Nikolova v. University of Texas At Austin

Opinion

IN THE UNITED STATES DISTRICT COURT FOR THE WESTERN DISTRICT OF TEXAS AUSTIN DIVISION

EVDOKIA NIKOLOVA, § § Plaintiff, § § v. § 1:19-CV-877-RP § UNIVERSITY OF TEXAS AT AUSTIN, § § Defendant. §

ORDER Before the Court is Defendant University of Texas at Austin’s (“UT”) Motion for Partial Summary Judgment, (Dkt. 39), and all responsive briefing.1 Having considered the parties’ submissions, the record, and the applicable law, the Court will deny UT’s motion. I. BACKGROUND This in an employment discrimination case concerning UT’s decision to deny tenure to Plaintiff Evdokia Nikolova (“Nikolova”). Nikolova is an assistant professor in UT’s Electrical and Computer Engineering (“ECE”) Department in the Cockrell School of Engineering (“Cockrell”). She claims that UT’s denial of tenure was impermissibly based on her gender and pregnancy. A. UT Tenure Process Tenure at UT is a “continuing appointment,” when an assistant professor is promoted to an associate professor. (UT Handbook: Academic Titles and Tenure, Dkt. 39-1, at 1). The assistant professor position is a temporary one, considered “probationary,” which a person can hold for a maximum of seven years. (Id. at 2). In the person’s sixth year, the “up or out” year, the budget

1 Plaintiff Evdokia Nikolova also filed a Motion for Leave to File Sur-Reply. (Dkt. 50). UT opposed the motion. (Opp., Dkt. 51). Given the Court’s present Order disposing of the underlying motion, the Court will deny the motion for leave to file sur-reply as moot. council and department chair recommend to the administration either that the person be promoted to associate professor with tenure or be placed on a terminal appointment for their seventh year.

(Id.; Shockley Dep., Dkt. 39-2, at 25). UT only considers full-time service for both semesters of the year at UT in calculating years of service. (UT Handbook: Academic Titles and Tenure, Dkt. 39-1, at 1). Years at another institution or on leave are not counted. (Id.). In some cases, UT will extend the probationary period based on personal circumstances, including being “the principal caregiver of a preschool child.” (UT Handbook: Extension of the Tenure Track Probationary Period, Dkt. 39-1, at 1). The extension cannot exceed two years. (Id.). UT will also consider tenure before the sixth year. In these cases, the school will promote, deny tenure permanently, or decide to “not promote, which is not make a decision now, without prejudice, and review the case later in the probationary period.” (Fenves Dep., Dkt. 39-2, at 41). Such cases are “accelerated and must be explained in the department chair’s and dean’s statements.” (General Guidelines for Promotion, Dkt. 39-1, at 3). As President Gregory Fenves (“Fenves”) described, the inquiry for early promotion asks, “Why now? . . . Given the record in teaching,

research, service, award—all the categories we look at in a promotion case—why is this case ripe for an affirmative decision to promote at this time?” (Fenves Dep., Dkt. 39-2, at 42). The tenure review “comprises an independent review at multiple levels: budget council/executive committee, department chair, college/school, dean, and central administration.” (General Guidelines for Promotion, Dkt. 39-1, at 1). If an assistant professor decides to seek early review, they must seek and obtain approval from the department chair and the budget council.2 (Fenves Dep., Dkt. 39-2, at 38). The person must then assemble a tenure dossier with documents provided for in UT’s general guidelines.

2 The budget council is made up of all full professors in a department and is the only body with authority to decide whether to put someone up for promotion outside of their sixth year. (Fenves Dep., Dkt. 39-2, at 54). (General Guidelines for Promotion, Dkt. 39-1, at 9). The assistant professor is responsible for compiling the dossier before the department considers it. (Id. at 5). The budget council then

evaluates the candidate and prepares a statement for each area of contribution: teaching, research, advising, university and professional service, and other evidence of merit. (Id. at 2, 7). The budget council then anonymously votes on whether to support the candidate. (Id. at 7). Separately, the department chair, who is present for the budget council’s discussion but does not vote, provides a written statement with an assessment for each of the areas of contribution. (Id.). The chair describes the budget committee’s rationale and summarizes members’ views (anonymously), explaining negative votes. They “identify the candidate’s strengths and weaknesses, provide context as needed, and address whether and how the candidate’s promotion would improve the quality of the department.” (Id.). The statement becomes a part of the dossier at the next level of review. (Id.). The application then moves onto the college advisory committee, or the Promotion & Tenure Committee (“P&T Committee”), comprising representatives from each department. After reviewing the dossier, the P&T Committee votes on the case, and votes are recorded on a

Recommendation for Change in Academic Rank/Status form. (Id. at 8; Fenves Dep., Dkt. 39-2, at 55). The P&T Committee also discusses the case with the Dean, who then writes a letter with their own evaluation and recommendation, informed by the prior discussion and vote. (Id. at 60–61). The Dean then presents the case, answers questions, and conducts a discussion with the President’s committee (the President, Provost, Vice President of Research, Dean of Undergraduate Studies, and Dean of the Graduate School). (Id. at 57, 61). The President then conducts an advisory poll of each member’s recommendations. (Id. at 65). However, the ultimate decision on whether to grant tenure rests with the president. (Id.). Although a candidate in their sixth year can seek to make final arguments to the President if denied, a candidate seeking early tenure has no opportunity to appeal a denial. (Shockley Dep, Dkt. 39-2, at 83). The candidate can also request a review by the Committee of Counsel on Academic Freedom and Responsibility (“CCAFR”), which is limited to procedural violations and questions of academic freedom. (Id. at 91; General Guidelines for Promotion, Dkt.

39-1, at 17). Nikolova notes that the ECE Department is “overwhelmingly male dominated.” (Resp., Dkt. 44, at 3). When she applied for tenure in 2019, “there were 53 tenured faculty members within the ECE Dept., 49 men, 92.5% and 4 women, 7.5%.” (Id.). In the five years between when she was hired at UT and when she was denied tenure, the ECE Department promoted all seven of the two men who applied for tenure and denied both women. (Id. at 3–4). B. Nikolova’s History and Tenure Application Nikolova received her undergraduate and master’s degrees from Harvard University in 2002, in applied mathematics and computer science, respectively. (Nikolova Dep., Dkt. 39-2, at 24). She received a second master’s degree in mathematics from Cambridge University, and her Ph.D. from Massachusetts Institute of Technology (“MIT”) in 2009. (Id. at 25). She also completed a postdoctoral position at MIT. (Id.). She was first employed as an assistant professor at Texas A&M

