JOHN DOE

v.

ROLLINS COLLEGE

JORDAN, Circuit Judge:

Jane Roe, a student at Rollins College, accused John Doe, a fellow student, of sexual assault. Following an investigation, Rollins determined that Doe violated its sexual misconduct policy. Doe was able to graduate and receive his undergraduate degree but was not allowed to participate in commencement/graduation ceremonies. Rollins imposed a sanction of dismissal, resulting in permanent separation of Doe without the opportunity for readmission; privilege restrictions, including a prohibition on participating in alumni reunion events on or off campus; and a contact restriction as to Roe.

Doe sued Rollins in federal court, asserting two claims under Title IX,²⁰ U.S.C. § 1681—one for selective enforcement and one for erroneous outcome—and a third claim under Florida law for breach of contract. Following discovery, the district court excluded the opinions proffered by Doe&#x²⁷;s expert as to Rollins&#x²⁷; purported gender bias. Then, on cross-motions for summary judgment, the district court (a) entered summary judgment in favor of Rollins on the Title IX claims and (b) entered partial summary judgment in favor of Doe on the breach of contract claim.

On the Title IX claims, the district court concluded that there was "no evidence by which a reasonable juror could conclude [that] Rollins[&#x²⁷;] conduct toward Doe was motivated by his gender."

D.E. 156 at 16. On the breach of contract claim, the district court concluded that Doe had shown that Rollins breached a provision of its Title IX sexual misconduct policy stating that the process for resolving a sexual misconduct claim would typically be completed within 60 days. The investigation began in November of 2017, but Rollins did not complete it until March of 2018. See id. at 17. [fn 1]

The district court did not enter judgment in favor of Doe on the contract claim because it concluded that a jury had to decide whether Rollins&#x²⁷; breach was material and caused harm. See id. at 26-27. The district court also ruled that there were factual questions as to whether Rollins&#x²⁷; inquiry into Doe&#x²⁷;s sexual history was relevant and whether Rollins responded fairly and equitably to Roe&#x²⁷;s allegations. See id. at 27–28.

After a three-day trial, the jury returned a verdict in favor of Rollins on the breach of contract claim. The jury found that Rollins&#x²⁷; breach of the 60-day provision was not material; that Rollins did not breach a contractual obligation to not have irrelevant prior sexual history considered in the investigation of Roe&#x²⁷;s allegations; and that Rollins did not breach a contractual obligation to reach a timely and fair resolution of Roe&#x²⁷;s allegations.

This is Doe&#x²⁷;s appeal. Following oral argument and a review of the record, we affirm. We conclude that the district court did

not abuse its discretion in precluding Doe&#x²⁷;s expert from presenting opinions about Rollins&#x²⁷; purported gender bias, and that it correctly granted summary judgment in favor of Rollins on Doe&#x²⁷;s two Title IX claims. On the breach of contract claim, we cannot review Doe&#x²⁷;s challenge to the district court&#x²⁷;s partial denial of summary judgment because materiality is not a pure legal issue under Florida law and was later resolved by the jury. Insofar as Doe contends that he was entitled to judgment as a matter of law on materiality, we disagree because the evidence on that issue was disputed and the jury could have found that the breach was not material.

I

In his Title IX claims, Doe asserts that Rollins—based on its gender bias—selectively prosecuted him and erroneously found him to be in violation of its sexual misconduct policy. To provide the necessary context for these claims, we set out the different versions of events presented by Roe and Doe.

A

From 2014 to 2017, Doe was a student at Rollins. During that time, he met Roe, another student at Rollins, and the two became friends. They "hung out around campus, did homework" and "spent time together.” D.E. 61-4 at 5. On February¹⁷, 2017, Doe invited Roe to his fraternity&#x²⁷;s “Grab-a-Date” event. The event was "champagne and shackles themed," which meant that dates were "zip-tie[d]" to one another until a shared “bottle of

champagne [was] empty." D.E. 61-2, Exh. B at 18. Roe accepted Doe&#x²⁷;s invitation, and the two went to the event together that evening.

Later, in the early morning hours of February¹⁸, Doe and Roe, along with others from the party, walked to a local bar. Eventually, Doe and Roe left the bar together and went to Doe&#x²⁷;s room. But Doe&#x²⁷;s and Roe&#x²⁷;s recollections of how they ended up at Doe&#x²⁷;s room, and what happened once there, differ.

Roe stated that her “intention was to go to [her] own room to go to sleep, by [herself], because [she] was incoherently drunk and incredibly tired" but "in between &#x²⁷;blackouts&#x²⁷; somehow [she] ended up in [Doe&#x²⁷;s] dorm and room.” Id. at 20–21. She told Doe "multiple times that [she] wanted to leave and just go to bed in [her] own room but [Doe] insisted [they] hang out and to just wait [there in his room] because he had to go to the bathroom.” Id. at 21. Roe "too tired to argue, and too confused to want to go elsewhere”—saw Doe&#x²⁷;s couch, laid down, and “fell asleep for what felt like⁵ or¹⁰ min[utes]." Id. When Doe returned, he picked a movie and Roe fell asleep again. While Roe was on the couch and as the movie was playing, Doe “started making out with [Roe] and touch-ing [her] near [her] chest/bra area,” then “started touching [her] underwear and moved it aside and began penetrating [her] with his finger." Id.

Roe recounted "saying [n]o&#x²⁷; and objecting [to] his advances through multiple parts of the beginning of this assault, but [her] right hand was trapped in the fold of the couch, and [her] left

trapped under him as he was on top of [her] and [she] was too tired, dizzy, and intoxicated to do anything other than try to say no.” Id. Roe continued to say “no” and fell in and out of sleep. She ex-plained:

Id.

As he was touching my thighs I said no, as he was touching my vagina I said no, and as he began penetrating me I said no[.] I do think during these no&#x²⁷;s I was kissing him (in the sense where kissing means my lips were on his) at this point but I was intoxicated, falling in and out of sleep, and felt as though I was unable to move my own body. I remember seeing his penis out as he was on top of me and my right hand at my sides, trapped between the couch and my body, and my left hand under him, which he was now moving towards his penis as I was saying no. This part of the contact was only for about⁴ seconds at most (that my hand was touching his penis).

According to Roe, Doe asked "what[&#x²⁷;]s wrong" multiple times and "discouraged [Roe] from leaving when [she] would ex-press that [she] wanted to be in [her] own room.” Id. When Doe had "shifted off of Roe to ask what was wrong, she sat up and "finally stood up in an attempt to get out and head towards the door." Id. at 21-22. Doe "stopped [Roe] and was holding [her] arms and was saying things like &#x²⁷;no, just stay&#x²⁷; and &#x²⁷;we don&#x²⁷;t have to have sex&#x²⁷; and &#x²⁷;you can just sleep here&#x²⁷; and would keep trying to kiss [Roe]

and move [her] over to the bed.” Id. at 22. Roe, wanting to go to her own bed, left Doe&#x²⁷;s room.

Doe provided a very different account of the evening. While at the local bar, he and Roe sat "in the outdoor section" by them-selves and “had normal conversation.” D.E. 61-2, Exh. C at 26. “It started getting late" and they "decided to return to [his] dorm room to watch a movie.” Id. They "began walking and talking on the way back, and ultimately started holding hands until [they] arrived to [their] building on campus." Id. Once they got to Doe&#x²⁷;s room, Doe went downstairs to the water fountain with his room-mate, who “asked if anything was going on between [Roe] and [Doe], as [Roe] appeared to be interested in [Doe] over the course of the night and was knowledgeable that the two of [them] had been talking regularly for quite some time.” Id. Doe “was not sure what to expect and was going to see where the night would take [him]." Id.

Afterwards, in Doe&#x²⁷;s room, Doe and Roe “played a movie on Netflix and sat next to each other." Id. Doe described what happened next as follows:

I put my arm around [Roe], in which she was comfortable with. we started kissing. We continued on and I started touching her intimate parts, in which she showed no resistance. I penetrated her digitally and there were no signs of her being uncomfortable. Eventually she paused, and I asked if everything was okay. I asked because I knew we had been friends for an extended period of time and perhaps getting

intimate was not worth losing a friendship over. She said that everything was fine and then continued to unbuckle my belt, unbutton my pants, and then ex-posing and touching my penis. In no way, shape or form did I cause her to touch my intimate body parts. We continued with kissing and forms of petting until [Roe] paused again. I asked if everything was okay, and this second time, she indicated that she wanted to stop. We stopped immediately and continued watch-ing the movie for a little while longer. It had gotten late (after 2 AM) and I wanted to go to bed. I said that she was welcome to spend the night on the futon or to go home (right next door) if she wanted to. She stated that she wished to go home and I walked her out.

Id. at 26-27.

B

A few days later, on February²², 2017, Roe went to Rollins&#x²⁷; Title IX office and met with the Title IX Coordinator, Oriana Jimé-nez Guevara, to discuss what happened on February¹⁷—an incident that she believed to be in violation of the Rollins Title IX sexual misconduct policy. See D.E. 56-5 at 2-3. Roe "chose not to exercise her right to request an investigation" at that time, but she "filed a report" and requested “interim measures” such as housing accommodations-moving to a different building and to a single room—as well as academic accommodations. See D.E. 56-2 at 117, 123–124.

