Cooper v. University of Texas at Dallas , 482 F. Supp. 187 ( 1979 )

482 F. Supp. 187 (1979)

Mary Weis COOPER
v.
UNIVERSITY OF TEXAS AT DALLAS.

Civ. A. No. CA-3-75-1510-G.

United States District Court, N. D. Texas, Dallas Division.

December 10, 1979.

Steven B. Thorpe, James C. Barber, Martin Frost, Law Offices of James C. Barber, Dallas, Tex., for plaintiff.

John Hill, Atty. Gen., Lonny Zwiener, Martha Allan, Asst. Attys. Gen., Austin, Tex., for defendant.

FINDINGS OF FACT AND CONCLUSIONS OF LAW

PATRICK E. HIGGINBOTHAM, District Judge.

This is a sex discrimination case brought pursuant to 42 U.S.C. § 2000e et seq. Plaintiff contends that she, as well as the class of female applicants and employees whom she claims to represent, have been subjected to discriminatory hiring practices by the University of Texas at Dallas ("UTD"). Dr. Cooper argues that although the hiring "system" may be on its face neutral, it is not in fact neutral in operation or effect; she claims that decisions are made subjectively by a predominantly male group, and have had a disproportionate and illegal impact on females. In addition to attacking hiring practices, Dr. Cooper alleges that UTD has discriminated against females in promotions and salary.

Dr. Cooper first applied to be a UTD professor in 1971. Her application was on file when in May, 1973, UTD hired a male as an assistant professor in Operations Research, a position in UTD's School of Management and Administrative Sciences for which she claims she was fully qualified. Thereafter, UTD filled nine additional positions. Dr. Cooper claims these positions were in her area of competence or in closely related areas. UTD told Dr. Cooper of her rejection on March 4, 1975, after which she timely filed a charge with the EEOC and her suit in this court.

As a class representative, Dr. Cooper has not established that sex discrimination played a part in the hiring, promotion, or salary practices of UTD. As to her individual claim, Dr. Cooper only demonstrated that she was a victim of a curtailed budget and bad luck; she did not establish that UTD discriminated against her on the basis of her sex.

I. The Scope of the Class

On January 25, 1977, plaintiff filed her motion to certify the class in this case as follows:

All past female applicants for full-time teaching positions at the University of Texas at Dallas who were rejected, all present full-time female faculty members at the University of Texas at Dallas, and all future female applicants for full-time teaching positions.

This court certified the following class on April 8, 1977:

All female applicants for full-time teaching positions at the University of Texas at Dallas who were rejected, and all present full-time faculty members at the University of Texas at Dallas.

Thereafter, the court modified the scope of the class to include only:

All female applicants for full-time teaching positions at the University of Texas at Dallas who were rejected after March 24, 1972.^[1]

Both parties filed motions to amend the class. Plaintiff moved that the class include employees as well as applicants; defendant moved that the class not include those female applicants who were rejected for employment more than 180 days before March 7, 1975, the date that Dr. Cooper filed her EEOC charge. Because of the relatively small faculty size, the still smaller number of female faculty members, and the overlap of proof of class and liability issues, the court carried those motions. Recognizing its continuing duty to monitor and modify the class according to the facts that develop, the court permitted plaintiff to introduce at trial evidence on not only the certified class, but also on plaintiff's proposed class. Before turning to the merits, the court will first address the question of the scope of the class.

A. The Class Cut-Off.

Defendant moved to limit the class to exclude those females who were hired more than 180 days before Dr. Cooper filed her EEOC complaint. It relies on cases holding that a class representative cannot represent those who, because of the statute of limitations, could not have filed charges with the EEOC at the time the representative filed her charge. Those cases are inapposite. The discriminatory acts of which plaintiff here complains—hiring, promotion, and salary — constitute continuing violations of Title VII. In Wetzel v. Liberty Mutual Insurance Co., 508 F.2d 239, 246 (3rd Cir.), cert. denied, 421 U.S. 1011, 95 S. Ct. 2415, 44 L. Ed. 2d 679 (1975), the court held that the 180-day class cut-off only applied to class members complaining of discrete acts of an employer such as a termination. The court stated:

[The representatives'] charges alleged the maintenance of discriminatory hiring and promotion policies up to the time of the filing of their complaint. Such policies are continuing violations of Title VII and would allow a filing of a charge at any time . . .. The only employees barred from the class are those who left the employ of the Company more than [180] days before the filing of the charges with the EEOC by [the representative]. Id.

