Dobbs-Weinstein v. Vanderbilt University

Full title: Idit DOBBS-WEINSTEIN v. VANDERBILT UNIVERSITY

Court: United States District Court, M.D. Tennessee, Nashville Division

Date published: Jan 20, 1998

1 F. Supp. 2d 783 (M.D. Tenn. 1998)

Opinion

No. 3:95-0560.

January 20, 1998.

Richard J. Braun, Gerard Thomas Nebel, Nashville, TN, for plaintiff.

John Shannon Bryant, William Ozier, John Claiborne Callison, Nashville, TN, for defendant.

MEMORANDUM

NIXON, Chief Judge.

Pending before the Court in the above-styled matter is Defendant Vanderbilt University's Motion for Summary Judgment (Doc. No. 35), to which Plaintiff Idit Dobbs-Weinstein filed a Response (Doc. No. 43). For the reasons outlined below, the Court grants the Motion and dismisses Plaintiff's state law claims.

I. Background

Plaintiff Dobbs-Weinstein alleges discrimination in employment on the basis of her gender and national origin in violation of Title VII of the Civil Rights Act of 1964, as amended, 42 U.S.C. § 2000e et seq., and the Tennessee Human Rights Act, Tenn. Code Ann. § 4-21-101 et seq. Plaintiff also brings state claims for breach of contract and breach of covenant of good faith.¹ Plaintiff is an Israeli national, and also holds Canadian citizenship. In the Fall of 1987, she was appointed to the position of Assistant Professor of Philosophy, a tenure-track position.² Plaintiff's claims arise out of Vanderbilt University's decision to deny her tenure in the Spring of 1994.

1 Plaintiff's Complaint also alleged a libel claim, but that claim has since been dropped.

2 Vanderbilt University is divided into several colleges and schools, including the College of Arts and Science ("the College"). Within the College are various departments, among them the Philosophy Department. The top academic officer of the College is the Dean, while the University's chief academic officer is the Provost. The Provost ultimately reports to the Chancellor.

After joining the faculty in 1987, Plaintiff's performance and her potential to qualify for tenure were reviewed at various intervals, in accordance with the Rules and Procedures for Appointments, Renewals, Promotions and Tenure in the College of Arts and Science ("College Rules," Mem.Supp.Mot.Summ.J., App. B). In order to be reappointed at each interval, the College Rules require approval by a majority of the tenured members of Plaintiff's department, in this case the Philosophy Department. At the end of her second year, in the spring of 1989, Plaintiff was reviewed and she was reappointed for an additional two years. At the end of Plaintiff's fourth year, in the spring of 1991, Plaintiff was again considered for reappointment, and again was approved, this time for a three-year extension. At the end of that three-year term, in the spring of 1994 — Plaintiff's seventh year of appointment — the Department considered whether to grant Plaintiff the tenured position of Associate Professor.

The University's Faculty Manual ("Manual," Mem.Supp.Mot.Summ.J., App. A) outlines the criteria by which each renewal and reappointment decision are to be made. The Manual provides for three general areas of excellence which must be fulfilled:

For the award of tenure, Vanderbilt requires (1) excellence in research, scholarship, or creative expression in one's discipline and (2) a high level of effectiveness in teaching. From discipline to discipline, the form taken by a candidate's contribution will vary. But, in each case, Vanderbilt expects the level and quality of achievement in research, scholarship, or creative expression and teaching equivalent to that required for tenure in leading departments or schools of other major research universities. In addition, Vanderbilt expects satisfactory performance in the area of (3) service.

(Manual, Chapter 1, Section E.)

The standard for assessing factor (1), "research, scholarship, and creative expression," is included in the Manual:

Successful candidates for tenure at Vanderbilt must be active scientists, scholars, critics, or artists. By the time of tenure review, they must have completed and made available research, scholarship, criticism, or artistic production of such high quality as to gain favorable recognition within their discipline and at a national level. The works may be available through the publication of books and articles, the circulation of manuscripts intended for publication, lectures and presentations, exhibits or performances. Both past achievements and future promise, both the quantity and quality of completed work, determine one's eligibility for tenure.

(Manual, Chapter 1, Section E, Subsection 1.) The evaluation of a candidate's work includes a consideration of "at least three letters solicited by the Chair [of the department] from a list suggested by the candidate of at least six qualified reviewers outside the University, and at least three letters from colleagues chosen by the Chair in consultation with those department members whose fields are closest to those of the candidate." (College Rules, Section V, A, 4.)

The Manual also defines the standard for assessing factor (2), effectiveness in teaching:³

Candidates for tenure must accept as career obligations the dissemination of knowledge and the nurturing of a spirit of inquiry. To meet tenure standards in teaching, candidates must demonstrate a high level of effectiveness in any of the numerous forms that teaching takes and in any of the settings in which it occurs.

Command of the subject, clarity in communication, and sensitivity to the needs of students are indispensable assets of effective teachers. Candidates for tenure must possess both the skills required to transmit the contents of their disciplines and the capacity to motivate an active pursuit of new knowledge or insight. Such skills and capacities spring from the same qualities that lead to successful scholarly inquiry.

3 Because it is only Plaintiff's success at meeting criteria (1) and (2) that is in dispute in this case, the standards for assessing factor (3) need not be outlined here.

(Manual, Chapter 1, Section E, Subsection 2.) Under the College Rules, conclusions regarding teaching effectiveness are based on "[e]valuations of teaching by members of the department and by graduate and undergraduate students (see Section V, F), or evidence concerning the candidate's teaching in other institutions, along with comments by Vanderbilt students who have met and heard the candidate;. . . ." (College Rules, Section V, A, 4.) Section V, F of the College Rules provides that each department "shall develop a standard procedure for obtaining student evaluations of candidates for renewal and promotion" and those evaluations should "normally include course evaluations made in accordance with faculty legislation and evaluations prepared by graduate and undergraduate majors in the department." (College Rules, V, F.)

As noted above, the tenured members of the Philosophy Department voted to reappoint Plaintiff at both her two- and four-year reviews. However, at Plaintiff's four-year review, the voting members expressed some concern with Plaintiff's performance as to factors (1) and (2) — scholarly contribution and teaching performance.⁴ For example, "varying evaluations of her scholarship" were given: "Some find her written work to be not clearly focused, to be a bit scattered, to be in need of editorial assistance. Others do not dispute these observations, but nevertheless find her ideas interesting and important. . . . On the other hand, others suggested that her work is good only if judged by the standards appropriate to book reports." (Venable Aff., Ex. 1 at 4.) Although the Department reported that Plaintiff was considered by all to be an extremely effective teacher of and role model to graduate students, "[i]t was also universally acknowledged that Dobbs-Weinstein's teaching at the undergraduate level is below par; in fact, that it has been quite unacceptable and must improve if she is to be considered tenurable down the road." (Id. at 3.) Plaintiff's reappointment at her four-year review was ultimately approved by a six to three margin, with two abstentions. (Id. at 1.) A positive vote for tenure, as provided by the College Rules, indicated that the voter believed the candidate "may in due time qualify for tenure," while a negative vote denoted the opinion that the candidate "[was] very unlikely to qualify for tenure." (College Rules, Section IV, 5 6.) The author of the Department's report on this vote noted that "[he] would judge that one reasonable understanding of an abstention would be something like, `possible, but unlikely' (as distinguished from `very unlikely')." (Venable Aff., Ex. 1, at 2.)

4 In regard to factor (3) — Plaintiff's service to the department, the University, and the field of philosophy — Plaintiff received very high marks as having done "a remarkable job of getting a foot-hold in that corner of the professional world that focuses on the issues central to her research" and as having "brought energy and enthusiasm" to her departmental duties. (Venable Aff., Ex. 1 at 4.)

Three years later, in the spring of 1994, the tenured members of the Department convened to decide whether to promote Plaintiff to the tenured position of Associate Professor. By a vote of five to four, tenure was approved. However, the concerns that were voiced at Plaintiff's four-year evaluation were again raised. For example, while it was noted that Plaintiff's book manuscript had recently been accepted by an academic press for publication, the departmental report on the vote reiterated that scholarship concerns persisted:

There is general agreement that by generally accepted standards Dobbs-Weinstein does not write as well as she might, though there is also general agreement that her recent writing . . . is better than her earlier writing. . . . It is [also] clear that some of her faculty colleagues have found conversations with her mystifying while other colleagues . . . have found conversations with her to be insightful and helpful.

Those who oppose tenure add the charge that, apart from the clarity issue, they do not find her presentation of materials always accurate. . . .

(Venable Aff., Ex. 2 at 4-5.)

Discussed "at even greater length" by the voting members of the Department was Plaintiff's success at teaching. The voting members struggled with the "range" of the teaching requirement, disagreeing as to whether quality of teaching should be equivalent in both graduate and undergraduate contexts, or whether one category of instruction was more important than the other; indeed, the faculty noted that "[o]ur criteria refer to teaching, not to the teaching of undergraduates." (Venable Aff., Ex. 2 at 7.) While it was concluded that Plaintiff's teaching of graduate students was exceptional, "[a]t the undergraduate level the data [were] complex." (Id.) The panel went on:

Evaluation numbers vary. Her performance in freshman seminars has clearly improved . . . and . . . her Medieval Philosophy course is on an upswing. But the undergraduate reaction is mixed and complex. Is the clear success of her teaching at the graduate level plus evidence of some improvement at the undergraduate level enough to satisfy the tenure requirement for excellence in teaching? All of us wish the undergraduate record were better but some of us feel that the graduate record plus the bright parts of the undergraduate record are adequate, while others do not.

(Id.)

Under the University's Faculty Manual and the College Rules, a recommendation for tenure by the Department is forwarded to the Dean of the College. The Dean then assesses the recommendation in light of the principles laid out in the Manual and the Rules. He or she has the authority to either concur in the recommendation, or to decline to concur. If a Dean refuses to concur, the Department, by a two-thirds vote, may appeal the decision to the University's Promotion and Tenure Review Committee ("PTRC"). Irrespective of the Dean's decision, the PTRC reviews the tenure file. If the PTRC accepts the Department's recommendation, it recommends tenure to the Provost (even if the Dean has declined to concur). The Provost, in turn, reviews the recommendation and the tenure file, and makes the decision whether to recommend tenure to the Chancellor. The Chancellor then makes a recommendation to the Board of Trust, which makes the final tenure determination. (See Manual, Chapter 1, Section E, Subsection 5.)

In this case, the 5-4 recommendation of the tenured members of the Philosophy Department was forwarded in January, 1994 to then-Acting Dean John Venable.⁵ Dean Venable decided that he was unable to concur with the recommendation. In his letter to Dr. Donald Sherburne, Chair of the Department, dated May 26, 1994,⁶ Dean Venable explained the various grounds for his refusal to concur:

5 Venable was Acting Dean of the College of Arts and Science at all times relevant to this case, unless otherwise indicated, from September 1, 1992 through July 31, 1994.

6 Dr. Sherburne was Chair of the Department at the time Dean Venable wrote his Memorandum. However, because Dr. Sherburne resigned on May 25, 1997, Dean Venable did not deliver his letter until a successor was appointed. When Professor William Race was appointed the Acting Chair, on June 14, 1994, Dean Venable delivered his Memorandum regarding his refusal to concur in the recommendation of Plaintiff for tenure. Plaintiff was made aware of Dean Venable's decision shortly after it was communicated to Professor Race.