University (“A&M”) in 2011. UT Professor Constantine Caramanis invited her to apply for an assistant professor position at UT, which she applied and interviewed for, and began in 2014. (Id. at 28, 34). As part of the interview process, she met with ECE Department Chair Ahmed Tewfik (“Tewfik”). Tewfik represented to her that she “would be able to go up at UT Austin on [her] normal tenure time clock that began at Texas A&M” and that the “standard offer” included tenure evaluation “after five years at UT,” but if she “need[ed] more time for whatever reason” she could have “more flexibility.” (Id. at 35–36). After discussions of her record, and after receiving 77% of the committee vote, the ECE Department offered Nikolova the assistant professor position. (Employment Emails, Dkt. 39-1, at 48). Tewfik sent her an offer letter which contained details on the tenure process; she accepted on July 18, 2013. (Offer Letter, Dkt. 39-1, at 51). Nikolova began working at UT in January 2014, meaning the first year counting towards her tenure application began that fall. (Nikolova Dep., Dkt. 39-2, at 28). Starting then, she taught three

semesters of classes at UT. (Id. at 37–38). In fall 2015, she worked at the University of California at Berkeley as a visiting researcher at the Simons Institute for the Theory of Computation (“Simons Institute”). (Id. at 39). Nikolova became pregnant, and in October 2015 she applied for Modified Instruction Duty (“MID”) so she would not have to teach that spring. (Id. at 39–40; MID Memo, Dkt. 39-1, at 53). Tewfik encouraged her to take the extension year, effectively assuring her “it will not hurt [her],” even though she expressed reluctance given than she did not “expect to wait until [her] up-or-out year at UT” and “was expecting to go to be considered for tenure on [her] normal . . . time clock that began at A&M.” (Nikolova Dep., Dkt. 39-2, at 67–68). Tewfik approved the request, noting that “she will not be teaching at all this academic year” as a result of her semester at the Simons Institute. (MID Approval, Dkt. 39-1, at 55, 58, 60). She also applied for and received an extension of the probationary period for her tenure application during that year. (Id. at 56). The paperwork noted that she could rescind the extension up to February 1 of the fall of her sixth

probationary year. (Id. at 60). Nikolova’s first child was born on March 10, 2016. (Nikolova Dep., Dkt. 39-2, at 30). Nikolova resumed teaching in fall 2016 and taught at UT for three semesters. (Id. at 39–40). In September 2017 she met with Sanjay Shakkottai (“Shakkottai”), her career mentor in the ECE Department, who told her she was ready to apply for tenure. (Id. at 89). Nikolova claims he told her that if she applied at that time, she would “not be considered early” although she would be “technically early.” (Id.). At that time, Nikolova had been an assistant professor for eight years: 2.5 years at A&M and 5.5 years at UT. Nikolova then met with Tewfik and Christin Julien (“Julien”), a colleague in the ECE Department, about the process. (Id. at 94). Twefik spoke to Sharon Wood, Dean of the Cockrell School of Engineering, before she began the process. Wood was “neutral” regarding Nilolova’s prospects. (Id. at 116). Nikolova spent the spring 2018 semester back at the Simons Institute. (Id. at 76). In May 2018, Shakkottai presented Nikolova’s case to the budget

council, which approved her early candidacy with ten votes in favor and one abstention. (Id. at 96– 97; Candidacy Votes, Dkt. 39-1, at 73). One comment noted “several areas of particular concern.” (Id. at 74). Nikolova underwent her third-year review at this time. (Nikolova Dep., Dkt. 39-2, at 81). Nikolova’s second child was born on June 13, 2018. (Id. at 30). Part of Nikolova’s tenure application process was to assemble her dossier, which would constitute the centerpiece of her review at each level. (Id. at 102; General Guidelines for Promotion, Dkt. 39-1, at 5). On July 18, 2018, about one month after Nikolova gave birth to her second child, an administrative assistant emailed Nikolova, at Tewfik’s instruction, requesting that she update her dossier; she did not respond. Her husband emailed a student requesting the student work on the dossier. (Dossier Emails, Dkt. 39-1, at 79, 84). The assistant emailed Shakkottai and Tewfik, noting that the request to the student was “inappropriate.” (Id. at 81). Tewfik then told Nikolova to finish her dossier without assistance. (Id. at 84). Nikolova’s dossier included many letters indicating strong

support from professors at peer universities. (See Resp., Dkt. 44, at 6). The budget council voted to approve Nikolova’s early promotion on September 10, 2018, with a vote of 38 in favor, one opposed, and five abstentions. (Promotion Votes, Dkt. 39-1, at 86– 87). She received some negative comments, including that her research was underdeveloped and her publication record “weak.” (Promotion Votes, Dkt. 39-1, at 91). Tewfik claimed the vote was unremarkable, as “Departments tend to support their own members.” (Tewfik Dep., Dkt. 39-2, at 55). This was particularly true after another woman left UT after a “negative gender experience” with students and a negative vote from the budget council on her early tenure application. (Julien Dep., Dkt. 39-2, at 53, 133). Tewfik stated that the Department “learned a lesson” after that experience and “would not have voiced their concerns.” (Tewfik Dep., Dkt. 39-2, at 55). Tewfik felt that Nikiolova’s case “was close to the bar” but believed “she was capable of delivering.” (Id. at 11). He “wanted to give her the chance to shine again and “was reasonably

confident that she was going to shine again” so he “wanted her to stay in the department.” (Id.). He sent the initial version of his letter of support for Nikilova to the P&T Committee, which was chaired by Shakkottai. (Id. at 196). The P&T Committee then requested additional information to “provide additional justifications for early promotion.” (Id. at 197). Tewfik then submitted a second version stating there is “zero correlation” between fitness for promotion and years of service: “we put somebody up for promotion. We said that it’s technically early. That’s the end of story.” (Id. at 198). Tewfik noted that the application was “technically early”” because she had been at UT for only five years and took an extension year. (Early Promotion Emails, Dkt. 39-1, at 96). He noted that she had spent two years as an assistant professor at A&M, and “[h]er case would not be early if these two years of service are taken into account.” (Id.). He stated that “[a]ll promotion cases should be evaluated by the well established metrics” and that dismissing Nikolova’s promotion as early “will do irreparable harm to our ability to continue to recruit highly promising assistant or associate

professors from other institutions,” and to “retain the star assistant professors that we have recently recruited from top peer departments . . . who had served 2 to 4 years as assistant professors at their initial home institutions.” (Id.). Tewfik also stated—or, as UT characterizes it, “quipped”—that he “hope[d] that this question” regarding Nikolova’s accelerated timeline “isn’t a reflection of gender bias.” (Id. at 92; Mot. Summ. J., Dkt. 39, at 8). Associate Dean Gerald Speitel (“Speitel”) asserted that “prior service at another university is a mitigating factor” but “is not in itself a justification for promoting her now” and that “UT is not obligated to consider her for promotion now.” (Early Promotion Emails, Dkt. 39-1, at 95). Comments placing her “below the median of the department” were “in conflict with the general expectation that early promotions be ‘well above the bar’ in all categories.” (Id.). In spite of this contentious exchange, the P&T Committee voted for Nikolova’s promotion. (Wood Dep., Dkt. 39-2, at 200).

C. Tenure Decision Wood reviewed Nikolova’s dossier and met with the P&T Committee to hear their thoughts in addition to their written evaluation. (Id. at 201). Wood had a “long discussion” with the P&T Committee about “where the bar should be for an early case” and found a “fundamental disconnect” in that they found her argument for not being evaluated under a higher, early bar to be “compelling.” (Wood Dep., Dkt. 39-2, at 208). Wood noted that the P&T Committee members “were not present at all of the meetings” nor “the briefings with the President’s Committee and the Department Chairs.” (Id. at 208–209). After those briefings, Wood felt that, because Nikolova was considered two years early, “we have a higher bar.” (Id. at 211). Whereas for a person “being considered at the normal time for promotion,” if there is a “weakness, a flat spot, . . . then that’s considered perfectly normal and . . . that would not lead to a negative decision.” (Id. at 168). However, for an early case, “the expectation is that we don’t have the flat spots” and the person

