NOT RECOMMENDED FOR PUBLICATION File Name: 25a0081n.06 No. 24-1092 UNITED STATES COURT OF APPEALS FILED FOR THE SIXTH CIRCUIT Feb 12, 2025 KELLY L. STEPHENS, Clerk ) CHEY DAVIS, ) ) Plaintiff-Appellant, ON APPEAL FROM THE ) UNITED STATES DISTRICT ) v. COURT FOR THE EASTERN ) DISTRICT OF MICHIGAN DELTA COLLEGE; JEAN GOODNOW, ) ) Defendants-Appellees. OPINION ) ) Before: SUTTON, Chief Judge; LARSEN and MURPHY, Circuit Judges.

LARSEN, Circuit Judge. Chey Davis taught English at Delta College. In 2019, Delta’s then-president, Jean Goodnow, denied Davis a promotion to full professor. After Goodnow retired, Delta’s new president, Michael Gavin, retroactively promoted Davis to full professor with backpay. Soon thereafter, Davis quit teaching and sued Goodnow and the school, claiming that the original promotion denial had been based on her race and her support for unionizing Delta’s faculty. The district court awarded Delta and Goodnow summary judgment, and Davis appealed.

For the following reasons, we AFFIRM the district court’s grant of summary judgment.

I.

From 2007 until August 2022, Chey Davis, a black woman, taught English at Delta College, a community college with several campuses in eastern Michigan. Davis primarily taught freshman and remedial English. Delta promoted her to assistant professor in 2010, awarded her tenure in 2012, and promoted her to associate professor in 2014.

No. 24-1092, Davis v. Delta College

In 2018, the Delta faculty initiated a vote on unionizing. Davis and two other faculty members met with then-President Goodnow to notify her of the faculty’s plan. Davis alleges that Goodnow became “extraordinarily angry and growled” at Davis and her colleagues. R. 25-3, Davis Dep., PageID 989. Goodnow pounded her fist on a binder, which Davis assumed contained the policies proposed by the pro-union contingent, and asked Davis and her coworkers whether they “want[ed] to give this up.” Id. at 993.

Davis alleges that Goodnow, a white woman, exhibited hostility towards Davis in other ways beyond opposing unionization. For instance, Davis says Goodnow spoke “aggressive[ly]” to Davis in public meetings, “purposefully ignore[d]” Davis and other black faculty in the hallway, and opposed some of Davis’s projects, including certain theatre productions and a visit to the Jim Crow Museum. Id. at 986, 989, 1038. Moreover, Davis alleges that Goodnow generally treated black faculty in a “stark[] and unsupportive” manner, which she says led to an “exodus” of black professors and staff. Id. at 986, 988.

During the fall 2018 semester, Davis began preparing her application for promotion to full professor. Under Delta’s promotion procedures for faculty hired before 2012, the first step is to secure approval from the applicant’s faculty division and the division chair who provides a written recommendation to the appropriate academic dean. The dean provides a recommendation to the chief academic officer, who in turn makes a recommendation to Delta’s president. The president then provides a recommendation to Delta’s Board of Trustees, which makes the final decision.

Throughout the process, the relevant promotion criteria are the candidate’s “[t]eaching effectiveness,” “[e]ducational and professional growth,” additional “[p]roductive activity for the College,” and leadership record. R. 17-8, Senate Handbook, PageID 519. The candidate must demonstrate “excellence” in teaching effectiveness. Id.

-2- No. 24-1092, Davis v. Delta College

Davis passed the first several stages of promotion review, but her application stalled with Goodnow. On April 5, 2019, Goodnow informed Davis that her promotion review would be delayed, consistent with Delta Senate policy. Goodnow cited concerns with Davis’s teaching effectiveness and leadership record. She noted that Davis’s application lacked student feedback— which Goodnow considered important to a candidate’s ability to demonstrate “excellence” in teaching effectiveness—and she believed Davis’s record did not reflect the level of leadership required for promotion. Goodnow asked Davis to submit student feedback from 2015 to 2019 and additional evidence of leadership. Two months later, in June 2019, Goodnow further delayed Davis’s promotion decision, noting that Davis had failed to provide the requested materials. Later that month, Goodnow received Davis’s “Student Feedback Summary,” though it included data only from 2015 to 2017.

