Therese Ballet Lynn was an assistant professor at the University of California at Irvine. She was denied merit salary increases and tenure. Lynn filed suit under Title VII of the Civil Rights Act of 1964, 42 U.S.C. Section 2000e, et seq., alleging sex discrimination. Following trial, the district court entered judgment for the University. In her complaint Lynn also alleged discrimination on the basis of national origin; however, she has not pursued that claim.

Lynn was first employed by the University as a lecturer in 1969. She was promoted to assistant professor in 1971, and held that rank until leaving the University. She was denied a merit salary increase in 1971, after extramural evaluators judged her scholarship to be deficient. One evaluator was selected by the University, and one by Lynn. She was warned of deficient scholarship after internal reviews of extramural evaluations in both academic years 1972-1973 and 1973-1974. The latter evaluation was Lynn's mid-career review, pursuant to the University's policy that tenure be achieved within eight years or termination will result. She was given the opportunity to discuss her problems with members of her department, and during the 1974-1975 academic year was granted a sabbatical "for the explicit purpose of improving her scholarly research." Tenure review began in 1975 and culminated in the official denial of tenure by the University in June 1976. Lynn accepted the University's offer of a terminal one year appointment to June 1977.

The district court described the University's tenure review process as follows:

The University has a tenure review system for promotion in which the excellence of a candidate's research, teaching, University and public service are judged. In order to receive candid evaluations from peers the reports of the various levels of review are confidential, as are the identities of an ad hoc review committee and the identities of extramural evaluators selected by the University. There are five levels of review in the tenure review process; each is independent of the other. The levels of review include that of the individual department of which the tenure candidate is a member, the dean of the school of which the department is a part, an ad hoc review committee, a review by the Budget Committee of the Academic Senate, review by the Office of Academic Affairs, and finally, review and decision by the Chancellor of the University of California, Irvine.

Lynn provided documentation and named extramural scholars for purposes of her tenure review; others were selected by her department chairman. The district court found, after in camera review of the tenure review files that "(h)er work was favorably commented upon by a majority of these scholars." Nevertheless, denial of tenure was recommended at each level of the review process. After the denial, reconsideration was denied by the Budget Committee. A claim by Lynn, alleging discriminatory treatment, was then rejected by a special committee, after an investigation.

Lynn claims that she received disparate treatment from the University in its promotion and tenure decisions. The standards generally applicable to claims of disparate treatment under Title VII were laid down in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 800, 93 S. Ct. 1817, 1823, 36 L. Ed. 2d 668 (1973). See also Furnco Const. Corp. v. Waters, 438 U.S. 567, 572-73, 98 S. Ct. 2943, 2947, 57 L. Ed. 2d 957 (1978). In McDonnell Douglas the Supreme Court set forth the four elements necessary to establish a prima facie case of race discrimination under Title VII; the plaintiff must show:

1. that he belongs to a racial minority;

2. that he applied and was qualified for a job for which the employer was seeking applicants;

3. that, despite his qualifications, he was rejected; and

4. that, after his rejection, the position remained open and the employer continued to seek applications from persons of complainant's qualifications.

411 U.S. at 802, 93 S. Ct. at 1824 (footnote omitted). The Court cautioned that these four elements might not be necessarily applicable in every respect in all Title VII cases. Id. at 802 n.13, 93 S. Ct. at 1824.

The courts of appeals have consistently approved the application of the McDonnell Douglas test to charges of discrimination in the academic context.1 The Fourth Circuit adopted the McDonnell Douglas elements and applied them to sex discrimination in the academic context. Smith v. University of North Carolina, 632 F.2d 316 (4th Cir. 1980). In Smith, the Fourth Circuit described the elements of a prima facie case of Title VII discrimination by an institution of higher learning; it said that the plaintiff must show:

1. that she is a member of a class protected by Title VII;

2. that she was qualified for the position or rank sought;

3. that she was denied promotion or reappointment;

4. that in cases of reappointment (or tenure) others (i. e., males) with similar qualifications achieved the rank or position.