Accord, Laffey v. Northwest Airlines, Inc., 185 U.S.App.D.C. 322, 365-367, 567 F.2d 429, 472-74 (D.C.Cir. 1976), cert. denied, 434 U.S. 1086, 98 S. Ct. 1281, 55 L. Ed. 2d 792 (1978); In re Consolidated Pretrial Proceedings, 582 F.2d 1142 (7th Cir. 1978), petition for cert. docketed sub nom. Zipes v. Trans-World Airlines, Inc. and Trans-World Airlines, Inc. v. Zipes, 442 U.S. 916, 99 S. Ct. 2834, 61 L. Ed. 2d 282 (1979).

Defendant's motion to limit the class is denied.

B. The Proposed Employee Class.

1. Standing. Dr. Cooper has moved this court to expand the class to include employees. UTD contends that Article III of the United States Constitution precludes this court from such redefinition. Essentially it argues that a class representative only has standing to raise those class claims that she would be able to assert individually. In other words, it submits that the class device cannot extend the court's jurisdiction to claims of class members; that the named plaintiff and the class members must assert the same claims. UTD's contention is overly broad and unsound. Its overbreadth stems from a confusion of the requirements of Article III and Rule 23. This court has already addressed this issue in depth in Vuyanich v. Republic National Bank of Dallas, 82 F.R.D. 420 (N.D.Tex.1979). In that case the court held that a class representative has standing if she alleges that she as an individual has suffered a concrete injury. If she passes this threshold constitutional test, then the next question — whether she may represent the class members — is not in the first instance one of standing. Instead, the court must first scrutinize the relationship between the representative and the class members to determine whether it satisfies the requirements of Rule 23. A not insubstantial argument can be made that as the relationship between a class representative and the class members becomes increasingly drawn out, Article III concerns start to rise. Even if this is so, this concern provides only a backdrop to Rule 23 construction and, moreover, is only a prudential limitation of Article III because the class representative at the least alleges an injury in fact. At the same time, Rule 23 should not be read in a vacuum. Rather, it must be viewed against the nature and purpose of the substantive law underlying the claims at issue. In an employment discrimination case, the court must apply Rule 23 in such a way as to fulfill Congress' purposes in enacting Title VII of the Civil Rights Act of 1974, 42 U.S.C. § 2000e-5.

Congress intended Title VII to be a broad, remedial measure aimed at eliminating the widespread and shameful discriminatory employment conditions that have a destructive impact not only on those "forced into a condition of marginal existence" but also on the nation as a whole deprived of the talents and creativity of the victims of discrimination. [1964] U.S.Code Cong. & Admin.News, pp. 2355, 2514 (1964); see Id. at 2515, 2516; House Rep. No. 914, Id. at 2393; Sen.Rep. No. 872, Id. at 2377. This court must read Rule 23's standards with sufficient breadth to carry out Congress' lofty goal. If courts proceed immediately to address class actions in the context of judicially crafted prudential limitations of Article III rather than Rule 23, they risk substituting their own concerns with judicial access and overcrowded courts for Congress' objectives. Believing that risk too great, the court now turns to Rule 23.

2. Rule 23. It is well established that in order for a suit to be maintained as a class action, each of the four requirements specified in Rule 23(a) must be satisfied. If any one of the four tests is not met, the court must refuse to certify the class. See Huff v. N. D. Cass Co. of Alabama, 485 F.2d 710, 712 (5th Cir. 1973); Eisen v. Carlisle & Jacquelin, 391 F.2d 555, 556 (2d Cir. 1968).

(i) Commonality. Many courts have glossed over the question of whether common questions exist under Rule 23(a)(2). See 3B Moore's Federal Practice ¶ 23.06-1, at 23-171 (2d ed. 1979). Where courts have considered the issue, they have found that Rule 23(a)(2) does not require that all questions of law and fact be identical, but rather that a question of law or fact be presented that inheres in the complaints of all the class members. See Escalera v. New York City Housing Authority, 425 F.2d 853 (2d Cir.), cert. denied, 400 U.S. 853, 91 S. Ct. 54, 27 L. Ed. 2d 91 (1970); Norwalk CORE v. Norwalk Redevelopment Agency, 395 F.2d 920, 937 (2d Cir. 1968).