In August 2019, Goodnow informed Davis that she would be denying Davis’s request for promotion to full professor, again citing a lack of “excellence in teaching effectiveness” and leadership. R. 17-17, Aug. 2019 Memo, PageID 583. Eight days later, Davis filed a grievance with an associate dean. In November 2019, Delta’s Grievance Committee recommended reconsideration of Davis’s application. The Grievance Committee stated that Goodnow had misapplied Senate policy by requiring Davis’s leadership “to relate to Delta College.” R. 17-19, Grievance Comm. Memo, PageID 588. In December 2019, after again reviewing Davis’s promotion packet and student-feedback forms, Goodnow denied Davis a promotion. Goodnow encouraged Davis to reapply for promotion in the future, and she offered suggestions on how Davis might strengthen her application. The record does not reflect that Davis further pursued a promotion.

-3- No. 24-1092, Davis v. Delta College

In August 2021, Goodnow retired, and Dr. Michael Gavin succeeded her. In January 2022, Gavin promoted Davis to full professor, deeming the promotion retroactive to July 1, 2019, and awarding her backpay. Six months later, Davis resigned to pursue a career in social work.

In August 2022, Davis sued Delta and Goodnow in state court. She claimed race discrimination under Michigan’s Elliott-Larsen Civil Rights Act (ELCRA) and First Amendment retaliation under 42 U.S.C. § 1983.1 After Delta and Goodnow removed the case to federal court, they sought summary judgment, which the district court granted. Davis timely appealed.

II.

We review a district court’s grant of summary judgment de novo. Dixon v. Gonzales, 481 F.3d 324, 330 (6th Cir. 2007). Summary judgment is warranted if the movant can show there’s no genuine dispute of material fact and the moving party is entitled to judgment as a matter of law.

Fed. R. Civ. P. 56(a). We construe the evidence and make all reasonable inferences in the nonmovant’s favor. Tingle v. Arbors at Hilliard, 692 F.3d 523, 529 (6th Cir. 2012). The party opposing summary judgment must produce enough evidence to show that “a reasonable jury could return a verdict” in her favor. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986).

Davis also brought an ELCRA claim alleging sex discrimination, but the parties stipulated to the claim’s dismissal in August 2023. Although Davis was retroactively promoted and awarded backpay, her claims remain live because she seeks non-economic damages, which are available under ELCRA even without underlying economic damages. See Hyde v. Univ. of Mich. Regents, 575 N.W.2d 36, 41 (Mich. Ct. App. 1997); Ronan v. Fam. Chiropractic & Wellness of Midland, PLLC, No. 352706, 2021 WL 2025182, at *2, *4–5 (Mich. Ct. App. May 20, 2021) (per curiam) (upholding a jury award of $150,000 in noneconomic damages under ELCRA, despite the jury awarding no economic damages). Emotional distress is also compensable under § 1983. See Bloch v. Ribar, 156 F.3d 673, 679 (6th Cir. 1998). -4- No. 24-1092, Davis v. Delta College

III.

A.

We start with Davis’s ELCRA claim. ELCRA says that an employer may not “[f]ail or refuse to hire or recruit, discharge, or otherwise discriminate against an individual with respect to employment, compensation, or a term, condition, or privilege of employment, because of . . . race . . . .” Mich. Comp. Laws § 37.2202(1)(a).