“would be above the ‘meeting expectations’ in each area.” (Id.). Wood also noted that UT had no written policy indicating a higher bar for early tenure consideration. (Id. at 212). Ultimately, Wood did not support Nikolova’s promotion, stating “I do not believe Dr. Nikolova’s performance meets expectations for early promotion to associate professor.” (Dean’s Assessment, Dkt. 39-1, at 111). She elaborated that she would likely agree with the P&T Committee’s recommendation of promotion if Nikolova were in her sixth probationary year, but because she was being considered two years early, Wood did “not believe that she has taken responsibility for improving her teaching,” although her early teaching evaluations were “quite strong for a new assistant professor, and indicate that she has the ability to engage her students in the classroom.” (Id. at 111, 108). Nikolova “attributed [the lower ratings] directly to the Teaching Assistants” which Wood found to be “the most direct deflection of taking responsibility” because “everything that happens in the classroom is the faculty . . . member’s responsibility.” (Wood Dep.,

Dkt. 39-2, at 224). Nikolova notes that math-heavy undergraduate courses, like the course Nikolova was teaching, traditionally receive lower teaching scores, but that Nikolova still had the third highest score of the 13 instructors who had taught the course in the past five years. (Resp., Dkt. 44, at 8; Chair Recommendation, Dkt. 39-1, at 98). Nikolova further asserts that her teaching scores only dropped several tenths of a point while she was teaching two large sections, and while she was pregnant. 3 Nikiolova also claims that other male professor who applied for and received tenure early had lower teaching scores. (Id. at 8–9). She claims that Wood failed to acknowledge this this context when she “compared Nikolova’s teaching scores with the scores of all engineering faculty for all courses.” (Id.). Wood also had “concerns about the sustainability of her research program,” concerns which were “compounded by the fact that both her teaching and her external funding have dropped since she spent the 2015 fall semester at UC Berkeley.” (Dean’s Assessment, Dkt. 39-1, at 111). Wood

also explained that she spent fall 2015 at the Simons Institute and “was scheduled to teach two classes in the 2016 spring semester. However, she became pregnant during the 2015 fall semester and was assigned modified instructional duties.” (Id. at 109). Nikolova “did not teach during the 2015-2016 academic year, and since then her instructor ratings have fallen.” (Id.). Wood was concerned that “approximately 70% of her funding was awarded during her first three years,” and

3 One letter of support in Nikolova’s dossier noted that “there is a perception of a slight drop in Nikolova’s student numerical evaluation scores, from quite strong to being moderately strong. This perceived drop occurred after the birth of her first child, while a course with the lowest evaluation score was taught while Nikolova was pregnant with another child.” (Graves Letter, Dkt. 44-1, at 143). The letter explained that “[t]he phenomenon of lesser evaluations for women professors in male-typed disciplines is so typical that, sadly, it is almost to be expected . . . . A wealth of studies of student evaluations . . . have revealed i) systematic bias against women professors in male-typed disciplines . . . [and] 2) lack of correlation between these evaluations and educational outcomes.” (Id. at 144). Wood acknowledged familiarity with these studies. (Wood Dep., Dkt. 39-2, at 42). “[o]nly one grant has been awarded in the past four academic years” raising “questions about the sustainability of her research funding. (Dean’s Assessment, Dkt. 39-1, at 109.). Nikolova notes that

Wood failed to include all of her reasons for her recommendation to deny tenure. (Resp., Dkt. 44, at 13). In an email, Wood noted that “[t]his is the most controversial case. Intentionally, I did not state all my reasons in the letter. I felt that some things were best discussed with the committee.” (Most Controversial Email, Dkt. 44-3, at 156). Given her concerns, Wood told Tewfik that she would not recommend promotion and asked him to discuss with Nikilova the possibility of withdrawing her case because she “didn’t want to lose Dr. Nikolova” but feared the President’s Committee could deny her tenure and give her a terminal appointment rather than an opportunity to reapply later. (Tewfik Dep., Dkt. 39-2, at 217, 219). Tewfik also told Nikolova that she could write a rebuttal to the Dean’s letter, which she did. (Nikolova Dep., Dkt. 39-2, at 121). She declined to withdraw her tenure application. Wood presented Nikolova’s case to the President’s Committee, where discussion focused on Nikolova’s funding, research trajectory, and publication record. (Wood Dep., Dkt. 39-2, at 269).

Starting in Nikolova’s year, instead of discussing every candidate, the President’s Committee would tell Wood which cases it wished to discuss. (Id. at 268). In the two years under that new policy, they did not ask about “the very strong positive cases”; they only spoke with Wood about “the ones that are closer to the bar and also anyone where there’s a recommendation against promotion.” (Id. at 270). The President’s Committee voted unanimously (five votes) against advising promotion. (Id. at 273). The members had “concerns about her record” and felt “it was just not a clear case.” (Shockley Dep., Dkt. 39-2, at 20). Fenves agreed, explaining that “we did not feel the performance, particularly in research, funding, and trajectory and publications was ready to make that affirmative decision to promote to tenure.” (Fenves Dep., Dkt. 39-2, at 119). Fenves was not required to make a final decision, however, because Nikolova was not in her sixth year. (Id.). Fenves found she “wasn't ready for a promotion decision or a terminal-appointment decision” and preferred to give her the additional years to demonstrate her “research funding will be sustainable and have a positive

trajectory.” (Id. at 119, 122). Fenves had two major concerns with Nikolova’s application: (1) research funding, and (2) publication record. First, he was concerned that Nikolova did not demonstrate sustainable funding of her research from external funders, nor the ability “to recruit and support graduate students.” (Id. at 130). Fenves needed assurance of funding sources “sufficient to be able to support a strong research program . . . that after approval of promotion and tenure that that will continue through evidence of submitting proposals and successful grants for a number of years past the promotion case. (Id. at 130–31). Nikolova listed three grants in her application. The first had effectively ended by the time of her review. (Id. at 131). The second she received while teaching at A&M and was set to expire around the same time. (Id.). The only active grant that would continue into her next year and was at a “very modest level of funding.” (Id.). The university also looks for “a good pipeline of pending grants that’s showing active research.” (Id. at 132). Nikolova had one pending grant that had

many coinvestigators and was not based at UT. (Id. at 132–133). In sum, Nikolova had “a modest level of current funding” and “only one grant pending at even a slightly lower funding level,” leaving “concerns about the level of funding and the trajectory.” (Id. at 133). Fenves noted that Nikolova’s research was theoretical, rather than applied, and so required generally required a lower level of funding than the latter. (Id.). Nikolova asserts that her funding and trajectory compared favorable to many candidates who had received support from Wood and Fenves, particularly when looking outside of ECE to other Cockrell departments, and when taking the theoretical versus applied distinction into account. (See Resp., Dkt. 44, at 16–17). Second, Fenves expressed concern regarding Nikolova’s publication record. She had published three papers as an assistant professor (in addition to prior publications) and had 18 peer reviewed conference proceedings as an assistant professor. (Id. at 137). He noted that in her field, “there is a strong outlet, important outlet in referee conference proceedings; but journal publications

are also important.” (Id.). According to Fenves, “[t]hree journal papers is a low number; and in combination with 18 referee conference papers is a generally . . . modest number . . . . [T]here’s nothing obvious that this is a strong publication record that was getting the results of research out into the field and the profession.” (Id. at 137–38). Fenves acknowledged that the review was “a qualitative . . . subjective assessment based on our years of experience of looking at cases” and that “in this case, [he] disagreed with” the budget council and P&T Committee, whose members also “had a lot of experience” reviewing these cases. (Id. at 141). Nikolova asserts that Fenves’s claims lack support, especially in light of comments by others regarding the strength of her research. (Resp., Dkt. 44, at 14; Julien Dep., Dkt. 39-2, at 163). Furthermore, she claims that her publication record was as strong as or stronger than male ECE faculty members at the time they were granted tenure— including the publication record of Tiwari himself.4 (Resp., Dkt. 44, at 18). Nikolova claims that her tenure denial is suspect given the statistics related to tenure

decisions around the same time. (Resp., Dkt. 44, at 6; see Tenure Data, Dkt. 44-1). According to Nikolova, in the nine years prior to her application, Cockrell considered 86 tenure cases. 62 received unanimous P&T Committee votes. Nikolova was the only one of these that did not receive tenure. (Id. at 7). She also claims that the average budget council vote for someone denied tenure was 67% in favor, whereas she received 97% of the vote. (Id.). Finally, she asserts that only three of the 29 people who were considered for tenure in the ECE Department between 2009 and 2020 were

