43 F.4th 784

7th Cir.

2022

JOHN DOE, Plаintiff-Appellant, v. UNIVERSITY OF SOUTHERN INDIANA, Defendant-Appellee.

No. 22-1864

United States Court of Appeals For the Seventh Circuit

Decided August 8, 2022

Argued July 27, 2022

Before HAMILTON, BRENNAN, and ST. EVE, Circuit Judges.

Appeal from the United States District Court for the Southern District of Indiana, Evansville Division. No. 3:21-cv-00144-TWP-MPB — Tanya Walton Pratt, Chief Judge.

OPINION

HAMILTON, Circuit Judge. A Title IX committee at the University of Southern Indiana found by a preponderance of evidence that one student, plaintiff-appellant John Doe, had sexually assaulted another student, Jane Doe. The committee imposed a three-semester suspension. After losing his appeal through the university‘s procedures, John brought this lawsuit. He alleges that the university violated Title IX of the Education Amendments of 1972, 20 U.S.C. § 1681(a), by discriminating against him on the basis of his sex. He sought a preliminary injunction to stop the university from imposing the suspension, but the district court denied his request. We affirm. To secure a preliminary injunction, John needs to show “that he is likely to succeed on the merits,” among other requirements. Winter v. Natural Resources Defense Council, Inc., 555 U.S. 7, 20 (2008). John has not shown he is likely to succeed on his claim оf sex discrimination, so we agree with the district court that he is not entitled to a preliminary injunction.

I. Factual and Procedural Background

The record in this case includes documents from the university‘s process for handling complaints of sexual assault, which included an outside investigator‘s report with exhibits; John‘s and Jane‘s written responses to the investigative report; an audio recording and transcript from the formal hearing on Jane‘s complaint; the appointed hearing committee‘s written decision on her complaint; documents from the appeal process, including the decision affirming the committee; and various correspondence from the process. The district court did not hold an evidentiary hearing on John‘s motion for a preliminary injunction. Neither party sought such a hearing. Our account of the facts is drawn from the extensive record.

John and Jane met as freshmen. The two quickly became “best friends” and spent time together almost daily. During the night of November 13 and early morning of November 14, 2020, John and Jane had been hanging out and drinking with friends. Jane acknowledges that she was very intoxicated that night and was repeatedly sick to her stomach. John maintains that he stayed sober throughout the night. At some point, John left, but around 2:00 am, Jane sent John a message inviting him to come back and join her and their friends in her room. He did.

The Title IX complaint against John concerned what happened after he returned. Jane complained that later that night, John got on her bed and the two started kissing. Jane acknowledged that they had kissed on prior occasions, and she did not tell him to stop kissing her that night. In a complaint filed three months later, however, she asserted that John went on to touch her breasts and digitally penetrate her without her consent. Later on November 14, Jane told her roommate about what she said was her sexual encounter with John the previous night. She did not tell her roommate then that it was without her consent.

After the night of November 13–14, 2020, Jane and John communicated regularly, at school and during the winter holidays. Jane did not tell anyone at school that she had been sexually assaulted, i.e., that the encounter was without her consent, until February 11, 2021. That day Jane had a panic attack and told her roommate and one of her suitemates that she had not consented to John‘s actions. The roommate reported the incident to the university‘s public safety office. Jane then blocked John‘s phone number and social media accounts. Jane filed a written complaint with the university‘s Title IX coordinator on February 25. John received formal notice of the complaint on March 26. At that point, under university рolicy, John and Jane were each entitled to an advisor to assist with the grievance process.

The university hired a lawyer from an outside law firm to investigate Jane‘s complaint. The investigator began by interviewing the parties. In her interview, Jane said that John touched her breasts and digitally penetrated her without her consent. When asked if she was “okay with the kissing,” Jane said, “I didn‘t like want to kiss him, but like I didn‘t, I didn‘t like tell him to stop at that point.” As for John, he acknowledged hanging out with Jane that night and into the early morning. But he denied any sexual contact with Jane and denied that they were ever in bed together that night. John told the investigator that he and Jane had kissed and cuddled before, “but that was, that was it.” The investigator also interviewed Jane‘s roommate and suitemate who were there the night in question. Both said they remembered seeing John with Jane in her bed at some point that night. The investigator compiled a preliminary report summarizing the interviews and documents he received from the parties. John and Jane both submitted written statements in response to the preliminary investigative report.¹

On July 26, 2021, the university notified John and his advisor that it would hold a hearing on Jane‘s complaint. (John‘s lawyer in this appeal and before the district court was his advisor during the grievance process.) As with the investigation, the university delegated the hearing process to outside contractors, a committee of two lawyers and an educator from an independent firm that specializes in Title IX services. Before the hearing, the university sent John and his advisor a copy of the agenda and the university‘s policy that would govern the procedures. John and his advisor also met with the school‘s interim Title IX Coordinator to discuss the procedures and any questions they might have.