(1) Research quality. Dean Venable found that the evidence in this area was "mixed," but that ultimately "the case for excellence [was] not made." (Venable Aff., Ex. 3 at 2.) Venable cited three concerns. First, he noted that the input from outside reviewers was far from unanimous. Of nine outside scholars, he found that two reviews were clearly negative and the remaining seven recommended tenure; of those seven, however, he found the content of the letters, on the whole, only "moderately favorable." (Id.) One positive letter he described as "rather superficial" and ultimately "not persuasive." (Id.) Another reviewer who recommended tenure also "offer[ed] the reservation that the candidate has not produced a `clear synthesis' of her various subjects." (Id.) A third positive review "present[ed] a well-argued case, though one with reservations." (Id.) On the whole, Dean Venable found only three of the seven positive recommendations to be "quite favorable," "glowing," and "strongly supportive." (Id.)

Dean Venable also cited the departmental report on the vote which indicated that "some of [the tenured professors] . . . express serious doubts about the quality of the candidate's scholarship," and that even "[s]ome supporters of tenure . . . acknowledge problems." (Id.) In this vein, Venable went on to articulate his specific concern with the weight to be given to the recent acceptance of Plaintiff's manuscript for publication by SUNY Press. "Although a publishable book is a virtual necessity for tenure in Philosophy . . . mere publication is not enough." (Id.) Venable noted that the readers for the SUNY Press were "supportive [of publication] but with reservations," and that the department was "apparently united in its perceptions of a lack of clarity." (Id. at 3.) Finally, Venable questioned the acceptance process at SUNY Press, suggesting that the process was unduly influenced by relationships between Vanderbilt faculty and the press. (Id.)

Dean Venable also indicated his concern that some of the scholarly papers included on Plaintiff's curriculum vitae did not undergo peer review, and that overall "[o]ne is left with the sense that this is not a body of work that can establish a strong scholarly reputation." (Id.)

(2) Teaching. In this area, Dean Venable disagreed with the Philosophy Department report, which had noted that the voting members felt there was some ambiguity as to whether the College considered the teaching of undergraduates to be as significant as other teaching by tenure candidates. Venable wrote, "There is . . . no ambiguity in the application of the [teaching effectiveness] criterion in the College of Arts and Science: to earn tenure, faculty members must teach undergraduates well." (Id. at 3.) In this regard, Venable was troubled by the fact that even though just eight of the twenty-four courses Plaintiff had taught during her Vanderbilt career had been graduate-level, she had taught approximately the same number of undergraduate as graduate students. Consequently, "[Plaintiff] escaped undergraduate teaching to an extent that would seem impossible, particularly so in the period during which she had been advised to demonstrate improvement in undergraduate teaching." (Id. at 4.) The disproportionate teaching of graduate students was due to the typically low undergraduate enrollment in Plaintiff's less advanced courses. Venable found this enrollment indicative of teaching quality, since "[t]he College's general experience is that medieval courses with good teachers (e.g., in English and History) draw well." (Id.) He felt that such inferences were borne out by the assessment given in the Department's report, and by student evaluations Plaintiff received for her undergraduate teaching, which Venable found "do not support a conclusion of high effectiveness." (Id. at 5.) While the Dean gave Plaintiff high marks for willingness to spend time with students outside of class, he concluded that while her undergraduate students "appreciate her efforts on their behalf, [they] do not understand her in class," and that "[t]he overall result is, in my judgment, mediocre undergraduate teaching." (Id.)

Finally, while Venable concurred in the Department's assessment of Plaintiff as an exceptional teacher of graduate students, he expressed concern that not all graduate students who had experiences to contribute about working with Plaintiff were free to do so; he suggested that some were "coerced" into not voicing opposition to Plaintiff's candidacy for tenure. (Id.)

Upon return to school for the new academic year in August 1994, the tenured members of the Philosophy Department met, in accordance with the Manual and the College Rules, to vote on the question of whether to appeal Dean Venable's decision to the PTRC. The Department's vote was the same as the original tenure vote, 5-4, and thus failed to yield the two-thirds margin necessary for such an appeal. Consequently, Plaintiff was informed that her contract with the College would end in May 1995.

In response to Dean Venable's action, and the subsequent Departmental vote, Plaintiff filed a grievance in October 1995 with the University's Senate Committee on Professional Ethics and Academic Freedom ("PEAF Committee"). Plaintiff's complaint alleged various improprieties in the process by which her tenure candidacy was reviewed by Dean Venable, and by which the Department subsequently considered an appeal.⁷ (Venable Aff., Ex. 4.) She also contended that both Dean Venable's decisionmaking, and the premises upon which such decisionmaking was based (such as student evaluations), were tainted by discriminatory animus on the basis of gender and national origin. (Id.)

7 The PEAF Committee summarized the violations of University procedure alleged by Plaintiff.

(a) the dean granted an interview with a known opponent of her tenure after receiving the Department's recommendation;

(b) two newly appointed and tenured Department members were excluded from the Department's decision in the appeal vote;

(c) materials were added to her file after the Department's recommendation.

(Venable Aff., Ex. 5 at 3.) These procedural irregularities are also cited by Plaintiff in the instant suit as both suggesting discriminatory motive and supporting her state claims for breach of contract and of covenant of good faith and fair dealing. As to allegation (a), Plaintiff contends that although the College Rules provide that the Philosophy Department speaks only through the Department Chair, rather than through its individual members (Mem.Supp.Mot.Summ.J., Ex. B at 6), Dean Venable met with two opponents of Plaintiff's candidacy after the Department's vote and before issuing his decision. (Pl.'s Resp. Summ.J. at 12.) Moreover, Plaintiff asserts that Dean Venable told students who asked to meet with him to express support of Plaintiff's tenure that the College Rules did not permit such meetings. (Id.) As to (b), Plaintiff points out that the new Chair of the Department, Professor William Race, as well as another member of the Department, Dr. Richard Zaner, was not permitted to take part in the departmental tenure vote, either initially or after Dean Venable's refusal to concur. As to allegation (c), Plaintiff avers that after the departmental vote and before Dean Venable's decision was issued, Dean Venable solicited and received ex parte communications from opponents to Plaintiff's tenure, (id. at 30), including a letter from Professor John Post which raised questions regarding SUNY Press' acceptance of Plaintiff's manuscript, questions later cited by Venable as partial grounds for his failure to concur.
Plaintiff also complains about three separate unwarranted procedural delays which amplified the hardship endured by Plaintiff in appealing her denial, and in subsequently winning tenure. First, she complains that Dean Venable refused to act as Acting Chair of the Philosophy Department, in contradiction of existing precedent, after the resignation of Professor Sherburne in May 1994, and instead waited until he had appointed an Acting Chair before communicating his refusal to concur to Plaintiff. Consequently, Plaintiff did not learn that she had been denied tenure until June 14, 1994, well after the time when most faculty had left campus and/or Nashville for summer vacations and research sabbaticals. Consequently, Plaintiff's ability to marshal support for an appeal was impeded, and any hearings on the matter had to be postponed until the fall semester. (Pl.'s Resp.Opp.Mot.Summ.J. at 14.) Second, Plaintiff contends that after her appeal was filed during the 1994-95 academic year, delays in the hearing process occurred at the behest of Dean Venable, who at that point had been appointed Associate Provost. (Id.) Finally, Plaintiff notes that the Chancellor did not forward Plaintiff's file to the PTRC for more than six weeks after the PEAF Committee recommended that he do so. (Id. at 15.)
Last, Plaintiff argues that University policies were breached when Plaintiff's tenure file was not forwarded to the Department for re-consideration when the vote was taken on whether to appeal Dean Venable's failure to concur.

In April 1995, the PEAF Committee issued its findings. While the Committee rejected Plaintiff's allegations that the denial of tenure was motivated by discrimination, it did conclude that there were "several areas of concern" (Venable Aff., Ex. 5 at 8) which warranted — in accordance with the Manual and College Rules — the PTRC's review of Plaintiffs file. In particular, the PEAF Committee referenced Dean Venable's letter to the Department and concluded, "[T]he committee is troubled at how simple `considerations' [about Plaintiff's scholarship and especially her book's publication] have acquired the status of `reasons' to support a conclusion regarding the merits of her scholarship, when there is so little substantial discussion [in the letter] about the quality of her work." (Venable Aff., Ex. 5 at 10.) The Committee also found that Dean Venable had placed undue emphasis upon Plaintiff's teaching of undergraduates, since her area of study was uniquely specialized, and would more naturally place her in greater contact with graduate students. (Id. at 13-14.)

Consequently, the PTRC reviewed the tenure file, and, in August 1995, issued its conclusions. In its brief report, the PTRC concluded that although Dean Venable's concerns were "legitimate, reasonable and weighty," a recommendation should be made to the Provost to promote Plaintiff to the position of Associate Professor with tenure. The Provost, Dr. Thomas G. Burish, accepted the PTRC's recommendation, which he in turn conveyed to Chancellor Joe B. Wyatt. Chancellor Wyatt concurred in the recommendation; consequently, on November 3, 1995, the University's Board of Trust voted to promote Plaintiff to Associate Professor with tenure. The appointment was made retroactive to the end of the 1993-94 academic year, the time at which Plaintiff would have been promoted but for Dean Venable's refusal to concur in the Philosophy Department's recommendation. Consequently, Plaintiff received back pay for the 1994-95 academic year that reflected the difference between her Assistant Professor salary level and that which she would have earned as an Associate Professor, as well as the full Associate Professor salary she would have received during the elapsed portion of the 1995-96 academic year had her contract not concluded in August 31, 1995.

Plaintiff filed this action in May 1995. She charges that Defendant's initial denial of tenure was illegally tainted by sex and national origin discrimination, in violation of Title VII and the Tennessee Human Rights Act, and that her subsequent instatement at the tenured level does not account for the lost interest which would have accrued on her salary or compensate her for the emotional and professional damages she suffered over the year following her denial of tenure. Plaintiff also brings state breach of contract and covenant of good faith and fair dealing claims stemming from alleged irregularities in the way the University's tenure procedures were implemented in her case.⁸ Finally, Plaintiff has added a claim that the University discriminates on the basis of sex in faculty salaries, for which she seeks both retroactive and prospective relief.

8 Plaintiff's Complaint also included a libel claim stemming from Dean Venable's statements regarding the acceptance of Plaintiff's book manuscript by SUNY Press, but, as indicated supra, that claim has been abandoned.

II. Legal standards

Under Federal Rule of Civil Procedure 56(c), summary judgment is appropriate when "the pleadings, depositions, answers to interrogatories, and admissions on file, together with affidavits . . . show that there is no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law." In order to grant summary judgment, the Court must determine that a reasonable jury would be unable to return a verdict for the nonmoving party. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). The moving party bears the initial burden of proof that no genuine issues of material fact are present. Celotex Corp. v. Catrett, 477 U.S. 317, 323, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). Moreover, the facts, and reasonable inferences to be drawn from those facts, must be viewed in the light most favorable to the non-movant. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986).

To defeat a motion for summary judgment, the opposing party must come forward with "more than just some metaphysical doubt as to the material facts." Street v. J.C. Bradford Co., 886 F.2d 1472, 1479 (6th Cir. 1989). Where the record in its entirety would not lead a jury to find for the nonmovant, a genuine issue of material fact does not exist. Id. at 1478. Material facts are those which, under the substantive law governing the issue, might affect the outcome of the suit. Anderson, 477 U.S. at 247-48. Mere allegations of a factual dispute are not sufficient to meet this standard. Id.