4 Nikolova states that at the time of their tenure applications, both she and Tiwari had 12 peer-reviewed proceedings at conferences while at UT, although she had 30 and he had only 22 across their whole careers. (Resp., Dkt. 44, at 18). Further, she published three journal articles at UT, while he published only two. (Id.). UT notes only that Tiwari did applied rather than theoretical research and was in his sixth year. (Reply, Dkt. 47, at 6). women; she was the only one pregnant, and the only one denied tenure. (Id.) She does not note how many of these were early applications.

D. Aftermath of Tenure Decision In February 2019, Tewfik told Nikolova that Fenves had decided against promoting her. (Nikolova Dep., Dkt. 39-2, at 137). She declined to meet with Wood to discuss the decision. (Text Messages, Dkt. 39-1, at 116). On February 18, she sent an email to the ECE Department regarding gender discrimination in her tenure denial—what she termed the “elephant in the room.” (ECE Email, Dkt. 44-4, at 74). She noted that five male candidates had received promotions, and she was both the only woman considered and the only denial. She did not note whether the men considered were in their sixth years. (Id. at 76). She concluded, “Anecdotally, it is my impression that women in this Dept. have a longer time to advancement if they pass through that hurdle at all.” (Id.). In a side exchange with another colleague in response to the email, Julien stated that the case “has been handled badly on so many levels. . . . I’m really worried about the impact of her email on the assistant professors . . . .” (Id. at 74). She elaborated that she thought the email “was going to have a

significant negative impact.” (Julien Dep., Dkt. 39-2, at 160). Nikolova spoke to Brian Evans (“Evans”), Chair of the CCAFR, regarding options for contesting the decision. Upon his recommendation, and after working with Tewfik, she submitted a Request for Reconsideration to the President and an appeal letter to the CCFAR on March 25, 2019. (Id. at 139–40; CCAFR Letter, Dkt. 39-1, at 124; Reconsideration Request, Dkt. 39-1, at 142). Her CCAFR letter noted that the denial of tenure “raises questions of equity and different treatment based on gender and pregnancy.” (CCFAR Letter, Dkt. 39-1, at 125). Her letter to the President raised concerns of “pregnancy discrimination if undue emphasis is given to [the] single lower” teaching evaluation score she received during the fall 2017 semester when she was pregnant. (Id. at 155). She “hypothesized that student bias based on pregnancy could have affected her student evaluation score, and stated that the denial of her tenure raised concerns of gender and pregnancy bias.” (Mot. Summ. J., Dkt. 39, at 13). She noted that she was experiencing daily morning sickness

and extreme fatigue,” and attached letters from professors at other institutions who also “experienced lower teaching scores during their pregnancies.” (Reconsideration Request, Dkt. 39-1, at 155–56). She also highlighted research showing that women systematically receive lower teaching evaluations than their male peers. (Id. at 156). No one reviewed her letter because final arguments are not allowed in early tenure applications. (Shockley Dep, Dkt. 39-2, at 83). A CCAFR subcommittee issued an advisory report on April 28, 2019. (CCAFR Report, Dkt. 39-1, at 169). Regarding the “higher bar” to which Nikolova was arguably held, the report noted, “[i]n our view, different standards are not justified in this case, if ever, since this is not an early case if all of Nikolova’s 7 years in rank as Assistant Professor are taken into account.” (Id. at 172). The Subcommittee was “concerned about the statistics regarding gender equity, diversity, and inclusion presented by [Nikolova]” and felt “any department or School with such a track record should also be concerned.” (Id. t 174). It recommended that the ECE Department and Vice Provost conduct a

review of the materials Nikolova submitted comparing candidates by gender but stated that it could not consider those materials thoroughly. (Id.). It found “a review of the climate with respect to gender and inclusion in the ECE Department is warranted.” (Id. at 175). It also expressed “serious concern over the ramifications of the decision” for both UT and Nikolova, given that both parties “proceeded with the usual practice and understanding that service at [A&M] would be counted in the promotion and tenure process, only to find out differently very late in the process.” (Id.). These circumstances “had a negative impact on [Nikolova’s] career, and could have a negative impact on the department’s ability to address its serious gender imbalance.” (Id.). The report expressed “concern[s] over the general implications for diversity and inclusion of such practices.” (Id.). Despite these concerns, the report made no explicit findings of bias. (Id.). Overall, the report noted that the “decision not to promote [Nikolova] to the rank of Associate Professor with tenure

was flawed by procedural errors.” (Id.). Nikolova asserts that no action was taken in response to the report, aside from a brief reply by Fenves. (Resp. at 13). Rather, in an email exchange, Speitel noted to Wood the recommendation that “serious consideration be given to the reversal of this decision.” (Report Emails, Dkt. 44-3, at 149). Wood quipped in response to one of the recommendations, “[B]ets on how many people would read a faculty handbook?” (Id.). Speitel responded, “Let’s just say I have better things to do with my time than write a faculty handbook.” (Id.). Nikolova filed a charge with the Equal Employment Opportunity Commission (“EEOC”) on May 10, 2019. (EEOC Charge, Dkt. 39-1, at 204). On May 13, Fenves sent a letter to the CCAFR to respond to its subcommittee report and reiterated that he would not change his decision in Nikolova’s case. (Fenves Response, Dkt. 139-1, at 209). On June 13, the EEOC issued a right to sue letter but did not make a finding in the case. (Right to Sue, Dkt. 39-1, at 211). Nikolova filed this lawsuit on September 9, 2019. (Compl., Dkt. 1).

Nikolova claims that she faced retaliation for her activities, in the form of negative evaluations by members of the ECE faculty.5 She asserts that “[u]nflattering or negative peer teaching evaluations for a tenure-track faculty are essentially never done because it seriously damages the professor’s chance to obtain tenure.” (Resp., Dkt. 44, at 19). Another professor, Yale Patt concurred in an April 2020 email discussion, stating that he “is not going to put anything negative on paper,” that he “bend[s] over backward to be positive in [his] Peer teaching evaluation and put nothing in writing that could eventually hurt a colleague’s promotion package,” and that senior

5 To the extent that Nikolova premises her retaliation claim on issues with her supervision of the senior design course, the Court considers that argument waived based on her failure to respond. (See Mot. Summ. J., Dkt. 39, at 14). colleagues at other universities had “[n]ever” “seen any negative comment on a teaching evaluation in a promotion package.” (Patt Email, Dkt. 44-4, at 135–36).