The hearing took place by video conference on August 4, 2021. Jane, John, the roommate, and the suitemate answered questions. The committee chair handled the direct questioning of each witness. Under the university‘s policy, John‘s and Jane‘s advisors were barred from objecting during questioning, but they could and did cross-examine the witnesses. During her turn, Jane acknowledged that she had not objected to kissing John, but she stood by her accusation that John had touched her breasts and digitally penetrated her without her consent in the early hours of November 14. Jane was also asked about potential inconsistencies in her account regarding her clothing, what she had been drinking that night, and just what time the assault occurred.²

John‘s story also changed in one important way. He continued to maintain that nothing sexual happened on the night in question and denied that he had lain in Jane‘s bed with her.³ He told the committee, however, that he had digitally penetrated her about a week before November 14, at a time when she was sober and consented. As the committee later wrote, that position was “directly and dramatically at odds with his statement during the investigation that he and [Jane] had only ever kissed and cuddled.” In John‘s new version of events, Jane was simply mistaken or confused as to what had happened the night of November 13–14.

A few weeks later, the committee issued its written decision. In the jargon of Title IX, the committee found by a preponderance of evidence that John was “responsible” for committing sexual assault. The committee imposed a three-semester suspension and mandated sexual harassment education to be completed upon his return. John filed an appeal, as authorized under the university‘s policies. In line with earlier proceedings, the university delegated John‘s appeal to an outside lawyer. Jane was entitled to respond to John‘s appeal, but her response was submitted after the deadline and was not submitted to the appeal officer or to John. On September 22, 2021, the appeal officer issued a written decision affirming the committee‘s decision.

John filed a complaint in state court two days later. He alleged, among other things, that the university violated Title IX by discriminating against him on the basis of his sex. The state court issued a temporary restraining order barring the university from imposing the suspension pending further proceedings. The university then removed the case to federal court based on federal-question jurisdiction. See 28 U.S.C. §§ 1331, 1441, and 1446. The parties agreed to extend the terms of the state court‘s temporary restraining order pending a decision on John‘s motion for a preliminary injunction, for which the district court set a briefing schedule. After briefing and oral argument, the district court denied his motion. Doe v. University of Southern Indiana, No. 21-cv-00144, 2022 WL 1471037 (S.D. Ind. May 10, 2022). Under 28 U.S.C. § 1292(a), John has appealed the denial of his motion. We denied his request for an injunction pending appeal, but we ordered expedited briefing and argument.

II. Analysis

A. Legal Standard

“A preliminary injunction is an extraordinary remedy never awarded as of right.” Winter v. Natural Resources Defense Council, Inc., 555 U.S. 7, 24 (2008). When evaluating a preliminary injunction decision, “[w]e review the district court‘s findings of fact for clear error, its legal conclusions de novo, and its balancing of the factors for a preliminary ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍injunction for abuse of discretion.” D.U. v. Rhoades, 825 F.3d 331, 335 (7th Cir. 2016). An error of law can cause an abuse of discretion. Common Cause Indiana v. Lawson, 978 F.3d 1036, 1039 (7th Cir. 2020).

To secure a preliminary injunction, John “must establish that he is likely to succeed on the merits, that he is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in his favor, and that an injunction is in the public interest.” Winter, 555 U.S. at 20. The first step in the analysis requires a plaintiff to “demonstrаte that [his] claim has some likelihood of success on the merits, not merely a better than negligible chance.” Mays v. Dart, 974 F.3d 810, 822 (7th Cir. 2020) (internal citation and quotation marks omitted). That first step “is often decisive,” Braam v. Carr, 37 F.4th 1269, 1272 (7th Cir. 2022), and it is decisive here. We need not address the remaining elements.⁴

