Appeal from the United States District Court for the Northern District of Illinois, Eastern Division. No. 90 C 5020--Brian Barnett Duff, Judge.

John H. Hager, Elaine K.B. Siegel (argued), Hager & Siegel, Chicago, IL, for plaintiff-appellee.

Charles C. Jackson (argued), Brenda H. Feis, Ellen E. McLaughlin, Seyfarth, Shaw, Fairweather & Geraldson, Chicago, IL, Hubert O. Thompson, Deborah A. Richards, Carney & Brothers, Chicago, IL, for defendant-appellant.

Before WOOD, Jr., FLAUM, and MANION, Circuit Judges.

FLAUM, Circuit Judge.

Carmelo Melendez filed a complaint in district court, alleging the defendant, Illinois Bell Telephone Company ("Illinois Bell"), violated 42 U.S.C. 1981, Title VII and the Age Discrimination in Employment Act ("ADEA") by engaging in discriminatory hiring practices. The district court entered summary judgment against the plaintiff on his ADEA claim, but the § 1981 and Title VII claims, based on race discrimination, proceeded to trial. As a sanction for discovery abuses, the district court barred defendant's sole expert witness from testifying. The § 1981 disparate treatment claim was tried before a jury, which returned a general verdict for the defendant. After the § 1981 jury trial, the district court conducted a hearing on plaintiff's Title VII claims, which alleged both disparate treatment of and disparate impact upon Hispanics. In particular, Melendez challenged the discriminatory nature of a standardized management test, the BSAT. The district court held that the § 1981 jury verdict controlled the Title VII disparate treatment claim, but ruled for the plaintiff on his disparate impact claim. The defendant appeals the judgment in favor of the plaintiff on the Title VII disparate impact claim. We affirm.

I.

In September 1988, Melendez, a Hispanic male, applied for the position of Manager of Urban Affairs at Illinois Bell. This job was a first-level management position that entailed analysis and investigation of issues in the urban environment that might have impacted Illinois Bell's service to its Chicago-area Hispanic customers. The vacancy for which Melendez applied was created when the incumbent in that job, an Hispanic woman, was promoted to a higher level position.

At the time, Illinois Bell screened management candidates through a three-part process administered by its personnel department. The personnel department required candidates to complete a formal application for employment and to pass both a structured interview, the Management Selection Interview ("MSI") and a standardized test, the Basic Skills Abilities Test ("BSAT"). Normally, candidates could interview with the particular department at Illinois Bell that was hiring only after successfully completing all three requirements. To accommodate schedules, however, Illinois Bell would occasionally have the hiring department interview candidates before they completed the tripartite screening process.

On September 21, 1988, before being screened by the personnel department, Melendez interviewed with two managers from the urban affairs department, John McDermott and Suzette Broom. Although Melendez did not particularly impress McDermott and Broom during his interview, McDermott directed Melendez to the personnel department for further processing. In the personnel department Melendez completed an employment application and passed the MSI. Melendez also had graduated in the top half of his class from college, which was another one of Illinois Bell's objective qualifications for management positions. Melendez, however, failed the BSAT. Illinois Bell required applicants taking the BSAT to receive a score of at least 196 to be eligible for employment.1 Melendez's score was 157. On September 25, 1988, Illinois Bell's personnel department advised Melendez that he had failed the BSAT and therefore was not qualified for the manager of urban affairs position. He was informed that he could retake the BSAT in six months. McDermott thereafter selected Henry Lara, an internal candidate who was also Hispanic, to become manager of urban affairs.

The BSAT is a standardized cognitive ability test, which Illinois Bell used to measure the learning potential of candidates for management positions. AT & T psychologists developed the BSAT in 1979 for telephone companies nationally. Local telephone companies, including Illinois Bell, implemented the BSAT in the early 1980's. The examination is "speeded," consisting of 100 multiple choice questions during a one hour time period. The BSAT tests four different skills: mathematics, grammar, reading comprehension, and following directions. The parties stipulated that the BSAT has a disparate impact against Hispanics.2