Faculty in the ECE Department undergo peer evaluations as part of a “constructive dialogue about teaching with faculty colleagues” to “encourage[] time for reflection and action.” (Peer Evaluation Example, Dkt. 39-1, at 192). Peer teaching evaluations play a role in the tenure promotion process, as a complement to other materials. (Id.). On April 24, 2019, Sujay Sanghavi (“Sanghavi”), a colleague in the ECE Department, received a notification that his peer teaching evaluation was due on April 15, but had not been submitted. (Teaching Evaluation Emails, Dkt. 39- 1, at 165). He responded that Nikolova “is teaching an advanced grad class” and “at this stage of the semester, the class now has students giving project presentations and answering questions. So I feel it may be better to do [Nikolova’s] evaluation in a future semester, maybe when she is doing the teaching and it is an undergrad class.” (Id.). Tewfik then asked Julien to do the evaluation that week. (Id. at 168). On May 2, Julien sent Nikolova her peer teaching evaluation. (Julien Emails, Dkt. 39-1, at

190). On May 5, Nikolova emailed Julien stating that she “fe[lt] there is nothing positive in” the evaluation and asked Julien to “add some positive things and tone down some of the negative.” (Id.). Julien revised the evaluation; Nikolova requested additional changes; and Julien again made revisions. (Id. at 188–90). Nikolova also completed a form for the evaluation and commented that the lecture observed was “in the last couple of weeks of class,” which was “a time when everyone is tired, students are extremely preoccupied with their final projects and exams and even the most engaged students sometimes are not as engaged.” (Id. at 200). She stated that the “observation should not be completed in the last weeks of class, which may not be an accurate representation of a typical lecture,” and that “evaluation scheduling is something that needs to be acted on at the start of semester and not as an after-thought in light of a deadline at the end of the semester, putting undue strain on both the peer evaluator and the instructor.” (Id.). Nikolova claims her negative evaluation was a response to her “oppos[ing] discrimination” in her email to the ECE department.

(Resp., Dkt. 44, at 19). She also claims that Julien failed to follow procedures requiring pre- and post-observation meetings. (Id. at 20). ECE Department faculty also undergo Faculty Annual Reviews (“FARs”). (Julien Dep., Dkt. 39-2, at 145). Evaluations for the 2018-2019 academic year were completed in spring 2020. (Id. at 147). Faculty members are reviewed by two independent reviewers—for Nikolova, Constantine Caramanis (“Caramanis”) and Michael Orshansky (“Orshansky”). The Chair of the FAR Committee, at that time Julien, synthesizes the review and then presents it to the committee. (Id. at 145, 147). Caramanis and Orshansky gave Nikolova a rating of “Meets Expectations” for 2018-2019. Caramanis wrote as bases for his review, “reasonable publication record. Teaching record is also ok. Service to the department is low, but perhaps this is on account of having been on leave for the first semester.” (Id. at 147). (FAR Emails, Dkt. 39-1, at 241). The anonymized comments similarly state, “Dr. Nikolova has a reasonable publication and teaching record. She maintains a modestly funded

research group. During this period, her service to the department was relatively low, but this is also reasonable, given the semester of leave.” (FAR Evaluation, Dkt. 44-4, at 143). On July 2, 2020, Nikolova amended her complaint to add retaliation claims. (Am. Compl, Dkt. 21, at 12). E. The Claims Nikolova asserts causes of action for sex and pregnancy discrimination and retaliation under Title VII of the Civil Rights Act of 1964, 42 U.S.C.§ 2000e et seq., as amended (“Title VII”) and for violation of the Equal Pay Act, 29 U.S.C. §206(d)(1). (Am. Compl., Dkt. 21, at 11–16). UT seeks summary judgment on Nikolova’s Title VII claims only. II. LEGAL STANDARDS Summary judgment is appropriate when there is no genuine dispute as to any material fact

and the moving party is entitled to judgment as a matter of law. Fed. R. Civ. P. 56(a); Celotex Corp. v. Catrett, 477 U.S. 317, 323–25 (1986). A dispute regarding a material fact is “genuine” if the evidence is such that a reasonable jury could return a verdict in favor of the nonmoving party. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986). “A fact is material if its resolution in favor of one party might affect the outcome of the lawsuit under governing law.” Sossamon v. Lone Star State of Tex., 560 F.3d 316, 326 (5th Cir. 2009) (quotations and footnote omitted). When reviewing a summary judgment motion, “[t]he evidence of the nonmovant is to be believed, and all justifiable inferences are to be drawn in his favor.” Anderson, 477 U.S. at 255. Further, a court may not make credibility determinations or weigh the evidence in ruling on a motion for summary judgment. Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 150 (2000). Once the moving party has made an initial showing that there is no evidence to support the nonmoving party’s case, the party opposing the motion must come forward with competent

summary judgment evidence of the existence of a genuine fact issue. Matsushita Elec. Indus. Co. v. Zenith Radio, 475 U.S. 574, 587 (1986). Unsubstantiated assertions, improbable inferences, and unsupported speculation are not competent summary judgment evidence, and thus are insufficient to defeat a motion for summary judgment. Turner v. Baylor Richardson Med. Ctr., 476 F.3d 337, 343 (5th Cir. 2007). Furthermore, the nonmovant is required to identify specific evidence in the record and to articulate the precise manner in which that evidence supports his claim. Adams v. Travelers Indem. Co. of Conn., 465 F.3d 156, 164 (5th Cir. 2006). Rule 56 does not impose a duty on the court to “sift through the record in search of evidence” to support the nonmovant’s opposition to the motion for summary judgment. Id. After the nonmovant has been given the opportunity to raise a genuine factual issue, if no reasonable juror could find for the nonmovant, summary judgment will be granted. Miss. River Basin All. v. Westphal, 230 F.3d 170, 175 (5th Cir. 2000).

III. DISCUSSION UT seeks summary judgment on all Nikolova’s Title VII claims: (1) sex and pregnancy discrimination resulting in disparate impact, (2) sex and pregnancy discrimination resulting in disparate treatment, and (3) retaliation. (Mot. Summ. J., Dkt. 39, at 2). In her response, Nikolova represents to the Court that she no longer wishes to pursue her disparate impact claim. (Resp., Dkt. 44, at 28). Accordingly, the Court considers this claim to have been waived, and as such UT’s motion is moot as to that claim. Thus, the Court will deny as moot the motion for summary judgment on Nikolova’s disparate impact claim. As to the other two claims, the Court finds summary judgment is improper at this time for the reasons set forth below. A. Sex and Pregnancy Discrimination (Disparate Treatment) Nikolova brings sex and pregnancy discrimination claims against UT under Title VII. Because Nikolova attempts to prove sex-based discrimination under Title VII through

circumstantial evidence, the Court must evaluate her claims under the burden-shifting framework set out in McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973), as modified by the Fifth Circuit. Paske v. Fitzgerald, 785 F.3d 977, 984 (5th Cir. 2015). “Under the modified McDonnell Douglas approach, the plaintiff must first demonstrate a prima facie case of discrimination; the defendant then must articulate a legitimate, non-discriminatory reason for its decision . . . ; and, if the defendant meets its burden of production, the plaintiff must then offer sufficient evidence to create a genuine issue of material fact that either (1) the employer’s reason is a pretext or (2) that the employer’s reason, while true, is only one of the reasons for its conduct, and another ‘motivating factor’ is the plaintiff’s protected characteristic.” Burrell v. Dr. Pepper/Seven Up Bottling Grp., Inc., 482 F.3d 408, 411–12 (5th Cir. 2007). To establish her prima facie case, Nikolova must show that she “(1) is a member of a protected class; (2) was qualified for the position; (3) was subject to an adverse employment action; and (4) was replaced by someone outside of the protected class, or, in the case of disparate

treatment, shows that other similarly situated employees were treated more favorably.” Standley v. Rogers, 680 F. App’x 326, 327 (5th Cir. 2017) (quoting Bryan v. McKinsey & Co., 375 F.3d 358, 360 (5th Cir. 2004)). Nikolova asserts a prima facie case of discrimination based on gender and pregnancy. (Resp., Dkt. 44, at 21–23). As UT acknowledges, Nikolova’s burden on this front is “not onerous.” Tex. Dep’t of Cmty. Affairs v. Burdine, 450 U.S. 248, 253 (1981). Nikolova points to several pieces of evidence to establish a prima facie case. First, in Wood’s assessment letter, she stated—allegedly unnecessarily—that Nikolova did not teach two courses, and was on modified instructional duties, because she was pregnant. (Wood Assessment, Dkt. 44-1, at 475). Nikolova claims that these statements establish pregnancy discrimination because they were part of the tenure denial decision, and Wood discussed these comments with Fenves as part of his decision. See Arismendez v. Nightingale Home Health Care, Inc., 493 F.3d 602, 608 (5th Cir. 2007) (stating under an analogous standard that