In assessing the merits, we do not accept John‘s allegations as true, nor do we give him the benefit of all reasonable inferences in his favor, as would be the case in evaluating a motion to dismiss on the pleadings. Cf. Alarm Detection Systems, Inc. v. Village of Schaumburg, 930 F.3d 812, 823 (7th Cir. 2019) (plaintiffs pled a “plausible” claim “because of the favorable inferences we afford[ed] to them under a Rule 12(b)(6) analysis,” but plaintiffs had not “demonstrated a likelihood of success on the merits“); see also In re Federal Bureau of Prisons’ Execution Protocol Cases, 980 F.3d 123, 134 (D.C. Cir. 2020) (preliminary injunction inquiry is a “decidedly far more searching inquiry” than motion to dismiss). We also do not give John the benefit of conflicting evidence, as we would in reviewing a grant of summary judgment. See Imaging Business Machines, LLC v. BancTec, Inc., 459 F.3d 1186, 1192 (11th Cir. 2006) (contrasting “significantly different burden” in moving for preliminary injunction with what is required to defeat summary judgment). In reviewing the district court‘s decision on a preliminary injunction, we approach the record from a neutral and objective viewpoint, assessing the merits as we think they are likely to be decided after more complete discovery and litigation.

A Title IX sex discrimination claim requires proof that “(1) the educational institution received federal funding, (2) [the] plaintiff was excluded from participation in or denied the benefits of an educational program, and (3) the educational institution in question discriminated against [the] plaintiff based on gender.” Doe v. Columbia College Chicago, 933 F.3d 849, 854 (7th Cir. 2019). The third element is the only one in dispute here. We emphasize, though, that the federal district and appellate courts do not provide third and fourth forums—after the university committee‘s hearing and the administrative appeal—to dеcide what actually happened between Jane and John on the night of November 13–14, 2020. In reviewing the denial of a preliminary injunction under Title IX, the question before us is whether John is likely to be able to show the university discriminated against him on the basis of his sex in suspending him on the basis of Jane‘s complaint and all available evidence.

John has not offered any direct evidence of sex discrimination. He instead relies on essentially three types of circumstantial evidence to support his claim: public pressure on the university to respond aggressively to complaints of sexual assaults by male students, “procedural irregularities” in the university‘s grievance process, and finally the weight of the evidence regarding Jane‘s complaint. We take each in turn, keeping in mind that the ultimate inquiry must consider the totality of the circumstances. Doe v. Purdue University, 928 F.3d 652, 667–68, 670 (7th Cir. 2019).

B. Public Pressure on the University

First, John contends that the university was undеr public pressure to find that he sexually assaulted Jane and that the desire to avoid further criticism motivated the university to act with an anti-male bias against him. John points to social media posts, an online petition, and a student newspaper article that all criticized the university for its inaction on Jane‘s complaint and others like it. He also directs us to university actions in response, including issuing a statement that the university takes sexual assault seriously.⁵

Evidence of public pressure on a university can be relevant in assessing sex discrimination claims under Title IX. See Purdue University, 928 F.3d at 668–69 (“The [Dear Colleague] letter and accompanying pressure gives John a story about why Purdue might have been motivated to discriminate against males accused of sexual assault.“); accord, Schwake v. Arizona Board of Regents, 967 F.3d 940, 948–49 (9th Cir. 2020); Doe v. Baum, 903 F.3d 575, 586 (6th Cir. 2018); Doe v. Columbia University, 831 F.3d 46, 56–58 (2d Cir. 2016). Public pressure is not enough on its own to support a claim of discrimination, however. Purdue University, 928 F.3d at 669; see also Columbia College, 933 F.3d at 855 (“A plaintiff cannоt rely on these generalized allegations alone, however, but must combine them with facts particular to his case to survive a motion to dismiss.“).

John‘s case is distinct from those cases where evidence of such public pressure could provide support for other circumstantial evidence of bias. Here, the university took significant steps to insulate the grievance process from any public pressure. It used independent contractors at each stage of the proceedings. The outside lawyer who did the original investigation, the committee members who heard the case, and the outside lawyer who acted as appeal officer were not affiliated with the university. No school officials were involved even in deciding John‘s sanction.

John‘s assertion that public pressure on the university supports his claim of sex discrimination is less cоnvincing when no university officials—who were the focus of the reported pressure—were responsible for compiling the evidence or assessing the merits of Jane‘s complaint. Cf. Purdue University, 928 F.3d at 657–58, 668–69 (alleged external pressure on school was relevant to plaintiff‘s case; university officials were involved in every step of the grievance process, and a Title IX coordinator, who “bore some responsibility for [the school‘s] compliance,” ultimately decided the merits and sanction in plaintiff‘s case, and the school was subject to a federal Department of Education investigation for Title IX compliance).⁶

C. Procedural Irregularities

Next John offers what he says were twelve procedural irregularities during the grievance process. Few trials in civil courts are error-free, but appellate courts do not quickly infer that procedural errors in a trial show the judge was biased. On the other hand, if procedural irregularities are sufficiently numerous, lopsided, and/or important, they can sometimes support an inference of sex discrimination. Purdue University, 928 F.3d at 669; accord, Doe v. University of Denver, 1 F.4th 822, 831–34 (10th Cir. 2021). But a plaintiff cannot prove gender discrimination by merely identifying mistakes or imperfections in the process. Doe v. Samford University, 29 F.4th 675, 688 (11th Cir. 2022) (“A deviation from a Title IX policy is not, in and of itself, a violation of Title IX.“).