To support the fairness and validity of the BSAT, AT & T psychologists conducted a validation study, which reported various statistical analyses of how the BSAT related to job performance. The validation study correlated test scores with actual job performance ratings and directly examined the BSAT by race for its ability to predict job performance. To review this validity evidence, Melendez introduced the expert testimony of Dr. Fred Bryant, Professor of Psychology at Loyola University of Chicago. Dr. Bryant received a Ph.D. in Social Psychology from Northwestern University and has numerous publications involving the statistical analysis of data. Dr. Bryant testified that, in the context of employment examinations, an examination is "valid" if it can predict how people will perform on the job. Dr. Bryant opined that the BSAT is very poor at predicting job performance and therefore lacks validity. In Dr. Bryant's opinion, the statistical evidence from the validation study convincingly showed that BSAT scores and job performance are unrelated for Hispanics. Moreover, Dr. Bryant concluded that the great weight of statistical evidence from the validation study demonstrated that there is no relationship between BSAT scores and job performance for whites. Dr. Bryant likewise found little evidence that the BSAT predicted job performance for African-Americans. Dr. Bryant also attacked the methodology of the validation study, asserting that the study improperly pooled data from different racial subgroups, which caused inflated validity data for the overall group. Even if one uses the inflated overall data, the BSAT predicts job performance only 3 percent better than chance alone. Indeed, prior to this litigation, psychologists hired by Illinois Bell to review its management hiring process concluded that "there is little or no support for the validity of BSAT scores in predicting the core areas of management performance...." Importantly, the data from the validation study illustrates that, despite a statistically significant difference in test scores between whites and Hispanics, there is no significant difference in job performance between the two groups.

In December 1988, Melendez filed a charge with the EEOC, alleging that the BSAT had an unlawful disparate impact on Hispanics. On August 28, 1990, Melendez filed the current action in district court. While his charge was pending before the EEOC, Illinois Bell formed a consortium with other local telephone companies to revise or replace the BSAT. Illinois Bell's representative to the consortium was Dr. Gary Morris.3 On July 11, 1990, Psychological Services, Inc. ("PSI"), a consulting firm, sent a proposal to Dr. Morris and other consortium representatives, which indicated the necessity of establishing a new test with "face validity." The consortium retained PSI to develop a replacement test, which was labelled the BSAT-Replacement ("BSAT-R").

Dr. Morris was personally involved with the development of the BSAT-R. On January 11, 1991 he wrote the following to various Illinois Bell department heads:

I am writing this letter to inform you that the [ ] Consortium to revise the BSAT is finally getting near the point of completing a contract and starting work. A primary goal of this project is to develop a test with item content that is much more job related than was the case with BSAT....

I will be sure to keep you posted on the status of the project as it progresses and on more details related to your involvement as soon as possible.

The contract for PSI's consulting services was executed on March 11, 1991 and identified Dr. Morris as a recipient of contract notices. In an April 4, 1991 letter, Dr. Morris again wrote to various department heads: "As you are all aware, we are involved in a consortium effort to build a replacement for the BSAT. The time frame for its introduction is mid-year 1992." The BSAT-R project included examining the validity evidence for the BSAT as well as collecting additional data relevant to the validity of the BSAT.

While the replacement test for the BSAT was being formulated, Melendez attempted to obtain information relevant to the BSAT's validity through discovery. On November 6, 1990, Melendez requested the production of documents "that constitute, refer or relate to the decision to use the BSAT, or to discussions or analyses concerning its use." Although the trial court ordered Illinois Bell to respond to this request by December 21, 1990, Illinois Bell produced no documents until January 14, 1991. The BSAT-R project was not disclosed, and Illinois Bell failed to produce many BSAT-related documents. When Melendez moved to compel production of these documents, Illinois Bell represented to the district court that production of the BSAT would cause irreparable harm because there was only one form.4 On March 15, 1991, upon discovering that there were at least two different versions of the BSAT, the district court imposed Rule 11 sanctions on Illinois Bell and ordered Illinois Bell to produce various BSAT-related documents. Illinois Bell produced additional BSAT-related materials on May 17, 1991, but did not reveal the BSAT-R project.