It has already been determined that common questions exist among the certified applicant class members. Plaintiff relies heavily upon statistics to demonstrate that employees also assert questions in common with applicants. First, plaintiff introduced statistics on the percentage of tenured male and female professors at UTD as compared to the national average over a period from 1972 to 1978. In "Category I" institutions which awarded at least 15 Ph.D.'s in three unrelated disciplines on the average each year for the last three years, 66% of the males and 41% of the females had tenure.^[2] In "Category IIA" institutions — all institutions offering graduate degrees not included in Category I — 69% of the males and 52% of the females held tenured positions.^[3] Fifty-eight percent of the males and 41% of the females in colleges and universities offering only the BA degree, or "Category IIB," were tenured. In "Category III" schools, or two-year institutions with academic ranks, 61% of the males and 48% of the females had tenure. At UTD, 48% of the males and 20% of the females were tenured. Plaintiff's statistical expert, Dr. William Roger Schucany, testified that the percentage of female faculty at UTD that were tenured was lower than the national average and that the probability of this disparity occurring by chance alone was .029, which was statistically significant at the .05 level.^[4]

At this point, it should be emphasized that in determining the scope of the class, the court must address the issue of whether there are common questions and not whether the defendant is actually guilty of discrimination. A plaintiff, of course, must do more to demonstrate the existence of the question than simply assert its existence. At the same time, a plaintiff need not make out a prima facie case of liability. Where between these poles a plaintiff must place his proof has not yet been articulated. Suffice it to say that plaintiff's statistical evidence on tenure demonstrated that class questions exist as to whether the defendant's practices, including promotion and hiring, had a grossly disparate impact to the injury of the members of the class.^[5] This question and defendant's rebuttal evidence are considered in the latter part of this opinion.

(ii) Typicality. Rule 23(a)(3) focuses the court's attention on the nature of the representative's claims. It requires a determination of whether, given the fact that the representative raises some question of law or fact that is common to the class, the remaining issues and possible defenses are also typical.

This court explored typicality in Vuyanich v. Republic National Bank of Dallas, supra, at 432-33. It found that the purpose of Title VII and the Fifth Circuit's repeated endorsement of across-the-board representation mandated the conclusion that a plaintiff complaining of one employment practice can represent an employee complaining of a different employment practice. See Satterwhite v. City of Greenville, 578 F.2d 987, 994 n. 8 (5th Cir. 1978), petition for cert. docketed, 439 U.S. 834, 99 S. Ct. 114, 58 L. Ed. 2d 129 (1979); Payne v. Travenol Laboratories, Inc., 565 F.2d 895, 900 (5th Cir.), cert. denied, 439 U.S. 835, 99 S. Ct. 118, 58 L. Ed. 2d 131 (1978). The court in Satterwhite stated, "It is not necessary that the representative suffer discrimination in the same way as other class members, but it is necessary that she suffer from discrimination in some respect." 578 F.2d at 994 n. 8. Under this reasoning, the fact that plaintiff complains of hiring practices while certain class members complain of tenure practices does not render Dr. Cooper's claim atypical.

(iii) Joinder Impracticable. The court has already determined that the certified class of applicants "is so numerous that joinder of all members is impracticable." Rule 23(a)(1). The addition of 25 employees, some of whom are no longer employed at UTD and have moved away, only reinforces the court's earlier conclusion that Rule 23(a)(1) is satisfied.

(iv) Adequacy. Adequacy of class representation is crucial to assure that the absent class members are not deprived of due process and that the final judgment is binding. A court, in assessing a representative's adequacy, must focus initially on the interrelated questions of attorney competence and conflicts of interest. In this case, counsel for the plaintiff and class is diligent, thorough, and skilled in Title VII litigation. Further, the evidence did not reveal the existence of any conflicts between counsel and the certified or proposed class members, or among the class members themselves. Additionally, in viewing the case as a whole, the court does not find that the plaintiff lacked either the incentive to prosecute or the ability to represent members of the certified or proposed class.^[6] Thus, plaintiff is an adequate representative.^[7]

Because all four requirements of Rule 23 have been satisfied, the court grants plaintiff's motion to redefine the class to include employees. It now addresses the merits of the case.