To put the point in arguably circular terms, procedural irregularities may support a finding of sex bias under Title IX if, in light of all the circumstances, a fact-finder is convinced that the defendant deviated from proper procedures not because of human error but by design, ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍to achieve covertly what it could not do openly: discriminate against the plaintiff on the basis of his sex. “[A]s the number of irregularities increases, or the irregularities become more serious,” it begins to look less likely that the errors wеre due to benign reasons. Samford University, 29 F.4th at 697–98 (Jordan, J., concurring) (collecting cases). We have often made essentially this point about departures from regular procedure in employment discrimination cases. E.g., Baines v. Walgreen Co., 863 F.3d 656, 664 (7th Cir. 2017) (reversing summary judgment); Gordon v. United Airlines, Inc., 246 F.3d 878, 891–92 (7th Cir. 2001) (same). The same reasoning can apply under Title IX.

The devil is in the details, though. Take the Purdue University case, for example. The plaintiff alleged what amounted to a sham grievance process. He alleged that the school‘s Title IX committee found that he committed sexual assault despite never hearing directly from the other party. 928 F.3d at 669. She did not provide her own account either in writing or orally. According to the plaintiff, however, the panel relied solely on a letter submitted on her behalf by a university employee to find that her charge was more credible than his denial. In addition, under the procedures used at the time, the Title IX coordinator did not give the plaintiff a copy of the investigative report or share its contents with him. The plaintiff allegedly received only a redacted version of the investigative report, and that only a few minutes before the hearing, when he learned it “falsely claimed that he had confessed to [the other party‘s] allegations.” Id. at 657. Further, the investigative report did not include favorable evidence that he had submitted, he was barred from presenting witnesses, and two of the three administrators on the panel admitted that they had not even read the investigative report. In reversing dismissal on the pleadings, we concluded that such significant and slanted procedural irregularities could, if proven, support an inference of sex bias. Id. at 669–70.

The purported procedural irregularities offered by John do not come close to those alleged in Purdue University or cases like it. See, e.g., Menaker v. Hofstra University, 935 F.3d 20, 34–35 (2d Cir. 2019) (plaintiff alleged, among other things, that school did not interview potential witnesses, give plaintiff the investigative report, allow him to file a written response, or produce a written decision of responsibility); see also Schwake, 967 F.3d at 951 (plaintiff alleged school provided him only an oral account summarizing the complainant‘s allegations, “failed to consider his version of the alleged assault,” did not follow up with any of his witnesses or the “evidence he offered in his defense,” found the plaintiff responsible without providing him “any access to evidence or considering his exculpatory evidence,” and finally suspended him based on additional accusations “to which he was not given an opportunity to respond“). In this case, by comparison, when we dig into the twelve errors that John presents, we find that some were not errors at all. Others were arguable errors that he invited, and a few applied equally to both John and Jane. The genuine and arguable errors that remain do not рersuade us that John is likely to prove bias against him because of his sex.

1. Non-Errors

Turning first to the non-errors, John says that the university violated 34 C.F.R. § 106.45(b)(2)(i)(B) (2020) by “first giving notice of Jane‘s Title IX charges to [him] immediately prior to the hearing and misrepresent[ing] the existence of her written complaint and the information therein.” But John received a notice more than four months before the hearing. Section 106.45(b)(2)(i)(B) requires only that a university give a respondent “sufficient details known at the time.” While John did not receive a verbatim copy of Jane‘s complaint, the March 26 notice included her name, the general allegation against him, and when and where the alleged wrongful conduct occurred. That‘s all the regulations demand. John and his lawyer had ample notice of Jane‘s accusations and ample time to prepare for the hearing, especially after receiving the investigator‘s report.

John also contends the university “suppressed” exculpatory evidence by withholding from him Jane‘s response to his appeal. It did not. Jane filed her appeal letter after the deadline. The university enforced that deadline against Jane by letting John‘s appeal go unanswered. That is not a procedural irregularity.