Illinois Bell intended to call Dr. Morris at trial as an expert witness to testify that the BSAT is a valid employment examination. Testifying for Illinois Bell as an expert witness was one of Dr. Morris's duties in his position at Ameritech Services. Melendez deposed Dr. Morris numerous times prior to trial. Despite specific deposition questions directed at any investigations of the BSAT, Dr. Morris failed to disclose the existence of the BSAT-R project. On November 7, 1991, Melendez subpoenaed Ameritech Services, requesting, among other records, "[a]ll documents, including drafts, that constitute, refer to or relate to validation studies in which the [BSAT] was reviewed and considered in any way." After Illinois Bell's motion to quash this subpoena was denied, Ameritech Services produced voluminous documents unrelated to the BSAT-R.

The § 1981 claim was originally scheduled for a jury trial beginning on September 13, 1993. The BSAT-R was implemented by Illinois Bell during the first week of September 1993. On September 9, two working days before trial, plaintiff sought a stipulation on the continued use of the BSAT. In response, counsel for Illinois Bell finally revealed the BSAT-R's existence and then produced the BSAT-R materials, after Melendez moved to compel their production. Melendez thereafter filed a motion for sanctions, alleging that Illinois Bell, along with its trial counsel and expert witness, intentionally concealed the BSAT-R and failed to disclose the test in response to repeated discovery requests. The district court postponed the trial and conducted a three-day hearing to determine the extent to which Illinois Bell, its trial counsel, and Dr. Morris were aware of the BSAT-R. At this hearing, Dr. Morris testified that he informed Illinois Bell's trial counsel about the development of a replacement test for the BSAT soon after the current litigation began. Dr. Morris also testified that he periodically advised Illinois Bell's trial counsel regarding the progress of the BSAT-R and the time frame for its implementation. After eliciting a great deal of testimony at this hearing, the district court concluded that Illinois Bell, its counsel, and Dr. Morris "knew or should have known" that the BSAT-R was relevant and responsive to Melendez's discovery requests. As a sanction under Fed.R.Civ.P. 37, the district court barred Illinois Bell's sole expert witness, Dr. Morris, from testifying at trial.

The § 1981 jury trial commenced immediately after the sanctions hearing. Plaintiff based his theory of intentional discrimination on Illinois Bell's continued use of the BSAT despite Illinois Bell's knowledge that the test was invalid. At the close of evidence, the district court instructed the jury that for Melendez to prevail, he must prove that the defendant intentionally discriminated against him. Tailoring the instructions to plaintiff's theory of the case, the district court informed the jury that it could "infer from Illinois Bell's knowing use of a seemingly neutral employment test that has a disparate impact against a minority group that Illinois Bell's discrimination was intentional...." Following deliberation, the jury returned a general verdict in favor of Illinois Bell on the § 1981 claim.

After the jury trial, the district court considered the merits of plaintiff's Title VII claims, which alleged both disparate treatment and disparate impact against Hispanics.5 Holding that it was bound by any factual issues decided by the jury, and that both § 1981 claims and Title VII disparate treatment claims require proof of intentional discrimination, the district court ruled that the jury verdict controlled plaintiff's Title VII disparate treatment claim. The court also ruled, however, that the jury verdict had no preclusive effect with respect to plaintiff's disparate impact claim. After determining that it had already garnered all of the relevant evidence during the jury trial, the court heard oral argument on the disparate impact claim. Following this hearing, the district court entered findings of fact and conclusions of law, holding that Illinois Bell violated Title VII through its use of the BSAT.

II.

On appeal, defendant presents four main grounds for reversing the district court's judgment for plaintiff on the Title VII disparate impact claim. With respect to the merits of the disparate impact claim, defendant contends first that plaintiff failed to establish a prima facie case of disparate impact because he was unqualified for a management position. Second, defendant argues that the district court's finding that defendant had no legitimate business reason to use the BSAT is clearly erroneous. Next, defendant asserts that the jury verdict on the § 1981 claim estopped the district court from ruling for the plaintiff on the Title VII claim. Finally, defendant maintains that the district court abused its discretion by inappropriately sanctioning Illinois Bell for its alleged discovery abuses. We address each of these arguments in turn.

A.