In indirect-evidence cases, like this one, Michigan courts apply the McDonnell Douglas burden-shifting framework. Hazle v. Ford Motor Co., 628 N.W.2d 515, 520–21 (Mich. 2001); see McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802–05 (1973). Under this framework, a plaintiff must first make out a prima facie case of discrimination. Hazle, 628 N.W.2d at 521. The employer must then present a legitimate, nondiscriminatory explanation for its action. Id. It then becomes the plaintiff’s burden to “demonstrate that [the employer’s] articulated reason was merely a pretext for unlawful discrimination.” Campbell v. Hum. Servs. Dep’t, 780 N.W.2d 586, 594 (Mich. Ct. App. 2009). Here, Goodnow presented a nondiscriminatory rationale for denying Davis a promotion: Davis failed to demonstrate “excellence” in teaching effectiveness. So we may proceed directly to the third step. Town v. Mich. Bell Tel. Co., 568 N.W.2d 64, 70 (Mich. 1997).

We ask whether Davis has produced sufficient evidence to permit the conclusion that race discrimination, rather than a lack of teaching excellence, was Delta and Goodnow’s “true motive” in denying her a promotion. Id. at 68.

A plaintiff can establish pretext by showing that the defendant’s explanation (1) lacked a factual basis; (2) did not actually motivate the decision; or (3) was “insufficient to justify the decision.” Dubey v. Stroh Brewery Co., 462 N.W.2d 758, 760 (Mich. Ct. App. 1990) (per curiam); see also Major v. Village of Newberry, 892 N.W.2d 402, 413 (Mich. Ct. App. 2016).

-5- No. 24-1092, Davis v. Delta College

First, Davis alleges that Goodnow’s proffered reason lacked a factual basis. This argument fails. Goodnow says she denied Davis a promotion because Davis scored poorly on student‑feedback metrics, thereby not demonstrating the required “excellence” in teaching effectiveness. Davis did, in fact, score poorly. Therefore, Goodnow’s explanation rests on solid factual ground.

Under Delta’s promotion policy, “student evaluation of faculty” was one measure of teaching excellence, along with “evaluation by division chair, . . . report by peer review team . . . , candidate’s personal appraisal, and any other evidence recognized by the division.” R. 17-9, Acad.

Promotion Guide, PageID 526. Two categories of data composed “student evaluation of faculty”: (1) semester-by-semester course evaluations completed by a candidate’s current students; and (2) former-student surveys completed during a candidate’s promotion process.

Course evaluations asked Delta students to answer questions like, “Did [the instructor] present course materials understandably?”; “As a result of [the instructor’s] teaching, how much did you learn?”; and “Would [you] recommend [this instructor] to other students?” R. 26-8, Lacina Survey, PageID 1439–40. Students could select from a set of responses, like “Definitely yes,” “Probably,” “Not sure,” and “Probably not” and assign a final rating of “Excellent,” “Good,” “Fair,” or “Average.” Id. at 1440. Instructors were given student-evaluation “scores,” which Delta calculated by adding the percentage of responses rating the instructor in either the first or second most positive categories. For example, if 45% of students rated an instructor as “excellent” and another 45% rated her as “good,” she’d receive a score of 90%. Candidates promoted to full professor generally scored above 90% on the bulk of questions asked.

For instance, Davis presented as comparators two non-black English professors whom Goodnow promoted to full professor: Ray Lacina and Betheen Glady-Teschendorf. Both scored

-6- No. 24-1092, Davis v. Delta College

above 90% on nearly all questions. For semester-by-semester feedback, Lacina scored above 90% on all but one question, and 95.6% of students said they would recommend him to others. And he scored above 90% on seven of eight questions sent to former students. Likewise, Glady-Teschendorf, in semester-by-semester feedback from 2013 to 2019, scored above 90% on twenty of twenty-two questions.

Davis, by contrast, scored much worse. Even though Goodnow had requested Davis’s student feedback through 2019, Davis only submitted four semesters’ worth of data, dating from 2015 to 2017. In Davis’s best year, she scored above the 90% threshold on just eleven of twenty-two questions; in her worst year, she hit the 90% target only twice; and her average across the years was seven.2 Davis fared even worse on the former-student survey, scoring above 90% on only one of nine questions. On the remaining eight questions, Davis scored above 80% only once, the rest between 67 and 79%. Davis also received a litany of negative student comments.3 What’s more, Davis’s scores (and qualitative student reviews) were trending downward.