632 F.2d at 340. We agree that these are the applicable elements in such cases.2

The test set forth above is not an exclusive one. Plaintiffs may under some circumstances establish a prima facie showing of unlawful discrimination in Title VII cases without satisfying the four specific elements of McDonnell Douglas or Smith. The Supreme Court has expressly rejected the argument that meeting the specific McDonnell Douglas requirements is the "only means" by which plaintiffs may make the requisite prima facie showing. Teamsters v. United States, 431 U.S. 324, 358, 97 S. Ct. 1843, 1866, 52 L. Ed. 2d 396 (1977). The key, the Court said, is simply that the "plaintiff must carry the initial burden of offering evidence adequate to create an inference that an employment decision was based on a discriminatory criterion illegal under the Act." Id. The Court subsequently explained that the plaintiff's initial burden is met where the plaintiff has shown that "it is more likely than not" that the employer's actions were based on unlawful considerations. Furnco Construction Corp. v. Waters, 438 U.S. 567, 576, 98 S. Ct. 2943, 2949, 57 L. Ed. 2d 957 (1978). Thus, in Title VII cases involving institutions of higher learning, plaintiffs may establish their prima facie case by a sufficient showing as to the four Smith elements or by offering other evidence which creates the inference that the complained of act was unlawful.

In McDonnell Douglas, the Court also established certain procedures governing the trial of Title VII cases. The "order and allocation of proof" are as follows: (1) the plaintiff must come forward with evidence sufficient to constitute a prima facie case of discrimination; (2) the defendant must then "articulate" a legitimate non-discriminatory reason for the adverse employment decision; and (3) the plaintiff must then be given the opportunity to show that the "assigned reason" was "a pretext or discriminatory in its application." McDonnell Douglas, 411 U.S. at 807, 93 S.Ct. at 1827.

The district court held, without discussing the four elements that establish a prima facie case under McDonnell Douglas and Smith, that Lynn failed to make the required showing. In the alternative, the district court held that the University "articulated legitimate and non-discriminatory reasons for not granting Lynn tenure." We assume, for purposes of our analysis, that the district court implicitly found that Lynn failed to show that the University's articulated reasons were "a pretext or discriminatory in (their) application."

The district court clearly erred in concluding that Lynn failed to establish a prima facie case of discrimination based on sex. "The burden of establishing a prima facie case is not onerous." Texas Dept. of Community Affairs v. Burdine, 450 U.S. 248, 253, 101 S. Ct. 1089, 1094, 67 L. Ed. 2d 207 (1981). Lynn obviously satisfied elements (1) and (3) of the McDonnell Douglas test as applied by Smith in the academic context. She also adduced sufficient evidence as to elements (2) and (4) to satisfy her initial burden. In addition to testimony and documentary evidence, Lynn submitted two types of statistical data; the first, which we refer to as "specific statistical data," tended to show that she met the objective criteria for tenure, and the second, which we refer to as "general statistical data," tended to show a general pattern of discrimination by the University in favor of men.

Lynn's specific statistical data, relating to the objective criteria for tenure, provided evidence that she had the same education, experience and number of published works as others who had been granted tenure. For purposes of deciding whether Lynn has established a prima facie case we find the specific statistical data submitted to be highly persuasive. It supports Lynn's contentions that she "was qualified for the position or rank sought," i. e., element (2), and that "others (i. e., males) with similar qualifications achieved the rank or position," i. e., element (4).

The general statistical data submitted by Lynn, as mentioned, provides evidence of a pattern of academic sex discrimination by the University. The district court described the University's past practices as follows:

Over the years University administrators have shown a lack of concern for the need for minority and female faculty members, and such indifference persists. Even though the University of California at Irvine maintains an Affirmative Action Program, statistical summaries still display an under utilization of these groups.

Later in its memorandum, the district court noted "since its founding, Irvine has granted tenure to 26 men and only two women the last woman tenured was in 1972."

The Supreme Court has stated that general statistical data is helpful in individual employment discrimination cases. McDonnell Douglas, 411 U.S. at 805, 93 S.Ct. at 1825. It is particularly helpful in the academic context, where the tenure decision is highly subjective. Cf. Rowe v. General Motors, 457 F.2d 348 (5th Cir. 1972).3 It is relevant despite the fact that the data may not be directly probative of any of the four specific elements set forth by McDonnell Douglas and Smith. Proof of a general pattern of sex discrimination is, in any event, evidence which tends to establish that it is "more likely than not" that a University's decision to deny tenure was based on sex, "a discriminatory criterion illegal under the Act." See Furnco Construction Corp., 438 U.S. at 576, 98 S.Ct. at 2949 (quoting Teamsters, 431 U.S. at 358, 97 S.Ct. at 1866).