Illinois Bell argues that a single plaintiff, such as Melendez, must prove as part of his prima facie case of disparate impact that he was qualified for the position he sought. Technically, after trial, the issue of prima facie case in the specialized Title VII burden-shifting sense becomes moot. Allen v. Seidman, 881 F.2d 375, 379 (7th Cir.1989). Once the lawsuit has been tried, the relevant inquiry simplifies to whether the evidence presented at trial supports a finding of a Title VII violation. Id. In a disparate impact case, the plaintiff must prove that the challenged practice is discriminatory because it has a disparate impact unjustified by the defendant's legitimate business needs. Id. Yet we cannot dismiss the defendant's challenge so quickly, for this challenge raises concerns broader then the establishment of a prima facie case. In order for an individual plaintiff to have constitutional standing to bring a Title VII action, he must show that he was personally injured by the defendant's alleged discrimination and that his injury will likely be redressed by the requested relief. See Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-61, 112 S.Ct. 2130, 2136-37, 119 L.Ed.2d 351 (1992); Allen v. Wright, 468 U.S. 737, 751, 104 S.Ct. 3315, 3324-25, 82 L.Ed.2d 556 (1984); see also Carpenter v. Board of Regents of Univ. of Wis. Sys., 728 F.2d 911, 915 (7th Cir.1984) (holding that plaintiff in disparate impact case must show injury resulted from policy alleged to have disparate impact). We have relied on these standing principles to require an individual Title VII plaintiff alleging disparate impact to establish that he was qualified for the position sought. See Gilty v. Village of Oak Park, 919 F.2d 1247, 1255 (7th Cir.1990) (reasoning that plaintiff "must establish that his individual circumstances entitle him to ... relief.").6 The basis for this qualification requirement is apparent. Absent direct evidence showing that a plaintiff was not hired or promoted because of a discriminatory employment practice, we assume that an unqualified plaintiff was not hired or promoted for the obvious reason-that he was unqualified. Such a plaintiff would have no standing to sue under Title VII, for he could not claim that he was injured, much less affected, by the defendant's use of an employment practice with an allegedly disparate impact. See Carpenter, 728 F.2d at 915 (holding that plaintiff's Title VII claim must fail if promotion was denied for reason unrelated to defendant's allegedly discriminatory test). In contrast, where a plaintiff demonstrates that he was not hired or promoted as the direct result of a discriminatory hiring practice, he has suffered a personal injury within the meaning of Title VII. To hold otherwise would unfairly narrow the language of Title VII, which makes it unlawful for an employer "to limit ... applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities ... because of such individual's race...." 42 U.S.C. § 2000e-2(a)(2) (emphasis added); see Connecticut v. Teal, 457 U.S. 440, 448, 102 S.Ct. 2525, 2531, 73 L.Ed.2d 130 (1982) (interpreting "opportunities" language to hold that Title VII prohibits employment practices having disparate impact despite employer's racially balanced "bottom line").

In the present case, Melendez presented extensive direct evidence that he was not hired because he failed the BSAT. Illinois Bell's personnel department informed Melendez that his failing score on the BSAT disqualified him from further consideration. Illinois Bell also suggested to Melendez that he could retake the BSAT in six months. Suzette Broom testified that once she and John McDermott discovered plaintiff had failed the BSAT, there was no need to pursue employment possibilities with Melendez. Although McDermott testified that he would not have hired Melendez even if he had passed the BSAT, the district court found McDermott's version of events to be an ex post facto justification for Illinois Bell's decision.7 After reviewing the entire record, we must agree with the district court's assessment. Because plaintiff has sufficiently demonstrated that he was not hired due to his failure of the BSAT, we need not examine whether plaintiff was qualified for a management position.8

Illinois Bell next contends that the district court erred in finding that the BSAT's discriminatory impact was unjustified by Illinois Bell's legitimate business needs, a finding which we review for clear error only. Daniels v. Pipefitters' Ass'n Local Union No. 597, 945 F.2d 906, 924 (7th Cir.1991), cert. denied, 503 U.S. 951 , 112 S.Ct. 1514, 117 L.Ed.2d 651 (1992). Contrary to Illinois Bell's assertions, there is ample evidence to support the district court's finding. Plaintiff's expert,9 after examining the validity evidence for the BSAT, testified that there is no significant correlation between an applicant's BSAT score and his or her ability to perform the duties of an entry-level manager. He also testified that even if one uses inflated correlations caused by grouping the entire test-taking sample, the BSAT can predict a person's job performance only 3 percent better than chance alone. In fact, Illinois Bell's hired psychologists observed that "there is little or no support for the validity of BSAT scores in predicting the core areas of management performance." Given the strong evidence of the BSAT's inability to predict job performance, the district court's finding that Illinois Bell had no legitimate business reason for using the BSAT was not clearly erroneous.