“remarks may serve as sufficient evidence of . . . discrimination if the offered comments are: 1) [pregnancy] related; 2) proximate in time to the terminations; 3) made by an individual with authority over the employment decision at issue; and 4) related to the employment decision at issue”). Although the Court tends to agree with UT that these comments alone do not constitute direct evidence of discrimination, (Reply, Dkt. 47, at 2), in combination with other circumstantial evidence they tend to support Nikolova’s prima facie case. In addition, Nikolova presents evidence that UT treated similarly situated men and one woman who was not pregnant more favorably when considering their records on the factors relevant to the tenure decision: research funding, publication record, and teaching. (Resp., Dkt. 44, at 23). The Fifth Circuit recently explained that when considering comparator evidence, a “plaintiff ‘must establish that the comparator was treated more favorably than the plaintiff under nearly identical circumstances.’” Ernst v. Methodist Hosp. Sys., 1 F.4th 333, 340 (5th Cir. 2021) (citing Rogers v.

Pearland Indep. Sch. Dist., 827 F.3d 403, 410 (5th Cir. 2016) (cleaned up)). “The employment actions being compared will be deemed to have been taken under nearly identical circumstances when the employees being compared held the same job or responsibilities, shared the same supervisor or had their employment status determined by the same person, and have essentially comparable violation histories.” Lee v. Kan. City S. Ry. Co., 574 F.3d 253, 260 (5th Cir. 2009). However, the Fifth Circuit has made clear that courts do not “interpret ‘nearly identical’ as synonymous with ‘identical.’” Lee v. Kansas City S. Ry. Co., 574 F.3d 253, 260 (5th Cir. 2009). This is because, when “applied to the broader circumstances of a plaintiff’s employment and that of his proffered comparator, a requirement of complete or total identity rather than near identity would be essentially insurmountable, as it would only be in the rarest of circumstances that the situations of two employees would be totally identical.” Id. The parties dispute whether the individuals Nikolova highlights are appropriate comparators

and are indeed similarly situated based on interpretation of characteristics including department, research area, and timing of application. (Reply, Dkt. 47, at 4–5). Nikolova claims that all comparators had the same decisionmakers on their tenure decisions (Fenves and Wood), were all assistant professors in the engineering school, all had the same professional responsibilities and expectations, and were all evaluated under the same tenure criteria. (Resp., Dkt. 44, at 23 n.58). UT counters with facts about each individual in an attempt to distinguish their specific circumstances from Nikolova’s. (Reply, Dkt. 47, at 4–5). This question turns on the level of specificity with which one defines their respective jobs and responsibilities, supervisors, employment status decisionmakers, and histories. (See Lee, 574 F.3d, at 260). In light of the low bar for plaintiffs at this stage of the inquiry, see Burdine, 450 U.S. at 253, the Court finds Nikolova’s evidence sufficient to establish a prima facie case of discrimination by UT and thus proceeds to the next stages of the Title VII test.

Next UT claims that, even if Nikolova can establish a prima facie case of discrimination, its tenure denial decision was not discriminatory because of two concerns Fenves highlighted with her application: (1) her research funding, and (2) her publication record. (Mot. Summ. J., Dkt. 39, at 22; Fenves Dep., Dkt. 39-2, at 121). Fenves explained that “[w]ith the facts presented, the timing of this particular case, it wasn’t ready for a promotion decision or a terminal-appointment decision.” (Id. at 122). However, Fenves’s decision was “without prejudice,” and could be revised at a later date. (Id. at 41). UT asserts that these facts constitute “a legitimate, non-discriminatory reason for its decision” sufficient to meet its burden of production. Burrell v. Dr. Pepper/Seven Up Bottling Grp., Inc., 482 F.3d 408, 411 (5th Cir. 2007). The Court agrees. Under the modified McDonnell Douglas approach, then, Nikolova must demonstrate that “either (1) [UT’s] reason is a pretext or (2) that [UT’s] reason, while true, is only one of the reasons for its conduct, and another ‘motivating factor’ is [Nikolova’s] protected characteristic.” Id. at 412; see Mayberry v. Vought Aircraft Co., 55 F.3d 1086,

1091 (5th Cir. 1995) (“The question is not whether an employer made an erroneous decision; it is whether the decision was made with discriminatory motive”). “To carry this burden, the plaintiff must produce substantial evidence indicating that the proffered legitimate nondiscriminatory reason is a pretext for discrimination. . . . The plaintiff must rebut each nondiscriminatory reason articulated by the employer.” Laxton v. Gap Inc., 333 F.3d 572, 578 (5th Cir. 2003). UT claims Nikolova lacks evidence that Fenves’s decision was based on her gender or pregnancies rather than his articulated concerns regarding research funding and publication. (Mot. Summ. J., Dkt. 39, at 22). However, UT’s conclusory assertion to this effect overlooks significant evidence put forth by Nikolova that gender and pregnancy may have been motivating concerns, even if not the only factors in the decision. Furthermore, to survive summary judgment, Nikolova need not establish that research funding and publication record played no role in the decision; she must only put forth evidence sufficient for a reasonable jury to find that her protected characteristics constituted another

“motivating factor” in UT’s decision. See Burrell, 482 F.3d at 412. “The issue at the pretext stage is whether [the defendant’s] reason, even if incorrect, was the real reason” for the employment decision—that the defendant’s reason was discriminatory “or at the least, that [the defendant’s] explanation of its motive is false.” Sandstad v. CB Richard Ellis, Inc., 309 F.3d 893, 899 (5th Cir. 2002). “Evidence demonstrating that the employer’s explanation is false or unworthy of credence, taken together with the plaintiff’s prima facie case, is likely to support an inference of discrimination even without further evidence of defendant’s true motive . . . because ‘once the employer’s justification has been eliminated, discrimination may well be the most likely alternative explanation. . . .’” Laxton v. Gap Inc., 333 F.3d 572, 578 (5th Cir. 2003) (citing Reeves v. Sanderson Plumbing Prod., Inc., 530 U.S. 133, 147–48 (2000)). Evidence from a plaintiff’s prima facie case of discrimination may still be relevant as, while a defendant’s articulation of a non-discriminatory basis for its actions “destroys the legally mandatory inference of discrimination arising from the