Title VII prohibits discrimination in the terms and conditions of employment on the basis of an individual's sex and national origin,⁹ and the Tennessee Human Rights Act imports Title VII's protections.¹⁰ To state a claim under Title VII, a plaintiff may provide direct evidence of discrimination (although such "smoking gun" evidence is rarely available), indirect evidence, or circumstantial evidence. Woythal v. Tex-Tenn. Corp., 112 F.3d 243, 246 (6th Cir. 1997) (quoting Gagne v. Northwestern Nat'l Ins. Co., 881 F.2d 309, 314-15 (6th Cir. 1989) (citation omitted)). If a plaintiff lacks direct evidence, a jury may still infer discriminatory intent under the burden-shifting framework adopted under McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973), and its progeny. See St. Mary's Honor Ctr. v. Hicks, 509 U.S. 502, 113 S.Ct. 2742, 125 L.Ed.2d 407 (1993); Texas Dep't of Comm. Affs. v. Burdine, 450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981).

9 The statute states, in relevant part:

It shall be unlawful for an employer . . . to fail or refuse to hire or to discharge any individual or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual's . . . sex, or national origin. . . .

42 U.S.C. § 2000e-2(a)(1).

10 The statute specifically provides: "[I]t is the purpose and intent of the General Assembly by this enactment to provide for execution within Tennessee of the policies embodied in the Federal Civil Rights Acts of 1964, 1968 and 1972, . . . as amended. . . ." Tenn. Code Ann. § 4-21-101(a)(1) (Michie 1991). Claims under the Tennessee law are analyzed according to the same standards as those under Title VII. Trentham v. K-Mart Corp., 806 F. Supp. 692 (E.D.Tenn. 1991), aff'd, 952 F.2d 403, 1992 WL 3716 (6th Cir. 1992).

According to this approach, a plaintiff may state a prima facie case of sex and national origin discrimination by presenting evidence that (1) she is a member of the protected class(es); (2) she is qualified to do the job;¹¹ (3) despite these qualifications, she suffered an adverse employment action; and (4) that "after [the adverse employment action], the position remained open and the employer continued to seek applicants from persons of complainant's qualifications."¹² McDonnell Douglas, 411 U.S. at 802. Once these four elements are proven, the burden shifts to the employer to articulate a legitimate, non-discriminatory reason for its actions. Id. If the defendant carries this burden, the plaintiff must then produce "sufficient evidence to allow a jury to reasonably reject [the employer's] explanation" as a pretext for discrimination, id. at 804, and to instead conclude that the plaintiff's gender and/or national origin, played a determinative role(s) in the adverse employment decision. Phelps v. Yale Sec., Inc., 986 F.2d 1020, 1023 (6th Cir. 1993). The Supreme Court in St. Mary's Honor Center made clear that "the ultimate burden of persuading the trier of fact that the defendant intentionally discriminated against the plaintiff remains at all times with the plaintiff." 509 U.S. at 518. Moreover, the Court held that "[i]t is not enough . . . to disbelieve the employer; the factfinder must believe the plaintiff's explanation of intentional discrimination." Id at 519.

11 In light of the fact that Plaintiff was ultimately granted tenure by the University, the Court finds that there is no doubt that Plaintiff satisfies this portion of the prima facie case. Defendant concedes this point, as well. (Def.'s Reply Mem. at 5.)

12 The Court notes that this fourth criterion is inappropriate in this case — as in most tenure cases — and considers it satisfied by the fact that the Philosophy Department was not barred by financial or other constraints from promoting an Assistant Professor to Associate Professor at the time Plaintiff was denied tenure. See Langland v. Vanderbilt Univ., 589 F. Supp. 995, 1003 (M.D.Tenn. 1984), aff'd, 772 F.2d 907, 1985 WL 13611 (6th Cir. 1985). It should be noted, however, that Plaintiff does allege that the tenured position in the Department for which she was applying was instead given to the husband of the College's incoming Dean. While the distinction is of no significance here, given the Court's conclusion that the fourth criterion of the prima facie case is satisfied in the tenure context irrespective of such evidence, the Court also accepts Defendant's response to this allegation: that the male Philosophy professor was an "add-on" to the Department, rather than someone hired to a tenured position "instead of" Plaintiff. This conclusion will be discussed further infra.

III. Discussion A. Sex and national origin discrimination — tenure denial

The Court rejects Defendant's argument that Plaintiff's discrimination claim is mooted by her subsequent receipt of tenure. If evidence of discrimination exists in this case, Plaintiff clearly still has an action for damages — for both emotional harm and damage to her professional reputation — and possibly for interest on the backpay she ultimately received. Thus, the Court will examine Plaintiff's discrimination claims.

The Court is satisfied that Plaintiff has made out a prima facie case of sex and national origin discrimination.¹³ Defendant's stated non-discriminatory reason for denial of tenure is Dean Venable's disagreement with the voting members of the Philosophy Department — based on legitimate academic considerations — that Plaintiff met University criteria for tenure in the areas of teaching and scholarship.

13 As indicated, the Court examines Plaintiff's discrimination claims because it agrees with Plaintiff that the subsequent grant of tenure by Vanderbilt does not moot Plaintiff's these charges. Whether the subsequent grant of tenure mooted Plaintiff's claims under the Tennessee Human Rights Act, and her state law breach of contract and covenant of good faith and fair dealing claims, is a question of what remedies are available for such actions; because this Court finds summary judgment appropriate as to Plaintiff's federal discrimination claims, mootness as to Plaintiff's state claims a matter for the state court, not this one, to decide.

Therefore, the Court turns to the evidence Plaintiff has put forth in support of her claims of pretext.

1. Plaintiff's evidence

Plaintiff has amassed a range of evidence which she alleges supports the conclusion that the decision to deny her tenure was discriminatorily tainted. The evidence may be divided into the following categories:

(1) Workforce data as evidence of gender discrimination. Plaintiff notes that at the time she was hired in 1987, she was the first woman employed in a tenure track position in the Philosophy Department at Vanderbilt, and, at all times relevant to this suit, she was the only woman in a department of eleven men. She also states in her deposition that during her employment at Vanderbilt, approximately 48 men received fellowships, while only 3 women did, (Dobbs-Weinstein Dep. at 93-95), although no documentation is provided for this assertion. Plaintiff notes that as of October 1, 1993, there were 138 male professors and 14 female professors at any level, including both tenure and tenure track positions, in the College of Arts and Science at Vanderbilt. At the same time, there were 69 male Associate Professors and 12 female Associate Professors (the tenured rank to which Plaintiff sought promotion) at the College.¹⁴ (Pl.'s Resp.Opp.Mot.Summ.J. at 29.) Finally, Plaintiff notes that "[o]ver 160 female faculty members in the College of Arts and Science between September 1, 1986 and August 1, 1995 are no longer employed by Vanderbilt University." (Id.)

14 Defendant disagrees with the first of these figures: It cites its Work Force Analysis as showing that as of October 1, 1993, there were 250 male professors and 55 female professors in tenured or tenure-track positions (tenured professor and associate professor, and untenured assistant professor level) (Def.'s Resps. to Addt'l Facts at 53). In other words, in October 1993, women professors of all ranks comprised 22 percent of the total number of professors, as opposed to the 10 percent cited by Plaintiff.

(2) Personal bias among Department members as evidence of gender and national origin discrimination. In addition to noting that opponents of her tenure were also opponents of the philosophical theories she espoused, Plaintiff claims that three of the four opponents also "made inappropriate comments of a sexual nature to either [Plaintiff] or students." (Pl.'s Resp.Opp.Summ.J. at 8.) Plaintiff does not provide any further support for these allegations. Plaintiff also asserts that members of the Philosophy Department were unwilling to accommodate absences of faculty members due to Jewish holidays.¹⁵

15 Defendant points out that Plaintiff's claim is one of national origin discrimination, not religious discrimination, and consequently that evidence of anti-Semitism is irrelevant to her claim. The Court considers such evidence, however, since Plaintiff's identities as an Israeli and as a Jew are sufficiently entwined that a separation of one kind of bias from another would be, as both a legal and practical matter, illogical. Cf. Jefferies v. Harris Co. Comm. Action Ass'n, 615 F.2d 1025, 1032 (5th Cir. 1980) (recognizing that plaintiff's identities as African-American and as woman could not be isolated for purposes of analyzing her discrimination claim); accord, Lam v. University of Hawaii, 40 F.3d 1551 (9th Cir. 1994) (applying same logic to Asian-American women).

(3) Male professor tenured "in place of" Plaintiff. Plaintiff argues that the incoming Dean of the College of Arts and Science, Madeline Goodman, conditioned her acceptance of Vanderbilt's job offer on the provision of a tenured Philosophy Department position for her husband, Lenn Goodman. This argument appears to be conjecture, based on the fact that Lenn Goodman's specialty is the same as Plaintiff's (medieval philosophy), and that he asked to take over her courseload after Plaintiff's denial of tenure. No further facts are provided to transform this correlation into a causal factor for denial of Plaintiff's tenure.

(4) Evidence of national origin discrimination at Vanderbilt generally. Plaintiff also argues that there is "substantial other evidence of bias at Vanderbilt based on national origin." (Pl.'s Resp.Opp.Mot.Summ.J. at 28.) Plaintiff cites unspecified damage to Jewish religious symbols across campus, which included anti-Semitic graffiti on Plaintiff's office door.

Plaintiff also submits the affidavit of Pascal Massie, a foreign-born teaching assistant pursuing her Ph.D. in the Philosophy Department. Massie contends that during her mandatory training at the Teaching Center — a supplemental training that all teaching assistants were required to fulfill — she, and other foreign students, were not "given the same depth of evaluation that foreign-born teaching assistants were given." (Massie Aff. at 1.) Moreover, she alleges that "[a]mong the advice given by the Teaching Center was advice to foreign students to bathe frequently, leaving the impression that those giving the advice did not understand that it was the practice of foreign-born students to bathe frequently." (Id. at 2.)

(5) Evidence of national origin discrimination in student evaluations. Plaintiff's primary evidence of national origin discrimination is Vanderbilt's reliance upon student evaluations in making tenure decisions, evaluations which she claims are inherently biased against accented speakers such as herself. Plaintiff presents the report of Dr. Donald L. Rubin, a professor of Speech Communication and Language Education at the University of Georgia. Dr. Rubin's area of expertise is the interaction between language and attitude, and specifically, "the impact of speakers' language variety on listeners' perceptions and comprehension of those speakers on listeners' perceptions and comprehension of those speakers." (Rubin Report at 1.) Dr. Rubin states that "[i]n general, my study of undergraduates' responses to international instructors leads me to the overall conclusion that college students tend to negatively judge the teaching proficiency of international instructors, and even to comprehend their classroom lectures poorly." (Id.) Dr. Rubin also suggests that negative student reactions and lowered comprehension rates occur in studies even when an instructor is merely identified as a non-U.S. native, but where the instructor's speech is unaccented. (Id.) These results, Dr. Rubin surmises, are the result of prejudices which lead students to "hear what they expect to hear," in other words, sub-standard teaching from non-native English speakers.¹⁶ (Id. at 2.) Interestingly, however, Dr. Rubin's research has not led him to recommend the elimination of student evaluations as a basis for employment decisions. Indeed, he notes that his research ultimately "indicates that such evaluation is in general a reasonably accurate measure of teaching effectiveness," (id.), though use of such evaluations should be tempered by an understanding of their shortcomings and potential biases.

16 Dr. Rubin's studies have also revealed that accented English is only one extraneous factor which may unduly influence students' reactions to their instructors. For example, he found that instructors who are unattractive, or who otherwise fail to conform to societal expectations (such as women who act or dress in stereotypically unfeminine ways) are also penalized by listeners. He notes in his report that as result of these other factors, "it may not be the case that international or non-native speaker status is the most powerful extraneous factor affecting student evaluations." (Rubin Report at 2.)