John nonetheless tries to frame this point as a subtle way for the university to slant the proceedings against him. He points to the part of Jane‘s response to his appeal where she explained that “my complaint against him was not that he kissed me without my consent. My complaint was that he touched my breasts and digitally penetrated me when I could not consent and did not consent.” The July 26 letter John received with notice of the hearing said that Jane‘s Title IX complaint had alleged non-consensual kissing. This inconsistency, says John, is evidenсe that the university knew Jane had filed a false complaint and withheld that evidence to protect her credibility under the cloak of enforcing the appeal deadline.

That is not a convincing series of inferences. It also does not align with what we typically consider “the suppression of exculpatory evidence” in criminal law, in which the prosecution fails to disclose material exculpatory information that the accused did not know about. See Giglio v. United States, 405 U.S. 150 (1972); Brady v. Maryland, 373 U.S. 83 (1963). Jane had told the investigator: “I didn‘t like want to kiss him, but like I didn‘t, I didn‘t like tell him to stop at that point.” Her response to the investigative report said definitively that John touched her breasts and digitally penetrated her without her consent, but she did not say that he had also kissed her without her consent. Before the hearing, John received a notice that the committee would hear evidence on whether John kissed Jane without her consent. Then at the hearing, Jane was equivocal on that point, saying she had not wanted to kiss him but had not told him to stop. At that point John was fully aware of Jane‘s apparent inconsistency on this issue. His advisor had the opportunity to cross-examine her on it and chose not to do so. The handling of Jane‘s response to the appeal does not reflect anti-male bias.

2. Invited Errors

Other alleged errors were invited by John. Take for instance his assertion that the university violated 34 C.F.R. § 106.45(b)(6) by “allowing evidence as to prior sexual relations between John and Jane.” But it was John who made comments about prior sexual relations with Jane (or the lack thereof) in his interview with the investigator. At the hearing, John then put the subject front and center by changing his story, without warning to Jane or the committee. He said for the first time in the entire proceedings that about a week before the night in question, he and Jane had engaged in similar sexual activity, but with her consent. John‘s argument faults the committee for allowing him to talk about his and Jane‘s prior sexual relations as part of his defense. If this was an error, violating the rules in John‘s favor does not show an anti-male bias against him.

Similarly, John argues the university violated § 106.45(b)(6)(i) by allowing witnesses to address statements of non-witnesses not subject to cross-examination who said that John had committed other sexual assaults. Issues of hearsay and evidence of other similar actions by the respondent have long been controversial in Title IX adjudications and the governing regulations and agency guidance. Regardless of the regulations, however, it was John who opened the door to this evidence. In his response to the investigative report, John said that Jane “may have been pressured by [her roommate] to make the false report against me because [her roommate] has made numerous online posts falsely accusing me of assaulting other [University of Southern Indiana] women.” That response called into question Jane‘s motives for filing the complaint. At the hearing, one question for Jane was of course why she waited three months to tell anyone she had been assaulted. She explained that she felt compelled to speak up after she realized “there was other girls in danger besides myself,” and “that needs to stop.” A hearing officer does not show anti-male bias by asking a complainant to address information the respondent has invoked to attack her credibility.

In addition, some of the statements arose at the hearing in response to text messages that John had provided to the investigator and that had been included in the investigative report. John had submitted close to 50 pages of text messages with Jane from the day after the alleged assault until her roommate first reported the incident three months later. John‘s advisor used those text messages at the hearing to try to impeach Jane‘s credibility. A few messages also referred to other accusations against John. The committee chair asked Jane to explain what those messages referred to, causing Jane to introduce non-witness statements to provide context for what she and John were talking about in the text messages. In that sense, the very text messages that John provided opened the door further to this type of evidence. We must also note that there is no indication in the committee‘s decision that it credited or relied upon accusations against John by anyone other than Jane. The handling of this information does not show bias against John, let alone based on his sex.

3. Errors that Applied Equally to Jane and John

Additional offered irregularities applied to both Jane and John. For instance, John says that the university violated § 106.45(b)(5)(iii) of the regulations by telling him that the investigative report “may not be copied or distributed in whole or in part,” and that “[v]iolation of this rule will result in disciplinary action up to and including dismissal from the University.” John contends that this notice amounted to a “gag” order that prevented him from “sharing information with his parents or any other person who could provide support or assistance relative to the Complaint process.”