plaintiff’s initial evidence . . . , this evidence and inferences properly drawn therefrom may be considered . . . on the issue of whether the defendant’s explanation is pretextual [and] . . . , combined with effective cross-examination of the defendant, [may] suffice to discredit the defendant's explanation. Texas Dep’t of Cmty. Affs. v. Burdine, 450 U.S. 248, 255 n.10 (1981). “Proof that the defendant’s explanation is unworthy of credence is simply one form of circumstantial evidence that is probative of intentional discrimination, and it can be quite persuasive.” Reeves v. Sanderson Plumbing Prod., Inc., 530 U.S. 133, 134 (2000). Nikolova claims that Fenves’s reasons--his concerns about her research funding and publication record--are pretextual. As to her publication record, Fenves explained that “both journal publications and conference publications were not at the level that we would really like to see for a tenured faculty member.” (Fenves Dep., Dkt. 39-2, at 99). For Fenves, Nikolova’s “teaching was not an issue.” (Id. at 193). Nikolova asserts that “[t]he ECE Dept., Chair, the College P&T Committee,

all external reviewers, and even Dean Wood found [her] publications were strong and merited tenure.” (Resp., Dkt. 44, at 23). Of these, all but Fenves and Wood believed that her research funding was strong. (Id.). Nikolova notes that only Wood cited her teaching record as a reason for denial of tenure, but that Wood’s reliance on her teaching creates a fact issue regarding the real reasons behind, and propriety of, the decision. (Resp, Dkt. 44, at 24). Moreover, the shift from teaching as a criterion for denial to Fenves’s identification of research and publication as the sole bases for denial may obscure the role of protected characteristics in the decision. (Id.). See Gee v. Principi, 289 F.3d 342, 347–48 (5th Cir. 2002) (an employer’s inconsistent explanations for its employment decisions at different times permits a jury to infer that the employer’s proffered reasons are pretextual). To the extent that UT claims Wood’s reasoning cannot be imputed to Fenves and UT and is therefore irrelevant, (Reply, Dkt. 47, at 2),6 Nikolova presents sufficient evidence at least to raise an issue of material fact that Fenves relied on Wood in making his decision. (Resp., Dkt. 44,

at 14). Wood made a recommendation to Fenves and met with the President’s Committee to discuss the case. (Id.). She explained to them why her recommendation to deny tenure diverged from the P&T Committee’s recommendation to grant. (Fenves Dep., Dkt. 39-2, at 101–02, 121). Fenves explained that “[b]ased on the dossier and the recommendation of the Dean to ‘do to not promote,’[sic.] [he and the President’s Committee] agreed with the Dean.” (Id. at 99). Thus, the Court finds that factual disputes remain that prevent it from deciding the issue of pretext.

6 For example, in an email Wood characterized Nikolova’s tenure candidacy as “the most controversial” and wrote that “I did not state all my reasons in the letter as I felt that some things were best discussed with the committee.” (Most Controversial Email, Dkt. 44-3, at 156). Wood later expressed no recollection as to what these matters might be. (Wood Dep., Dkt. 39-2, at 286). While the Court does not wish to overstate the significance of this evidence, it agrees with Nikolova that it creates a genuine issue of material fact about which “a jury should have the opportunity to decide.” In light of this conflicting evidence at multiple points in the inquiry, the Court finds that material issues of fact remain as to whether UT discriminated against Nikolova in violation of Title

VII by denying her tenure. Accordingly, UT’s motion for summary judgment on these claims is denied. B. Retaliation Nikolova also brings retaliation claims under Title VII. (Am. Compl., Dkt. 21, at 12). Title VII’s anti-retaliation provision prohibits employers from “discriminat[ing] against” an employee “because he has opposed any practice made an unlawful employment practice” by Title VII or “because he has made a charge, testified, assisted, or participated in any manner in an investigation, proceeding, or hearing” under Title VII. 42 U.S.C. § 2000e-3(a). Because Nikolova has once again only offered circumstantial evidence of retaliation, the Court will similarly apply the modified McDonnell Douglas burden-shifting paradigm to her retaliation claims. Under this approach, Nikolova must first establish a prima facie case of retaliation by showing that (1) she engaged in a protected activity, (2) an adverse employment action occurred, and (3) a causal link existed between the

protected activity and the adverse employment action. Hernandez v. Metro. Transit Auth. of Harris Cnty., 673 F. App’x 414, 419 (5th Cir. 2016). An employee engages in a protected activity by (1) opposing an employment practice made unlawful by Title VII, or (2) by making a charge or testifying, assisting, or participating in an investigation, proceeding, or hearing under Title VII. See Long v. Eastfield Coll., 88 F.3d 300, 304 (5th Cir. 1996) (quoting 42 U.S.C. § 2000e-3(a)). The underlying practice need not actually be illegal, but the employee must at least reasonably believe that it is. See id. The parties do not dispute that Nikolova engaged in a protected activity by filing a discrimination charge with the EEOC. (See Mot. Summ. J., Dkt. 39, at 23). A plaintiff must establish she suffered an adverse employment action to sustain a retaliation claim. “An action qualifies as materially adverse if it might dissuade a reasonable worker from making or supporting a charge of discrimination.” Murillo v. Travis Cty., Texas, No. AU-17-CA- 00113-SS, 2018 WL 1514094, at *6 (W.D. Tex. Mar. 26, 2018) (citing Burlington Northern. & Santa Fe

Ry. Co. v. White, 548 U.S. 53, 67–68, (2006)). “Title VII’s anti-retaliation provisions do not protect employees from petty slights, minor annoyances, and simple lack of good manners.” Welsh v. Fort Bend Indep. Sch. Dist., 941 F.3d 818, 827 (5th Cir. 2019) (internal quotations omitted). However, an adverse employment action in the retaliation context is not limited to “ultimate employment decisions” like hiring and firing, but instead includes actions that might dissuade a reasonable worker from protected activity. See Teague v. Williamson Cty., No. 1:18-CV-1098-RP, 2020 WL 2542869, at *12 (W.D. Tex. May 19, 2020); Garvin v. Sw. Corr., L.L.C., 391 F. Supp. 3d 640, 653 (N.D. Tex. 2019). Nikolova cites three examples that she claims constitute adverse employment actions: (1) an unfavorable peer teaching evaluation, (2) an unfavorable Faculty Annual Review (“FAR”), and (3) her supervisory assignment, all of which she received after filing a charge of discrimination with the EEOC. (Am. Compl., Dkt. 21, at 13–14).7 Nikolova does not address the supervisory assignment, so

the Court considers that argument against summary judgment to have been waived. As to the two evaluations, she explains that “in the context of an assistant professor attempting to obtain tenure at a tier one university, poor evaluations as shown here . . . would absolutely discourage other similar professors from speaking out against discrimination.” (Resp., Dkt. 44, at 27; see Fenves Dep., Dkt. 39-2, at 144–45).

7 While the peer evaluation occurred around the same time as her EEOC charge, it occurred after Nikolova’s email to the ECE Department charging gender discrimination, for which she claims she faced retaliation. While alone this evidence may not be sufficient, the Court views it in combination with the negative FAR, which unquestionably occurred after she filed her EEOC charge and the instant action. (See Resp., Dkt. 44, at 27). UT claims that neither of these actions constitute adverse employment actions within the meaning of a retaliation claim. Regarding the peer teaching evaluation, UT claims that a “lower-than-

expected job performance review” does not qualify as an adverse employment decision. UT cites Mitchell v. Snow, 326 F. App’x 852, 856 (5th Cir. 2009), for the proposition that “an employment review lower than [the plaintiff] expected would not have dissuaded a reasonable employee from asserting discrimination.” However, the specific circumstances of that case make it distinguishable from the facts here. The Court noted there that the review was “still quite strong,” that the accompanying comments were “quite critical,” and the review followed resolution of an EEOC complaint specifically addressing low performance reviews. Id. Thus, the Court finds that case is not dispositive here. Similarly, Jackson v. Honeywell Int'l, Inc., 601 F. App’x 280, 286 (5th Cir. 2015), is distinguishable because there, the plaintiff had documented performance deficiencies for years prior to the reviews in question, whereas here Nikolova cites her prior positive performance reviews. (Resp., Dkt. 44, at 27; Chair Recommendation, Dkt. 39-1, at 98; Chase Evaluation, Dkt. 44-4, at 127). UT further claims that Julien’s comments were a “legitimate, non-retaliatory,” “bona fide