II. The Merits of the Class Claim

A. The Employee Class Members.

Since the passage of the Civil Rights Act, courts have often relied on statistical evidence as proof of a violation. Indeed "``[i]n many cases the only available avenue of proof is the use of racial [or sex] statistics to uncover clandestine and covert discrimination by the employer or union involved.' United States v. Ironworkers Local 86, 443 F.2d [544], at 551 [CA 9]. See also, e. g., Pettway v. American Cast Iron Pipe Co., 494 F.2d 211, 225 n. 34 (CA 5); Brown v. Gaston County Dyeing Machine Co., 457 F.2d 1377, 1382 (CA 4) . . .." International Brotherhood of Teamsters v. United States, 431 U.S. 324, 340 n. 20, 97 S. Ct. 1843, 1857 n. 20, 52 L. Ed. 2d 396 (1977). See also Senter v. General Motors Corp., 532 F.2d 511, 527 (6th Cir.), cert. denied, 429 U.S. 870, 97 S. Ct. 182, 50 L. Ed. 2d 150 (1976). Statistics have considerable probative force because "absent explanation, it is ordinarily to be expected that nondiscriminatory hiring practices will in time result in a work force more or less representative of the racial and ethnic composition of the population in the community from which the employees are hired." Hazelwood School District v. United States, 433 U.S. 299, 307, 97 S. Ct. 2736, 2741, 53 L. Ed. 2d 768 (1977), quoting International Brotherhood of Teamsters v. United States, supra, 431 U.S. at 340 n. 20, 97 S. Ct. 1843. See also Castaneda v. Partida, 430 U.S. 482, 496-97 n. 17, 97 S. Ct. 1272, 51 L. Ed. 2d 498 (1977). The Supreme Court thus has held that strong statistical proof may be sufficient to establish a prima facie case both in jury selection and in employment discrimination cases. See, e. g., Hazelwood School District v. United States, supra, 433 U.S. at 307, 97 S. Ct. 2736; Turner v. Fouche, 396 U.S. 346, 90 S. Ct. 532, 24 L. Ed. 2d 567 (1970); International Brotherhood of Teamsters v. United States, supra, 431 U.S. at 399, 97 S. Ct. 1843.

The court in Castaneda v. Partida set forth the steps through which the parties must proceed in order to establish and rebut a prima facie statistical case. Although Castaneda is an equal protection jury selection case, its analysis is apposite in the present case. Applying Castaneda to the promotion issue in this case, the plaintiff first must compare the proportion of females in the relevant labor market to the proportion of females actually tenured by UTD. The idea behind this "rule of exclusion" is simple — "[i]f a disparity is sufficiently large, then it is unlikely that it is due solely to chance or accident, and, in the absence of evidence to the contrary, one must conclude that [sexually discriminatory] factors entered into the selection process." Id., 430 U.S. at 494 n. 13, 97 S. Ct. at 1280 n. 13. If the plaintiff adduces evidence of "gross statistical disparities," that alone may constitute a prima facie case. Hazelwood School District v. United States, supra, 433 U.S. at 307, 97 S. Ct. 2736. To support the statistical showing, the plaintiff may also introduce evidence of a selection process that is susceptible of abuse.

A finding of statistical significance merely rejects the null hypothesis — that only random chance is operating. It does not necessarily establish the counter hypothesis—in this case, that UTD discriminates against women in promotions. If plaintiff's evidence is strong enough to establish a prima facie case, defendant must establish that factors other than the counter hypothesis influenced the result. If the defendant fails to meet this explanatory burden, however, the court should accept the counter hypothesis as established. Otherwise stated, a defendant against whom a prima facie case is established must meet its burden of going forward with evidence. Williams v. Tallahassee Motors, Inc., 607 F.2d 689, 692 (5th Cir. 1979).

As discussed above, Dr. Cooper introduced evidence demonstrating that the percentage of women at UTD who were tenured differs from the percentage of women at other institutions who are tenured. To reiterate, at Category I, IIA, IIB, and III institutions, the percentage of females who received tenure were respectively as follows: 41%; 52%; 41%; and 48%. At UTD, however, only five out of 25, or 20%, of the female faculty were tenured. As a Category IIA institution, UTD would be expected to have 52%, or 13, of its female faculty in tenured positions. The probability of five females receiving tenure out of 25 is .029. That figure is statistically significant at the .05 level. With a test of statistical significance we learn the probability that the observed value could have happened by chance, i. e., the probability that in a random sample of an appropriate test population the variable would exhibit a value as extreme as that observed. A test of statistical significance is thus itself based on the results of a hypothetical experiment. We suppose that an infinite number of samples of the same size are drawn from the test population. The probability of the observed value occurring by chance is equal to the proportion of the samples in which the value is at least extreme as the observed value. It has become a convention in social science to accept as statistically significant values which have a probability of occurring by chance 5% of the time or less. See generally N. Nie, C. Hull, J. Jenkins, K. Steinbrenner, & D. Bent, Statistical Package for the Social Sciences 222 (2d ed. 1975).

The Supreme Court has not stated whether the .05 level of statistical significance is sufficient to establish a prima facie case. It has only stated that "gross disparities" must be shown. Hazelwood School District v. United States, supra, 433 U.S. at 316, 97 S. Ct. 2736. See Williams v. Tallahassee Motors, Inc., supra, at 692. Assuming arguendo that plaintiff's statistics have established a prima facie case [and also assuming that UTD's tenure decisions would be indicated by Rowe v. General Motors Corp., 457 F.2d 348 (5th Cir. 1972)], we turn now to UTD's rebuttal to determine the extent to which factors other than the counter hypothesis—that UTD discriminates against women in promotions — resulted in UTD's tenure pattern.