Later that year, in November of 2017, “an anonymous caller reported . . . that an individual with a name similar to [Doe&#x²⁷;s] sexually assaulted three girls." D.E. 56-5 at 3. Ms. Jiménez Guevara called Roe “to see if [Roe] had additional information about the caller." Id. They also discussed whether Roe wanted to move for-ward with an investigation, and ultimately she decided to do so.

On November²⁰, 2017, Ms. Jiménez Guevara called Doe and advised him that she needed to meet with him regarding a Title IX case he was involved in. Doe met with Ms. Jiménez Guevara that day and was informed that a complaint alleging sexual misconduct had been made against him by Roe. Doe was also advised of "some of the rights that [he] had[,]" including the right to an advisor. See D.E. 61-4 at 24. According to Doe, during the meeting Ms. Jiménez Guevara said “something to the effect of, &#x²⁷;[w]hen it&#x²⁷;s all said and done, I will make sure that you appeal this case.” Id. Doe understood that to mean “that the decision of [the Title IX] case had already been predetermined before [he] even had a single question asked of [him]." Id.

Rollins selected Deena Wallace as the outside independent investigator. Ms. Wallace, an attorney, had previously served as a sex-crimes prosecutor at the state attorney&#x²⁷;s office. See D.E. 228 at 88-89.

Ms. Wallace interviewed Roe and²² other witnesses (in-cluding²⁰ currently-enrolled Rollins students) as part of her investigation. She interviewed Doe on December⁷, 2017. During the interview, Doe had a lawyer hired by his mother present as his

advisor. Doe also provided two written statements to Rollins. Fol-lowing the interviews, Ms. Wallace prepared a 71-page report set-ting out what each of the witnesses had said. See D.E. 56-5 & D.E. 56-6.

In the report, Ms. Wallace summarized each witness&#x²⁷; testimony, considered additional evidence (e.g., text messages), made credibility determinations, explained those determinations, and applied Rollins&#x²⁷; sexual misconduct policy to the evidence. Using a preponderance of the evidence standard, she found that Roe&#x²⁷;s account was more credible than Doe&#x²⁷;s, and concluded that Doe had violated Rollins&#x²⁷; sexual misconduct policy by digitally penetrating Roe without her consent.

Ms. Jiménez Guevara reviewed the report, made some stylistic edits, and concluded that Ms. Wallace&#x²⁷;s “process, and what she wrote in the report, was permissible" and "followed guide-lines.” D.E. 56-2 at 183–84. Doe then submitted an appeal to Rol-lins. Mamta Accapadi, Rollins&#x²⁷; Vice-President for Student Affairs, reviewed the appeal and determined that Rollins had "followed [its] process.” D.E. 61-9 at 15. Because the investigation was completed after Doe had already earned his undergraduate degree, it did not impact that degree. See D.E. 61-9 at 17. [fn 2]

II

We begin with Doe&#x²⁷;s challenge to the district court&#x²⁷;s order limiting the testimony of his expert, Professor Robert K.C. John-son, to the history of Title IX and excluding his opinions on “the gender bias and fairness of Rollins[&#x²⁷;] procedures.” D.E. 109 at 7. This is our starting point because the summary judgment record on the Title IX claims changes if the district court erred in excluding Professor Johnson&#x²⁷;s opinions.

A

We review the exclusion of Professor Johnson&#x²⁷;s expert opinions for abuse of discretion. See General Elec. Co. v. Joiner, 522 U.S. 136, 143 (1997). The abuse of discretion standard allows for a range of choice, and that means that sometimes we will affirm even though we might have decided the matter differently in the first instance. See Gall v. United States, 552 U.S.³⁸, 51 (2007); United States v. Frazier, 387 F. 3d 1244, 1259 (11th Cir. 2004) (en banc). In-herent in this “deferential standard is a certain &#x²⁷;play in the joints&#x²⁷; that permits divergent results on the same evidence, depending on the [court&#x²⁷;s] explanation for the exercise of discretion." Thomas D. Schroeder, Toward a More Apparent Approach to Considering the Admission of Expert Testimony, 95 Notre Dame L. Rev. 2039, 2043 (2020).

The admissibility of expert testimony is governed by Federal Rule of Evidence 702, which has three basic requirements: the ex-pert must be qualified; his methodology must be reliable; and his testimony must be helpful to the trier of fact. See Sorrels v. NCL (Bahamas) Ltd., 796 F. 3d 1275, 1281 (11th Cir. 2015). The proponent of the expert testimony bears the burden of establishing each requirement by a preponderance of the evidence under Rule 104(a). See Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 592 & n.10 (1993); Frazier, 387 F. 3d at 1261.

These Rule 702 requirements are necessarily set out in general terms. As we have observed, the rules relating to expert testimony "are not precisely calibrated and must be applied in case-spe-cific evidentiary circumstances that often defy generalization.” United States v. Alabama Power Co., 730 F. 3d 1278, 1285 (11th Cir. 2013) (internal quotation marks and citation omitted).

B

Professor Johnson holds a bachelor&#x²⁷;s degree from Harvard, a master&#x²⁷;s degree from Chicago, and a Ph.D. from Harvard. All three degrees are in history, and he teaches that subject at Brooklyn College and the City University of New York Graduate Center. See D.E. 58-6 at 2; D.E. 109 at 6. He is, “first and foremost, a professor of history." D.E. 228 at 65. He does not have any experience or education in higher education administration, human resources, student affairs, gender studies, psychology, sociology, behavioral science, neurological science, statistical analysis, mental health counseling, forensic psychology, or psychiatry. See id. at 61–62.

In his initial expert report, Professor Johnson explained that his experience regarding Title IX "began in 2006-2007, when [he] wrote a daily blog on the Duke lacrosse case.” D.E. 63-1 at 2. He co-authored a book about that case, and “served as a consultant to ABC&#x²⁷;s Law & Justice Unit for the final five months of the case.” Id. A decade later, he co-authored another book which “examin[ed] due process and fairness in Title IX adjudications.” Id. He has been invited to speak on this topic—at the State Risk and Insurance Man-agement Association annual conference and before college and university audiences—and has also written on the topic. His writing includes a law review article which, at the time of his expert report, was forthcoming, two reported articles, two legal commentaries, and op-eds in certain publications.

The district court allowed Professor Johnson to testify about the history of Title IX. But it excluded three of his opinions because Doe "failed to show Professor Johnson [wa]s qualified to reach [his] conclusions on fairness or gender bias by comparing past lawsuits against the facts in this Title IX case. See D.E. 109 at 6–9. The first excluded opinion was that Rollins failed to provide Doe with a fair process. The second was that gender bias was a motivating factor behind Rollins&#x²⁷; erroneous findings. And the third was that Rollins acted in a manner that favored female reporting parties and disadvantaged male responding parties. See Br. for Appellant at 55. [fn³]

As the district court saw things, the “critical question” with respect to admissibility was “fit[:] Does Professor Johnson&#x²⁷;s education or experience fit the subjects in his expert report?" D.E. 109 at 7. The district court concluded that Professor Johnson could not opine that the "way Rollins treated Doe was unfair and gender-bi-ased” by "comparing past lawsuits [against other academic institutions] against the[ ] facts” in this case. Id. “Without training or experience in these areas," he was "not qualified to testify about them." Id.

C

The "principles set out in Daubert apply to soft-science ex-pert testimony” of the kind offered by Professor Johnson; “[s]ocial science testimony, like other expert testimony . . ., must be tested to be sure that the person possesses genuine expertise in a field and that her court testimony adheres to the same standards of intellectual rigor that are demanded in her professional work.” Carrizosa v. Chiquita Brands Int&#x²⁷;l, Inc.,⁴⁷ F. 4th 1278, 1317 (11th Cir. 2022) (internal quotation marks and citation omitted).

Professor Johnson sought to opine on Rollins&#x²⁷; discriminatory conduct and behavior (i.e., its purported Title IX gender bias). Generally speaking, federal courts do not allow experts to testify about such motives or biases in discrimination cases unless there is some rational connection between the methodology and the opinion-something that explains how the application of the method-ology permits the conclusion(s) reached. See, e.g., Chadwick v. WellPoint, Inc., 561 F. 3d 38, 48 (1st Cir. 2009) (affirming the

exclusion of the plaintiff&#x²⁷;s expert testimony on sex stereotyping in a gender discrimination case in part because of a “mismatch between the expert&#x²⁷;s knowledge and qualifications and her ability to helpfully opine on the specifics of th[e] case").

Consider Downing v. Abbott Laboratories,⁴⁸ F. 4th 793 (7th Cir. 2022), an employment discrimination case. The plaintiff in that case sought to use an expert, who was a "legal academic with a background in social psychology," to opine that the employer&#x²⁷;s conduct was "consistent with how stereotyping and biases (either implicit or explicit) manifest and affect people in employment set-tings." Id. at 802. The Seventh Circuit upheld the district court&#x²⁷;s decision to exclude the expert&#x²⁷;s testimony on various grounds, in-cluding lack of reliability. It explained that the expert had reviewed "literature concerning stereotyping and discrimination” and then, without discussing her methodology, pivoted to her opinion that the employer&#x²⁷;s conduct was "consistent with the possibility of stereotyping or bias.” Id. at 809. The district court properly exercised its discretion with respect to reliability because the expert had not ""bridge[d] the analytical gap&#x²⁷; by showing a &#x²⁷;rational connection&#x²⁷; between the data and the . . . contested conclusion." Id.