Dr. Rubin's observations about the applicability of these findings to Plaintiff's case are based upon his viewing of a videotape of Plaintiff delivering a simulated lecture. (Id.) He categorizes Plaintiff's speech as "extremely intelligible," though certain pronunciations make it clear that she "is not a native speaker of mainstream North American English" and "is an international." (Id. at 3.) Most notably, Dr. Rubin found that while "it is plausible that these identifications could result in students activating stereotyped expectations of this speaker," he concludes:

To more adequately assess whether this speaker is indeed subject to negative student expectancies based on language and national origin, one would need to conduct an experiment in which this particular speaker contributes one set of stimulus materials. Under controlled conditions, a simulated lecture delivered by this instructor would be judged by a sample of undergraduate students. A similar sample of students would under the same controlled conditions respond to a lecture delivered by a person of similar physical attractiveness who would be identified as mainstream North American. For further comparison, vocal tracks could be switched between the two speakers.

(Id. at 3-4.) (Emphasis added.) No such study was conducted by Dr. Rubin or by any other expert.

(6) Suggestion of bias by Dean Venable. The bulk of the evidence presented by Plaintiff comprises statements and actions attributed to Dean Venable which allegedly illustrate his bias against women and foreign-born persons. Each incident shall be explained in turn:

(a) Plaintiff submits the affidavit of English Professor Margaret Doody, who approached Venable in the Fall of 1993 regarding the hiring of a temporary visiting professor to teach comparative literature. Professor Doody's affidavit states that Venable, in granting such permission, told Doody, "Whatever you do, don't hire a foreigner. All our troubles come from foreigners." (Doody Aff. at 1.)

(b) Plaintiff also submits the affidavit of Assistant Professor Denise Kirschner of the Mathematics Department. In the spring of 1994, Professor Kirschner complained to Dean Venable about a comment made to her by a male Mathematics professor: Kirschner says that this professor told her that "in order to hire a woman in the Department of Mathematics at Vanderbilt, the woman must be better than any of the male candidates." (Kirschner Aff. at 1.) Kirschner contends that when she told Dean Venable about this comment, Venable "was patronizing to me and simply advised me that I must have misunderstood the Department Chair," (id.), as did Pat Pierce, a woman whose position at Vanderbilt is not identified by either Kirschner or Plaintiff. (Id. at 2.) The male professor's alleged remarks were not investigated, to Kirschner's knowledge, nor was he disciplined. (Id.) Kirschner also describes the "agitat[ion]" of the aforementioned math professor at a seminar held in the spring of 1994 regarding gender discrimination in the physical sciences.

(c) The only other woman in whose tenure recommendation Venable failed to concur during his tenure as Acting Dean was Dr. Phyllis Frus, an Assistant Professor in the English Department. Plaintiff notes that when Frus appealed her denial, the PEAF Committee found evidence that two male professors in the Department, during the departmental meeting at which Frus' tenure was considered, spoke of Frus' gender and age in a derogatory manner. This discussion was included in the minutes of the meeting, which Dean Venable saw when he deliberated on whether to concur in the Department's tenure recommendation. Dean Venable "failed to discipline, counsel or investigate the inappropriate statements relating to age and gender." (Pl's Resp.Opp.Mot.Summ.J. at 27.)

(d) Dr. Sabine Cramer, an Assistant Professor in the German and Slavic Department from 1991 through 1996, submitted an affidavit about her experience with Dean Venable. Dr. Cramer was asked to testify before a committee investigating charges of sexual harassment brought against the male Chair of the Department. Dr. Cramer states in her affidavit that she was reluctant to testify because she feared endangering her future chances for contract renewal and for tenure, and that she conveyed these concerns to Dean Venable. (Cramer Aff. at 2.) Dean Venable, she says, assured her that the professor against whom Dr. Cramer was scheduled to testify would not be permitted to participate in Dr. Cramer's renewal evaluation. (Id.) Contrary to these assurances, the professor was permitted to participate in the vote regarding Dr. Cramer, and he (along with the other two male professors in the Department) voted against her. (Id.) Dr. Cramer's contract was not renewed.

2. Assessment of Plaintiff's evidence

The Court undertakes its examination in light of the fact that tenure decisions are unlike virtually any other kind of employment decision, and consequently, they warrant a slightly different kind of scrutiny from the courts. See Zahorik v. Cornell Univ., 729 F.2d 85, 92-93 (2d Cir. 1984) (noting that the tenure decision gives rise to a lifetime personal service contract, is often non-competitive and thus scrutiny of other candidates for comparisons is of limited value, is made through a decentralized procedure, considers multiple subjective criteria, and sparks unusually high levels of disagreement). While the tenure decision's unique context certainly does not insulate universities from the reach of anti-discrimination statutes, the Court must also be cognizant of the well-established principle that a court considering a tenure decision must refrain from substituting its subjective judgment about a teacher's qualifications for those of professional scholars. See, e.g., Brousard-Norcross v. Augustana College Ass'n, 935 F.2d 974, 975-76 (8th Cir. 1991) (courts must not "sit as a `super personnel council' to review tenure decisions"); Zahorik, 729 F.2d at 93 ("triers of fact cannot hope to master the academic field sufficiently to review the merits of such views and resolve the differences of scholarly opinion"); Smith v. University of North Carolina, 632 F.2d 316, 345 n. 26 (4th Cir. 1980) (noting the "anti-interventionist policy" of courts regarding tenure decisions); Kunda v. Muhlenberg Coll., 621 F.2d 532, 548 (3d Cir. 1980) ("[d]eterminations about such matters as teaching ability, research scholarship, and professional stature are subjective, and unless they can be shown to have been used as a mechanism to obscure discrimination, they must be left for evaluation by the professionals"). Rather, a court's inquiry must focus on what actual evidence directly or indirectly implicates discrimination as the motive behind an institution's stated academic reasons for denying tenure. Smith, 632 F.2d at 345-46.

The Court finds that the evidence before it fails to create a genuine question of whether the decision to deny Plaintiff tenure was motivated by sex and/or national origin bias. Although Plaintiff's denial of tenure was ultimately found to be improper, this finding, and the other evidence presented by Plaintiff, do not sufficiently erode the substantial evidence supporting the legitimacy of Defendant's stated academic reasons for that denial.¹⁷ Regardless of whether Plaintiff believes the criticisms of her teaching and scholarship to be fair or warranted, they are criticisms which were raised throughout her employment at Vanderbilt by a number of different parties who had varying degrees of investment in her future. These concerns emerged as early as 1991, the time of Plaintiff's fourth-year reappointment vote: as discussed supra, "varying evaluations of her scholarship" were given at that time by her colleagues within the Department, and "[i]t was also universally acknowledged that Dobbs-Weinstein's teaching at the undergraduate level is below par; in fact, that it has been quite unacceptable and must improve if she is to be considered tenurable down the road." (Venable Aff., Ex. 1, at 3-4.) Ultimately, Plaintiff's reappointment was approved by a six to three margin, with two abstentions. (Id. at 1.) Similarly, the later 5-4 tenure vote within the Department indicates that Dean Venable's objections — while ultimately overruled by the PEAF committee and the PTRC — were rooted in academic concerns over which reasonable scholarly minds could differ. Additionally, the Department's denial of tenure and Venable's failure to concur were in part based on evaluations received from scholars at SUNY Press and at other institutions, who gave a range of assessments of Plaintiff's qualifications. Moreover, the negative student evaluations which were received echoed the concerns over intellectual clarity cited by Plaintiff's colleagues themselves, indicating to this Court that the criticism from Plaintiff's peers and the Dean was rooted more in substance than in pretext:

17 Plaintiff argues that the University's ultimate award of tenure itself provides "the most powerful proof" of pretext, because it proves that Plaintiff "had, in fact, met the standards set forth in the Faculty Manual." (Pl.'s Resp. Opp.Mot.Summ.J. at 23 .) This assertion is misguided. It is true that Plaintiff's case would be demonstrably weaker if two review boards had agreed that tenure was inappropriate. However, the fact that she was ultimately deemed qualified for tenure does not warrant the conclusion that the initial decision of "unqualified" was based on discriminatory motive. To read pretext into every overruled tenure decision would substantially alter the roles of both the university bodies, such as the PEAF Committee and PTRC, and courts such as this one.

It is clear from the letters of students that classes taught by Dobbs-Weinstein often begin by generating a certain amount of confusion and puzzlement — some of the graduate students, and undergraduates, who ended up feeling enormously indebted to her teaching remark that their initial exposure to her in classes left them confused.

(Venable Aff., Ex. 2 at 4-5.) Commenting on this phenomenon, Professor Sherburne, one of Plaintiff's ultimate supporters, wrote, "I fully understand the experience from which [Plaintiff's] critics are speaking; I have sometimes been mystified [by her comments and questions] and yet sometimes have felt that indeed there is something there she is trying to get at, even if I can't quite put my finger on it any more than she can." (Id. at 5.)

Indeed, the degree to which Plaintiff's supporters agree with her detractors is striking in this case. Professor Sherburne summarized the Department's narrow tenure vote in this way:

[I]t is not a simple, straightforward matter to determine whether [Plaintiff makes philosophical sense and contributes to the field]. Five of us, all of whom freely acknowledge that there are problems with clarity and connecting links . . ., feel that the scholarly accomplishment . . . merits tenure, especially when one focuses on the work done in medieval philosophy. Four of us did not. . . . The four who voted against tenure agreed that in their conversations with the candidate and in their reading of her manuscripts, they did not encounter the qualify of mind that they feel a candidate has to exhibit to be worthy of tenure. The five who voted for tenure were, at least some of them, quite willing to agree that the candidate does not have the kind of mind and style that is viewed as standard in some traditions, but they concluded that in its way it gets the philosophical job done and that therefore the candidate merits tenure.

(Id. at 6.) (Emphasis added.)

Finally, the Vanderbilt academicians charged with review of Plaintiff's appeal also found significant room for debate about tenuring her. The PEAF Committee indicated in its report that "all members of our Committee considered this a close and unusual case" and that "[t]he concerns raised by Dean Venable were legitimate, reasonable and weighty." (Id., Ex. 6 at 1.) The PTRC noted that its own members who dissented from the tenure recommendation considered Plaintiff's case "a close one." (Id. at 3.)

Of particular note in this regard is the PEAF committee's conclusion that while it ultimately found Dean Venable's reasoning conclusory rather than supported by the realities of Plaintiff's scholarly record, it did not find that his failure to concur was a "product of discrimination." (Id., Ex. 5 at 10.) In fact, both the PEAF committee and the PTRC indicated that if any improper motivation was to be attributed to the Dean, it was his unwillingness to take into account the fact that a vigorous ideological debate had divided the Philosophy Department and the discipline generally: As noted in the Department's tenure vote report, opponents to Plaintiff's tenure fell within the "Anglo-American empirical tradition" (otherwise known as the "analytic" tradition) which "wants to get perfectly clear on all the details of the parts and thinks that with that done the overall picture will fall into place." (Id., Ex. 2 at 5.) Plaintiff, on the other hand, fell into the "continental tradition" of philosophy, which "thinks that one proceeds from the whole down to the parts because only in the light of the overall picture does the meaning and significance of the parts come into view. Work produced from within this latter tradition . . . often looks fuzzy and disjointed to those working within the assumptions of the former tradition. . . ." (Id.)