B.

Illinois Bell further asserts that even if plaintiff presented sufficient evidence to support his Title VII claim, the jury's § 1981 verdict estopped the district court from finding a violation of Title VII. Indeed, we have consistently held that a "jury's verdict, when § 1981 and Title VII claims are tried simultaneously, binds the judge on factual issues common to both claims." Artis v. Hitachi Zosen Clearing, Inc., 967 F.2d 1132, 1137 (7th Cir.1992) (quoting Daniels, 945 F.2d at 923); see also Dombeck v. Milwaukee Valve Co., 40 F.3d 230, 236 (7th Cir.1994); Snider v. Consolidation Coal Co., 973 F.2d 555, 559 (7th Cir.1992), cert. denied, 506 U.S. 1054 , 113 S.Ct. 981, 122 L.Ed.2d 134 (1993); McKnight v. General Motors Corp., 908 F.2d 104, 113 (7th Cir.1990), cert. denied, 499 U.S. 919 , 111 S.Ct. 1306, 113 L.Ed.2d 241 (1991).10 The substantive standards governing liability for § 1981 claims and Title VII disparate treatment claims are identical. See, e.g., North v. Madison Area Ass'n for Retarded Citizens, 844 F.2d 401, 408 (7th Cir.1988). To prevail under 42 U.S.C. 1981, a plaintiff ultimately must prove that he has been a victim of intentional discrimination. Randle v. LaSalle Telecommunications, Inc., 876 F.2d 563, 569 (7th Cir.1989). To prevail on a disparate treatment claim under Title VII, a plaintiff must meet the identical element of proof: that the defendant has intentionally discriminated against him. Hong v. Children's Memorial Hosp., 993 F.2d 1257, 1261 (7th Cir.1993), cert. denied, --- U.S. ----, 114 S.Ct. 1372, 128 L.Ed.2d 48 (1994). This identical requirement in proving § 1981 and Title VII disparate treatment claims has led our court to conclude that the trial judge, when ruling on a Title VII disparate treatment claim, must adhere to a jury's prior § 1981 general verdict. See Artis, 967 F.2d at 1137-38; Daniels, 945 F.2d at 923. Since both § 1981 violations and Title VII violations under the disparate treatment model necessarily hinge on a factual finding of intentional discrimination, and the jury's factual findings bind the judge, the judge's Title VII decision must follow the jury's § 1981 general verdict.11

On the other hand, the elements of proof for disparate treatment and disparate impact claims under Title VII are significantly different. A plaintiff may demonstrate a violation of Title VII under the disparate impact theory without proving discriminatory intent. Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 645-46, 109 S.Ct. 2115, 2118-19, 104 L.Ed.2d 733 (1989); Gilty v. Village of Oak Park, 919 F.2d 1247, 1254 (7th Cir.1990). Thus, at first glance, the § 1981 jury verdict in Illinois Bell's favor, from which a finding of no intentional discrimination can be extracted, does not appear to dictate the result of Melendez's Title VII disparate impact claim. In fact, we recognized as much in Daniels, where, despite a prior § 1981 jury verdict for the plaintiff, we implicitly held that it was necessary for the district court to make additional findings to support a disparate impact violation. 945 F.2d at 923. Illinois Bell attempts to avoid this caselaw by observing that, in this case, Melendez tried to prove his § 1981 claim by demonstrating the invalidity of the BSAT. In essence, Illinois Bell wishes to read the jury's general verdict as including a special verdict finding that the BSAT is valid. Concededly, Melendez did base his theory of intentional discrimination on Illinois Bell's knowing and continued use of an allegedly invalid test. The jury instructions stated that the jury could infer intentional discrimination under such a set of facts. The instructions, however, properly noted that for the plaintiff to prevail, he must have proven that Illinois Bell intentionally discriminated against him through its use of the BSAT. The jury verdict for the defendant, therefore, was necessarily based on a finding of no intentional discrimination. In contrast, a finding of BSAT validity was not necessary for the jury to reach its verdict. The jury could have believed the BSAT was invalid, yet also found that Illinois Bell had no discriminatory intent in using it. The general verdict of a jury can only bind the judge with respect to issues that were essential to the jury's decision. See Gentry v. Duckworth, 65 F.3d 555, 560 (7th Cir.1995) (holding that party claiming issue preclusion must prove resolution of issue was essential to prior judgment). We see no reason to pluck from the verdict speculative findings not necessarily reached by the jury. The § 1981 verdict therefore did not preclude the judge from ruling for plaintiff on his Title VII disparate impact claim.