Goodnow reported that the negative comments constituted a pattern that she did not see in other candidates for full professor; this led her to believe that Davis was not achieving her courses’ learning objectives.

Davis does not dispute the accuracy of either her numerical student-feedback scores or the negative comments. And Gavin, Delta’s subsequent president who promoted Davis,

During the winter 2015 semester, Davis scored above 90% on only ten of twenty-two questions.

In fall 2016, Davis scored above 90% on eleven of twenty-two questions. In winter 2016, Davis met the 90% threshold on only five of twenty-two questions. In fall 2017, Davis went two for twenty-two.

Negative student comments included concerns about course content (“Teach more English, how to properly write an essay.”) and course management (e.g., poor organization and failure to follow the schedule). To be sure, she received some positive qualitative feedback as well (“Chey Davis, you are amazing!”). -7- No. 24-1092, Davis v. Delta College

acknowledged that Davis’s scores were low. In sum, Goodnow’s proffered reason for denying Davis a promotion did not lack a factual basis. Just the opposite: Davis scored poorly and received substantial negative feedback. No reasonable juror could reject this conclusion.

Next, Davis alleges that Goodnow’s proffered reason did not actually motivate her decision. Davis points to several pieces of evidence which, she says, suggest that Goodnow was “looking for a reason” to reject her application and show that her lack of teaching excellence did not actually motivate Goodnow’s decision. Appellant Br. at 39. For instance, Davis alleges that Goodnow’s initial decision unfairly discounted her off-campus leadership. But Davis presents no evidence suggesting that Goodnow considered other candidates’ off-campus leadership; to the contrary, Dr. Reva Curry, a Delta vice president, testified that for promotion to full professor Goodnow routinely demanded “that leadership . . . be to Delta College.” R. 17-6, Curry Dep., PageID 456.

Next Davis contends that Goodnow’s rationale changed over time. See Miles v. S. Cent.

Hum. Res. Agency, Inc., 946 F.3d 883, 890 (6th Cir. 2020) (“[A]n employer’s shifting termination rationales are evidence that the proffered rationale may not have been the true motivation for the employer’s actions.”). But Goodnow’s rationale did not meaningfully shift. Goodnow’s initial decision noted both Davis’s lack of teaching excellence and lack of leadership. After Davis contested the leadership determination, that rationale dropped away, but Goodnow’s student-feedback rationale remained constant throughout. And Delta policy required applicants for full professor to demonstrate teaching excellence.

Davis alleges that Goodnow’s focus on Davis’s student-evaluation scores was “inconsistent” with Delta’s promotion policy and that Goodnow ignored everything else.

Appellant Br. at 41. But Delta’s Academic Promotion Guide lists “student evaluation of faculty”

-8- No. 24-1092, Davis v. Delta College

as a measure of teaching effectiveness. R. 17-9, Acad. Promotion Guide, PageID 526. Certainly, another president might have emphasized this measure less than Goodnow—in fact, Gavin appears to have done so by promoting Davis despite her low scores. But Delta policy doesn’t dictate how an evaluator must weigh student-feedback scores against other measures of teaching effectiveness.

Goodnow reasonably viewed student-feedback scores as an important metric of teaching excellence. And, critically, she consistently applied the same promotion standards to Davis and other applicants.

Davis takes issue with Goodnow for not considering that she taught developmental English, a subject that sometimes produced less positive student feedback. But Davis presents no evidence suggesting that Goodnow viewed other candidates’ scores in a subject‑specific context. That includes Glady‑Teschendorf, who taught the same courses as Davis but scored markedly higher.

So Davis’s assertions don’t give rise to an inference that Goodnow treated Davis less favorably than non-black candidates. Davis might take umbrage with how Goodnow reviewed all promotion applications, but that casts no doubt on Goodnow’s explanation for why she denied Davis a promotion.

