hourly rates. In Blum v. Stenson, 465 U.S. 886, 104 S.Ct. 1541, 79 L.Ed.2d 891 (1984) the Supreme Court determined that, "`reasonable fees' under § 1988 are to be calculated according to the prevailing market rates in the relevant community. . . ." Id. at 895, 104 S.Ct. at 1547. See also Kelley v. Metropolitan County Bd. of Education, 773 F.2d 677, 683 (6th Cir. 1985) (en banc). The Court recognized in Blum that, "determining an appropriate market rate for the services of a lawyer is inherently difficult" and that "[t]he type of services rendered by lawyers, as well as their experience, skill, and reputation, varies extensively — even within a law firm." Id. at 895 n. 11, 104 S.Ct. at 1547 n. 11. Using the "prevailing market rates" is a method to ensure that the proposed hourly rate requested in an application for fees is "reasonable." However, in this case the district court did not have to engage in this inquiry because he had previously been given the actual billing rates of plaintiff's attorneys. "The legislative history explains that `a reasonable attorney's fee' is one that is `adequate to attract competent counsel, but. . . . [that does] not produce windfalls to attorneys.'" Blum, 465 U.S. at 897, 104 S.Ct. at 1548 (citing S.Rep. No. 94-1011, p. 6 (1976)). We find that the district court did not abuse its discretion in relying upon the hourly rates provided by plaintiff's attorneys in determining their appropriate hourly rates. See Coulter v. Tennessee, 805 F.2d 146, 149 (6th Cir. 1986) (affirming district court's reduction of hourly rate to an amount which the attorney previously claimed in a prior application before the court).

The district court found that Scales had failed to prevail on four of her five significant claims and because of this limited success, reduced the lodestar amount by eighty percent. Because of our reversal of the district court's finding as to Scales' disparate impact claim, we must reevaluate this reduction.

In Hensley, the Court held that, "the extent of a plaintiff's success is a crucial factor in determining the proper amount of an award of attorney's fees. . . ." Hensley, 461 U.S. at 440, 103 S.Ct. at 1943. In this case, Scales claimed that Bradford violated Title VII by denying her promotions because of her gender and by retaliating against her for filing a gender discrimination charge with the EEOC. Scales prevailed on both of these claims. However, Scales also claimed constructive discharge and a violation of the Equal Pay Act. Scales was unsuccessful as to these claims. In light of these results, we find that the lodestar should be reduced by fifty percent.

The district court awarded an upward revision in the amount of five percent based upon the applicability of several of the factors discussed in Johnson v. Georgia Highway Express, Inc., 488 F.2d 714 (5th Cir. 1974). Scales contends that the district court abused its discretion in only awarding a five percent enhancement. However, Scales has failed to prove, at least to our satisfaction, that the district court failed to consider any relevant factors. See Kelley, 773 F.2d at 683.

Because we reject the district court's conclusion with respect to Scales' Title VII claim, we find that the amount originally awarded by the district court to be inadequate. We award attorney fees in the amount of fifty-five percent of the lodestar amount or $35,263.41 plus the costs of preparing the application, $2,745.00, for a total award of $38,008.41, with interest from the date of the original judgment.

Therefore, for the reasons stated, we affirm in part and reverse in part.

ORDER [33] March 19, 1991.

We have before us two motions submitted by the plaintiff, Deniece Scales: (1) a motion requesting the award of costs for her appeal under Rule 39(a) of the Rules of Appellate Procedure, and (2) a petition for rehearing in order to recalculate damages.

Both motions are denied. All these issues have been raised before and have been decided by this Court. We need not rehash them again.