The circumstances here are similar. Professor Johnson based his unfair process and gender bias opinions on a review of hundreds of Title IX lawsuits against colleges and universities brought by persons who had been accused of sexual misconduct. See D.E. 228 at 12-14, 67. But he never explained why that comparative review allowed him to reach his gender-bias opinions.

There were also other problems. We highlight some of them below.

First, in conducting his review of post-2011 Title IX cases by accused students against colleges and universities, Professor John-son assumed that complaints that survived the motion-to-dismiss stage were indicative of gender discrimination. See, e.g., D.E. 58-6 at 17-18, 24. But, as he later acknowledged, the courts in those cases had merely ruled that the factual allegations pled in the com-plaints, accepted as true, had made out plausible claims of gender bias. See D.E. 228 at 72-73.

Second, Professor Johnson determined that Rollins&#x²⁷; single-investigator model was not the "best practice” or the optimum method for getting at the truth, and that schools were moving away from such a model. See id. at 26–27, 29, 84-86. Yet disagreement with the single-investigator model, no matter how sincere or well-grounded, says nothing about why the use of that model suggests gender bias on the part of Rollins in conducting Title IX in-vestigations generally or in conducting the investigation of Roe&#x²⁷;s allegations specifically.

Third, Professor Johnson pointed to sex stereotypes in some of Rollins&#x²⁷; Title IX training materials. See id. at 22–23. But he acknowledged that the training materials he criticized were pre-pared for students and not for investigators like Ms. Wallace. Moreover, he conceded that Ms. Wallace had not been shown any of the stereotyped training materials and disavowed any suggestion

that gender stereotyping by Ms. Jiménez Guevara should be attributed to Ms. Wallace. See id. at 75–80, 88, 98–99.

Fourth, Professor Johnson relied on Rollins&#x²⁷; failure to investigate two possible cases of sexual misconduct by females when both the females and males were intoxicated. See id. at 34-36. In these two cases, however, the males did not file Title IX com-plaints. See id. at 36–37.

Doe argues that Professor Johnson&#x²⁷;s “academic research makes him fully qualified to comment on gender bias in Title IX matters" because a “witness&#x²⁷; expertise generally allows the witness to offer an expert opinion reasonably within the confines of that expertise." Br. for Appellant at 56–57. Even where experience is the basis for an expert&#x²⁷;s qualification, however, there can some-times be "too great an analytical gap between the data and the opinion proffered.” Joiner, 522 U.S. at 146. “If an expert is relying solely or primarily on experience, then [he] must explain how that experience leads to the conclusion reached, why that experience is a sufficient basis for the opinion, and how that experience is reliably applied to the facts.” Carrizosa,⁴⁷ F. 4th at 1322 (internal quotation marks omitted).

It is not enough for Doe to point to Professor Johnson&#x²⁷;s academic research and review of other Title IX cases—he “must put forth sufficient indicia of reliability anddemonstrate that [Professor Johnson&#x²⁷;s] expertise permits the opinion(s) rendered.” Id. Doe does not explain in his brief how Professor Johnson&#x²⁷;s expertise allowed him to opine that Rollins&#x²⁷; Title IX investigation was infected

with gender bias. And that is fatal under abuse of discretion review. See id. (affirming exclusion of an expert who "didn&#x²⁷;t explain how his experience or sources led to or supported the conclusion he reached"); Hughes v. Kia Motors Corp., 766 F. 3d 1317, 1331 (11th Cir. 2014) (holding that the district court did not abuse its discretion in concluding that "the leap from data to opinion was too great, and therefore exclud[ing] [the] testimony”).

According to Doe, “because [Professor Johnson] main-tain[ed] a database of litigation documents involving Title IX, he has on record this enormous array of material of how other col-leges and universities handled these matters and could compare Rollins to how other institutions had done so." Reply Br. for Ap-pellant at 46 (internal quotation marks omitted). We still cannot discern any abuse of discretion. Though this collection of data may demonstrate how Professor Johnson could opine on how Rollins&#x²⁷; handling of Title IX matters differed from that of other colleges and universities and maybe why the single-investigator model was not the optimal system under current industry practices—Doe does not explain how the data provided a basis for concluding that Rollins had a Title IX process that was based on or infected with gender bias. Indeed, when asked about how his “education, train-ing, and experience as a historian assist[ed] [him] in drawing conclusions from [his] database," Professor Johnson answered only that "[t]his is the kind of research that [he’d] always done, which is [to] study primary sources and especially primary sources that help to illustrate how institutions make procedural decisions behind the

scenes." D.E. 228 at 97. An expert testifying on the basis of his experience cannot support an opinion merely by saying that he reached it after doing what he has always done. He must also ex-plain why the method he has long employed permits or supports the opinion at issue.

III

Our review of the district court&#x²⁷;s grant of summary judgment on Doe&#x²⁷;s Title IX claims is plenary. See Marbury v. Warden, 936 F. 3d 1227, 1232 (11th Cir. 2019). Summary judgment is war-ranted "when the evidence, viewed in the light most favorable to the nonmoving party, presents no genuine issue of material fact and compels judgment as a matter of law in favor of the moving party." Owusu-Ansah v. Coca-Cola Co., 715 F. 3d 1306, 1307 (11th Cir. 2013) (citation omitted).

A

Under Title IX, “[n]o person in the United States shall, on the basis of sex, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any education program or activity receiving Federal financial assistance."²⁰ U.S.C. § 1681(a). "Title IX bars the imposition of university discipline where gender is a motivating factor in the decision to discipline." Yusuf v. Vassar Coll.,³⁵ F. 3d 709, 715 (2d Cir. 1994).

Last year, in a case arising at the Rule 12(b)(6) stage, we addressed the "framework for analyzing Title IX challenges to university disciplinary proceedings.” Doe v. Samford Univ.,²⁹ F. 4th 675,

686 (11th Cir. 2022). We examined the tests set forth by the Second Circuit in Yusuf,³⁵ F. 3d at 715—the “erroneous outcome” test and the "selective enforcement" test—as well as "a test first developed by the Seventh Circuit: ‘do the alleged facts, if true, raise a plausible inference that the university discriminated against [the plaintiff] &#x²⁷;on the basis of sex&#x²⁷;?" Samford Univ.,²⁹ F. 4th at 686 (quoting Doe v. Purdue Univ., 928 F. 3d 652, 667–68 (7th Cir. 2019)).

Under the “erroneous outcome” test, “a student must show both that he was innocent and wrongly found to have committed an offense&#x²⁷; and that there is a causal connection between the flawed outcome and gender bias.” Doe v. Valencia Coll., 903 F. 3d 1220, 1236 (11th Cir. 2018) (quoting Yusuf,³⁵ F. 3d at 715). And under the “selective enforcement” test, “a student must allege and ultimately prove &#x²⁷;that, regardless of the student&#x²⁷;s guilt or innocence, the severity of the penalty and/or the decision to initiate the proceeding was affected by the student&#x²⁷;s [sex].” Samford Univ.,²⁹ F. 4th at 686 (quoting Yusuf,³⁵ F. 3d at 715). We explained in Sam-ford University that these two tests "do not capture the full range of conduct that could lead to liability under Title IX," but instead "simply describe [two] ways in which a plaintiff might show that sex was a motivating factor in a university&#x²⁷;s decision.” Id. at 687 (internal quotation marks omitted).

Because "[t]he Seventh Circuit&#x²⁷;s test hew[ed] more closely to &#x²⁷;the text of the statute and binding precedent&#x²⁷; than . . . Yusuf," we agreed with that approach in Samford University, albeit with one modification. See id. at 686. Namely, rather than asking

whether the facts alleged by the plaintiff raise a plausible inference, we determined that the inquiry ought to be whether the facts, if true, permit a reasonable inference of a Title IX violation. See id. at 687.

Although the procedural posture of the appeal in Samford University was different—that case involved the dismissal of a com-plaint at the pleading stage—the analysis on the appropriate frame-work still holds. See, e.g., Doe v. Univ. of Denver, 1 F. 4th 822, 830 (10th Cir. 2021) (reframing the Seventh Circuit&#x²⁷;s test for summary judgment review). Here we simply modify the inquiry to fit review of a grant of summary judgment: could a jury presented with the record evidence, viewed in Doe&#x²⁷;s favor, reasonably find that Rol-lins discriminated against Doe on the basis of sex? See id. (“[W]e reframe the operative question for summary judgment and ask: Could a reasonable jury-presented with the facts alleged-find that sex was a motivating factor in the University&#x²⁷;s disciplinary decision?").

B

To state a Title IX claim for selective enforcement, a plaintiff "must plausibly allege &#x²⁷;an inconsistency&#x²⁷; between his treatment by the university and the university&#x²⁷;s treatment of a similarly-situated member of the other sex.” Samford Univ.,²⁹ F. 4th at 693. Applied at the summary judgment stage, this standard requires a plaintiff to present sufficient evidence for a jury to find that the institution&#x²⁷;s treatment of similarly-situated male and female students was in-consistent. See Univ. of Denver, 1 F. 4th at 830.

