IN THE UNITED STATES COURT OF APPEALS
FOR THE FIFTH CIRCUIT
No. 96-50764
BETTY TRAVIS,
Plaintiff-Appellee,
versus
THE BOARD OF REGENTS OF THE UNIVERSITY
OF TEXAS SYSTEM; UNIVERSITY OF TEXAS,
at San Antonio,
Defendants-Appellants.
Appeal from the United States District Court
for the Western District of Texas
September 8, 1997
Before POLITZ, Chief Judge, and HIGGINBOTHAM and SMITH, Circuit
Judges.
PATRICK E. HIGGINBOTHAM, Circuit Judge:
Betty Travis prevailed in a jury trial against her employer,
the University of Texas at San Antonio and the Board of Regents of
the University of Texas System. After reviewing the record, we
conclude that as a matter of law Travis did not prove a violation
of Title VII by a preponderance of the evidence. Thus, we agree
with the university and the regents that the lower court should
have entered judgment against Travis.
I.
The University of Texas at San Antonio hired Betty Travis on
a tenure track as an assistant professor of mathematics in 1980.
The university awarded her tenure and promoted her to associate
professor in 1985. Her specialty is mathematics education.
During the 1993-1994 academic year, she applied for promotion
to full professor. The university puts promotion applications
through several tiers of reviews and recommendations. An
applicant’s file, which includes teaching evaluations,
publications, letters of recommendation, evidence of service to the
university, and so forth, moves from the applicant’s division to
the applicant’s college to the provost to the president and
ultimately to the board of regents. Each stage involves an
independent review. The president’s decision to promote or not to
promote — a decision that the board of regents virtually always
adopts — comes in light of the recommendations from the lower
levels, but is not dictated by them.
Travis received favorable recommendations from a committee of
the Division of Mathematics and Statistics, from the division’s
director, from a committee of the College of Sciences and
Engineering, and from the dean of the college. But the provost,
Raymond T. Garza, recommended against promotion. At trial, Garza
explained that although Travis’s record in teaching and service was
excellent, her research was meager and was not published in the
better academic journals. He noticed that reviewers on the
division and college levels had failed to make detailed comments on
2
her publications and concluded that they had not scrutinized them
carefully. Although the dean of the college praised Travis for her
teaching skills and her success in landing grants for the
university, his report indicated that Travis’s research was only
“marginally adequate.” After investigating Travis’s publications,
Garza concluded that only one co-authored article had appeared in
a “premier” journal and that she had placed only three articles in
what he called “category 2" journals. She had also published two
chapters in books and three pieces in “regional journals,” but in
Garza’s view those, along with her many conference presentations
and invited talks, were of marginal scholarly significance. Garza
decided that, compared to other faculty members applying for status
as full professor, Travis had not yet made a sufficient
contribution to scholarship in her field. The university’s
president, Samuel Kirkpatrick, concurred in Garza’s analysis.
Travis received notification in March of 1994 that her application
for promotion had been denied.
Travis immediately scheduled a meeting with Provost Garza in
early April to discuss the reasons for the denial. Garza explained
that in order to earn a promotion, her research would need to be
more substantial. He indicated that, in combination with her
outstanding teaching and service, she was very close to meeting the
university’s expectations for a full professor. Several weeks
later, Travis told Garza that a journal had expressed interest in
one of her papers, and Garza promised to bring that fact to the
attention of the president, although he did not know whether it was
3
too late to reverse the denial. The president informed Travis by
letter that if she had additional material to include in her file
she would have to submit another promotion application in the 1994-
1995 academic year.
On May 11, 1994, Travis filed a charge of discrimination with
the Equal Employment Opportunity Commission. She amended her
charge on August 5 to add an allegation of unequal pay after
learning that Jerry Keating, a male colleague hired in 1981, had
been appointed Acting Division Director and earned $12,000 more
than she did. On September 24, she filed this lawsuit, which was
removed to federal court. The petition alleged causes of action
against the university and the board of regents and also against
Garza and Kirkpatrick in their individual capacities. Travis
alleged that the university had breached a contract and violated
the Fair Labor Standards Act and the Texas Equal Rights Amendment
by failing to honor a memorandum in which it notified Travis of her
salary as Interim Associate Dean. According to the university, a
clerical error caused the memorandum to include the salary of
Keating, who preceded Travis in the associate deanship, rather than
Travis’s salary, which was nearly $12,000 less. Travis also
alleged that Garza and Kirkpatrick violated the First Amendment by
retaliating against her for positions she took in the faculty
senate.