In addition, testimony at trial revealed that the University's evaluation of Lynn's scholarship was due, in part, to its view that women's studies is not a substantial topic for scholarly work.4 The district court stated:

The criticism leveled at her work by scholars and administration officials appears to reflect their disdain of this (women's studies) as a topic of substance in a scholarly work. I find their lack of enthusiasm for her effort may stem from their belief that woman's [sic] studies was an unworthy topic to pursue.

The district court concluded, however, that the University's lack of enthusiasm towards women's studies was not evidence of discrimination because the University would have had the same objection if a man concentrated his studies on women's issues. We do not agree. A disdain for women's issues, and a diminished opinion of those who concentrate on those issues, is evidence of a discriminatory attitude towards women.5 The existence of a discriminatory attitude, like general statistical data, tends to establish that it is more likely than not that the University's decision was based on an impermissible criterion, and therefore tends to establish Lynn's prima facie case.

After consideration of both types of statistical proof, and the other evidence submitted at trial, we believe it is clear that the district court's finding that Lynn failed to satisfy her initial burden of establishing a prima facie case was erroneous.

Next, we turn to whether the University "articulated" reasons for its denials of merit salary increases and of tenure. The University argues that its denials were based on deficiencies in Lynn's scholarship.

A prima facie case "raises an inference of discrimination." Furnco Construction Corp., 438 U.S. at 577, 98 S.Ct. at 2949. "To dispel the adverse inference from a prima facie showing under McDonnell Douglas, the employer need only 'articulate some legitimate, non-discriminatory reason for the employee's rejection.' " Id. at 578, 98 S.Ct. at 2950, quoting McDonnell Douglas, 411 U.S. at 802, 93 S. Ct. at 1824. The Court, in Board of Trustees of Keene St. Col. v. Sweeney, 439 U.S. 24, 25, 99 S. Ct. 295, 58 L. Ed. 2d 216 (1978), made it clear that the burden of "articulating some legitimate, non-discriminatory reason" is significantly less than proving the absence of discriminatory motive.6 Accordingly, the court held in Burdine that the defendant-employer need not prove by a preponderance of the evidence that the plaintiff was rejected for a legitimate, nondiscriminatory reason, and the court there stated that "(i)t is sufficient if the defendant's evidence raises a genuine issue of fact as to whether it discriminated against the plaintiff." 101 S. Ct. at 1094 (emphasis added) (footnote omitted).7

The evidence offered by the University shows that throughout Lynn's career she was warned that her scholarship was suspect. As mentioned earlier, she was given the opportunity to discuss her problems with members of her department, and after her mid-career review, she was granted sabbatical leave for the explicit purpose of improving her research and writing skills. It was Lynn's deficient scholarship that the tenure review committee claims was the basis of its decision not to recommend merit increases or tenure. Without doubt, deficient scholarship is a legitimate, nondiscriminatory reason to deny salary increases or tenure. In view of the standard set forth in McDonnell Douglas, Sweeney and Burdine, and the relative ease with which the courts of appeals have found the employer's burden to be satisfied in the academic context, see, e. g., Smith v. University of North Carolina, 632 F.2d 316, we agree that the University successfully "articulated" reasons that justify its actions, i. e., it offered evidence supporting its reasons for denying Lynn tenure.

Under some views of the McDonnell Douglas analytical process, the University's articulated reason and supporting evidence might be considered at step one of that process, i. e., in connection with plaintiff's prima facie showing that she possesses the requisite qualifications for tenure. See Leiberman v. Gant, 630 F.2d 60, 64 (2d Cir. 1980). We think it preferable, however, to consider the University's arguments at steps two and three. In our view, objective job qualifications are best treated at step one and subjective criteria, along with any supporting evidence, are best treated at the later stages of the process. To do otherwise would in many instances collapse the three step analysis into a single initial step at which all issues would be resolved. This would defeat the purpose underlying the McDonnell Douglas process. See Sweeney, 439 U.S. at 24 n.1, 99 S. Ct. at 295 n.1. In addition, addressing the University's arguments at the first step of the analysis would increase the possibility that courts will be required to engage in evaluations of the performance of faculty members, a task to which others are better suited.8