Doe argues that Rollins engaged in selective enforcement by treating Roe more favorably based on sex. Doe contends that even though he and Roe were similarly situated in terms of their con-duct, Rollins treated Roe more favorably by "fail[ing] to initiate a disciplinary process” to determine whether she violated Rollins&#x²⁷; sexual misconduct policy. See Br. for Appellant at 31. According to Doe, his conduct was “nearly identical” to Roe&#x²⁷;s because “(i) they both engaged in the exact same sexual conduct; (ii) they both admitted to drinking; and (iii) they both claimed the other initiated sexual activity." Id. at 32.

The district court concluded that these similarities were not enough because Doe, unlike Roe, did not allege any misconduct and did not file a complaint. It explained that "[c]omparing . . . Roe and Doe does not eliminate a non-discriminatory reason for Rol-lins[&#x²⁷;] decision not to investigate . . . Roe – Doe&#x²⁷;s failure to allege misconduct.” D.E. 156 at 24. As explained below, we agree with the district court&#x²⁷;s conclusion that Doe was not similarly situated to Roe, and as a result "[n]othing about the different[ial] treatment of [Doe and Roe] suggests [that sex] had anything to do with it." Anthony v. Georgia, 69 F. 4th 796, 808 (11th Cir. 2023) (Hinkle, J., concurring).

It is undisputed that Doe (unlike Roe) did not allege any misconduct. He acknowledged in his deposition that he did not make a complaint that Roe had sexually assaulted him.

Although Doe does not claim that he lodged a complaint, he asserts that Roe never initiated a complaint either. This is simply

incorrect. Roe did not initially "request an investigation," but days after the incident she went to Rollins&#x²⁷; Title IX office, "filed a report" with Rollins&#x²⁷; Title IX Coordinator, and sought "interim measures”—“academic accommodations and other support services.” D.E. 56-2 at 117, 123–24. See also D.E. 56-5 at 2-3 (Rollins College Report Part 1: “On February 22nd, 2017, [Roe] . . . appeared in person at the Office of Title IX and met with [the] Title IX Coor-dinator . . . to discuss an incident that she believed to be in violation of the Rollins Title IX Policy."). Thus, unlike Doe, Roe reported what she believed to be a Title IX violation to Rollins&#x²⁷; Title IX Co-ordinator.

Doe asserts that "[t]he obligation of Rollins to investigate turned on the school&#x²⁷;s knowledge of potential misconduct, not on whether someone filed a formal complaint.” Br. for Appellant at 33. According to Doe, Rollins "failed to initiate a disciplinary pro-cess with respect to Roe after receiving credible information that Roe may have violated the [s]exual [m]isconduct [p]olicy by initiating sexual activity with Doe while [he] was incapacitated due to alcohol." Id. at 31. On this record, that contention fails.

Under Rollins&#x²⁷; policy, "incapacitation constitutes a circumstance where a person would not be able to consent.” D.E. 56-2 at 58–60. See also D.E. 56-1 at 7–8 ("An individual who is physically incapacitated from alcohol or other drug consumption (voluntarily or involuntarily), or is unconscious, unaware or otherwise physically helpless is considered unable to given consent. For example, one who is asleep or passed out cannot give consent.”). Rollins&#x²⁷;

policy states that "incapacitation is a state beyond drunkenness or intoxication” and “indicators of incapacitation may include, but are not limited to, lack of control over physical movements, being un-aware of circumstances or surroundings, or being unable to communicate for any reason." D.E. 56-1 at 8. See also D.E. 56-2 at 60 (Deposition of Ms. Jiménez Guevara: “Q. Okay. If a person is in-toxicated, are they able to provide affirmative consent, under the Rollins policy? A. According to our policy, the statement is, where alcohol is involved, incapacitation is a state beyond drunkenness or intoxication.”). [fn⁴]

For Doe&#x²⁷;s argument to succeed, the record must contain evidence sufficient for a jury to find that Rollins learned that Doe was incapacitated—and therefore unable to consent—but nonetheless chose not to investigate the matter or initiate disciplinary proceedings against Roe. The record does not contain such evidence. Alt-hough Roe said that she believed Doe was drunk at certain points during the evening, she thought he was not as intoxicated as she was, and she did not think that he was incapacitated. See D.E. 56- 5 at 8-9.

Ms. Wallace, the Title IX investigator, testified similarly:

Q. Did you have any – after you received information that the respondent may have been drunk, did you

A. No. There w[ere] no allegations that – and even I believe reporter or responding party himself told me that he wasn&#x²⁷;t incapacitated. I don&#x²⁷;t want to misquote him, but I believe he told me he wasn&#x²⁷;t even drunk at the time of the incident.

Q. But the reporting party told you he was drunk, right?

A. No. The reporting party told me that he was drunk earlier in the night during the – I believe like the sorority serenade event.

D.E. 56-9 at 106.

Critically, Doe never claimed that he was incapacitated. When asked whether she made "any efforts to determine whether [Doe] was able to consent to [ ] sexual activity,” Ms. Wallace answered that "[t]here was absolutely no mention of [Doe] being victimized by [Roe.]” Id. at 107–108. According to Ms. Wallace, Doe "never brought that up. Had he brought that up, had he said – had he told me [he] was in an even if he didn&#x²⁷;t allege incapacitation, even if he said intoxication, [she] would have inquired further as to his intoxicated state; however, that is not what he said at all.” Id. at 108.

Indeed, Doe&#x²⁷;s own statements belie any claim of incapacitation. Doe submitted two statements to Ms. Wallace in response to allegations made against him. See D.E. 61-2, Exh. C at 25–31. In the first statement, Doe did not say or suggest that he was

physically incapacitated from alcohol, or unconscious, or unaware, or otherwise physically helpless. He did not even claim that he was intoxicated. Doe&#x²⁷;s only reference to his alcohol consumption that evening was the following: "When we entered the bar, someone purchased me a beer (Corona), while Jane Roe did not consume any drinks there.” Id. at 26. Although Doe specifically “address[ed] the term incapacitation and it[s] definition” in [Rollins&#x²⁷; policy], he did not at any point state that he was incapacitated that evening. See id. at 28. When he submitted a second statement to Ms. Wal-lace with additional thoughts after he "processed all the information that [she] gave [him]," Doe again did not say anything to indicate or suggest that he was incapacitated in any way. See id. at 30-31.

The record, as we have noted, contains some evidence of Doe drinking alcohol on the night in question. But this was not enough to create a genuine issue of material fact as to whether Rol-lins discriminated against Doe on the basis of sex by failing to in-vestigate Roe. Given that Doe did not file a complaint against Roe, and that Doe disclaimed any incapacity, no reasonable jury could find that Rollins&#x²⁷; failure to investigate Roe was based on sex.

Doe&#x²⁷;s reliance on Doe v. Amherst College, 238 F. Supp. 3d 195 (D. Mass. 2017), falls short for this very reason. In Amherst College, the operative complaint alleged that the college learned that the alleged female victim “may have initiated sexual activity" with the male student subject to the disciplinary proceedings "while he was &#x²⁷;blacked out,&#x²⁷; and thus incapable of consenting” but

"did not encourage him to file a complaint, consider the information, or otherwise investigate." Id. at 223. Those allegations, the district court concluded, were “sufficient to survive a motion for judgment on the pleadings.” Id. Here, the record—viewed in the light most favorable to Doe does not suggest or show that Rollins learned that Doe was incapacitated or otherwise unable to consent. As a result, a reasonable jury could not find that Rollins&#x²⁷; failure to investigate Roe was improperly based on her sex or Doe&#x²⁷;s sex.

C

To state a Title IX erroneous outcome claim, a plaintiff must plausibly allege “both that he was innocent and wrongly found to have committed an offense and that there is a causal connection between the flawed outcome and [sex] bias.” Samford Univ.,²⁹ F. 4th at 686 (internal quotation marks omitted). At the summary judgment stage, a plaintiff must present sufficient evidence for a jury to make both of these findings. See Univ. of Denver, 1 F. 4th at 830.

Doe denied that he engaged in sexual activity with Roe with-out her consent. We agree with the district court that his first-hand version of events, if believed, was sufficient to create an issue of fact on the first prong of the erroneous outcome standard. See United States v. Stein, 881 F. 3d 853, 857–58 (11th Cir. 2018) (en banc).

That leaves the second prong-whether there is sufficient evidence for a jury to find a causal connection between the flawed outcome and gender bias. Doe claims that he "presented evidence of procedural deficiencies and patterns of enforcement which, when combined with evidence of external pressure from [the U.S. Department of Education, Office for Civil Rights,] was sufficient to show that Rollins employed outdated and discriminatory views of gender and sexuality." Br. for Appellant at 37 (internal quotation marks omitted). Like the district court, we conclude that this evidence-viewed collectively is not sufficient for a jury to find there was gender bias in Rollins&#x²⁷; investigation and determination, such that Doe was disciplined on the basis of sex.