As a result of this suit, the university investigated Travis’s
salary and discovered that a grant from the Office of Naval
Research included a salary supplement that seemed to violate OMB
4
guidelines. Richard Dawson, the university’s director of internal
audit, concluded that OMB Circular A-21 prohibited government grant
money from going toward salary supplements above a faculty member’s
base salary. Although the university had approved the grant that
included Travis’s supplement, there was no evidence that university
officials were aware of any potential violation prior to the audit.
Once he learned about the violation, Kirkpatrick was concerned
enough to order an audit of all salary supplements at the
university. This audit revealed that an untenured faculty member
was also receiving a salary supplement that exceeded her base
salary. The university terminated both supplements in an effort to
comply with federal regulations.
Travis applied again for full-professor status during the
1994-1995 academic year. The only significant change was the
acceptance of the article she had mentioned to Provost Garza after
the first denial. She also had a new article under submission,
three new grants, and talks at two national conferences. She got
the same result: the lower levels recommended promotion, but Garza
recommended denying the application, and, in spite of Travis’s
lawsuit, Kirkpatrick followed that recommendation. Garza based his
recommendation on the fact that Travis’s recently accepted article
was to appear in a journal based in India with a circulation of
only about 300. Without any significant new research, he was
unwilling to reach a different result in 1994-1995 than he had in
1993-1994.
5
Travis had served as assistant director of her division since
1989. But a few weeks after accepting the post of division
director in January of 1995, Don Allen decided that the position of
assistant division director was no longer necessary. The
university eliminated the position of assistant division director
in February of 1995, although it paid Travis her administrative
supplement through the end of the academic year.
In an April 10, 1995, amended complaint, Travis sought damages
for the removal of her salary supplement and her termination as
assistant division director. These actions, Travis alleged, along
with the denial of her 1994-1995 application for a promotion, were
retaliation for her 1994 EEOC complaint and lawsuit.
The parties agreed to trial before a magistrate judge. After
five days of trial, the jury found that the university’s denial of
Travis’s promotion was sexually discriminatory and that the
university had retaliated against her for filing a discrimination
suit. It also found by special interrogatory that Garza and
Kirkpatrick did not violate her First Amendment rights. No other
theories of recovery went before the jury. The court ordered the
regents to promote Travis as of September 1, 1994, and to pay back-
and front-pay for the failure to promote, for the termination as
assistant division director, and for the failure to pay
supplemental grant salary. It also granted Travis’s request in
full for $91,088.75 in attorneys’ fees. The university and the
regents appeal the court’s denial of their motion for judgment as
6
a matter of law on both the sex-discrimination claim and the
retaliation claims.1
II.
We review de novo the lower court’s ruling on a motion for
judgment as a matter of law under Fed. R. Civ. P. 50(a). Omnitech
Int’l, Inc. v. Clorox Co., 11 F.3d 1316, 1322-23 (5th Cir.), cert.
denied, 513 U.S. 815, 115 S. Ct. 71, 130 L. Ed. 2d 26 (1994). A
court should grant a Rule 50(a) motion not only when the non-movant
presents no evidence, but also when there is not a sufficient
“conflict in substantial evidence to create a jury question.”
Foreman v. Babcock & Wilcox Co., 117 F.3d 800, 804 (5th Cir. 1997)
(quoting Boeing Co. v. Shipman, 411 F.2d 365, 375 (5th Cir. 1969)
(en banc)).