C.

The defendant argues that the district court abused its discretion by barring Dr. Morris from testifying under Fed.R.Civ.P. 37(b)(2)(B).12 As we have often stated, Rule 37(b) sanctions provide the district court with an effective means of ensuring that litigants will timely comply with discovery orders. See, e.g., Parker v. Freightliner Corp., 940 F.2d 1019, 1024 (7th Cir.1991). Both the decision to sanction and the choice of an appropriate sanction are reviewed for an abuse of discretion. Godlove v. Bamberger, Foreman, Oswald & Hahn, 903 F.2d 1145, 1148 (7th Cir.1990), cert. denied, 499 U.S. 913 , 111 S.Ct. 1123, 113 L.Ed.2d 230 (1991). Because the district court is in the best position to determine if a party has complied with its discovery orders, we will not reverse the imposition of sanctions unless "it is clear that no reasonable person could concur in the trial court's assessment...." Patterson v. Coca-Cola Bottling Co., 852 F.2d 280, 283 (7th Cir.1988). Similarly, the district court has primary responsibility for selecting an appropriate sanction, and we will not reverse the district court's selection "absent a clear abuse of discretion." Id. We review the factual findings underlying the district court's imposition of sanctions under the deferential clearly erroneous standard. Govas v. Chalmers, 965 F.2d 298, 301 (7th Cir.1992).

Illinois Bell argues that, at most, it made an honest mistake when it failed to produce the BSAT-R, and therefore the district court should not have visited sanctions upon it. Illinois Bell asserts that the district court erred by ascribing bad faith motives to it. Bad faith, however, is not required for a district court to sanction a party for discovery abuses. Sanctions are proper upon a finding of wilfulness, bad faith, or fault on the part of the noncomplying litigant. Marrocco v. General Motors Corp., 966 F.2d 220, 224 (7th Cir.1992); Patterson, 852 F.2d at 283. The district court found that Illinois Bell, through its trial counsel and Dr. Morris, knew or should have known that disclosure of the BSAT-R was required by the court's discovery orders. We interpret this finding to mean that Illinois Bell acted in bad faith (i.e., that Illinois Bell knew disclosure was required) or, alternatively, was at fault (i.e., that Illinois Bell should have known disclosure was required). After reviewing the facts, we cannot deem the district court's finding to be clearly erroneous.

The testimony at the sanctions hearing demonstrates that Illinois Bell's trial counsel was aware of the BSAT-R project's existence. Dr. Morris was Illinois Bell's representative to the consortium, which was formed to develop a replacement for the BSAT. At the beginning of this litigation, Dr. Morris informed Illinois Bell's trial counsel that such a project was underway. Dr. Morris also periodically updated Illinois Bell's counsel regarding the time frame for the completion of the BSAT-R. Moreover, the plaintiff's discovery requests clearly encompassed the BSAT-R. The BSAT-R project involved a collection of data relevant to the validity of the original BSAT, and, almost by necessity, entailed a reexamination of the previously assembled validity evidence for the BSAT. Given the trial's focus on the invalidity of the BSAT, the BSAT-R project was patently relevant to plaintiff's § 1981 and Title VII claims. On November 6, 1990, Melendez requested that Illinois Bell produce any documents that related "to the decision to use the BSAT, or to discussions or analyses concerning its use."13 On November 7, 1991, Melendez subpoenaed Ameritech Services, seeking documents that referred to or related to "validation studies in which the [BSAT] was reviewed or considered in any way." (emphasis added) Additionally, Dr. Morris avoided mentioning the BSAT-R project at numerous depositions, where he was repeatedly asked about investigations into the validity of the BSAT. Given these simple inquiries, Illinois Bell's non-responsive conduct was unreasonable and easily falls within the concept of "fault." Moreover, Illinois Bell had previously been sanctioned for attempting to mislead the district court by claiming that only one form of the BSAT existed. In light of Illinois Bell's earlier evasive discovery tactics, there is ample evidence on the record from which to infer bad faith in Illinois Bell's non-disclosure of the BSAT-R project. We find it difficult to conceive that Illinois Bell reasonably and in good faith believed that the BSAT-R project was not covered by plaintiff's straightforward discovery requests.