It is difficult to see how preventing John from copying or distributing the report itself also meant, as he claims, that he ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍could not share information about the case with his parents, so we are not convinced this was an error at all. Section 106.45(b)(5)(iii) bars schools from “restrict[ing] the ability of either party to discuss the allеgations under investigation or to gather and present relevant evidence.” We recognize that provision is in some tension with § 106.71, which requires that schools protect the confidentiality and identity of the parties in Title IX proceedings, but the two regulations do not necessarily conflict. Without deciding whether the university‘s approach to confidentiality in this case amounted to an error at all, the restriction on distributing the report applied to Jane as well. The possible procedural error that applied equally to both Jane and John does not support an inference of anti-male bias against John.

Along this line, John next contends that the university violated his rights under § 106.45(b)(8)(i) by restricting the subjects he could raise on appeal. That section permits complainants and respondents to file appeals discussing “[p]rocedural irregularit[ies] that affеcted the outcome of the matter,” among other things. § 106.45(b)(8)(i)(A). The university‘s general policy, by comparison, restricts the “procedural irregularities” inquiry to those affecting the “investigation.” John says that policy limitation “restricted his protected rights,” presumably implying that he could not appeal at least some procedural irregularities that affected the hearing. It is not self-evident that the change in terms (affected the outcome v. affected the investigation) was intended to change the scope of a party‘s appeal rights. Even if it might have, John did appeal aspects of the proceedings that he argues here were off limits, and the appeal officer addressed those points in his decision. Moreover, the appeal policy applied to every respondent and every complainant, regardless of sex. The university did not act with an anti-male bias against John by enforcing a generally applicable policy that also applied to Jane.⁷

4. Arguable Errors

What we have left are procedural choices that could arguably be considered mistakes. They are not enough to show a likely bias against men. For example, as noted, Jane‘s complaint against John asserted a charge of non-consensual kissing. It also contained charges that John touched her breasts and digitally penetrated her, all without her consent. The committee found that it was more likely than not that John committed the second and third acts, but it did not make a finding as to the first. Under § 106.45(b)(7)(ii)(A) and (E) of the regulations, Title IX committees are supposed to render determinations of responsibility for each allegation in a complaint. Failing to do so here on the non-consensual kissing allegation was an error.

John asserts this procedural violation was intentional and shows the university‘s bias. His theory is that if the committee had found for him on that allegation, it would have been inconsistent with the committee‘s finding that Jane‘s story was consistent and credible. The decision not to make that finding, he says, is evidence that the committee was biased against him because of his sex. We are not convinced. At the hearing, Jane was equivocal about whether the kissing was consensual. The question of kissing then dropped out of the hearing as everyone turned their focus to the more serious allegations that were in dispute. Given the formal accusations, the committee should have made a finding about the question of non-consensual kissing, but we have no difficulty understanding how the focus would shift from that apparently minor issue to the more serious accusations and more serious issues affecting both parties’ credibility. The failure to make a finding on the abandoned non-consensual kissing point does not seem like convincing evidence of bias against men in general or John in particular.

John also says that the committee violated § 106.45(b)(1)(ii), which provides that “credibility determinations may not be based on a person‘s status as a complainant, respondent, or witness.” Here, the committee said that John had a motive to falsify information because he could face a sanction if he were found responsible. Given the history of Title IX regulations over the last ten to twelve years and some schools’ alleged adoption of strongly pro-complainant procedures and even presumptions, that warning in the regulations is important. At the same time, however, § 106.45(b)(1)(ii) is in tension with the “Preamble” to the regulations, which explains that “Title IX personnel are not prevented from understanding and tаking into account each party‘s interests and the ‘stakes’ at issue for each party.” Nondiscrimination on the Basis of Sex in Education Programs or Activities Receiving Federal Financial Assistance, 85 Fed. Reg. 30026, 30247 (May 19, 2020). A witness‘s motives and biases are often at the center of decisions about credibility. Recognizing those factors does not support an inference of sex bias.⁸

Finally, in making its decision, the committee referred to time stamps from two photographs on cell phones to find a “plausible” time window during which John could have assaulted Jane. The first photo was taken at 4:36 am and showed Jane on the floor in her room. That photo was referenced in the investigative report. The second photo was taken at 4:50 am and showed Jane in another room. That photo came up for the first time at the hearing. Jane said the assault lasted five to ten minutes. The committee concluded that the assault occurred and reasoned that it could have occurred between the taking of those two photos. The committee, which met with the witnesses through a video conference, never saw those photos, however. It relied solely on what the witnesses said about them in rendering its decision.⁹

John now says that the committee violated § 106.45(b)(1)(ii) and (b)(5)(i), (vi), and (vii) of the regulations by relying on the photos to create an “unreliable” timeline and by not securing the original photos from the parties. But John did not object to the use of these photos at any point. In fact, he referred to the 4:36 am photo in his response to the investigative report as support for his account. And neither he nor his advisor questioned the photos or their time stamps at any point. At the hearing, John‘s advisor asked the committee members if they had seen the 4:36 am photo and one additional photo (but not the 4:50 am photo) because if they had, he asked that John and Jane have the chance to see them too. The committee chair said that they had not seen the photos and that “[w]e don‘t have any information that you don‘t.” No issue of reliability was raised.