Defendant's evidence convincingly refuted any inference that UTD's tenure decisions are infected with illegal considerations of sex. UTD is a young university; it was officially designated a university and a part of the University of Texas system on September 1, 1969. Its predecessor was the Southwest Center for Advanced Studies. The Southwest Center was a nonprofit, privately-endowed highly specialized scientific research center with no academic program. It awarded no degrees although it occasionally gave a Ph.D. student of another institution laboratory work which his own degree-granting institution credited to his degree. When the Southwest Center became UTD, UTD inherited the science-oriented staff, virtually all of whom received tenure. That entire staff had only one female employee. Because UTD inherited its core staff from a male-dominated faculty, it was inevitable that at least in the short run, a proportionately higher percentage of male professors at UTD would be tenured than female professors. UTD is still aging out of this short run period with many of its female personnel lacking sufficient longevity to achieve tenure. Any meaningful comparison of races to tenure must assume similar starting points. At UTD, females started later than the groups to which they are compared. There is no contention that UTD's later starting point is illegal.

UTD also points out that any statistical significance of plaintiff's figures is explained by UTD's unique program that draws from male-dominated disciplines. UTD is intended to be a quality research institution with a strong emphasis on graduate programs suited to the Dallas-Fort Worth business and industrial community. Only recently has UTD begun to offer the last two years of undergraduate study in addition to its graduate programs. UTD persuasively argued that these characteristics have contributed to its faculty make-up. Unlike many of the institutions to which UTD was compared in plaintiff's statistics, UTD does not offer study in disciplines with historically higher female participation such as nursing and home economics programs. This is not to typecast females by discipline; it is to say that UTD's employment force can only reflect the makeup of the population of qualified potential job holders. No statistical significance using the true relevant labor market has been shown. Indeed, it is thus questionable whether the prima facie case has been established. As Justice Stewart observed in Hazelwood: ". . . What the hiring figures prove obviously depends upon the figures to which they are compared.. ." 433 U.S. at 310, 97 S.Ct. at 2743.

Plaintiff, to bolster her equivocal statistics and to establish that UTD discriminates against women in granting tenure, argues that UTD uses vague and subjective standards for promotion which are implemented by predominately male committees. Dr. Alex Clark, UTD's academic vice president, testified that tenure decisions at UTD are based on teaching ability, on research potential or productivity, and on service to the community, and that the recommendation of a university official such as a department chairperson on a committee is the most important element in the decision-making process. Because evidence of such subjective standards is on its face neutral, it is meaningless when viewed alone. It could only be probative when viewed against statistically significant evidence that suggests that the null hypothesis — here, that at UTD women are just as likely to receive tenure as they are at other institutions — is invalid. Initially, it bears emphasis that the court has found that the one statistically significant result was attributable to nondiscriminatory factors such as UTD's unusual history and programs. Beyond that, however, it is inescapable that at institutes of higher learning such as UTD, certain subjective judgments must be made in promoting. The qualities of a good university professor simply cannot be reduced to a rigid formula to be applied uniformly to each candidate. This observation does not apply only to universities and is not intended to carve a special exception for them. It is made simply to emphasize that subjectivity is not in itself illegal. While obviously presenting a potential for discrimination, subjective evaluations still have an important role in decision-making processes. This is another way of saying that we know that there are prediction factors which as yet we have not quantified. Of course such subjectivity must withstand the test of nondiscriminatory impact. In this case, defendant passes that test.

B. The Applicant Class Members.

Plaintiff introduced certain statistical evidence concerning defendant's female hires from March, 1972, through 1977. The statistics were broken down by schools and compared UTD's hires to "national availability" figures, based on a survey of recent doctorates in 1975. The evidence was as follows:

%FEMALES          TOTAL UTD          ACTUAL       EXPECTED
UTD SCHOOL           AVAILABLE[8]   HIRES, '72-'77    FEMALE HIRES   FEMALE HIRES
Arts & Humanities       38.3             48                14            18.38
Human Development       38.5             32                12            12.32
Management and
Administration         4.3             26                 0             1.12
Natural Science
and Mathematics       13.8             38                 1             5.24
Social Sciences         20.9             34                 6             7.11

Statistical analysis anticipates some variation from the precise expected result in any given random sampling. The chi-square test which plaintiff's expert used on the above data compares actual or observed numerical distribution with a theoretical or expected distribution. The statistical model defines the degree of variation that is predicted if only random chance is operating and the measure of the predicted fluctuations from the expected result is the standard deviation. The Supreme Court has stated that if the difference between the expected value and the observed number is greater than two or three standard deviations, then the null hypothesis would be suspect. Hazelwood School District v. United States, supra, at 309 n. 14, 97 S. Ct. 2736.