Finally, Davis points out that Curry recommended her for promotion. At Goodnow’s deposition, Davis’s counsel asked Goodnow whether she thought “Curry wasn’t being objective” in evaluating Davis’s application because “Curry is also black.” R. 26-1, Goodnow Dep., PageID 1367. Goodnow responded, “I think that thought crossed my mind.” Id. When asked why, Goodnow said: Well, because of how [Curry] talked with me about it, about how she was struggling. And because she and I had some wonderful conversations in the past about diversity and our shared commitment to diversity, and I think it was a struggle for her in this particular situation, so I think it was hard, I think it was – I think it was real difficult for her.

-9- No. 24-1092, Davis v. Delta College

Id. But Curry herself testified that she did not think Goodnow’s evaluation of Davis’s application was at all race based. Moreover, Goodnow invited Davis to reapply for a promotion later and suggested how Davis could improve her application. This too suggests a lack of pretext.

In sum, Davis has not raised a genuine issue that race, rather than a lack of teaching excellence, motivated Goodnow’s decision to deny Davis a promotion. Davis’s qualitative and quantitative teaching evaluations were dramatically lower than those of other candidates who were promoted, even those who taught the same classes as Davis. And there is no evidence suggesting that Goodnow applied different criteria to judge Davis’s application compared to her peers. It is true that many people at Delta felt Davis deserved a promotion, including Goodnow’s successor.

But the bare fact that Goodnow disagreed with her colleagues about Davis’s application does not establish that Goodnow constructed her student-feedback rationale as a cover for racial discrimination. All it shows is a difference of opinion between Goodnow and others at Delta on how much weight to place on student evaluations. To succeed on an ELCRA claim, however, “[t]he plaintiff cannot simply show that the employer’s decision was wrong or mistaken, since the factual dispute at issue is whether discriminatory animus motivated the employer, not whether the employer is wise, shrewd, prudent, or competent.” Town, 568 N.W.2d at 72 (citation omitted).

B.

We turn now to Davis’s First Amendment retaliation claim. Davis alleges that Goodnow refused to promote her because she supported unionizing Delta’s faculty. To establish a First Amendment retaliation claim, a plaintiff must show (1) she engaged in protected conduct; (2) the defendant took an adverse action against her; and (3) a causal connection. DeCrane v. Eckart, 12 F.4th 586, 594 (6th Cir. 2021). We assume that Davis’s pro-union advocacy constituted

-10- No. 24-1092, Davis v. Delta College

protected conduct4 and that the promotion denial was an adverse action. Even so, Davis cannot establish a retaliation claim because she cannot show causation.

A First Amendment plaintiff may recover under § 1983 only if her protected speech was a but-for cause of the defendant’s adverse action. See Lemaster v. Lawrence County, 65 F.4th 302, 309 (6th Cir. 2023). To establish causation, a plaintiff must first show that her speech “was a substantial or motivating factor of” the adverse action. Id. (internal quotation marks and citations omitted). She may do so by presenting direct or circumstantial evidence “reasonably linking” the protected speech and adverse action. Rodgers v. Banks, 344 F.3d 587, 602 (6th Cir. 2003) (citation omitted). If Davis successfully establishes each element, the burden shifts back to the defendants to show that the decision to reject Davis’s application “would have been the same absent the protected conduct.” Sowards v. Loudon County, 203 F.3d 426, 431 (6th Cir. 2000). In other words, the defendant “must prove the absence of but-for causation.” Lemaster, 65 F.4th at 309 (emphasis omitted).

The district court held that Davis failed to show a causal connection between her pro-union activity and Goodnow’s decision to deny Davis a promotion. We agree.

We typically begin our causation inquiry by analyzing the time gap between the protected speech and the adverse action. Id. at 310. The longer the gap, “the more evidence that a plaintiff must offer to show that the speech motivated the [adverse] action.” Id. A long gap alone may doom a plaintiff’s ability to show causation. Fuhr v. Hazel Park Sch. Dist., 710 F.3d 668, 676 (6th Cir. 2013) (stating that a “multi-year gap proves fatal” to establishing causation). On the other

The First Amendment protects public-employee speech only if the employee (1) addresses a matter of public concern and (2) speaks as a private citizen rather than pursuant to her job duties.