Before discussing the evidence Doe relies on, we briefly summarize the allegations presented, and found to be insufficient to make out a claim of gender bias, in Samford University. Alt-hough Samford University arose in a Rule 12(b)(6) context, we mention the allegations there because, if they were deemed inadequate to survive a motion to dismiss, then the same allegations, when established by evidence at summary judgment, would like-wise not be enough to get to a jury.

The plaintiff in Samford University alleged that "procedural irregularities at the investigation and hearing stages, &#x²⁷;public pres-sure[] to comply with Title IX,&#x²⁷; public statements by university officials, and &#x²⁷;statistics revealing numerous allegations against male students&#x²⁷; raise[d] a reasonable inference of sex discrimination."²⁹ F. 4th at 687. Specifically, the plaintiffs allegations of procedural

irregularities included the following: “the decision of [t]he [university&#x²⁷;s] appellate board . . . [to] reject[ ] all medical evidence showing the falsity of [Roe&#x²⁷;s] claims”; “the university&#x²⁷;s failure to redact the portions of the investigative report containing statements from witnesses who did not testify at the hearing, as the [university&#x²⁷;s] Title IX policy requires"; the "complaint against [the plaintiff] was [the investigator&#x²⁷;s] first Title IX investigation"; and "prior to lead-ing the investigation . . . [the investigator] received little to no train-ing" and "lacked the experience necessary to conduct a Title IX in-vestigation on her own in accordance with the [p]olicy." Id. at 688–89 (internal quotation marks omitted). As for statistics revealing numerous allegations against male students, the plaintiff in Sam-ford University alleged that “there ha[d] been at least seven re-ported rapes and nine reported cases of fondling; and . . . all the accused students were . . . males.” Id. at 692 (internal quotation marks omitted). The plaintiff further alleged that “the university promoted an It&#x²⁷;s On Us Initiative&#x²⁷; tout[ing] [that] every²¹ hours there is a rape on an American college campus; and that the university was subjected to . . . pressure from federal regulators to comply with Title IX and appear tough on sexual assault.” Id. at 691 (internal quotation marks omitted). According to the plaintiff, "the university favor[ed] all complainants and disfavor[ed] all re-spondents in Title IX proceedings." Id. at 689 (internal quotation marks omitted).

These and other allegations, accepted as true and viewed in the light most favorable to the plaintiff, fell short in Samford

University. They did not permit a reasonable inference that the university discriminated against the plaintiff on the basis of sex. See id. at 687-93. We therefore do not write on a clean slate in review-ing Doe&#x²⁷;s Title IX erroneous outcome claim.

Doe identifies four types of evidence which he says shows (or creates an issue of fact on) Rollins&#x²⁷; gender bias: (1) evidence of stereotyped views of gender; (2) evidence of procedural flaws in the investigation; (3) evidence of external pressure from the Depart-ment of Education; and (4) evidence on patterns of decision-mak-ing at Rollins. We assess this evidence collectively because "the whole is often greater than the sum of its parts.” District of Co-lumbia v. Wesby, 138 S. Ct. 577, 588 (2018) (discussing the "totality of the circumstances" standard for assessing probable cause). Cf. Lewis v. City of Union City, 918 F. 3d 1213, 1220 n.6 (11th Cir. 2019) (en banc) (explaining that a plaintiff may show discriminatory in-tent by presenting a “convincing mosaic" of circumstantial evidence that would allow a jury to infer intentional discrimination). But to understand what Doe&#x²⁷;s evidence means collectively, we first discuss each component of the mosaic and its limitations. See Sam-ford Univ.,²⁹ F. 4th at 687–92.

Stereotyped Views of Gender. Doe points to two emails sent by Rollins&#x²⁷; Title IX Coordinator in October and November of 2015, as well as certain information set out in some of Rollins&#x²⁷; train-ing materials. According to Doe, the email from October of 2015 “misrepresented the definition of consent . . . indicating a female-centric view where consent could not be implied by a few too

many drinks or a short skirt," while the email from November of 2015 "stat[ed] a common stereotype about males: &#x²⁷;Men act in a hyper-masculine and sexually aggressive manner to gain approval from their male peers . .” Br. for Appellant at 40. As for the training materials, Doe argues that they “reinforced the stereotype of men as predators." Id. Doe cites documents that “included the false claim that &#x²⁷;99% of people who rape are men." Id. (citing D.E. 74-9).

The October 2015 email includes the “short skirt” comment, but the rest of its content is gender neutral with respect to the matter of consent: “To make a mutual, voluntary, informed decision between clear-minded, of age participants before ANY and EVERY sexual act. Consent can be withdrawn at ANY time and you are NOT ENTITLED.” D.E. 74-8 at 1. It also states, with bullet points, that "[c]onsent is an enthusiastic, unequivocal, voluntary, verbal and passionate – YES!”; that “[c]onsent is not an interpretation”; that "[t]here are no grey areas"; that consent "is not unsure, or implied, after a few too many drinks or by a short skirt, it&#x²⁷;s not drugged, it&#x²⁷;s never a &#x²⁷;maybe&#x²⁷; or an &#x²⁷;okay&#x²⁷; and it sure isn&#x²⁷;t coerced or convinced"; and that a "person cannot consent if they are asleep or incapacitated by alcohol, drugs, or any other cause!" See id.

The November 2015 email states that “[m]en are an integral part of moving campus culture away from tolerance of sexual violence," and that “[w]hile most men do not commit acts of violence, males perpetrate the majority of sexual assaults on women and other men.” D.E. 74-7 at 1. It also says that “[m]en act in a hyper-

masculine and sexually aggressive manner to gain approval from their male peers, not to impress women, research suggests.” Id. at 2.

One set of training materials from the Office of the Title IX Coordinator, which were used for student orientation, defines sexual assault as “[h]aving or attempting to have sexual intercourse or sexual contact with another individual without consent” and in a bullet point mentions “[t]hreat, force, coercion, incapacitation." D.E. 74-12 at 14. The slide for “consent" states the following: "The decision to engage in sexual activity must be informed, knowing, and voluntary. It exists when all parties exchange mutually under-standable affirmative words or behavior indicating their agreement to freely participate in mutual sexual activity." Id. at 17. Additional slides about "consent” and “coercion” are gender-neutral. See, e.g., id. at 18 (explaining that a person who is “physically incapacitated" is “considered unable to give consent"). One of the “scenario” discussions employs a hypothetical in which a male student harasses a female student; another uses a hypothetical in which one male student engages in the sexual assault of another male student; and a third uses a hypothetical in which a male student touches the breast of a female student without her consent. See id. at 36-38. [fn⁵]

Another set of training materials, this one from August of 2014, is entitled "Peer Education Title IX Training." See D.E. 74-9. It contains a number of slides on sexual assault statistics. One of the slides states that "1 in⁵ college women will be sexually assaulted during their college years," id. at 3 (citing "USDOJ, 2005" as the source), and another states that "99% of people who rape are men," id at 7 (citing "Greenfeld, 1997" as the source). [fn⁶]

Doe contends in his brief that the former statistic is "questionable” and that the latter is “false.” Br. for Appellant at 40. But he does not cite to any evidence in the record indicating the respective statistics are questionable or false, and it is not appropriate for us on our own to try to figure out if the figures are accurate or not even assuming we could. If Doe asserts that Rollins used allegedly questionable or false statistics about sexual assaults to suggest or show gender bias on the part of Rollins, it is his responsibility to point us to a source which questions the validity of those statistics.

"What do individuals who identify as women do on a daily basis to prevent themselves from being raped of [sic] sexually assaulted?” D.E. 74-11 at 2. An-other slide from a different training presentation states that one of the four "Things to Know About Title IX” is that it “applies to all genders.” D.E. 74-12 at 9. In a later slide, “Sexual Assault (Rape)” is defined as “[h]aving or attempt-ing to have sexual intercourse of sexual contact with another individual with-out consent." Id. at 15.

Evidence of Procedural Flaws in the Investigation. Doe identifies a number of purported deficiencies in the investigation. These are Rollins&#x²⁷; failure to adhere to the 60-day timeline; Ms. Wal-lace&#x²⁷;s alleged failure to consider evidence that could have negatively affected the credibility of Roe, such as her delay in pursuing an investigation or her receipt of academic and housing benefits; Ms. Wallace interviewing Doe only once but Roe multiple times; Ms. Wallace taking Roe&#x²⁷;s account at face value; the belief of Doe and his advisor that Ms. Wallace supposedly made up her mind before his interview; Ms. Wallace&#x²⁷;s reliance on the stereotype that men are constantly seeking sex and are responsible for obtaining consent, even when Doe indicated that Roe had initiated the sexual activity; and Ms. Wallace&#x²⁷;s inquiry into Doe&#x²⁷;s prior sexual history, but not Roe&#x²⁷;s prior sexual history. See Br. for Appellant at 42–44. Most of these alleged flaws are not flaws at all or demonstrate little connection to gender bias.

First, although Rollins exceeded its own 60-day deadline, "[a] deviation from a Title IX policy is not, in and of itself, a violation of Title IX.”Samford Univ.,²⁹ F. 4th at 688. Ms. Jiménez Gue-vara explained that she did not open up an investigation at the time of Roe&#x²⁷;s initial report in February of 2017 because she "didn&#x²⁷;t have enough information to supersede [Roe&#x²⁷;s] wishes at that time." D.E. 56-2 at 117. Choosing not to begin an investigation against Doe for dearth of information (and/or because of Roe&#x²⁷;s decision not to proceed) is not evidence of sex discrimination.