Both committees found that this pedagogic division may have accounted for much of the contentiousness of Plaintiff's candidacy: the PEAF committee criticized the Dean for overestimating the importance of the close departmental vote on Plaintiff's candidacy because "the split vote may not be an accurate reflection of her merit" but rather of "the deep division currently represented in the department," (id., Ex. 5 at 15), while the PTRC noted that poor reviews of Plaintiff's scholarship were difficult to evaluate "in part because of disagreements within [Plaintiff's] own department, and in part because of an apparent division within the discipline of philosophy generally about what constitutes quality research." (Id., Ex. 6 at 2.) Indeed, Plaintiff herself also suggests that this ideological rift fueled the opposition to her candidacy, given her observation that "[t]he four professors who voted against recommending tenure for [Plaintiff] were all analytic professors." (Pl.'s Resp.Opp.Mot.Summ.J. at 7.) However, while it is unfortunate and even unfair that such disputes may have clouded the tenure decision here, "academic and intellectual bias is not evidence of sex discrimination." Ezold v. Wolf, Block, Schorr and Solis-Cohen, 983 F.2d 509, 541 (3d Cir. 1992) (reviewing denial of law firm partnership).

In sum, such vigorous debate among Plaintiff's peers about her merits strongly indicates that Defendant's initial denial of tenure was motivated by a truly legitimate nondiscriminatory reason for denial of tenure. The strength of the evidence supporting this contention places the responsibility for creating a material question of pretext squarely at Plaintiff's feet. Ultimately, this is a burden that Plaintiff fails to carry.

As a threshold matter, the Court agrees with Plaintiff that evidence of skewed gender ratios can provide important context to allegations of sex discrimination and may help establish pretext in individual disparate treatment cases. See McDonnell Douglas, 411 U.S. at 804-05. Thus, the Court assesses Plaintiff's other evidence of discrimination with cognizance of the fact that she was the first woman hired into a tenure-track position within the Department, that there were no women graduate students in the Department at that time, and that she remained the only woman in the Department through the time of her tenure vote. Plaintiff's other data, however, about the number of tenured and non-tenured women in the entire College of Arts and Sciences and about the number of women who left the College from 1986 to 1995, however, are less helpful. Such "raw numerical comparisons," when not accompanied by any particularized discussion of qualified application pools or comparisons to the number of men who left, are of minimal probative value and do not allow definitive conclusions regarding bias at the College. Ezold, 983 F.2d at 543 (and cases cited therein); see also Shaw v. Monroe County, No. 95-70107, 1996 WL 426483, at *8 (E.D.Mich. Apr. 18, 1996) (plaintiff's citation of low rates of minority employment at defendant employer not probative because "statistics without correlation are indicative of no meaningful inference or conclusion") (quoting Long v. City of Saginaw, 911 F.2d 1192, 1201 (6th Cir. 1990)); Zahorik, 729 F.2d at 95. This is especially true in the academic setting, where departmental needs for adding faculty may vary dramatically, both from department to department and from year to year, and where decisions about hiring faculty — as noted supra — are highly subjective. See Zahorik, 729 F.2d at 95. Without contextual analysis of the statistics themselves, the statistics' value in creating context in individual cases is substantially undermined. Moreover, Plaintiff ignores the Supreme Court's important caveat to the principle that statistical data may aid an individual's case: "We caution that such general determinations, while helpful, may not be in and of themselves controlling as to an individualized hiring decision, particularly in the presence of an otherwise justifiable reason for refusing to rehire." McDonnell Douglas, 411 U.S. at 805 n. 19 (quoted in Lieberman v. Gant, 630 F.2d 60, 69 (2d Cir. 1980)). As discussed supra, the Court believes that such a reason — albeit later overturned — has been articulated by Defendant.

As to Plaintiff's allegation that incoming Dean Madeline Goodman required the Philosophy Department to hire her husband, whose specialty placed him in competition with Plaintiff, Defendant has denied that Lenn Goodman either shared Plaintiff's specialty or that he was hired instead of her; rather, Defendant states that it hired Lenn Goodman as an additional member of the faculty, irrespective of the tenure decision regarding Plaintiff. (Def.'s Resp. Addt'l Facts at 24-25, ¶ 65.) The PEAF Committee also concluded: "Although the appointment of Professor [Lenn] Goodman overlapped with the time in which the [Plaintiff's] tenure decision occurred, we accept Mr. Venable's assurances that there was no connection between the two decisions and recognize that the University's practice of offering positions to spouses of appointed persons occasionally results in additional persons within departments." (Venable Aff., Ex. 5 at 7.) The Court finds no reason to question the Committee's assessment of Professor Goodman's hiring.

The Court is thus left to consider the anecdotal evidence that Plaintiff claims is suggestive of gender and national origin bias on the part of her colleagues and Dean Venable, and her contention that the evaluations by her students were biased and should not have been the basis for her tenure denial.

It is certainly relevant to this Court's inquiry to know how women and Jews were treated within the Philosophy Department, and by Dean Venable. Any statements or behavior by any of the decisionmakers that were derogatory on the basis of any protected characteristic may provide evidence of discrimination. Wilson v. Wells Aluminum Corp., No. 95-2003, 1997 WL 52921, *5 (6th Cir. Feb. 7, 1997) (citing Cooley v. Carmike Cinemas, Inc., 25 F.3d 1325 (6th Cir. 1994)); Ezold, 983 F.2d at 546 ("[P]roof of a discriminatory atmosphere may be relevant in proving pretext since such evidence `does tend to add "color" to the employer's decisionmaking processes and to the influences behind the actions taken with respect to the individual plaintiff.'") (quoting Conway v. Electro Switch Corp., 825 F.2d 593, 597 (1st Cir. 1987)); Merrick v. Farmers Ins. Group, 892 F.2d 1434, 1438 (9th Cir. 1990) (comments indicating bias are "clearly relevant" to a disparate treatment claim).

There is more to this principle, however: without evidence that illegal animus affected the contested employment decision itself, such "atmospheric" evidence is not enough to withstand summary judgment. See, e.g., Gartman v. Gencorp, Inc., 120 F.3d 127, 131 (8th Cir. 1997) ("Not all comments that reflect a discriminatory attitude will support an inference that an illegitimate criterion was a motivating factor in an employment decision.") (citations omitted); Ezold, 983 F.2d at 547 (gender-based comments by single person not sufficient to show "such a pervasive hostility to women at [law firm] . . . was more likely the result of discriminatory bias than [law firm's] perception of [plaintiff's] legal ability"). This is not a hostile environment sexual or national origin harassment case. To prove that her colleagues and the Dean denied her tenure because of her gender and national origin, she must prove not only how she (and other women) perceived the climate for women and Jews in the Department — and at the College generally — but how that climate actually infected the tenure process in her case, and played a determinative role in denying her tenure. See Price Waterhouse v. Hopkins, 490 U.S. 228, 277-78, 109 S.Ct. 1775, 104 L.Ed.2d 268 (1989) (O'Connor, J., concurring) ("[S]tatements by . . . decisionmakers unrelated to the decisional process itself [cannot] suffice to satisfy the plaintiff's burden [of persuasion]."); see also Wilson, 1997 WL 52921 at *5 ("[W]orkplace remarks furnished as evidence of discrimination [must be] clear, pertinent, and directly related to decisionmaking personnel or processes in order to prove discriminatory intent."); Venters v. City of Delphi, 123 F.3d 956, 973 (7th Cir. 1997) ("Evidence of discriminatory motives must . . . have some relationship with the employment decision in question; inappropriate but isolated comments that amount to no more than `stray remarks' in the workplace will not do.") (citation omitted); Cooley v. Carmike Cinemas, Inc., 25 F.3d 1325, 1333 (6th Cir. 1994) (anecdotal evidence of discriminatory comments must "go[ ] beyond . . . vague, ambiguous, and isolated comments").

No such nexus is established by the anecdotal evidence presented here by Plaintiff. She has not compared the criticisms levied at her qualifications to those levied at male tenure candidates, in order to expose them as illegitimate. She has failed to provide a single comment or incident, contemporaneous or otherwise, which would link alleged gender or ethnic animus on the part of the Department and/or Dean Venable to the tenure decisionmaking process in her case. Rather, her evidence is too removed, either temporally or in substance, to indicate that Plaintiff's gender and/or national origin played a determinative role in her tenure denial.

The Court notes that the evidence regarding Dean Venable's alleged insensitivity to the concerns of women faculty and to the hiring of foreign instructors,¹⁸ and the alleged irregularities in the tenure review process, are together the only portions of Plaintiff's case which come close to suggesting improper motive. Dean Venable's alleged behavior toward Doody, Kirschner, Frus, and Cramer, if true, could be perceived as evidence of the kind of "deep-rooted, ongoing pattern [of discriminatory attitude] that is anything but isolated." Cooley, 25 F.3d at 1331. Moreover, Plaintiff is correct in arguing that an institution's deviation from its established policies can "discredit the legitimacy of the [institution's] articulated reason for . . . the denial of tenure." Kunda, 621 F.2d at 545; see also Brousard-Norcross, 935 F.2d at 977 (quoting Zahorik, 729 F.2d at 93).

18 Dean Venable has denied that he made assurances to Dr. Sabine Cramer that a male professor against whom she testified would not be involved in her tenure decision. (Venable.Supp.Aff. at 5, ¶ 7.) He claims that he did question the male mathematics professor who allegedly told Professor Denise Kirschner that women had to be "better" than male candidates in order to be hired by the department. (Id. at 4, ¶ 5.) Venable also asserts that he had no knowledge of allegedly biased comments made by English Department members during the tenure vote regarding Professor Phyllis Frus, and only learned of the comments nearly a year later, when the PEAF Committee issued its report on the matter. He notes that the PEAF Committee concluded that such comments were "entirely speculative." (Id. at 4-5, ¶ 6.) Dean Venable has not offered a refutation of Professor Margaret Doody's account of their conversation regarding the hiring of a visiting professor in comparative literature.

Plaintiff's evidence is insufficient, however, to create a genuine issue of pretext here. First, Dean Venable's alleged insensitivity to the concerns of women faculty has not been shown to have any link to his tenure decisions, either in general or in relation to Plaintiff. During his appointment as Acting Dean, Venable acted on fifteen tenure recommendations. (Venable Aff. at 1-2, ¶ 2.) In the case of male candidates, Venable concurred in eight of nine departmental recommendations; in the case of female candidates, Venable concurred in four of six departmental recommendations. The two recommendations of women in which Venable did not concur were Plaintiff and Phyllis Frus, discussed supra. Such statistics are drawn from too small a sample to suggest a pattern of bias in Venable's tenure decisions.¹⁹ See Langland, 589 F. Supp. at 1005 (where dean had concurred in 29 of 37 male tenure recommendations, and 1 of 3 female tenure recommendations, statistical sampling insufficient). Furthermore, Plaintiff's complaint that Dean Venable required an excellence in undergraduate teaching that he did not ask of male tenure candidates is flawed for two reasons: first, as discussed extensively supra, there appears to be genuine contention among all academicians involved in Plaintiff's case as to how much weight undergraduate teaching should have been accorded, and to what extent Plaintiff's undergraduate teaching should be balanced with other criteria. This contention undermines a conclusion that Venable's focus on the undergraduate teaching was motivated by bias. Second, Plaintiff does not provide evidence regarding the tenure decisions involving male candidates; consequently, the Court is unable to assess for itself whether the undergraduate teaching criterion was imposed selectively or in an otherwise biased manner.

19 These statistics are even less probative of discrimination when it is considered that Dean Venable's successor, Madeline Goodman, concurred in Venable's conclusion that Professor Frus should not be granted tenure. (Venable Supp.Aff. at 5, ¶ 6.)