Travis’s sex-discrimination and retaliation theories are
subject to the burden-shifting analysis expounded in McDonnell
Douglas Corp. v. Green, 411 U.S. 792, 93 S. Ct. 1817, 36 L. Ed. 2d
668 (1973), Texas Department of Community Affairs v. Burdine, 450
U.S. 248, 101 S. Ct. 1089, 67 L. Ed. 2d 207 (1981), and St. Mary’s
Honor Center v. Hicks, 509 U.S. 502, 113 S. Ct. 2742, 125 L. Ed. 2d
407 (1993). But we need not parse the evidence into discrete
segments corresponding to a prima facie case, an articulation of a
legitimate, nondiscriminatory reason for the employer’s decision,
and a showing of pretext. “When a case has been fully tried on the
merits, the adequacy of a party’s showing at any particular stage
1
The defendants have also appealed evidentiary rulings and
the imposition of attorneys’ fees. We have no occasion to consider
those matters.
7
of the McDonnell Douglas ritual is unimportant.” Molnar v. Ebasco
Constructors, Inc., 986 F.2d 115, 118 (5th Cir. 1993) (citation
omitted).
A Title VII plaintiff bears the burden of proving not only
that the employer’s purported reasons for taking an adverse
employment action are pretextual, but also that the employer
engaged in illegal discrimination. Hicks, 509 U.S. at 511. In
this case, Travis’s burden is crucial for two reasons. First, it
means that “[to] satisfy the statutory burden, the plaintiff must
offer some evidence, whether direct or circumstantial, that permits
the jury to infer that the proffered explanation was a pretext for
illegal discrimination.” Swanson v. General Services Admin., 110
F.3d 1180, 1185 (5th Cir.), petition for cert. filed, 66 U.S.L.W.
3129 (U.S. July 23, 1997) (No. 97-163). Second, although evidence
of pretext, in conjunction with a prima facie case, usually creates
a jury question on the ultimate issue of discrimination, it does
not always do so. We engage in “traditional sufficiency-of-the-
evidence analysis” to determine whether reasonable jurors could
find discriminatory treatment. Rhodes v. Guiberson Oil Tools, 75
F.3d 989, 993 (5th Cir. 1996) (en banc). In other words, it is
possible for a plaintiff’s evidence to permit a tenuous inference
of pretext and, by extension, discrimination, and yet for the
evidence to be insufficient as a matter of law to support a finding
of discrimination. See Walton v. Bisco Industries, ___ F.3d ___,
___, 1997 WL 433984, at *3 (5th Cir. Aug. 19, 1997) (“Separate from
her pretext evidence, Walton has offered nothing to suggest that
8
impermissible discrimination underlies her termination.”);
Ontiveros v. Asarco, Inc., 83 F.3d 732, 734 (5th Cir. 1996) (“There
may barely be enough evidence to sustain a finding of pretext.
However, there is insufficient evidence to support a reasonable
inference of discrimination.”).
III.
We turn first to Travis’s claim that the denial of her
promotion was caused by sex discrimination. Surprisingly little of
the trial involved any reference to Travis’s sex. A casual
observer would have thought the jury had been asked to decide
simply whether the university should have promoted Travis. During
their extensive testimony, Travis, Garza, and Kirkpatrick discussed
primarily topics such as the prestige of various academic journals,
the legitimacy of publications based on a doctoral dissertation,
the scholarly value of oral presentations to gatherings of
academics or groups of local educators, the place of education
specialists within a division of mathematics and statistics, and so
forth. If nothing else, the trial made it clear that measuring the
value of academic work is sticky business.
Given the legitimate controversy over the quality and
importance of Travis’s research, the jury could reasonably have
concluded that U.T.—San Antonio should have promoted her to full
professor. Indeed, we assume for the purposes of this analysis
that a reasonable jury could even have concluded that the adequacy
of Travis’s research was not the real reason that the university
twice denied her promotion. The jury could have drawn the
9
conclusion that the administration disagrees with Travis’s vision
of the university. Trial testimony suggested that she might oppose
the university’s efforts to emulate major research institutions by
focusing on developing doctoral programs. The administration might
have thought that her energetic pursuit of improvements in
education are incompatible with the division’s mission of
conducting pure research. Or perhaps Garza and Kirkpatrick were
simply jealous of her popularity among students.
The dispositive question, however, is not whether the
university made a mistake or whether it gave forthright
explanations for its failure to promote. Instead, we must
determine whether Travis has presented sufficient evidence that her
sex was the reason for the failure to promote. We can find no more
than a sliver of a suggestion that sex had anything to do with this
employment dispute. We conclude that this sliver, viewed against
the background of the university’s evidence, is inadequate to
produce an evidentiary conflict strong enough to survive a Rule
50(a) attack.