As for the delay from November of 2017 to March of 2018, Ms. Wallace interviewed Roe, Doe, and²² other witnesses, and then prepared a 71-page report. When asked why it took her so long to finish the report, Ms. Wallace explained that although her investigation concluded on December¹², 2017, she was admitted into the hospital on December²² and "didn&#x²⁷;t really come out until March¹⁹" and during that time Ms. Jiménez Guevara was re-viewing her first draft. See D.E. 61-7 at 40.

Second, Doe asserts that Ms. Wallace interviewed Roe four times—three in-person interviews and one phone call—but inter-viewed him only once. But Doe submitted two written statements in addition to his interview. So Ms. Wallace still heard from Doe several times, albeit in a different format. Moreover, Ms. Wallace testified that Doe “was interviewed at the end” and after she “in-terviewed [Doe] and re-interviewed [Roe], there were no outstand-ing questions in [her] mind that gave [her] any reason to re-inter-view him for any reason.” Id. at 34. Additionally, Ms. Wallace in-terviewed the witnesses Doe submitted and they did not provide any additional information or leave any outstanding questions in her mind "that would have called for him to be re-interviewed." Id. [fn⁷]

Third, Doe says that he and his advisor believed Ms. Wallace had made up her mind before the interview. Lay witnesses may sometimes testify, based on their own observations and perceptions, about the state of mind of another person, and we assume that the opinions of Doe and his advisor were admissible. See, e.g., John Hancock Mut. Life Ins. Co. v. Dutton, 585 F. 2d 1289, 1293–94 (5th Cir. 1978). Nevertheless, we note that any inference of bias by Ms. Wallace in favor of Roe may be a bias in favor of report-ing/complaining parties, and not a bias in favor of females. See Samford Univ.²⁹ F. 4th at 690 (“discrimination against respondents is not discrimination on the basis of sex,&#x²⁷; . . . and does not permit a reasonable inference of an anti-male bias because both men and women can be respondents") (internal quotation marks and citation omitted). See also Doe v. Stonehill Coll., Inc., 55 F. 4th 302, 334 (1st Cir. 2022) (“[D]eference to Roe, without more, does not show that her treatment or Doe&#x²⁷;s -- is attributable to sex rather than to some other reason, such as Roe&#x²⁷;s status as the complain-ant."). As the district court put it, though the record “may demonstrate a victim-centered approach to the investigation of the alleged sexual misconduct, [ ] conspicuously absent is evidence of discriminatory bias motivated by gender.” D.E. 156 at 23.

Fourth, it is true that Ms. Wallace inquired about Doe&#x²⁷;s past sexual history but not about Roe&#x²⁷;s past sexual history. In a vacuum, that might permit some inference of gender bias. But the evidence cannot be viewed in isolation, and here there was an anonymous allegation that Doe had sexually assaulted three other

female students. Given that allegation, and the lack of any similar allegation about Roe, the decision to inquire about Doe&#x²⁷;s past sexual history is not suggestive of gender bias.

Pressure from the Department of Education. Doe points to the Department of Education&#x²⁷;s issuance of its 2011 “Dear Col-league" letter—which discussed, among other things, schools&#x²⁷; obligations to respond to sexual harassment and sexual violence, pro-cedural requirements pertaining to such harassment and violence, and steps to prevent and correct its discriminatory effects on the complainant and others—and maintains that Rollins changed its Title IX policies in response to “pressure” from the Department. See Br. for Appellant at 44–45; Reply Br. for Appellant at 36-38. Doe also cites to the testimony of Ms. Jiménez Guevara, who said that "[p]art of [her] job was ensuring that [Rollins was] following the mandates of Title IX, as an institution” and that the “Dear Col-league letter and the government was very clear on what the con-sequences are for any institution that doesn&#x²⁷;t follow the mandates of Title IX.” D.E. 56-2 at 114.

In Samford University, we explained that the Department of Education&#x²⁷;s 2011 “Dear Colleague Letter" was not indicative of gender bias on the part of the university because it was rescinded in September of 2017 and because new regulations, e.g.,³⁴ C.F.R. § 106.45, provided more procedural protections for those accused of sexual misconduct. See Samford Univ.,²⁹ F. 4th at 691–92. Here, Rollins began its investigation of Roe&#x²⁷;s complaint in Novem-ber of 2017, after the “Dear Colleague Letter” had been rescinded.

See also Doe v. Columbia Coll. Chi., 933 F. 3d 849, 855–56 (7th Cir. 2019) (cited with approval in Samford University, and holding that "generalized allegations" about “the ‘Dear Colleague&#x²⁷; Letter," even when combined with allegations of procedural impropriety, did not permit a "plausibl[e] infer[ence] that [the] investigation or adjudication was tainted by an anti-male bias”).

Patterns of Decision-Making at Rollins. Doe maintains that that Rollins "never investigated a female for sexual misconduct while Doe was a student at the school despite the fact that the Rol-lins student body at that time was 60% female." Br. for Appellant at 47. He also argues that Rollins&#x²⁷; investigations “show[ ] actions consistent with gender bias"-for example, "[m]any of the reports delved into the sexual history of the male respondents." Id. at 48.

Statistics are often illuminating, but they sometimes do not tell the whole story. Maybe that is why Benjamin Disraeli, the Brit-ish statesman, once remarked that there are three kinds of lies: "lies, damned lies, and statistics.” The Yale Book of Quotations 208 (Yale Univ. Press 2006).

Doe&#x²⁷;s statistical presentation is misleading. Of the¹² re-ported cases of alleged sexual misconduct at Rollins between 2011 and 2018, none included complaints against a female student—11 were female students complaining about male students, and the other was a male student complaining about a male student. Doe therefore has not cited any evidence of Rollins choosing not to in-vestigate complaints against female students. See Samford Univ.,²⁹ F. 4th at 692.

As the district court explained, “schools are not responsible for which students choose to report sexual misconduct.” D.E. 156 at 22 (quoting Rossley v. Drake Univ., 342 F. Supp. 3d 904, 929 (S.D. Iowa 2018), affd, 979 F. 3d 1184 (8th Cir. 2020)). In any event, the fact that⁵ of the¹² accused male students were ultimately found to not have engaged in sexual misconduct, see D.E. 74-1, demonstrates that Rollins&#x²⁷; investigations didn&#x²⁷;t invariably yield anti-male results. And that, in turn, weighs against a finding of anti-male bias. As for looking into the sexual history of the male stu-dents accused of sexual misconduct, in Doe&#x²⁷;s case the inquiry was due to the fact that an anonymous caller reported that Doe had assaulted three female students.

The Evidence Viewed Collectively. As noted, Doe&#x²⁷;s evidence of procedural deficiencies and patterns of enforcement must be viewed collectively. But even when taken as a whole, and viewed in Doe&#x²⁷;s favor, the evidence does not create a jury issue as to whether there was a causal connection between the purported erroneous outcome in Doe&#x²⁷;s case and gender bias on the part of Rollins. See Yusuf,³⁵ F. 3d at 715 (explaining that the record must contain enough evidence to create a genuine issue of fact as to whether "gender bias was a motivating factor behind the erroneous finding" in the plaintiff&#x²⁷;s case). First, although some of the 2015 emails and the various training materials Doe highlighted do con-tain some problematic language-for example, as Doe notes, the hypothetical example scenarios included only male perpetrators— they were geared toward student orientation and training and

some were years removed from Doe&#x²⁷;s investigation. And, as Pro-fessor Johnson acknowledged, Ms. Wallace—the Title IX investigator in Doe&#x²⁷;s case-did not see any of those materials and there was no claim that any gender stereotyping on the part of Ms. Jiménez Guevara was imputed to Ms. Wallace. Second, the alleged procedural deficiencies are either not flaws at all or matters that have a gender-neutral explanation. Cf. Samford Univ.,²⁹ F. 4th at 688 ("Doe&#x²⁷;s bare assertion&#x²⁷; that the procedural irregularities are attributable to his sex does not make his speculation plausible."). Third, the pressure from the Department of Education had dissipated by the time of Doe&#x²⁷;s investigation, as the 2011 “Dear Col-league" letter was rescinded in September of 2017. See Univ. of Denver, 952 F. 3d at 1192–1193 (explaining that the “Dear Col-league" letter is gender-neutral on its face and evidence of the pres-sure the university felt to comply with its guidance “cannot support [the student&#x²⁷;s] summary judgment burden unless combined with a particularized &#x²⁷;something more&#x²⁷;") (internal quotation marks omit-ted). Fourth, on this record there are no discernable patterns of gender-biased decision-making.

The cases Doe relies on exemplify the shortcomings of this record. For example, in Doe v. Marymount University, 297 F. Supp. 3d 573, 586 (E.D. Va 2018), the district court had before it allegations specifically regarding the adjudicator of the sexual misconduct proceedings—the person "ultimately responsible for determining [the student&#x²⁷;s] guilt or innocence.” Because the district court was considering a motion to dismiss, it accepted as true the

allegations, which “reveal[ed] that [the adjudicator&#x²⁷;s] decision-making was infected with impermissible gender bias.” Id.