The alleged procedural deviations by Dean Venable also are insufficient, standing alone, to support an inference of sex and/or national origin discrimination. As to Dean Venable's meeting with or accepting ex parte letters from faculty members after his receipt of the Philosophy Department's report, but before he issued his decision on Plaintiff's file, Venable notes that the rules outlined in the Faculty Manual do not preclude such meetings. He also told the PEAF Committee that in his one meeting with an opponent of Plaintiff's tenure, he "suggest[ed] to this person that he would not consider any matters he might raise in conversation; he then invited the person to submit a letter, which he also did with other members of the department." (Venable Aff., Ex. 5 at 16.) These letters were then submitted to Plaintiff's tenure file. The PEAF Committee, while noting that the propriety of accepting such communications is an ongoing matter of dispute within the university, did not find that their acceptance in Plaintiff's case tainted her tenure decision.²⁰ (Id. at 16-17.) Particularly in light of the vagueness of Plaintiff's description of the content of such communications, this Court is in no position to disagree. The PEAF Committee also found that it was fair for Professor Race, the new Chair of the Philosophy Department who oversaw the vote on whether to override Dean Venable's failure to concur, to prohibit two new senior members of the department from participating in the vote.²¹ One of these faculty members did not have tenure in the department, and the other had not taken part in the original tenure discussion and vote.²²

20 The fact that Plaintiff's full tenure file was not forwarded to the Department for consideration before its vote cannot be found to have improperly skewed the Department's vote, since the same parties were voting as in the first vote, and were consequently well appraised of the issues before them. Plaintiff does not cite any portion of the Faculty Manual or the College Rules which mandates a re-evaluation of the tenure file before an appeals vote is taken, nor does she present evidence that members of the Department asked to see such letters, or that the letters would have changed the breakdown of the departmental vote. The Court notes that at the Department's meeting to discuss and vote on the appeal, Professor Race did read portions of Dean Venable's letter outlining his reasons for failing to concur.

21 Plaintiff does not indicate why the exclusion of these faculty members from the vote contributed to an unfair result in her case, but the Court presumes it is because Plaintiff believes these faculty to be her supporters, and thus that they could have altered the 5-4 vote against appealing Dean Venable's decision. The Court notes that even if this were the case, and the vote had instead been 7-4 in favor of pursuing an appeal, such a vote still does not provide the necessary two-thirds margin.

22 Plaintiff also complains that, on the same logic, Professor Race should not have been allowed at the department's appeal meeting. However, the Court is satisfied that Professor Race only convened the meeting, but did not express an opinion about the vote or participate in the vote itself.

Finally, the Court does not find that Plaintiff's allegations regarding the temporal delay in the transmission of Dean Venable's decision to her (through the new Chair, Professor Race) and in the tenure appeal process raise the specter of unlawful discrimination. While neither the PEAF Committee nor the PTRC addressed these issues, the Court notes that the Faculty Manual only provides that "written notices of renewal or non-renewal normally will be made by June 1 of the year preceding their final year." (Manual, Chapter 1, Section D.) (Emphasis added.) Given the fact that Professor Sherburne resigned as Chair of the Department the day before Dean Venable authored his letter regarding his failure to concur — prompting Dean Venable to begin a search for a new Chair — the Court is inclined to find that the delay in informing Plaintiff of the Dean's decision until after Professor Race was named Chair is not probative of discrimination.²³

23 Because Plaintiff offers no evidence in support of alleged interference by Dean Venable in the appeals process — which Plaintiff claims added to the delay in resolving her case — the Court does not consider that allegation.

The Court now turns to Plaintiff's allegations regarding national origin bias in the student evaluations, and finds that no material question of fact exists on this question. It is well-established that "student reaction is a legitimate, nondiscriminatory factor on which to evaluate tenure candidates." Brousard-Norcross, 935 F.2d at 976. Such an assumption is apparently not questioned by the Plaintiff; indeed, Plaintiff's own expert, Dr. Rubin, endorses the use of student evaluations in the aggregate, although he recommends that each case be studied individually to prevent hidden bias from infecting the tenure process.

Plaintiff has failed to show that any such bias was present in the evaluations. No question has been raised as to the framing of the evaluation form's questions. No evidence of biased student comments on the forms has been presented. Plaintiff's only evidence of bias is the report of Dr. Rubin, which merely outlines the phenomenon by which accented instructors may be penalized by their students. However, Dr. Rubin specifically notes that he would have to conduct further study to conclude whether Plaintiff herself was penalized in this way.

Plaintiff does not even provide the Court with the actual evaluations she received; rather, the Court must rely on the summaries of the evaluations offered in the Department's and Dean Venable's reports, and on the expert report of Dr. William E. Cashin submitted on behalf of Defendant. Dr. Cashin's report indicates a significant fact: first, Plaintiff "in general did not receive low ratings." (Cashin Aff., Ex. 2 at 2.) In comparison with other members of the Philosophy Department, it appears that Plaintiff received lower ratings than her colleagues in the 100- and lower 200-level courses, but higher ratings in the more advanced 200- and 300-level courses. (Id.). Indeed, when Plaintiff's ratings in all her courses are averaged, she is found to have received higher average ratings than her colleagues. (Id.) In light of these statistics, it becomes clear that it is not all student evaluations to which Plaintiff objects, but only the fraction submitted by those students in her lowest-level courses, courses in which the students are presumably younger undergraduates and/or are receiving their first exposure to philosophy. Indeed, as the PEAF Committee noted, Plaintiff's negative student evaluations came from "a relatively small number of undergraduate students — at most, roughly half of the small pool of undergraduate students she has taught." (Venable Aff., Ex. 5 at 14.)

As discussed supra, Plaintiff's own colleagues in the Philosophy Department were themselves split about Plaintiff's quality of thought and of oral communication, with some finding her insights provocative and others finding them confusing. (Venable Aff., Ex. 2.) If tenured faculty members had such disparate reactions, the Court finds it quite plausible that younger or less sophisticated philosophy students would have more difficulty with Plaintiff's courses, leading to less positive evaluations.

Of course, it is also possible that such students are also more susceptible to extraneous influences, such as Plaintiff's accent. The PEAF Committee indicated that this was the flaw in allowing such a small group of evaluations to control the tenure process for Plaintiff, given the "multiple possibilities for prejudice that could occur in the evaluations, just because of the person in question and the character of her subject matter: [Plaintiff] is Jewish, female, foreign, active in issues regarding sexual orientation, and she challenges students' views of religious truth and reality in the context of her courses on medieval philosophy. She has been the target of ethnicity-religion-based attacks in the form of vandalism and graffiti while at Vanderbilt." (Venable Aff., Ex. 5 at 14.) (Emphasis added.) Such trepidation is the responsible approach, in light of studies such as Dr. Rubin's, as well as related work by legal academicians regarding the unconscious psychological processes by which race negatively affects perceptions. See, e.g., Charles R. Lawrence III, The Id, the Ego and Equal Protection: Reckoning with Unconscious Racism, 39 STAN.L.REV. 317 (1987). However, the conclusion that the Department and Dean Venable did not consider the potential for discrimination in a small number of negative evaluations does not translate into the conclusion that such discrimination actually occurred in those evaluations. Indeed, the PEAF Committee itself found that "we have not received evidence of prejudice in the evaluations." (Venable Aff., Ex. 5 at 14.)

The anecdotal data provided by Plaintiff regarding her students' attitudes does not alter this conclusion. Plaintiff states that she was "told by some undergraduate students that she misrepresented Christian philosophers because she is a Jew" and that "some undergraduates have asked Dobbs-Weinstein how a Jewish woman could teach Christian philosophy." (Pl.'s Resp. Opp. Mot. Summ. J. at 41.) According to Defendant's filings, however, there were evaluations for twenty-four courses which formed the basis of Plaintiff's teaching record. Any link between the negative evaluations drawn from this large sampling, and the unspecified anti-Semitic graffiti at Vanderbilt and remarks made to Plaintiff by a few students about her teaching of Christian philosophy, is simply speculative. It is certainly a connection too tenuous to warrant a finding of national origin discrimination in all, or even a portion, of the student evaluations in Plaintiff's case.

Plaintiff's evidence regarding alleged national origin discrimination within the Philosophy Department suffers from the same defect. The remark allegedly made to Pascal Massie at the Teaching Center is in no way linked to members of the Department or to Dean Venable, and will not permit the inference that Plaintiff was hindered in the tenure process because of her national origin. While unwillingness of unspecified faculty members to accommodate Jewish holidays could conceivably be considered, along with other evidence related to Plaintiff's tenure consideration specifically, to raise an inference of hostility to an Israeli colleague, it cannot, taken alone, create such an inference. Again, this is particularly so in light of the subsequent findings by both the PEAF Committee and the PTRC that Plaintiff's tenure candidacy presented a "close case" (Venable Aff., Ex. 5 at 1, Ex. 6 at 3) on the basis of the "legitimate, reasonable and weighty" concerns raised by her detractors within the Department. (Id., Ex. 6 at 1.)

B. Sex discrimination — wages

Plaintiff makes both individual and classwide allegations of wage discrimination. She contends that there is a "systematic and wide-spread disparity" in Defendant's salaries for male and female professors of all levels, including her own. (Pl.'s Resp. Opp. Def.'s Mot. Summ. J. at 35.) She does not specify whether this disparity is the result of intentional discrimination, or whether Vanderbilt's salary policies disparately impact women faculty. Plaintiff also argues that her own salary increases were dramatically reduced during the period that Venable served as Acting Dean. (Id. at 38.) Presumably, then, Plaintiff argues that any alleged disparity in her salary is the result of intentional discrimination by Dean Venable or the University itself.

Plaintiff contends that during the period of Dean Venable's deanship, from September 1, 1992 to July 31, 1994, Plaintiff's salary increases declined, from being the second highest cumulative salary increase in the Philosophy Department to being only the eighth highest. (Pl.'s Resp. Opp. Def.'s Mot. Summ. J. at 38.) Plaintiff provides no support for this contention. Defendant rebuts Plaintiff's assertion in two ways: first, Dean Venable notes that he was not involved in setting salaries for the 1992-93 academic year, but rather, outgoing Dean Vogeli set such salaries. Therefore, the 1993-94 year is the only one in question in Dean Venable's case.

Defendant argues that Plaintiff's contention must fail based on both the cumulative salary increases she received during her employment in the Department, as well as on a yearly basis during Dean Venable's tenure as Dean. Defendant submits data showing that, from the 1988-89 academic year through the 1993-94 year, Plaintiff received the third highest cumulative percentage raises in the Department. The two professors receiving higher cumulative increases were (1) Michael Hodges, who received a promotion to full Professor in the 1992-93 year and thus was awarded a supplemental salary increase; and (2) John Lachs, who holds the College's prestigious Centennial Professorship and was awarded a higher salary as an inducement not to accept offers from other institutions. (Def.'s Reply Mem. at 11.) Moreover, if the cumulative increases are measured through the 1994-95 year (the year for which Dean Venable did have authority to set salaries), Plaintiff ranks second in the Department, behind only Professor Lachs. (Id.) Plaintiff has not presented any evidence that Professors Hodges and Lachs are similarly situated to her, and consequently, that their higher cumulative increases are discriminatory.