First, Travis claimed that the university treated several male
professors more favorably. Lawrence Williams was the only other
candidate from the College of Sciences and Engineering who applied
for promotion from associate to full professor in the 1993-1994
cycle. In 1994-1995, three other males applied for and received
promotions to full professor. Their success, according to Travis,
indicates gender bias. Provost Garza, however, explained to the
jury in detail his system for evaluating scholarly accomplishments
10
and why these four males had stronger records than Travis’s. His
calculations indicated that one of the males had 47 articles in
“premier” journals, one had 15 such articles, one had six, and one
had three. According to Garza, Travis had published only one
article in a top-tier journal. The male with only three articles
in a premier journal had also published a book with a university
press, an accomplishment on which Garza placed significant weight.
The university could have promoted Travis along with these
males; unlike many positions, there is no artificial limit on the
number of full professors the university can sustain. Thus, even
if the four males were indeed better candidates than Travis, the
denial of promotion could still conceivably have been caused by her
sex. But Travis did not give the jury any basis for concluding
that sex played a role in these five promotion decisions. For the
most part, the four males had stronger records, measured by the
standard academic criteria outlined by Garza. We cannot turn an
attack on those standards, however outmoded they might be, into a
Title VII case. By themselves, minor, reasonable disagreements
about scholarly qualifications do not raise an inference that the
disagreements have their roots in sex discrimination.
Second, Travis urges that the jury could find disparate
treatment based on a comment her dean made in 1986 that she was not
“tough enough” to serve as acting director of her division. That
dean, however, is no longer employed by the university.
Kirkpatrick and Garza did not come to university until 1990 and
1991 respectively. This isolated comment, made more than seven
11
years before the disputed employment decisions by a person with no
connection to Travis’s promotion applications, has no evidentiary
force. See Ray v. Tandem Computers, Inc., 63 F.3d 429, 434 (5th
Cir. 1995) (“[A] single comment, made several years prior to the
challenged conduct, is a stray remark too remote in time to support
an inference of sex discrimination in later employment actions.”);
Mooney v. Aramco Services Co., 54 F.3d 1207, 1221 (5th Cir. 1995)
(finding irrelevant testimony of “sporadic and isolated” anecdotes
of discrimination committed by supervisors not involved in the
employment decisions at issue in the plaintiffs’ suit).
Third, Travis indicated to the jury that females seeking
promotion have faced discriminatory delays. Since 1990, the
university has denied promotions to two female associate professors
and to one male associate professor within Travis’s college.
During that time, six applicants have received promotions to full
professor on their first application, and all of them have been
male. This extremely small sample hardly establishes
discrimination. Moreover, the two women denied promotions were
granted promotions when they applied the following year. Travis’s
evidence is weak on this score because it is geared to her
division, which recommended Travis for promotion at each stage on
both of her applications. The more relevant statistic is
university-wide, and there the university seems to be on solid
ground. Since 1990, 12 of 25 male applicants have received
promotions from associate to full professor, while 7 of 13 female
applicants received the same promotions. Because Garza and
12
Kirkpatrick must approve all of these promotions, these numbers
suggest that they have not been inclined to turn women down based
on their sex.
Fourth, Travis attempted to draw the jury’s attention to pay
disparities between her salary and Keating’s salary. The
university, however, accounted for these disparities by explaining
that other universities had expressed interest in hiring Keating at
higher salaries. To stay competitive, the university had to make
it worth Keating’s while to remain in San Antonio. Travis did not
effectively rebut this explanation.
Travis’s other references to sex discrimination during the
trial were of no significance. In sum, the relative absence of
evidence of sex discrimination could not have given the jury
insight into whether the university allowed Travis’s sex to play a
role in the decision not to promote her. The jury may not have
liked the academic criteria in use at the University of Texas at
San Antonio, but that was not a legitimate ground for finding the
university liable for a Title VII violation. When we view all of
the evidence and resolve any doubts in favor of Travis, we cannot
find sufficient evidence to support the jury’s finding that
Travis’s being female caused the university to deny her a
promotion.