Plaintiff's own expert concluded that with the exception of one school — Natural Science and Mathematics — the results as to hiring were not statistically significant. In Natural Science and Mathematics, however, the result was .025; that is, the probability that only one female would have been hired in Natural Science and Mathematics if chance alone were at work was 25 out of 1,000. This is significant at the .05 level.^[9] The standard deviation for that school was 2.13; that is, the figure of actual hires could deviate from the expected value as much as 2.13 in either direction and a significant probability would continue to exist that the observed result would be attributable to chance. The number of actual hires was -1.99 standard deviations from the expected.^[10] Plaintiff's statistician also testified that while it is true that when the schools are viewed separately, the level of hiring is statistically significant only in the School of Natural Science and Mathematics, the fact that defendant's hiring was below the expected when all schools are viewed as a whole is statistically significant at the .032 level.

Defendant challenges plaintiff's figures on two grounds. First, UTD contends that plaintiff's proposed market is improper and that the market should consist only of females who actually applied to UTD for a position. Second, even assuming plaintiff's market is proper, defendant contends that plaintiff's statistics suffer from other weaknesses that undermine their probative value. The court need not resolve the question of whether defendant's proposed alternative market is proper.^[11] Instead, the court finds that because of the flaws in plaintiff's proposed market, plaintiff failed to establish a prima facie case.^[12]

The use of the national availability market is based on an unproven assumption. Plaintiff's expert's use of nationally available new Ph.D.'s assumed that newly graduated Ph.D.'s were able to meet UTD experience qualifications (not attacked) for all positions — senior as well as junior. The testimony of Drs. Clark and Lutz established that in fact newly graduated Ph.D.'s could only fill a small proportion of the positions for which plaintiff's figures assumed them to be available. UTD was hiring not only assistant professors with little or no teaching experience, but also professors with some years of teaching and research behind them who could guide the more advanced students in their work.

It should be noted that defendant's rebuttal of plaintiff's statistical showing as to applicants was qualitatively different from its rebuttal as to employees. With respect to employees, defendant successfully demonstrated that statistically significant results were not due to discrimination but to neutral factors such as the youth of the university. In contrast, defendant's approach to plaintiff's hiring statistics was not to offer some other external factor as an explanation for the results, but to reveal the weakness in plaintiff's model and methodology, thus challenging plaintiff's prima facie case in a more direct manner. Indeed, it may be argued that the flaws in plaintiff's analysis were sufficiently severe that no prima facie case was ever established.

Plaintiff tried to bolster her statistics by pointing to the subjective hiring standards which are essentially the same as the standards for promotion. The court's observations about the promotion standards also apply to hiring. Subjectivity is not in itself illegal and in a situation such as a university, is probably inevitable. Absent an unexplained showing of discriminatory effect, evidence of subjective standards does not provide enough help to plaintiff's case. Plaintiff's related contention that a number of women who were rejected were equally or better qualified than the males selected is also unpersuasive. The court finds that plaintiff failed to establish that sex was a factor in these decisions.^[13]

Plaintiff further asserts that the facially neutral Ph.D. requirement works disproportionately to exclude women from employment as professors at UTD. The record in this case reflects that the plaintiff, after raising this argument, devoted little attention to its development at trial. Counsel for the class, a sophisticated advocate, undoubtedly anticipated a finding of job-relatedness with respect to the Ph.D. requirement. The market that plaintiff claimed is relevant for comparison to actual number of hires is the market of recent Ph.D.'s. Plaintiff did not come forward with statistical calculations based on a market of non-Ph.D.'s or a combined market of Ph.D.'s and non-Ph.D.'s. Plaintiff did introduce certain raw data on the numbers of men and women who earned bachelor's and master's degrees in 1975. These figures were broken down by discipline and were not related to UTD's more inclusive schools. Furthermore, plaintiff performed no statistical calculations on this data to enable the court to assess the impact of the Ph.D. requirement even where there is a significant disparity between the proportions of female MA's and of female Ph.D.'s.