From a practical standpoint, and despite our determination that evidence of the nature here involved should be considered in connection with steps two and three rather than step one, it should make little difference to the outcome which way the evidence is analyzed. If evidence relating to subjective criteria is treated as relevant to step one, the plaintiff would be entitled, in connection with that aspect of the case, to show that the alleged criteria are inherently discriminatory or discriminatorily applied, or to show that notwithstanding the University's contentions, she meets the criteria. If offered by the University to meet its step two burden, the plaintiff must attempt to show that the reason offered is not the real explanation for the University's decision, but is merely an excuse for a discriminatory act. While it is true that the University need offer little evidence in support of its articulated reason in order to require plaintiff to proceed to step three, the more substantial the University's evidence the more persuasive plaintiff's step three evidence must be in order to establish that the articulated reason is pretextual or was discriminatorily applied. Regardless of the stage at which we consider evidence of the nature here involved, the plaintiff's ultimate burden remains the same, to show by a preponderance of the evidence that the University's action was based on unlawful discriminatory conduct.9

Since, as we concluded ante, the district court properly found that the University "articulated (a) legitimate, non-discriminatory reason," we would ordinarily next determine if the reason "articulated" by the University was "a pretext or discriminatory in its application." However, we do not decide that issue here.

Throughout the proceedings below, Lynn was denied access to her tenure review file. The materials contained in the file were those upon which the tenure review committee claims that it based its denial of tenure, and, as such, are highly relevant to the issues in this case. At the discovery stage, when Lynn requested that the University produce the file, the district court issued a protective order. At trial, the University submitted the file to the court; the court reviewed it in camera but refused to disclose the contents of the file to Lynn. Lynn asserts that the file was submitted by the University, and used by the district court, as evidence, rather than for the purpose of determining whether the contents of the file were privileged.10 Thus, Lynn contends that the refusal to disclose the contents of the file violated due process. We agree.11

During direct examination of Lynn, her attorney offered the minority report of the tenure review committee as evidence. In response, the following statements were made:

MS. HELWICK (attorney for the University): The problem, you Honor, is that obviously, if this is admitted into evidence then we have a problem by way of a response, and then we get into the whole scholarly debate of the quality of Dr. Lynn's scholarship ...

THE COURT: I think I have to permit it to come in, even though it may place you in the position of having to counterbalance it. It is sort of like the position the government is put in when it wants to prosecute a person for some type of intrusion into government records, and in order to do it must bring forth those records ...

MS. HELWICK: ... and I think that by submitting this document we will necessarily have to further violate the confidentiality of those same extramural scholars.

THE COURT: Is it possible for you to meet this by showing content of the majority report but not authorship?

MS. HELWICK: I think not, ...

THE COURT: Why can't it be submitted, then for in camera inspection of the Court and not of the plaintiff.

MS. HELWICK: I have no objection an in camera review by your Honor of the entire tenure review file, if you deem that appropriate.

The minority report was then admitted into evidence.

Later in the trial, the University presented the district court with the tenure review file for in camera inspection:

MR. LUNDBERG (attorney for the University): ... And you have a portion of that which has been admitted as Exhibit 23 (the minority report). But there is material you do not have, and we feel for purposes of your making a complete review that you ought to be provided with the file that was before the ad hoc review committee. This is what we propose to provide you.

The record leaves little doubt that the University submitted the tenure review file to counterbalance the effect of the minority report and that the district court acceded to this use of the file.

The receipt and review by the district court of the tenure review file for the purpose of assisting it to make factual determinations or to evaluate other evidence violated principles of due process upon which our judicial system depends to resolve disputes fairly and accurately. The system functions properly and leads to fair and accurate resolutions, only when vigorous and informed argument is possible. Such argument is not possible, however, without disclosure to the parties of the evidence submitted to the court. Thus, the district court's receipt and review of the file, without disclosure of its contents to Lynn, requires reversal of the order of the district court.

In view of our holding, we need not decide the question whether tenure review files, and more particularly peer evaluations, are privileged in academic Title VII cases generally. We do wish, however, to provide the district court with some guidance on the question since that court will be required, once again, to consider the issue of Lynn's right to obtain her tenure review file, upon remand.13

When determining whether tenure review files, including peer evaluations, are privileged, courts have balanced the university's interest in confidentiality, i. e., in maintaining the effectiveness of its tenure review process, and the need which Title VII plaintiffs have for obtaining peer evaluations in their efforts to prove discriminatory conduct. Jepsen v. Florida Bd. of Regents, 610 F.2d 1379, 1384-85 (5th Cir. 1980); Keyes v. Lenoir Rhyne College, 552 F.2d 579, 581 (4th Cir.), cert. denied 434 U.S. 904, 98 S. Ct. 300, 54 L. Ed. 2d 190 (1977). In making that determination it is necessary to consider the importance of enabling plaintiffs to prove that discriminatory conduct has occurred, the difficulty of obtaining direct proof of discriminatory motivation and the strong national policy against discrimination in educational employment.