Defendant next contends that the district court's choice of sanctions was draconian and an abuse of discretion. Illinois Bell argues that the district court could not have formulated a more severe sanction than barring the expert testimony of Dr. Morris, which Illinois Bell equates with the entry of default judgment against it. Illinois Bell then attempts to rely on caselaw indicating that a district court's discretion under Rule 37 is more limited when it dismisses a case or enters a default judgment. See Marrocco v. General Motors Corp., 966 F.2d 220, 223-24 (7th Cir.1992); Crown Life Ins. v. Craig, 995 F.2d 1376, 1381 (7th Cir.1993). This court has recognized that a de facto default judgment can occur when "a district court prevents a defendant from presenting any evidence whatsoever on a claim." Id. In this case, however, the district court did not prevent Illinois Bell from presenting evidence other than Dr. Morris' testimony. Although the district court's sanction was indeed severe, the result on the Title VII claim was not predetermined once Dr. Morris could no longer testify.14

In our review of the propriety of Rule 37 sanctions, we do not require district courts to select the "least drastic" or "most reasonable" sanction. See Marrocco, 966 F.2d at 225; Newman v. Metropolitan Pier & Exposition Auth., 962 F.2d 589, 591 (7th Cir.1992). District courts, however, may only impose sanctions that are "just," that is, proportionate to the circumstances surrounding a party's failure to comply with discovery rules. Crown Life Ins., 995 F.2d at 1382; Newman, 962 F.2d at 591. In light of the entire record in this case, the district court's choice of sanctions was just and therefore did not constitute an abuse of discretion. We acknowledge the severity of the sanctions that the district court visited upon Illinois Bell. Given the extensive morass of relatively indecipherable statistical data from the validation study, Dr. Morris' testimony would have been crucial to demonstrating the BSAT's validity. Without Dr. Morris' testimony, Illinois Bell could only rely on the validation study data itself to support the validity of the test. The validity of the BSAT is extremely probative of its "business necessity" and thus was critical to plaintiff's Title VII claim. Additionally, the validity of the BSAT was central to plaintiff's theory of intentional discrimination on his § 1981 claim. Certainly, district courts should be cautious before barring the introduction of such extremely relevant testimony into evidence.

Ironically, the importance of the BSAT's validity to plaintiff's case is also the justification for the severity of the district court's sanction. Because the BSAT-R project involved an examination of evidence relevant to the validity of the BSAT, plaintiff was greatly prejudiced by Illinois Bell's non-disclosure of the BSAT-R project until the eve of trial. Moreover, the district court was under no obligation to postpone the trial to allow plaintiff to wade through the quagmire of statistical data that was finally produced. Given Illinois Bell's earlier failure to produce the BSAT and the testimony indicating Dr. Morris' complicity in Illinois Bell's wrongful failure to disclose the BSAT-R project, the district court did not abuse its discretion in barring Dr. Morris from testifying at trial. Able to reflect on the procedural history of the case as well as the prejudice suffered by plaintiff, the district court was in the best position to select a fitting sanction. Although we might not have chosen so severe a sanction, Illinois Bell has, quite simply, not met its demanding burden of demonstrating that no reasonable person could agree with the district court's assessment of the appropriate sanction.

For all the foregoing reasons, the judgment of the district court is AFFIRMED.