assessment” rather than pretext for retaliation. (Mot. Summ. J., Dkt. 39, at 23). Nikolova counters that the teaching evaluation is contrary to prior positive evaluations and is inconsistent with two male assistant professors who received higher evaluations for poorer performances than hers. (Chair Recommendation, Dkt. 39-1, at 98; Chase Evaluation, Dkt. 44-4, at 127; Resp., Dkt. 44, at 21). Furthermore, she claims that UT departed from its own procedures by conducting her evaluation after its deadline had passed, with a graduate student presenting in her place, and without the required pre- and post-observational meetings. (Resp., Dkt. 44, at 27 n.66). Finally, Nikolova asserts that Julien’s criticism of Nikolova for her “elephant in the room” comment, and Julien’s statement that the comment would have a “negative impact” on other women in the department, are evidence of pretext for both evaluations. (Julien Dep., Dkt. 39-2, at 160).). See Laxton v. Gap Inc., 333 F.3d 572, 583 (5th Cir. 2003) (“An oral statement exhibiting discriminatory animus may be used to demonstrate pretext or . . . it may be used as additional

evidence of discrimination” when it “demonstrate[s] discriminatory animus and [is] made by a person primarily responsible for the adverse employment action or by a person with influence or leverage over the formal decisionmaker.”). As to the FAR, Nikolova states that her “Meets Expectations” evaluation also constitutes an adverse employment action. (Resp., Dkt. 44, at 26). UT attempts to classify this evaluation as the type of “minor annoyance” that is insufficient to sustain a retaliation claim. See Welsh, 941 F.3d at 827. It points to Tewfik’s testimony that such evaluations are “typically not given a lot of weight.” (Tewfik Dep., Dkt. 39-2, at 242). The two professors who gave Nikolova her rating, Caramanis and Orshansky, wrote that she has a “reasonable publication and teaching record,” that she has a “modestly funded research group,” and that “her service to the department was relatively low, but this is also reasonable, given the semester of leave.” (ECE Performance Reviews, Dkt. 39-2, at 780). In isolation these comments seem neutral, even unremarkable. However, the Court cannot presume

to understand their significance when placed in their proper context. The evidence suggests that professors tend to be cautious in their evaluations, such that tepid language such as this may indicate a negative review. For example, in an email, Yale Patt states that he “is not going to put anything negative on paper,” that he “bend[s] over backward to be positive in [his] Peer teaching evaluation and put nothing in writing that could eventually hurt a colleague’s promotion package,” and that senior colleagues at other universities had “[n]ever” “seen any negative comment on a teaching evaluation in a promotion package.” (Patt Email, Dkt. 44-4, at 135–36). As various deposition testimony demonstrates, even such seemingly innocuous language can in context be a “serious red flag.” (See Fenves Dep., Dkt. 39-2, at 160; Tewfik Dep., Dkt. 39-2, at 118–119 (analyzing negative valence of letter in which “nothing . . . to the uneducated, would identify as subpar performance”). And Tewfik made clear that the performance evaluation rating is “one input” that he—the department chair—as well as the Dean and Provost use in formulating recommendations and

making the ultimate tenure decision. (Id. at 245). As Nikolova notes, determining whether an action constitutes an adverse employment action is a highly context-specific determination, grounded in professional norms of the industry. “[T]he significance of any given act of retaliation will often depend upon the particular circumstances. Context matters. The real social impact of workplace behavior often depends on a constellation of surrounding circumstances, expectations, and relationships.” Burlington Northern. & Santa Fe Ry. Co. v. White, 548 U.S. 53, 67 (2006). The Court recognizes that there are many subtleties of the interpersonal relationships, language, and tone that have significant implications for a decision as personalized and subjective as a tenure determination. Those implicit messages, which may be clear to the participants in these exchanges, are not easily reduced to paper. When dealing with disputed evidence of this nature, the importance of a jury’s role in weighing credibility cannot be overstated. Particularly when taken together, the two evaluations present little factual clarity. Given the

significance of context in understanding the weight of the comments and evaluations at issue, and the questions that Nikolova raises as to their meaning, the Court cannot accept UT’s assertion that no facts are in dispute on this matter. Having found at least a genuine factual dispute on the adverse employment action prong, the Court turns to whether Nikolova has established a causal connection between those actions and her protected activity. To demonstrate a causal nexus, the plaintiff’s evidence must establish that adverse employment activity was taken “at least in part on the knowledge of the plaintiff’s [protected] activity.” Standley v. Rogers, 202 F. Supp. 3d 655, 670 (W.D. Tex. 2016), aff’d, 680 F. App’x 326 (5th Cir. 2017); see Ackel v. Nat’l Commc'ns, Inc., 339 F.3d 376, 385–86 (5th Cir. 2003). However, “mere knowledge” alone of the protected activity is insufficient. See Hutto v. Univ. of Houston Sys., Civ. Ac. No. V–05–70, 2008 WL 4453427, at *7 (S.D. Tex. Sept. 28, 2008). If the plaintiff can establish temporal proximity between this knowledge and the adverse employment action, this proximity can

support a finding of a causal link so long as the knowledge and the adverse action are “very close” in time. Clark Cty. Sch. Dist. v. Breeden, 532 U.S. 268, 273 (2001); see Strong v. Univ. Healthcare Sys., L.L.C., 482 F.3d 802, 808 (5th Cir. 2007). Nikolova asserts that the temporal proximity between her negative peer evaluation and her email charging discrimination—only two months—satisfies this requirement. Absent other facts tending to show causation, “plaintiffs may rely solely on temporal proximity between protected activity and an adverse employment action only if the two are very close.” Zamora v. City of Houston, 798 F.3d 326, 335 (5th Cir. 2015) (citing Clark Cty. Sch. Dist., 532 U.S. at 273–74). While Nikolova is correct that the nine-month gap between her teaching evaluation and her EEOC charge is likely too long to support her claim, (Resp., Dkt. 44, at 27), the two months between her peer evaluation and her protected activity does demonstrate such proximity. See Clark Cty. Sch. Dist. v. Breeden, 532 U.S. 268, 273–74 (2001) (noting that while a three-month or four-month period may be close

enough in time to support an inference of causation, twenty months was too long); Washburn v. Harvey, 504 F.3d 505, 511 (5th Cir. 2007) (finding that a two-year separation between participation in a protected activity and an alleged adverse employment action insufficient to support an inference of causation). There may very well be nonretaliatory reasons for the evaluations that Nikolova received. However, because material disputes of fact exist as to whether Nikolova experienced an adverse employment action, and whether the action in question was causally connected to her protected activity, the Court finds that these issues are more properly resolved by a jury. As such, the Court will deny UT’s motion for summary judgment on this basis. IV. CONCLUSION For the reasons stated above, IT IS ORDERED that UT’s Motion for Partial Summary Judgment, (Dkt. 39), is DENIED. IT IS FURTHER ORDERED that Nikolova’s Motion for Leave to File Sur-Reply, (Dkt. 50), is DENIED AS MOOT. SIGNED on February 15, 2022.

ROBERTPITMAN UNITED STATES DISTRICT JUDGE