The Fourth Circuit has suggested, Keyes v. Lenoir Rhyne College, 552 F.2d 579, and the Fifth Circuit has held, Jepsen v. Florida Bd. of Regents, 610 F.2d 1379, that when evaluations serve as the alleged basis for the University's decision to deny tenure or promotion, the plaintiff's interest in proving his case outweighs the University's interest in protecting the confidentiality of a file and that in such cases the evaluations must be provided to the plaintiff. In Keyes, the Fourth Circuit denied plaintiff's request for all peer evaluations for the entire faculty of the College. The court noted, however, that "if the College had sought to justify any male-female disparity on the basis of these evaluations the plaintiff should have been granted the opportunity to use them" to prove her case. 552 F.2d at 581.14 The Fifth Circuit, in Jepsen, found the reasoning in Keyes to be "persuasive," and held that where "the university defends a claim of discrimination on the ground that promotional decisions were based solely on unbiased faculty evaluations which involved criteria unrelated to sex," a plaintiff is entitled to obtain the evaluations. 610 F.2d at 1384.15

The University claims that Lynn was denied tenure because of deficient scholarship. Since its view of Lynn's ability is based, in large part, on the content of the tenure review file, including peer evaluations, the University is defending, in essence, on the ground that its tenure decision with respect to Lynn was based on non-discriminatory peer evaluations. Under Jepsen and Keyes, disclosure of the evaluations would be required. We agree fully with the views expressed by the Fourth and Fifth Circuits in that respect. However, in Lynn's case it is not necessary to rely on the right of academic plaintiffs generally to obtain evaluations when a university has used the documents in reaching its decision. The record before us reveals that the University has far less interest in the confidentiality of Lynn's tenure review file than in the ordinary case in which disputes over such files arise.16

Lynn received summaries of the comments of her evaluators, as do all University academic personnel when they so request.17 Unlike the ordinary plaintiff, however, Lynn knew the names of her evaluators due to the fact that Dr. Nicole Marzac, apparently in violation of the University rule, provided her with a copy of the minority report.18 Moreover, Marzac provided Lynn with copies of letters written by Marzac to Chancellor Aldrich and to Professor Stephen, as well as the minutes of the meeting of the tenure review committee.19 The final pieces of information undermining the confidentiality of the file were provided, at trial, by Professor Slim, chairman of the tenure review committee. Professor Slim was cross-examined on the proceedings that were conducted by the tenure review committee, and on the majority report. His testimony provided Lynn with further knowledge as to the personal views of the committee members.

Although it cannot be determined with certainty which members of the committee, or evaluators, made which comments, the documentary evidence in Lynn's possession, in addition to Marzac's description of that evidence in her deposition filed with the district court and Slim's testimony at the trial make it possible to attribute specific views and attitudes to specific evaluators. Accordingly, the confidentiality of Lynn's file has been substantially diminished as a result of the information already in her possession.

The district court's determination as to whether and when Lynn may have access to her tenure review file for purposes related to a new trial should be reached in light of the strong interest which Lynn continues to have in determining whether the contents of the file will assist her in proving her case (and particularly in establishing that the University's articulated reason for denial of tenure was "a pretext or discriminatory in its application"), and the minimal interest that the University now has in preserving the vestiges of confidentiality in the file.20

The order of the district court is reversed and remanded for proceedings consistent with this opinion.

ALARCON, Circuit Judge, concurring.

I concur in the opinion of the court insofar as it appears to hold that the district court may have violated Lynn's right to due process by admitting the contents of the tenure review file into evidence "for the purpose of assisting it in making factual determinations or to evaluate other evidence", supra at 1346, after denying Lynn's request to examine these records because they contain privileged and confidential matter.

The balance of the court's opinion is obiter dictum which is not only unnecessary to the disposition of this matter, but fails to give due consideration to the problems which will flow from a requirement that the confidentiality of peer review evaluation must be breached in every case in which a teacher's ability is based, in large part, on the content of the peer review file, without regard to protection of the privacy of the commentators.

I would exercise judicial restraint and leave to another day the consideration of these seductive issues, when the questions are squarely before this court on a record which is otherwise free of reversible error.