Furthermore, Defendant's evidence indicates that in the 1993-94 year — the first year for which Dean Venable set faculty salaries — Plaintiff received a 3.85% increase, the sixth highest in the Department (out of seventeen faculty members). This was higher than the increase she had received the previous year, 1992-93, under Dean Vogeli's leadership — 3.66%, only the eighth highest increase. For the 1994-95 year, also set by Dean Venable, Plaintiff's increase was 4.08%, the third highest raise in the Department, behind Professors Jeffrey Tlumak and Victoria Lynn McGeer. Again, Plaintiff has not offered any proof to show that Professors Tlumak, McGeer, or both were similarly situated to Plaintiff and undeserving of the higher increases. In light of the evidence offered by Defendant, and the failure by Plaintiff to rebut these findings, the Court cannot conclude that Plaintiff's own salary increases evidence discrimination. Consequently, summary judgment as to this portion of Plaintiff's wage discrimination claim is appropriate.

As to her class-wide claims of salary bias, Plaintiff relies on statistical analysis of Vanderbilt's own salary data. Statistical evidence purporting to show pervasive discrimination "ha[s] served and will continue to serve an important role" in such cases, Kincade v. Firestone Tire Rubber Co., 694 F. Supp. 368, 376 (M.D.Tenn. 1987) (quoting Mayor of Philadelphia v. Educational Equality League, 415 U.S. 605, 620, 94 S.Ct. 1323, 39 L.Ed.2d 630 (1974)); see also Coates v. Johnson Johnson, 756 F.2d 524, 539 (7th Cir. 1985) ("Statistical analysis appears to be a science and, when properly and fairly used, it can be of great assistance to the factfinder.") A common method for assessing statistics in discrimination cases is regression analysis, which isolates the impact of certain variables in effecting a certain outcome. By identifying all possible legitimate factors contributing to a certain disparate outcome, the statistician "can make predictions about what . . . job benefits similarly situated employees should ideally receive, and then can measure the difference between the predicted treatment and the actual treatment of those employees." Ottaviani v. State U. of New York at New Paltz, 875 F.2d 365, 367 (2d Cir. 1989). In this way, a plaintiff may seek to prove that any "residual" disparity in the salaries of men and women could only be attributed to an illegitimate factor, such as gender. Id.

Statistical disparities alone, however, do not end a court's inquiry. A court must also examine whether such disparities reach such levels as to be "statistically significant," which essentially "is a measure of the probability that a disparity is simply due to chance, rather than any other identifiable factor," such as sex. Id. (and cases cited therein). One mechanism for measuring such statistical significance is the "standard deviation." Where the standard deviations — or dispersions — among numbers measuring expected and actual results are low, the greater the likelihood that such dispersion is due to chance. If the standard deviations between numbers are greater, however, the greater the likelihood that the disparity is not due to chance. Id.

Social scientists, and in turn the courts, have adopted two standard deviations as a threshold measure of statistical significance. See Hazelwood School District v. United States, 433 U.S. 299, 312 n. 17, 97 S.Ct. 2736, 53 L.Ed.2d 768 (1977); Castaneda v. Partida, 430 U.S. 482, 497 n. 17, 97 S.Ct. 1272, 51 L.Ed.2d 498 (1977); see also E.E.O.C. v. American National Bank, 652 F.2d 1176, 1192 (4th Cir. 1981) ("courts should be extremely cautious in drawing conclusions [of law] from standard deviations in the range from one to three"). Two standard deviations translates, approximately, into a one in twenty, or five percent (.05), chance²⁴ that a particular disparity is due to chance, and not to a particular factor such as sex. Consequently, statistical analysis revealing probability levels of .05 or smaller indicates larger numbers of standard deviations — suggesting that the reasons for a disparity are questionable. Put another way, the higher the number of standard deviations, the lower the probability (measured as a percentage), that the disparity was random. The Court notes, however, that this threshold measure of statistical significance is not an absolute, either for plaintiffs or defendants. See Watson v. Fort Worth Bank Trust, 487 U.S. 977, 995 n. 3, 108 S.Ct. 2777, 101 L.Ed.2d 827 (1988) ("We have emphasized the useful role that statistical methods can have in Title VII cases, but we have not suggested that any particular number of `standard deviations' can determine whether a plaintiff has made out a prima facie case in the complex area of employment discrimination."); accord Ottaviani, 875 F.2d at 373 (and cases cited therein). Rather, statistical analysis' "evidentiary value `depends on the magnitude of the disparity it reflects, the relevance of its supporting data, and other circumstances in the case supportive of or in rebuttal of a hypothesis of discrimination.'" Kincade, 694 F. Supp. at 377 (quoting E.E.O.C. v. Federal Reserve Bank of Richmond, 698 F.2d 633, 646-47 (4th Cir. 1983)).

24 In reality, a .05 probability margin equals 1.96 standard deviations.

In support of her allegations, Plaintiff submits the report of Paula England, Ph.D., a Professor of Sociology at the University of Arizona. Dr. England analyzed the data presented in a report issued in September 1994 by Vanderbilt's own Ad Hoc Committee to Study Gender Equity in Faculty Salaries in Vanderbilt University's College of Arts and Science.²⁵

25 The Court notes that its analysis of this portion of Plaintiff's case is somewhat constrained by the fact that neither Plaintiff nor Defendant submitted a copy of Vanderbilt's report to the Court.

Dr. England performed a regression analysis to assess whether sex played a role in the salaries of professors of varying ranks — Assistant Professor, Associate Professor (Plaintiff's rank), and full Professor — both at the College and within the Humanities departments in particular, for the 1993-94 and 1994-95 academic years. Dr. England adjusted her figures according to the legitimate factors of rank, the national average salary in one's discipline and rank, time since terminal degree, time at Vanderbilt, and time at current rank.

All told, Dr. England performed sixteen analyses. For the 1993-94 and 1994-95 academic years, at the Assistant Professor level²⁶ college-wide, she found women earning less than men by probability margins — or "gender coefficients" — of .16 and .14, respectively; within the Humanities, she found women earning less than men by a margin of .33 in 1993-94, but earning more than men by a margin of .89 in 1994-95. At the Associate Professor level²⁷ college-wide, Dr. England concluded that women earned less than men by probability margins of .30 and .28, for 1993-94 and 1994-95, respectively; among Humanities faculty, female Associate Professors made less than male colleagues by margins of .29 and .46, respectively. The disparities uncovered by Dr. England were greatest at the full Professor level: college-wide, women earned less than men by probability margins of .22 and .15 in 1993-94, respectively, and within the Humanities, women earned less than men by margins of .18 and .05 for each of the respective years.

26 This was Plaintiff's category for the 1993-94 academic year prior to her tenure denial.

27 This was Plaintiff's category for the 1994-95 year following her successful appeal of her tenure denial, and her consequent receipt of back-pay.

As is made clear by the statistics above, women generally earned less than men both at the College and within the Humanities specifically for each of the two years measured. However, only at the full Professor level in the Humanities, for the 1994-95 year, did the disparity reach the "statistically significant" level of .05; for all other years and at all other faculty levels — including Plaintiff's — the probability margins were greater than .05, resulting in standard deviations too small to reach the "statistically significant" threshold.²⁸

28 Dr. England also calculated the average standard deviations when data for each rank is pooled for the two measured years. Consequently, the standard deviations for all Full Professors, all Associate Professors, and all Assistant Professors for both years are .10, .21, and .09 respectively. For those ranks within the Humanities faculty, the standard deviations are .03, .32, and .84 respectively. Thus, the only statistically significant disparity is again at the full Professor level within the Humanities.

Dr. England also noted that controlling for years at current rank in performing a regression analysis in salaries may mask discrimination in salaries at the Associate and full Professor level because women at Vanderbilt generally take longer than men to be promoted from Associate to Full Professors.²⁹ Dr. England performed a second study, and found that the "gender coefficients" for years spent at the Associate Professor level, before promotion, were .18 and .28 for all faculty in the 1993-94 and 1994-95 years, respectively, and within the Humanities, the gender coefficients were .22 and .31, respectively. Again, the Court notes that these disparities do not meet the statistically significant probability threshold of .05.

29 The Court notes that Plaintiff does not allege that the length of time before she herself was considered for tenure — beyond the delays created through her tenure denial and the subsequent appeal — was unduly long or biased in any way.

In response to Dr. England's report, Defendant filed the report of David W. Peterson, Ph.D., the President of PRI Associates, Inc., a consulting firm providing statistical analyses for employment discrimination cases. Dr. Peterson's paramount observation is that — as indicated supra — none of Dr. England's findings, save one, reach the statistically significant level identified by the Supreme Court as most probative of discrimination. Dr. Peterson goes on to also question Dr. England's other — albeit statistically insignificant — findings of gender disparity by questioning her methodology.

A principal threshold concern of Dr. Peterson's is that Dr. England's comparisons of male and female faculty members' salaries are based upon incomplete information; consequently, he contends that Dr. England's statistics fail to actually compare women faculty with "similarly situated" male faculty. Dr. Peterson notes that Dr. England has omitted from her analysis most of the measures of professional achievement — such as research productivity, speaking engagements, participation in professional organizations and other service to the field, and success in obtaining grants — which contribute to disparities in salary. Dr. Peterson also suggests that salary levels in the university faculty context are especially decentralized, and are highly dependent upon the subjective value placed upon certain achievement measures — indeed, much like tenure decisions themselves. Consequently, argues Dr. Peterson, Dr. England's more formulaic approach, such as lumping departments together to obtain average salaries and comparing salaries at Vanderbilt with salaries at other universities, ignores other legitimate reasons for deviations, yielding inflated "gender coefficients."³⁰ For more accurate results, Dr. Peterson proposes a more individualized approach:

30 Dr. Peterson notes that the University itself was aware of these deficiencies when it published its salary Report, and noted them. He quotes the Report: "[I]n the absence of independent measures of teaching effectiveness, scholarly research contributions, or service, rank might be included to provide a rough indication of these otherwise unmeasured academic contributions." (Peterson Report at 4 n. 1.) The Report noted that it had not undertaken such an individualized analysis because "it is quite costly to do so, and standardizing across disciplines is difficult, if not impossible." (Id.)

At the first stage, one might identify people who are in the same department and rank, and display their salaries as scatter plots against their time in rank, their number of years at the University, and their number of years since achieving their highest degree. If female faculty are represented by symbols different from those used for males, these scatter plots depict the extent to which females differ in pay from males who have similar rank, experience and academic discipline. This picture . . . will show whether the female faculty are generally among the best or least well paid faculty with similar seniority within that rank and department, and they will also show whether there are individuals, male or female, who appear to be over or under paid. Someone intimately familiar with the relative qualifications of these individual faculty members then needs to pore over these charts to determine whether, in light of the qualifications not captured on the charts, the members' relative pay rates are appropriate.

(Peterson Report at 3.)³¹

31 In support of his conclusion that Dr. England's failure to incorporate more individualized performance information indeed leads to an inflated coefficient, Dr. Peterson makes another observation: When one looks at various "measurable" qualities of male and female faculty, such as faculty chairs, numbers of male versus female faculty at different ranks, and length of time on the Vanderbilt faculty, men generally outstrip women — 6.4% of male professors are department chairs, as opposed to 5.45% of women, for example, and male faculty have been at Vanderbilt an average of 15.6 years, as opposed to women's 6.3 average years. Consequently, Dr. Peterson posits that in light of these overall disparities, it is likely that the unmeasured qualities of professional and teaching achievement discussed above would also reflect higher average levels of achievement by male faculty, which would in turn legitimize higher pay.