Not having introduced such evidence, plaintiff failed to meet her initial burden of demonstrating that the facially neutral Ph.D. requirement "select[s] applicants for hire in a significantly discriminatory pattern." Dothard v. Rawlinson, 433 U.S. 321, 329, 97 S. Ct. 2720, 2726, 2727, 53 L. Ed. 2d 786 (1977). It is not enough that plaintiff merely introduce statistics indicating discriminatory hiring and other evidence that a facially neutral requirement exists. There must also be evidence causally linking the requirement to the adverse impact of females. Thus in Griggs v. Duke Power Co., 401 U.S. 424, 91 S. Ct. 849, 28 L. Ed. 2d 158 (1971), a high school diploma requirement and aptitude tests were shown to disproportionately exclude blacks. Similarly, in Albemarle Paper Co. v. Moody, 422 U.S. 405, 95 S. Ct. 2362, 45 L. Ed. 2d 280 (1975) the court held that the defendant has the burden of showing job relatedness only after the plaintiff shows that the tests in question selects applicants for hire or promotion in a significantly discriminatory pattern. Plaintiff in this case made no such showing.

Because of the lack of evidence presented on the Ph.D. issue, those applicants and employees, if any, who lack Ph. D.'s were not adequately represented. Thus that portion of the class consisting of applicants and employees not holding Ph.D. degrees is hereby decertified. Plaintiff's evidence failed to establish that UTD discriminates against women in hiring or promotions. All class claims therefore are DENIED.

III. The Individual Claim

In 1973 when Dr. Raymond Lutz became dean of UTD's School of Management and Administrative Science, he had no desk, no phone, and no faculty. It was his job to develop a curriculum and build a faculty for his school's degree program within the confines of an uncertain budget. He testified that the School of Management and Administrative Science was to include such disciplines as Operations Research, Information Systems, Finance, and Accounting. Dr. Lutz filled the first position in 1973 with Dr. Klaus Truemper, a new Ph.D. who was expected to teach Operations Research. Operations Research was also Dr. Cooper's area of expertise. Dr. Lutz testified convincingly that he and other members of the UTD administration considered Dr. Cooper a serious candidate and were interested in hiring her. Very shortly after hiring Dr. Truemper, however, Dr. Lutz was informed that he could make no more job offers for an indefinite period because of a budget cut. That Dr. Truemper was hired rather than the plaintiff was simply a matter of timing. Lutz testified that when he extended the offer to Dr. Truemper, he believed he had another position available for Dr. Cooper and would have hired her in 1973 had his budget not been unexpectedly restricted.

When the budgetary restriction began to be lifted, Dr. Lutz gradually tried to fill positions with faculty qualified to teach the numerous courses listed in the catalogue. He was looking for someone to teach in each area. Dr. Larry J. Merville was hired, for example, to teach Finance, Dr. Lawrence Madeo to teach Information Systems, and Dr. David Raddock to teach International Management Studies — not areas of Dr. Cooper's expertise. The Operations Research position had already been filled by Dr. Truemper.

Finally, in 1975, Dr. Lutz was able to hire a second person in the Operations Research area. He needed a senior professor who would be able to take over the Ph.D. programs, and who had experience in guiding dissertations. Dr. Chandrasekarian, an assistant and associate professor at Case-Western University for six years, had guided 13 Ph.D. dissertations and had been on the committees of 14 others. On the other hand, Dr. Lutz had no assurance that Dr. Cooper's dissertation would even be published. She had worked for several years at LTV and had published one article which Dr. Lutz testified was not as rigorous as it might have been. He added that her overall work "output" since receiving her Ph.D. was not as good as it should have been. In short, her experience was not comparable to Dr. Chandrasekarian's. She was simply not as qualified.

It is true that Dr. Chandrasekarian had been Dr. Truemper's Ph.D. instructor, that Dr. Chandrasekarian was looking for a job because he had not received tenure at Case-Western, and that Dr. Truemper strongly recommended Dr. Chandrasekarian. Plaintiff suggests that she was not hired because of this relationship and not because Dr. Chandrasekarian was more qualified than she. Even if this were true, however, it would not establish that Dr. Cooper was a victim of sex discrimination.

Plaintiff also contends that UTD discriminated against her by hiring Dr. Carillo instead of her. Dr. Carillo, however, was hired primarily to teach Production and, secondarily, to bolster the International Management faculty, although he also taught some Operations Research. Dr. Lutz testified that Dr. Cooper was not qualified in Production.