Here, the record is markedly different. Doe has not pointed to any evidence showing that Ms. Wallace, the Title IX investigator in his case, had or acted on gender bias. Ms. Wallace was selected as a Title IX investigator for Rollins around April of 2017 and the first training she attended was on May³¹, 2017. See D.E. 61-7 at 6–7. Doe does not argue that Ms. Wallace attended any of the problematic training sessions he cites to, and it is unclear from the record whether they even took place while Ms. Wallace was serving as an investigator—some of the presentations appear undated, but at least one appears to be from August¹⁴, 2014, which was years before Ms. Wallace was even selected by Rollins. For the same reason, Ms. Wallace could not have received the October and November 2015 emails Doe references.

We are likewise unpersuaded by Doe&#x²⁷;s reliance on Doe v. Purdue University, 928 F. 3d 652 (7th Cir. 2019). The allegations in Purdue University—which, again, had to be accepted as true at the Rule 12(b)(6) stage of that case—included that the university center posted an article during the same month the male student was disciplined titled “Alcohol isn&#x²⁷;t the cause of campus sexual assault. Men are." Id. at 669. Moreover, the university chose to credit the female student&#x²⁷;s account without actually "hearing directly from her." Id. The record here does not contain such facts.

Finally, we have considered Doe v. William Marsh Rice Uni-versity, 67 F. 4th 702, 714 (5th Cir. 2023) (2-1 decision reversing

grant of summary judgment in favor of university on Title IX claims by male student who was found to have engaged in sexual misconduct), which Doe has submitted as supplemental authority. Title IX cases are inherently fact-specific, see Samford Univ.,²⁹ F. 4th at 701 (Jordan, J., concurring), and though Rice University involved the same summary judgment posture as this case, the un-derlying accusation and the handling of the Title IX investigation are markedly different.

In Rice University, a student—we&#x²⁷;ll call her Student A—alleged that "she became infected with herpes after a consensual sexual encounter with another student who knew he had herpes" but failed to sufficiently inform her of that during their relationship. See 67 F. 4th at 705. After Student A submitted a written complaint to Rice University, the other student—we&#x²⁷;ll call him Student B— was placed on an interim suspension. See id. at 706. The investigation against Student B was to determine whether he violated Rice University&#x²⁷;s prohibition from “intentionally inflicting or attempting to inflict mental or bodily harm on any person" and "taking reckless disregard, from which mental or bodily harm could re-sult to any person” and “whether his conduct may qualify as dating violence under the [u]niversity&#x²⁷;s [s]exual [m]isconduct [p]olicy." Id. (internal quotation marks omitted).

Roe and Student A are alike in that each submitted a com-plaint to their university alleging sexual misconduct, and Doe and Student B are alike in that each filed lawsuits against their university alleging gender bias. But that&#x²⁷;s really where the relevant

similarities end. First, in Rice University, the issue was not inability to provide consent due to alleged incapacitation. Instead, Student A accused Student B of "failing to inform her of his herpes diagnosis, and this was the charge [Student B] defended against. Rice ultimately found that [Student B] had informed [Student A] of his herpes diagnosis but sanctioned him anyway for failing to go further by informing [Student A] of the risks of having sex with a her-pes carrier, despite the fact that no such rule appears in Rice&#x²⁷;s stu-dent code." Id. at 708. Second, Student B&#x²⁷;s lawyer—unlike Doe&#x²⁷;s lawyer during Rollins&#x²⁷; investigation—“was not allowed to participate in the process or view any documents in the disciplinary file in order to counsel” Student B. See id. Third, Student B—unlike Doe-"was kicked off campus on²⁴ hours&#x²⁷; notice because of the risk that he might have sex with other students and infect them with herpes." Id. Fourth, Student B “was ultimately sanctioned with what amounted to expulsion for failing to inform a sex partner of the potential consequences of having sex with a herpes carrier. Yet when [Student B] offered the [u]niversity evidence that [Stu-dent A] was doing precisely the same thing, and [Student A] ex-pressly informed the [u]niversity of her intent not to tell future sex partners of her herpes diagnosis, the [u]niversity did nothing.” Id. As discussed, Doe did not complain to Rollins during the investigation that he was incapacitated and unable to consent, or that Roe may have engaged in sexual misconduct, or that she was otherwise engaging in sexual misconduct with other students. In contrast, in Rice University, Student B alleged Student A "was guilty of the

same conduct of which he was charged: failure to disclose the risk of STD transmission.” Id. at 712.

Based on these facts, a divided panel of the Fifth Circuit concluded that the district court did not view the record in the light most favorable to Student B, and that material fact issues remained as to whether the university&#x²⁷;s proceedings were motivated by gender bias. Rice University does not establish that the district court here erred; it shows how a different record can lead to a different result. [fn⁸]

IV

Doe argues that he was entitled to summary judgment on the liability aspect of his breach of contract claim. To recap, the district court partially granted Doe&#x²⁷;s motion for summary judgment, concluding that Rollins breached the 60-day provision of its Title IX sexual misconduct policy. That provision typically re-quired the process for sexual misconduct and harassment reports to be completed within 60 days. Doe challenges the district court&#x²⁷;s

ruling that a genuine dispute of fact existed on whether Rollins&#x²⁷; breach of the 60-day provision was material—an issue that pro-ceeded to trial and was resolved by the jury in Rollins&#x²⁷; favor.

There was "a full trial and judgment on the merits” of the materiality issue. As a result, Doe&#x²⁷;s challenge to the district court&#x²⁷;s denial of summary judgment is not properly before us. See Lind v. United Parcel Serv., Inc., 254 F. 3d 1281, 1286 (11th Cir. 2001) (“We hold that this Court will not review the pretrial denial of a motion for summary judgment after a full trial and judgment on the merits.); Am. Builders Ins. Co. v. S.-Owners Ins. Co., 56 F. 4th 938 (11th Cir. 2023) ("[W]e have repeatedly and broadly held that &#x²⁷;this Court will not review the pretrial denial of a motion for summary judgment after a full trial and judgment on the merits.”) (quoting Lind, 254 F. 3d at 1286)); Carrizosa,⁴⁷ F. 4th at 1339 (“In our circuit . . . a party may not appeal an order denying summary judgment after there has been a full trial on the merits.”) (citation omitted).

The Supreme Court recently held in Dupree v. Younger, 143 S. Ct. 1382, 1389 (2023), that a "post-trial motion under Rule 50 is not required to preserve for appellate review a purely legal issue resolved at summary judgment.” As a panel of our court just ex-plained, under Dupree “arguments denied at summary judgment are appealable after a trial on the merits if they raise &#x²⁷;purely legal issues." Am. Builders Ins. Co. v. Southern-Owners Ins. Co., 71 F. 4th 847, 859 (11th Cir. 2023).

Dupree does not apply here. The reason is that under Flor-ida law-which governs Doe&#x²⁷;s breach of contract claim—

materiality is an issue of fact and not a pure question of law. See Covelli Fam., L.P. v. ABG⁵, L.L.C., 977 So. 2d 749, 752 (Fla. 4th DCA 2008) (“The issue of whether an alleged breach is vital or material is reviewed as a question of fact."). Accord²³ Williston on Contracts § 63:3 (4th ed & May 2023 update) ("The determination whether a material breach has occurred is generally a question of fact, dependent on the circumstances of the case.”) (footnotes omit-ted). We therefore cannot review the district court&#x²⁷;s denial of sum-mary judgment on liability for Doe&#x²⁷;s breach of contract claim. See Am. Builders Ins. Co., 71 F. 4th at 859–61.

V

Doe also challenges the district court&#x²⁷;s denial of his Rule 50 motions for judgment as a matter of law on the issue of materiality. Our review is plenary, and we view the evidence in the light most favorable to Rollins, which prevailed at trial. See Combs v. Planta-tion Patterns, 106 F. 3d 1519, 1526 (11th Cir. 1997).

During trial, Doe moved for partial judgment as a matter of law under Rule 50(a), arguing that Rollins&#x²⁷; breach of the 60-day provision was material. Specifically, Doe asserted that he and his advisor in the Title IX investigation both testified that his "ability to defend himself was impacted because of the delay in informing [him] of the allegations and initiating an investigation.” D.E. 191 at 3. Doe further claimed that “the ability to obtain physical evidence, such a[s] surveillance videos or electronic communications, was lost," and that “memories fade over this period of time.” Id. According to Doe, Rollins did not present "any contrary evidence"

and "no reasonable jury would have a legally sufficient evidentiary basis to find for [Rollins] on the issue of materiality related to the 60-day provision." Id.

The district court reserved ruling on Doe&#x²⁷;s partial Rule 50(a) motion, and the jury returned a verdict in Rollins&#x²⁷; favor, finding that Rollins&#x²⁷; breach of the 60-day provision was not material. Fol-lowing trial, the district court denied Doe&#x²⁷;s partial motion for judgment as a matter of law. Doe renewed his motion under Rule 50(b), and in the alternative requested a new trial pursuant to Rule 59.