To be sure, under the university‘s policy, advisors could not raise objections during the live hearing. But in the lead-up to the hearing, John and his advisor never indicated that they had a problem with the authenticity of photos or videos that were only described by the parties. If authenticity had been at issue, John‘s advisor could have asked witnesses about them even if he could not make formal objections. He did not do so. The committee included these two photos in its list of facts that were “undisputed in that they [were] not points of contention.”

The regulations do not impose the Federal Rules of Evidеnce on Title IX hearings, and we appreciate the challenges of handling physical and electronic evidence in a hearing conducted by video conference. Still, we agree with John that it would have been a better practice to secure physical copies of the photos before rendering a decision. The committee‘s failure to do so here, however, cannot support a finding of sex discrimination. The committee relied on the relevant evidence that the investigator, Jane, John, and the other witnesses discussed, and John never disputed the reliability of the photos. Everyone had the same information. The committee‘s failure to secure the physical copies did not amount to evidence of anti-male bias against John.

At bottom, John has failed to show that any procedural irregularities, taken one at a time or together, were nearly as clear or nearly as serious as the errors alleged in Purdue University. He has not shown that any irregularities were evidence of gender bias rather than at most simply mistakes.

D. Weight of the Evidence

Finally, John asserts that the committee‘s decision was so clearly against the weight of the evidence that the only way the committee could have found for Jane was through bias against him because of his sex. Again, trials in civil courts sometimes produce erroneous outcomes as well as procedural errors. Appellate courts do not quickly infer that an erroneous result must have been caused by unlawful bias, especially in difficult credibility contests. In a sufficiently lopsided Title IX case, however, an erroneous outcome can support an inference of gender bias. Columbia University, 831 F.3d at 57 (“When the evidence substantially favors one party‘s version оf a disputed matter, but an evaluator forms a conclusion in favor of the other side (without an apparent reason based in the evidence), it is plausible to infer (although by no means necessarily correct) that the evaluator has been influenced by bias.“). That is not what occurred here.

Throughout the process, John and Jane told very different stories about what occurred on the night of November 13–14, 2020. The outcome turned on whom the committee found more credible, and both sides had credibility problems to some degree. John has not convinced us that he is likely to succeed in showing that the decision to credit Jane was made because of his sex.

At the hearing, the committee worked through the investigative report, asking questions of both John and Jane on factual issues that were still not clear from the report and their written responses. The committee also gave both Jane and John a chance to respond to some of the statements made at the hearing by others that ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍contained arguably new information. The committee even asked John to walk through many of the alleged contradictions in Jane‘s account that he had identified in his written response to the investigative report. The committee let him discuss social media posts and threats he had received that were not mentioned in the investigative report to explain how Jane‘s accusations had affected him. Also, of course, as required under the regulations and university policy, both advocates had the chance to cross-examine the witnesses.

In applying a preponderance-of-the-evidence standard, the committee ultimately believed Jane‘s account. It looked unfavorably on John‘s change in narrative from the investigation to the hearing and concluded that Jane‘s story did not have a comparably serious inconsistency. It also found that Jane‘s account was corroborated in two ways: first, by her roommate‘s consistent statement that Jane told her on November 14 that John had digitally penetrated her that night (albeit, not that it was without her consent), which contradicted John‘s new account that the sexual encounter happened a week before. Second, Jane‘s suitemate said John was lying with Jane in bed that evening, bolstering Jane‘s version of events and flatly contradicting John‘s denials that he was ever in bed with Jane that night.

It is also true that there were potential inconsistencies in Jane‘s accounts about whether the kissing was or was not consensual, as well as what Jane was wearing that night, what she was drinking, and what time the assault occurred. Jane also originally told a detective that she did not communicate with John after November 14, which was not true. These issues were aired at the hearing, however. The committee did not engage with those inconsistencies and others in the written report as much as it could have, but that does not mean its findings were against the weight of the evidence, let alone evidence of bias.