IV.
The jury found that the university retaliated against Travis
for filing her discrimination claim in three ways: (1) by denying
her a promotion in the 1994-1995 cycle, (2) by removing her as
13
assistant division director, and (3) by terminating her salary
supplement from the Office of Naval Research grant.
A.
As we have explained, the university satisfied its burden of
producing evidence of a nondiscriminatory reason for denying
Travis’s second application for a promotion. Outside of disputing
the merits of the promotion decision, Travis did not present
evidence of retaliation. Garcia explained that, because of the
pending lawsuit, he made a special effort to confirm his conclusion
that Travis’s file was not significantly different from the
application she filed in 1993-1994. Contrary to Travis’s
suggestion, Garza did not testify that her suit caused him to try
to find weaknesses in the 1994-1995 application. A few days after
the university denied Travis’s 1994-1995 promotion, the dean of the
college remarked that it was because Travis “hit them with a
lawsuit.” But there was no evidence that the dean, who supported
Travis’s promotion, had any special knowledge of retaliation. In
context, the remark was mere speculation.
Because of the marginal change in Travis’s credentials between
the unsuccessful 1993-1994 application and the 1994-1995
application, a different decision on the merits of her promotion
would have been unusual. Under these circumstances, the mere fact
that Travis filed an EEOC charge cannot support the jury’s
conclusion that the denial was retaliatory.
14
B.
The university explained the termination of Travis’s position
as assistant division director by offering the testimony of Linda
Whitson, vice president of administration and planning. Whitson
stated that Don Allen became division director in January of 1995
and that he quickly decided that he did not need an assistant.
Travis did not contradict Whitson’s claim that there were no
difficulties in the relationship between Allen and Travis. More
importantly, Travis did not attempt to refute the university’s
evidence that Allen made his decision for legitimate administrative
reasons. There was no testimony that Allen even knew about
Travis’s EEOC charge. In other words, Travis did not carry her
burden of producing evidence of retaliatory intent. The jury might
speculate that Allen was cooperating with Garza and Kirkpatrick to
punish Travis, but without any evidence to that effect, we cannot
uphold the jury’s findings.
C.
Finally, the jury found that the university retaliated against
Travis when it cut off her salary supplement. According to the
university, it did not discover that the supplement was in excess
of Travis’s base salary until it investigated her allegation of
unequal pay. When the problem came to light, President Kirkpatrick
took swift action: he notified the director of internal audit and
asked him to make the matter top priority. Kirkpatrick’s memo
mentioned Travis by name so that the auditor would at least have a
place to begin his investigation. Kirkpatrick also testified that
15
he contacted the general counsel of the U.T. system, the executive
vice chancellor of the system, and presidents of other Texas
universities to determine whether his understanding of the OMB
regulation was accurate. After concluding that Travis’s supplement
was a violation of federal law, he decided to end it prospectively
only, assuming that the federal government would not be concerned
about payments Travis had already received. One other faculty
member was receiving a similar grant, and the university ended it,
as well.
Again, aside from the fact that the university cut off the
salary supplement after it knew of Travis’s suit, Travis did little
to rebut the university’s account of its reasons for acting. On
cross-examination, defense witnesses admitted that the OMB
regulation is difficult to interpret and that one might be able to
make a colorable argument that Travis’s grant satisfied the
regulation. But Travis did not present testimony that Kirkpatrick
or the university misunderstood the regulation, much less that
their effort to adhere to their understanding of the regulation was
pretextual. Notwithstanding Travis’s accusations to the contrary,
Kirkpatrick’s procedures were consistent with his stated concerns.
Without evidence, Travis’s assertion that retaliation caused the
termination of her salary supplement is merely her own subjective
belief, which is insufficient to create a jury question. See
Armendariz v. Pinkerton Tobacco Co., 58 F.3d 144, 153 (5th Cir.
1995) (reversing a jury verdict of discrimination and collecting
16
cases), cert. denied, ___ U.S. ___, 116 S. Ct. 709, 133 L. Ed. 2d
664 (1996).
V.
The judgment based on the jury verdict below is REVERSED, and
a take-nothing judgment is RENDERED in favor of the university and
the board of regents.
17