See DeCrane, 12 F.4th at 594. If both prongs are met, we balance the employee’s interest in speaking against the employer’s countervailing operational interests. Id. Because we hold that Davis cannot show causation, we need not address whether she engaged in protected conduct. -11- No. 24-1092, Davis v. Delta College

hand, a very short gap (i.e., mere days or weeks) “can sometimes (if rarely)” be enough to permit an inference of causation. Lemaster, 65 F.4th at 310; see also Sensabaugh v. Halliburton, 937 F.3d 621, 630 (6th Cir. 2019). When neither extreme applies, we ask whether the plaintiff has presented enough additional evidence of retaliation to survive summary judgment. See Lemaster, 65 F.4th at 310.

Here, the gap was more than a year. Davis informed Goodnow of the faculty’s intent to unionize in summer 2018. Goodnow did not deny Davis a promotion until August 2019, and that decision did not become final until December 2019.5 A year-long gap between protected conduct and adverse action is certainly far too long for a court to infer causation based on temporal proximity alone. And while it may not be so long as to automatically doom Davis’s claim,6 see

Davis argues that the relevant date of the adverse action is March 2019, making the gap between protected conduct and adverse action seven months. In March 2019, Goodnow told Curry that she did not plan to promote Davis, and she notified Davis in April 2019 that she was delaying the decision whether to recommend her for promotion. But Davis’s proffered date—March 2019— cannot possibly be the relevant one. A supervisor who informs a subordinate third-party of her intention to reject another employee’s promotion application does not take any adverse action against the employee. An “adverse action” in the context of a First Amendment retaliation claim is “one that would ‘deter a person of ordinary firmness’ from the exercise of the right at stake.”

Thaddeus-X v. Blatter, 175 F.3d 378, 396 (6th Cir. 1999) (en banc) (opinion of Moore, J.) (citation omitted). Examples in the employment context include a “discharge, demotion[], refusal to hire, nonrenewal of [a] contact[], and [a] failure to promote.” Id. These actions are all formal decisions carrying negative consequences for an employee—unlike an employer either musing to a colleague about what action she may later take or informing an employee of her intent to delay a promotion decision. Thus, March 2019 is not the relevant date.

Our caselaw does not make clear exactly when, and under what circumstances, a temporal gap is so long that it entirely fails as proof of causation. In Fuhr, we held that a two-year gap “prove[d] fatal.” 710 F.3d at 675. But otherwise, Fuhr offered little guidance on where to draw the line. In George v. Youngstown State University, we considered a Title VII retaliation claim brought by a professor who sued his employer alleging retaliation and age discrimination after he was terminated from his position as an assistant professor in 2015. 966 F.3d 446, 453 (6th Cir. 2020).

George had previously sued the school in 2006, after he was denied tenure, and that 2006 lawsuit constituted protected conduct. Id. at 452. Despite the nine-year gap between the original lawsuit and the 2015 termination, we concluded that the gap did not automatically preclude George from demonstrating causation. Id. at 460–61. Over a dissent citing Fuhr, the panel counseled against -12- No. 24-1092, Davis v. Delta College

Fuhr, 710 F.3d at 675–76, the long gap increases Davis’s burden to offer additional evidence “to show that the speech motivated the action.” Lemaster, 65 F.4th at 310; see Vereecke v. Huron Valley Sch. Dist., 609 F.3d 392, 400 (6th Cir. 2010).

Davis resists this conclusion, arguing that the lengthy gap between her speech and the promotion decision should not cut against a finding of causation because Goodnow retaliated at her first meaningful opportunity. The first-meaningful-opportunity doctrine functions as an exception to the general rule that a lengthy delay vitiates a finding of causation. See Kirilenko-Ison v. Bd. of Educ., 974 F.3d 652, 665 (6th Cir. 2020). But the exception applies only when the alleged retaliator had no prior opportunity to retaliate. For example, in Dixon v. Gonzales, we held that the plaintiff, Dixon, could not show causation because he was within Reutter’s (the alleged retaliator) chain of command for several years, yet Reutter took no adverse action against him. 481 F.3d 324, 335–36 (6th Cir. 2007). During that time, Reutter “could have easily exercised his authority . . . as Dixon’s superior,” as Reutter signed off on Dixon’s performance evaluations. Id. at 336.