Trial judges sometimes credit part of a witness‘s story even if that witness was not consistent with other aspects of her story. E.g., United States v. Tate, 822 F.3d 370, 374 (7th Cir. 2016) (“Discrepancies or inconsistent prior statements are of course relevant in assessing witness credibility, but they ‘do not, as a matter of law, render a witness‘s testimony incredible.‘” (citation omitted)). Here, Jane stayed consistent with the core version of events: the non-consensual touching of her breasts and digital penetration of her vagina, and key aspects of her account were corroborated by others. We do not know just what happened between John and Jane on the night of November 13–14, 2020, but that is not the question for federal courts on the merits of John‘s Title IX claim. On the record before us, the committee‘s choice to credit Jane‘s account over John‘s appears reasonable and falls well short of proof that the committee was biased against men. Accord, Samford University, 29 F.4th at 691 (Title IX committees are not held “to a higher standard than we hold district courts“).

* * *

Just as few trials are perfect, the hearing in this case was not perfect. There is room to criticize the proceedings. But John has not shown that the imperfections or the final decision against him were likely the products of intentional bias against his sex. That is the standard. He has not shown a likelihood of success on the merits that would support a preliminary injunction, and the district court did not abuse its discretion in denying one. Proft v. Raoul, 944 F.3d 686, 693 (7th Cir. 2019). We recognize that the consequences of the committee‘s decision are very serious for John: a three-semester suspension disrupting his college education at a formative stage of life, and loss of his scholarships for at least that time. Yet even if John were found to have shown a threat of imminent irreparable harm and a balance of hardships that would weigh in his favor, he still must show a likelihood of success on the merits, not merely “a ‘better than negligible’ chance.” Mays, 974 F.3d at 822. He has not done so. The district court‘s denial of a preliminary injunction is

AFFIRMED.

Notes

Jane also spoke with a Vanderburgh County Sheriff‘s detective about the November 13–14 incident. She later declined to pursue criminal charges. Her statements to the detective were included in the investigative report.

In our comments about the kissing, we do not mean to imply that this situation would have met the legal definition of consent. Jane was very intoxicated that night. From a legal standpoint, it is possible that she was incapable of consenting to any sexual activity. The committee made no such finding, however, and did not make its decision on that basis.

When pressed on this point, John conceded that he sat on the bottom of Jane‘s bed at one point that evening when everyone was hanging out in her room. He continued to deny ever lying in bed with her that night.

This circuit uses a “sliding scale” approach to preliminary injunctions: “the more likely the plaintiff is to win on the merits, the less the balance of harms needs to weigh in his favor, and vice versa.” . But even if we were persuaded that the balance of harms weighed overwhelmingly in John‘s favor (and we are not saying it does), he would still need to show a likelihood of sucсess on the merits, “not merely a ‘better than negligible’ chance.” , citing .

Jane‘s complaint is the only formal Title IX complaint against John that the university received.

The university‘s reliance on outside contractors reminds us of the role that outside experts can play under ERISA, for example, to help mitigate conflicts of interest faced by employee benefit plan administrators or other fiduciaries deciding about paying benefits. See, e.g., , citing . We do not mean to adopt any general rule of law addressing the point under Title IX. As a matter of fact here, though, it is difficult to see public pressure about sexual assault on campus in general or the accusations against John in particular causing the outside investigator, hearing officers, or appeal officer to be biased against John based on his gender.

The university‘s new policy issued in August 2021 limits procedural irregularities to those affecting the investigation. This is the policy the committee used to inform John on the grounds he could appeal. The appeal officer applied the old policy to John‘s appeal, which did not contain a similar limitation.

Some federal courts have reasoned that evidence of a school‘s anti-respondent bias does not necessarily support an inference of anti-male bias since both men and women can be victims of sexual assault and can commit sexual assault, even though most complainants are women and most respondents are men. See, e.g., (collecting cases; “Most courts to have addressed the issue have concluded that evidence of a school‘s anti-respondent bias does not create a reasonable inference of anti-male biаs.“); see also . At least one court has permitted an inference of anti-male bias where statistical evidence cast some doubt on whether the university‘s “practices were uniformly pro-complainant and anti-respondent.” . We need not resolve this issue here. This arguable error does not support an inference of sex bias even if general anti-respondent bias could support such a claim.

This practice was consistent with the handling of ‍‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‌‍other photos and videos. The committee explained:

Throughout the hearing, the parties and witnesses referred to photos and videos on their phones that had been described during their interviews but had not been submitted during the investigation, as well as to additional photos or videos that had not been reference[d] in the investigation report. The Decision-makers have not seen these photographs or videos and this impacted their ability to establish a reliable chronology of events.

Read the detailed case summary