* The Honorable Walter J. Cummings, Circuit Judge did not participate in the consideration of this petition for rehearing

1 Illinois Bell did not require internal candidates for first level management positions to take the BSAT

2 In fact, Illinois Bell statistics show that in 1979, 77 percent of whites, but only 22 percent of African-Americans and 47 percent of Hispanics passed the test. Similarly, 1987-88 statistics reflect a pass rate of 87 percent for whites, but only 57 percent for African-Americans and 53 percent for Hispanics. Illinois Bell presented no evidence that these statistical disparities could be explained by educational differences. With respect to the 1979 sample, 97 percent of those tested had college degrees. Melendez's expert also testified that the 1987-88 sample included people of similar background and education

3 Dr. Morris was actually employed by Ameritech Services, Inc., a subsidiary of Illinois Bell. He was Director of Selecting Standards and Testing and was responsible for the personnel selection processes for Illinois Bell and other local telephone companies

4 Illinois Bell professed their need to avoid dissemination of particular test questions

5 Because both the conduct at issue and the filing of the complaint occurred prior to the passage of the Civil Rights Act of 1991, the Title VII claims could not be tried before the jury. Landgraf v. USI Film Products, --- U.S. ----, 114 S.Ct. 1483, 128 L.Ed.2d 229 (1994). In addition, pre-1991 Civil Rights Act standards apply on this appeal

6 Other circuits have imposed this same qualification requirement on individual plaintiffs who allege a violation of Title VII under a disparate impact theory. See Coe v. Yellow Freight Sys., Inc., 646 F.2d 444, 451 (10th Cir.1981) (relying on constitutional standing doctrine); Robinson v. Polaroid Corp., 732 F.2d 1010, 1016 (1st Cir.1984) (following Coe)

7 The district court questioned why McDermott would have steered Melendez to the personnel department for a further interview and test if McDermott had already decided not to hire him

8 In any event, we could not deem the district court's finding that Melendez was qualified to be clear error. Other than failing the BSAT, Melendez satisfied all of the objective requirements for a management position-passing the MSI and graduating in the top half of his class from college

9 Illinois Bell cursorily argues that the district court abused its discretion by relying on the testimony of Dr. Bryant. Illinois Bell maintains that Dr. Bryant was not qualified to testify as an expert because he had never previously validated an employment test. Dr. Bryant, however, has a doctorate in social psychology and has published in the area of statistical analysis. Thus, "[h]is expertise came from experience, education, and training; all acceptable bases of expertise under Rule 702." United States v. Stevenson, 6 F.3d 1262, 1267 (7th Cir.1993). The district court did not abuse its discretion in qualifying Dr. Bryant as an expert witness

10 Although the rule itself is clear, the source of the rule is not completely apparent. Compare Artis, 967 F.2d at 1137 (basing rule on 7th Amendment) with Snider, 973 F.2d at 559 (grounding rule in collateral estoppel and not 7th Amendment). Our analysis here is not dependent on the rule's origin

11 Concretely, a § 1981 jury verdict for the defendant translates to a finding of no intentional discrimination, which dictates the failure of a Title VII disparate treatment claim. Analogously, a § 1981 jury verdict for the plaintiff translates to a finding of intentional discrimination, which establishes disparate treatment under Title VII

12 This rule provides that if a party fails to obey an order to permit discovery, the district court may sanction the party by, among other alternatives, prohibiting the introduction of certain matters into evidence. Fed.R.Civ. P. 37(b)(2)(B). Other portions of Rule 37 grant the district court discretion to impose similar sanctions for various types of discovery abuses, including failure to produce documents and failure to supplement prior discovery responses. The defendant, quite understandably, neglects to argue that its actions are not encompassed by Rule 37

13 Although this request for documents may have preceded any significant progress on the BSAT-R project, Fed.R.Civ. P. 26(e)(2) requires parties to supplement prior responses that are discovered to be incomplete due to later events

14 In the context of the other arguments that Illinois Bell has presented on appeal, we find its insistence that the district court's sanction was tantamount to the entry of default judgment somewhat disingenuous. The jury returned a verdict in favor of Illinois Bell on the § 1981 claim, which dictates, in Illinois Bell's view, a decision in its favor under Title VII. Illinois Bell has also argued that the evidence at trial (including evidence that Illinois Bell presented) cannot support the Title VII verdict in the plaintiff's favor. Illinois Bell's contention that the district court's sanction was truly equivalent to a default judgment is factually inconsistent with its alternative arguments that the evidence presented at trial required the district court to rule in its favor