Dr. Peterson also asserts that Dr. England's computational models are also deficient. In isolating the effect of gender on salary, Dr. England corrected for the legitimate factors of years since terminal degree, years of service at Vanderbilt, and years spent at current rank. Dr. Peterson notes, however, that Dr. England assumes that the percentage increase for each of these years is the same "whether the year is the first, the twenty-fifth, or any other year." (Id. at 4.) Moreover, Dr. Peterson argues that the interaction of these three factors also affects the percentage increase — for example, an Assistant Professor moving from her eighth to ninth year since terminal degree while moving from her third to fourth year at Vanderbilt might receive less of an increase than an Assistant Professor moving from her sixth to seventh year since terminal degree but moving from her fifth to sixth year at Vanderbilt, or vice versa.³²

32 Dr. Peterson also challenges Dr. England's pooling of the two measured years to obtain "average" probability margins. He argues that such a method contravenes generally accepted principles for calculating statistics:

Since the salary of a faculty member in one year is generally about the same as it is in the next, this mixing of the two files produces a jumble that is ill-suited to regression analysis, one of whose tenets is that each of the records in the file be independent of the others. Since this requirement is obviously not met by the mixed file, the connection between the fact situation and the regression model is severed, and there is nothing one can logically infer from the results.

(Peterson Report at 7.)

Dr. Peterson also challenges Dr. England's use of "one-tailed" probabilities as the yield of her regression analysis. Dr. Peterson asserts that in employment discrimination cases, standard deviations are commonly expressed in terms of "two-tailed" probabilities.³³ Thus, Dr. Peterson claims that Dr. England's standard deviation results should be doubled before being analyzed. After such doubling, he notes, the .05 gender coefficient for full Professors in the Humanities — the only statistically significant disparity uncovered by Dr. England's research — becomes a .10 gender coefficient, thereby eliminating its statistical significance altogether.

33 These terms will be discussed with greater specificity infra.

Finally, Dr. Peterson challenges the methodology underpinning Dr. England's analysis of tenure delays for women professors at Vanderbilt. First, he again argues that her study is based upon insufficient information because it only uses membership in a certain department as the criterion for promotion decisions, rather than the myriad professional, as well as subjective, factors which go into the tenure decision. Second, Dr. Peterson contends that Dr. England's data treats equally those promotion decisions made recently with those made many years ago, failing to take account of the fact that "[i]t is possible, even probable, that the standards for promotion or the timetable for promotion has changed over the past twenty or twenty five years, at least in some departments." (Peterson Report at 8.) Indeed, Dr. Peterson notes that because Dr. England's data covers just two years, it does not reflect tenure decisions made in the past (even relatively recent ones) for those faculty who may have left Vanderbilt. Dr. Peterson also argues that Dr. England's results are skewed because faculty who are about to be tenured are not included in her figures. Finally, Dr. Peterson again raises the inaccuracy of stating results in terms of one-tailed probabilities, indicating that the disparities found by Dr. England would be even less pronounced if two-tailed probabilities are used.

Based on the analyses provided by Drs. England and Peterson, the Court concludes that summary judgment on Plaintiff's wage discrimination claim is appropriate. The most compelling reason for granting summary judgment is that Plaintiff's statistics themselves, irrespective of Dr. Peterson's contentions regarding their incompleteness,³⁴ do not rise to the level of statistical significance which would indicate widespread discrimination. See Hazelwood, 433 U.S. at 307-08 ("Where gross statistical disparities can be shown, they alone may in a proper case constitute prima facie proof of a pattern or practice of discrimination.") (emphasis added). The Court is cognizant, and indeed has acknowledged supra, that the standard deviation model is not intended to be applied rigidly. However, in the absence of any other evidence of gender bias in salaries in this case — whether through anecdotes, critiques of Defendant's mechanisms for setting salaries, or other means — the Court has no choice but to rely solely on the statistics to find discrimination. See, e.g., Griffin v. Board of Regents, 795 F.2d 1281, 1293 (7th Cir. 1986) (in absence of any individual examples of disparate treatment in salary to "buttress a class claim," statistical evidence alone insufficient); cf. Coates, 756 F.2d at 539 ("[S]tatistics are so manipulable that some skepticism may be justified when set in the evidentiary context of the whole case."); Kincade v. Firestone Tire Rubber Co., 694 F. Supp. 368, 377 (M.D.Tenn. 1987) (value of statistical analysis as evidence of discrimination "depends on the magnitude of the disparity it reflects, the relevance of its supporting data, and other circumstances in the case supportive of or in rebuttal of a hypothesis of discrimination") (citation omitted).

34 The Court notes that its conclusion in this regard would be entirely different were Dr. England's calculations to yield standard deviations more suggestive of bias. While Dr. Peterson's critiques of Dr. England's studies are certainly persuasive, the Court believes that Defendant has not gone far enough to survive summary judgment in only pointing out the potential shortcomings in Plaintiff's calculations. The Supreme Court has held that a defendant must go beyond such criticism and actually demonstrate that its method of statistical analysis would yield statistics undermining the disparity shown by the plaintiff. Bazemore v. Friday, 478 U.S. 385, 403 n. 14, 106 S.Ct. 3000, 92 L.Ed.2d 315 (1986); accord Berger v. Iron Workers Reinforced Rodmen Local 201, 843 F.2d 1395, 1416 (D.C. Cir. 1988) ("Mere conjecture or general assertions of inadequacies in the opponent's statistical case, without demonstrating their effect on the results, will not suffice."); Sobel v. Yeshiva Univ., 839 F.2d 18, 34 (2d Cir. 1988) ("We read Bazemore to require a defendant challenging the validity of a multiple regression analysis to make a showing that the factors it contends ought to have been included would weaken the showing of a salary disparity made by the analysis."). Because Plaintiff's results so resoundingly fail to demonstrate statistical significance, however, summary judgment may be granted on those grounds.

Consequently, if the Court accepts the one-tailed probabilities presented by Plaintiff, it must conclude that only female full Professors in the Humanities are potentially affected by wage discrimination in a statistically significant way. Plaintiff's claim that she "can expect such discrimination in the future in the absence of injunctive relief" — that is, if she ultimately wins promotion to full Professor from her current Associate Professor position — is entirely unsupported and is simply too speculative a harm to preclude summary judgment in this case, at this time. The College Rules do not provide much guidance on the question of how far in the future Plaintiff would be considered for such a promotion, but it is clear that a full Professorship is hardly a foregone conclusion: "Although there is no time limit for this promotion nor even a requirement to make it, departments are responsible for regularly determining which members are eligible for a full external review." (College Rules at IV, E.) (Emphasis added.)

Moreover, the Court is skeptical of denying summary judgment solely on the basis of Dr. England's calculations because it is skeptical that the one-tailed analysis is even appropriate in this case. While Plaintiff provides a single statistical treatise citation in support of her endorsement of the one-tailed approach, she provides no further explanation nor any caselaw approving the use of one-tailed probabilities in the Title VII context; Defendant provides neither statistical nor legal support for the two-tailed approach. Nor has the Supreme Court has offered any guidance on the matter. However, the Court's research indicates that "one-tailed analysis tests whether a group is disfavored in hiring decisions while two-tailed analysis tests whether the group is preferred or disfavored." Hartman v. Duffey, 88 F.3d 1232, 1238 (D.C. Cir. 1996).³⁵ Put another way, one-tailed analysis is more appropriate "if the possibility of intentional discrimination favoring the protected group represented by plaintiff [e.g., women in this case] can be ruled out as defying logic, i.e., the available evidence excluding the statistic in question gives strong support to the conclusion that the system is either nondiscriminatory or disadvantageous to the plaintiff's group." D. BALDUS J. COLE, STATISTICAL PROOF OF DISCRIMINATION 129-30 (1986 Cumulative Supp.) (quoted in Palmer v. Shultz, 815 F.2d 84, 95 (D.C. Cir. 1987)) (emphasis added). The end result, in terms of the two tests' statistical conclusions, is that the one-tailed test makes it easier to claim "statistically significant" evidence of discrimination: while a probability level of .05 under two-tailed analysis translates into 1.96 standard deviations, under one-tailed analysis the .05 statistically significant threshold becomes just 1.65 standard deviations.³⁶

35 In light of the failure of either of the parties to enlighten the Court about the distinctions between one- and two-tailed tests, the Court relies heavily upon the excellent explanation of these methodologies provided in Palmer v. Shultz, 815 F.2d 84, 92-97 (D.C. Cir. 1987).

36 Indeed, at least one appeals court, after an examination of the statistical literature, as well as the transcript a statistical expert's testimony in the district court below, concluded that the use of the one-tailed test is most accurately seen as "data mining" designed to "beef up" statistics of discrimination, making it acceptable only in those Title VII cases where "independent evidence indicates the presence of discrimination of the type being challenged." E.E.O.C. v. Federal Reserve Bank of Richmond, 698 F.2d 633, 655, 656 (4th Cir. 1983). The Court notes, however, that Federal Reserve Bank received some criticism from two leading statisticians as being "unnecessarily strict." See D. BALDUS J. COLE, STATISTICAL PROOF OF DISCRIMINATION 129 (1986 Cumulative Supp.) (footnote omitted) (quoted in Palmer, 815 F.2d at 95).

Given the fact that two-tailed analysis was the methodology approved by the Supreme Court in the seminal statistics-as-evidence-of-discrimination cases, Castaneda and Hazelwood, and given the absence of any argument on this point by the parties, the Court is inclined to reject Plaintiffs assertion that "[i]n discrimination cases, the issue is always whether one group is favored over another," and consequently, that a one-tailed test is appropriate because "the only question of interest is the likelihood of a difference in one direction, i.e., when only a positive disparity between the two numbers is being tested." (Pl.'s Resp. Opp. Def.'s Mot. Summ. J. at 39.) Rather, the Court follows the ultimate conclusion of the D.C. Circuit in Palmer v. Shultz:

[A]lthough we by no means intend entirely to foreclose the use of one-tailed tests, we think that generally two-tailed tests are more appropriate in Title VII cases. After all, the hypothesis to be tested in any disparate treatment claim should generally be that the selection process treated men and women equally, not that the selection process treated women at least as well as or better than men. Two-tailed tests are used where the hypothesis to be rejected is that certain proportions are equal and not that one proportion is equal to or greater than the other proportion.

Palmer, 815 F.2d at 95-96 (citing W. CURTIS, STATISTICAL CONCEPTS FOR ATTORNEYS 119-22, 133-37 (1983)); accord Hartman, 88 F.3d at 1238.

Thus, the Court finds it is appropriate to double Dr. England's gender coefficients — as suggested by Dr. Peterson — to yield two-tailed probabilities. This calculation results in probabilities of discrimination ranging from .92 to .10 — in other words, no statistically significant disparities between the salaries of male and female faculty. A similar result obtains in relation to Dr. England's calculations of delays in tenure awards for women faculty, yielding probabilities ranging from .36 to .62, all well beyond the .05 guideline established by the courts.

C. State law claims

Given the Court's finding that Plaintiff's federal claims of sex and national origin discrimination are insufficient to withstand summary judgment, Plaintiff's only remaining claims are those brought under state law: her breach of contract and of covenant of good faith and fair dealing claims, and her Tennessee Human Rights Act discrimination claim. Because the Court no longer retains jurisdiction over these claims,³⁷ they are dismissed, although Plaintiff is free to refile them in state court.

37 Under 28 U.S.C. § 1367(c)(3), federal district courts may decline to exercise supplemental jurisdiction over related state claims if "the district court has dismissed all claims over which it has original jurisdiction."

IV. Conclusion

In light of the foregoing findings, Defendant Vanderbilt University's Motion for Summary Judgment is granted, and Plaintiff's remaining state law claims are dismissed. An Order consistent with the findings contained herein is filed contemporaneously.