Finally, plaintiff points to the hiring of a graduate student in a lecturer's position as an example of discrimination against her. Dr. Lutz testified that the graduate student was already at UTD when it was discovered that a lecturer would be needed to handle the unexpectedly high enrollment. He further testified that it was common practice to hire available graduate students in order to give them some financial assistance and teaching experience. Additionally, the graduate student was only hired on a part-time basis. Dr. Lutz testified that he believed Dr. Cooper was applying for a full-time faculty position and stated that he did not review any outsider's applications for this job.

Defendant successfully rebutted any inference that after plaintiff was rejected the defendant hired another person of her qualifications at UTD. See McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802, 93 S. Ct. 1817, 36 L. Ed. 2d 668 (1973). The only person who was comparably qualified or less qualified than plaintiff was the graduate student. Defendant convincingly established that his was not the sort of position for which Dr. Cooper or any outside applicant seeking a professional position would be considered.

Plaintiff's individual claim against UTD is therefore DENIED.

NOTES

[1] March 24, 1972 is the date on which Title VII became applicable to the University. Plaintiff conceded this fact in her brief filed August 10, 1977.

[2] These and the following figures were taken from the Bulletin of the American Association of University Professors (AAUP).

[3] AAUP lists UTD as a Category IIA institution.

[4] As explained further infra, statistical significance reflects the probability that the test statistic—in this case, the percentage of female professors at UTD who were tenured — would have been at least as extreme as its measured value if it were the result of the null hypothesis and random deviations. In this instance, the null hypothesis is that males and females receive tenure at the same rate; or otherwise stated, that the achievement of a tenured professorial rank does not depend on the candidate's sex.

[5] Plaintiff also introduced statistical evidence comparing salaries of males and females at UTD with such salaries at other institutions. Plaintiff's own expert testified that there was no significant statistical disparity at the .05 level in any of the salary figures for men and women. Thus those figures do not raise a question as to the existence of sex discrimination in salary.

[6] The question of whether plaintiff can represent employees is not one of adequacy, but as discussed above, one of commonality and typicality.

[7] As part of its decision on the merits, the court has determined that the portion of the class consisting of those not holding Ph.D. degrees should be decertified for lack of adequate representation. See section II(B), infra.

[8] This column represents the percentage of females in the pool of all people who received their doctorates in 1975 in disciplines included in the relevant UTD school. Plaintiff chose the year 1975 because roughly 70% of all of UTD's hiring occurred that year.

[9] The other probabilities are as follows: Arts & Humanities, .097; Human Development, .452; Management & Administrative Sciences, .319; Social Sciences, .319. None of these probabilities is significant at the .05 level.

[10] In the other schools the number of standard deviations between the expected hires and actual hires was not particularly noteworthy. In Arts and Humanities it was -1.30; in Human Development it was -.12; in Management and Administrative Sciences it was -1.09; in Social Sciences it was -.47.

[11] The court therefore need not address plaintiff's motion to strike defendant's applicant flow figures. Defendant has moved to strike plaintiff's evidence of UTD's affirmative action plans; because the court has not found it necessary to rely on that evidence, it also does not reach defendant's motion to strike.

[12] The court observes, without deciding the validity thereof, that plaintiff's computations are based on a one-tailed approach. Plaintiff only allowed for the possibility of hiring an equal or fewer number of women than would be expected; her expert's calculations excluded the possibility that a greater number of women might have been hired at UTD than would be expected on a random basis. Had plaintiff used this two-tailed approach, the result in the School of Natural Science and Mathematics would not have been statistically significant.

[13] The Second Circuit was faced with similar contentions of subjective standards and rejection of qualified women in Faro v. New York University, 502 F.2d 1229 (2d Cir. 1974). The court wrote:

Of all fields, which the federal courts should hesitate to invade and take over, education and faculty appointments at the University level are probably the least suited for federal court supervision. [The plaintiff] would remove any subjective judgments by her faculty colleagues in the decision-making process . . .. Such a procedure, in effect, would require a faculty committee charged with recommending or withholding advancements or tenure appointments to subject itself to a court inquiry at the behest of unsuccessful and disgruntled candidates as to why the unsuccessful was not as well qualified as the successful. This decision would then be passed on by a Court of Appeals or even the Supreme Court. The process might be simplified by a legislative enactment that no faculty appointment or advancement could be made without the committee obtaining a declaratory judgment naming the successful candidate after notice to all contending candidates to present their credentials for court inspection and discussion. This would give "due process" to all contenders, regardless of sex, to advance their "I'm just as good as you are" arguments. But such a procedure would require a discriminating analysis of the qualifications of each candidate for hiring or advancement, taking into consideration his or her educational experience, the specifications of the particular position open and, of great importance, the personality of the candidate.

502 F.2d at 1232.