In his renewed motion, Doe argued that the district court had "misapplied the material breach rule” by focusing on whether the delay impacted Rollins&#x²⁷; investigator, rather than himself, the non-breaching party. See D.E. 210 at 4. The district court denied Doe&#x²⁷;s Rule 50(b) motion, and noted that “[h]ow the delay affected the investigator goes directly to [Doe&#x²⁷;s] injury – it was Ms. Wallace [the investigator] who found [Doe] violated the policy." D.E. 222 at 4. The district court explained that “if Ms. Wallace testified that the delay negatively affected her investigation, it would be evidence that the breach of contract was not minor. Conversely, her lack of testimony that the delay affected her investigation relates to whether the breach was material." Id. The district court concluded that, "[i]n the light most favorable to [Rollins] as the non-moving party, there [was] sufficient evidence to support [the jury&#x²⁷;s] verdict," and "judgment as a matter of law [was] not appropriate." Id. at 5.

We will reverse the district court&#x²⁷;s denial of the Rule 50 motions only if "the facts and inferences point overwhelmingly in favor of [Doe], such that reasonable people could not arrive at a contrary verdict." Miller v. Kenworth of Dothan, Inc., 277 F. 3d 1269, 1275 (11th Cir. 2002) (internal quotation marks and citation omit-ted). Viewing the evidence in the light most favorable to Rollins, we hold that there was sufficient evidence for a jury to reasonably find that its breach of the 60-day provision was not material.

Under Florida law, “[t]o constitute a vital or material breach, a defendant&#x²⁷;s non-performance must be such as to go to the essence of the contract.” Marchisio v. Carrington Mortg. Servs., LLC, 919 F. 3d 1288, 1313 (11th Cir. 2019) (internal quotation marks omitted). See also Burlington & Rockenbach, P.A. v. L. Offs. of E. Clay Parker, 160 So. 3d 955, 960 (Fla. 5th DCA 2015) (“To establish a material breach, the party alleged to have breached the contract must have failed to perform a duty that goes to the essence of the contract and is of such significance that it relieves the injured party from further performance of its contractual duties."). "A party&#x²⁷;s failure to perform some minor part of his contractual duty cannot be classified as a material or vital breach.”Covelli Fam., L.P., 977 So. 2d at 752 (internal quotation marks omitted). See also JF & LN, LLC v. Royal Oldsmobile-GMC Trucks Co., 292 So. 3d 500, 509 (Fla. 2d DCA 2020) (“A trivial noncompliance or minor failure to perform is not a material breach.”).

As the district court explained, “the jury had to determine whether [Rollins&#x²⁷;] failure to comply with the -day provision

(i.e., taking longer than days to complete the process) was minor or trivial.” D.E. 222 at 4. The evidence introduced at trial shows that "reasonable and fair-minded persons exercising impartial judgment might reach different conclusions” on this issue. See Mee Indus. v. Dow Chem. Co., 608 F. 3d 1202, 1211 (11th Cir. 2010).

Doe points to testimony from Eric Barker, the lawyer who served as his advisor during the Title IX proceedings, to support his claim that the breach was material. Mr. Barker testified that Doe was placed at a disadvantage, that witnesses “may forget over nine months all of the details or some of the important details,” and that surveillance videos may disappear. See Br. for Appellant at 59 (citing D.E. 230 at 215–16). Doe also relies on his own trial testimony, where he "noted he would have taken &#x²⁷;several different acts&#x²⁷; had he been informed of the investigation in a timely manner.” Id. (citing D.E. 234 at 33). Specifically, Doe testified that he "probably would have thoroughly looked through any text messages that [he] may have had over that nine-month period . . . and collect any posts from social media from the time that could have been relevant[.]" D.E. 234 at 33. He also stated that he "[c]ould have [gone] to any local businesses with security footage to help provide more con-text" and "could have written down a lot of [his] thoughts from that evening, from that current time period instead of letting this sit over the course of nine months and potentially allowing or risk-ing the ability to recover a lot of that information." Id.

This testimony, if credited by the jury, could have supported a finding in Doe&#x²⁷;s favor on materiality. But the jury did not have to credit the evidence Doe presented, and in any event the evidence on materiality was not as one-sided as Doe suggests.

For example, Rollins cites to testimony from Ms. Jiménez Guevara that “immediately following . . . Roe&#x²⁷;s report in February [of] 2017, she attempted to obtain footage of Doe and . . . Roe on campus from the night in question, but no surveillance footage ex-isted." Br. for Appellee at 57-58 (citing D.E. 234 at 281–85). Ms. Jiménez Guevara explained that she "contacted Campus Safety to ask if there was any video footage of any of the areas where [Roe] described that she had been located walking with . . . Doe.”D.E. 234 at 284. Ms. Jiménez Guevara also testified that she asked Roe "to provide [her] with any documentation, with a statement, with any text messages, or anything that she could provide [her] with at that point in time.” Id. at 283.

There was also the investigation report, which was admitted into evidence. The report shows that, as part of her investigation, Ms. Wallace was able to review text messages from February of 2017, including messages Roe sent to different persons—some as immediate as³:29 a.m. and³:31 a.m. the morning of the incident that was the subject of the Title IX investigation—as well as the text exchange between Roe and Doe on February¹⁸, 2017, which Roe reported to be the last time she spoke with Doe. See D.E. 201, Exh.¹⁶ at 10–11. And although Ms. Wallace acknowledged that the memories of witnesses could fade over time, she also said that

they don&#x²⁷;t always do so: "Sometimes traumatic incidents are for-ever imprinted in a person&#x²⁷;s mind.” D.E. 234 at 162–63.

In sum, although Rollins breached the 60-day provision, "reasonable and fair-minded persons in the exercise of impartial judgment might reach different conclusions" on whether that breach was material. See Christopher v. Florida, 449 F. 3d 1360, 1364 (11th Cir. 2006). Cf. Universal Prop. & Cas. Ins. Co. v. Motie, 335 So. 3d 205, 206-07 (Fla. 5th DCA 2022) (reversing directed verdict and holding that jury had to decide whether homeowner&#x²⁷;s failure to report storm damage for 103 days was a material breach of the insurance contract). We therefore affirm the district court&#x²⁷;s denial of Doe&#x²⁷;s motions for judgment as a matter of law.

VI

The district court did not abuse its discretion in excluding the gender bias and unfair process opinions of Doe&#x²⁷;s expert, and properly granted summary judgment in favor of Rollins on Doe&#x²⁷;s Title IX claims for selective enforcement and erroneous outcome. On the breach of contract claim, the district court did not err in denying Doe&#x²⁷;s Rule 50 motions because the evidence on the issue of materiality was disputed and the jury could have reasonably found that Rollins&#x²⁷; breach was not material.

AFFIRMED.

[fn 1]: Doe submitted an appeal of the investigation, which was resolved in April of 2018. See D.E. 61-4 at 22. The appeal resulted in a modification of sanctions such that Doe&#x²⁷;s dismissal was made effective in the Spring of 2018. See id.

[fn 2]: As noted, Doe was not permitted to participate in commencement/graduation exercises. Following the appeal, Dr. Accapadi modified the sanctions to make clear that they were not impacting Doe&#x²⁷;s undergraduate degree and the dismissal from Rollins “was in effect post the earning of his degree.” D.E. 61- 9 at 17.

[fn³]: Doe addresses the excluded opinions together in his brief, and we follow suit here.

[fn⁴]: We are not called upon to decide whether Rollins&#x²⁷; policy—which expressly distinguishes being “drunk” from being “incapacitated" for purposes of con-sent—correctly assesses one&#x²⁷;s ability to provide consent.

[fn⁵]: Doe says in his brief that “all of the training scenarios included male perpetrators." Br. for Appellant at 40. Yet review of the training slides in the record shows largely gender-neutral guidance. For example, one “Exercise” slide pre-sented two questions: “What do individuals who identify as men do on a daily basis to prevent themselves from being raped or sexually assaulted?" and

[fn⁶]: Ms. Jiménez Guevara, who was not the Title IX Coordinator in 2014, testified she “never used” the 99% statistic and did not use the 2014 training presentation for any purpose. See D.E. 56-2 at 93.

[fn⁷]: As to Doe being interviewed at the end, Ms. Wallace “believe[d] the date kept getting pushed back because [Doe] was trying to line up his counsel . . . and so it took him some time to do that” and “the process [was] moving from obviously interviewing everyone else in the meantime while [they were] wait-ing for his to come along.” D.E. 61-7 at 34.

[fn⁸]: The dissent in Rice University maintained that Student B&#x²⁷;s "experience with the [u]niversity&#x²⁷;s disciplinary process reveal[ed] no evidence of gender discrimination." 67 F. 4th at 714 (Graves, Jr., J., dissenting). Specifically, the dis-sent believed that the selective enforcement claim failed because there was "no evidence that similarly situated women were either not investigated or punished in the same way as” Student B, and the erroneous outcome claim failed because Student B “produced no evidence that the [u]niversity&#x²⁷;s decision was wrong or that it was based on gender bias.” Id. at 718–721 (Graves, Jr., J., dissenting). Given how different the facts in Rice University were, we need not decide whether the majority or the dissent got it right.

[fn⁹]: Given our conclusion on Doe&#x²⁷;s breach of contract claim, we need not address Rollins&#x²⁷; cross-appeal.