Goodnow served as President for more than a year after Davis’s pro-union advocacy yet took no action against Davis. Davis asserts that Goodnow had no prior opportunity to retaliate against her, but she does not support that assertion with any factual detail. This lack of record evidence cuts against Davis. She bears the burden of establishing that her protected speech motivated Goodnow’s decision to deny her a promotion. See Lemaster, 65 F.4th at 309. Thus, she must present enough factual material to permit the inference that Goodnow retaliated at her

“draw[ing a] bright line rule[].” Id. at 461. This court noted that the employer had financial incentives to employ George until the end of the 2014–15 academic year because a previous settlement agreement between the parties required Youngstown State to provide George with medical coverage if they fired him before spring 2015. Id. at 460. We noted that the close temporal proximity to the end of George’s settlement coverage suggested that the firing was retaliatory. Id. -13- No. 24-1092, Davis v. Delta College

first meaningful opportunity. In other words, she must present some evidence to show that Goodnow couldn’t have retaliated earlier. She has not done so. Davis does not present any factual detail showing that Goodnow acted at her first meaningful opportunity.

In sum, the year-long gap between Davis’s speech and her promotion denial is too long to infer causation based on temporal proximity alone, and the first-meaningful-opportunity exception does not apply here. Therefore, Davis must “supplement h[er] claim with other evidence of retaliatory conduct.” Vereecke, 609 F.3d at 400 (internal quotation mark and citation omitted); see also Mickey v. Zeidler Tool & Die Co., 516 F.3d 516, 526 (6th Cir. 2008) (noting that additional evidence of retaliatory conduct “has commonly included evidence of additional discrimination” post-protected activity and pre-adverse action). And because the gap here is long, that evidence must be fairly strong to survive summary judgment.

Davis’s circumstantial evidence of causation is quite weak. She first points out that when Davis first informed Goodnow of the faculty’s plan to unionize, Goodnow became “extraordinarily angry and growled” at Davis, pounding her fist on a binder. R. 17-3, Davis Dep., PageID 219.

And Goodnow referred to Davis and her fellow union activists as “troublemakers.” R. 25-10, Randolph Declaration, PageID 1322. Karen Randolph, Delta English professor and chair of Davis’s promotion committee, said she “strongly believe[d]” this label stemmed from the trio’s pro-union activity. Id. While these events provide some indicia of how Goodnow viewed faculty unionization, they are weak evidence of causation because they occurred at least seven months before Goodnow first denied Davis a promotion in August 2019 (and at least eleven months before

-14- No. 24-1092, Davis v. Delta College

the denial became final in December 2019).7 Beyond this, Davis offers only a December 2018 email exchange between Goodnow, Curry, and an adjunct faculty member, which Davis says reveals Goodnow’s intolerance of dissent. But that exchange does not show causation because it is eight-to-twelve months removed from the non-promotion decision and completely unrelated to Davis.

Davis presents no other evidence from which a reasonable juror could conclude that the union issue motivated Goodnow’s promotion decision. Again and again, Goodnow’s communications both to Davis and others focused on her low scores and lack of leadership.

In sum, Davis fails to produce enough circumstantial evidence for a reasonable juror to conclude that her pro-union efforts were a substantial or motivating factor in Goodnow’s decision.

Davis was not denied a promotion for at least a year after she engaged in protected speech. That gap is far too long for temporal proximity to alone establish causation. And Davis’s additional circumstantial evidence is too weak for a reasonable juror to conclude that her pro-union advocacy motivated Goodnow’s decision.

*** We AFFIRM.

Because the faculty voted to unionize in January 2019 and Randolph says she heard Goodnow use the “troublemakers” nickname while she and Davis worked to organize the faculty union, presumably Randolph heard Goodnow use the nickname between summer 2018 and the unionization vote in January 2019. -15-