FOR PUBLICATION

UNITED STATES COURT OF APPEALS

FOR THE NINTH CIRCUIT

JUN YU, No. 20-35582 Plaintiff-Appellant, D.C. No. v. 4:15-cv-00430-REB

IDAHO STATE UNIVERSITY,

Defendant-Appellee. ORDER AND

OPINION

Appeal from the United States District Court for the District of Idaho Ronald E. Bush, Magistrate Judge, Presiding

Argued and Submitted June 7, 2021 Seattle, Washington

Filed October 20, 2021

Before: Ronald M. Gould, Richard R. Clifton, and Eric D. Miller, Circuit Judges.

Order; Opinion by Judge Gould; Concurrence by Judge Miller SUMMARY *

Title VI

The panel filed: (1) an order withdrawing the opinion filed on August 31, 2021, and replacing it with a superseding opinion; and (2) a superseding opinion affirming the district court’s judgment after a bench trial in favor of Idaho State University in an action brought under Title VI by Jun Yu.

Yu, a Chinese international student, alleged that the university intentionally discriminated against him based on his race or national origin when it dismissed him from a doctoral program in clinical psychology. At trial, Yu relied in part on expert testimony that he was a victim of aversive racism, a theory of prejudice that the parties, the district court, and the expert compared to unconscious or implicit racial bias.

The panel held that the district court did not clearly err in finding that Yu failed to show that the university intentionally discriminated against him. The panel declined to address whether implicit bias may be probative or used as evidence of intentional discrimination under Title VI because resolution of this issue was not necessary to the disposition of this appeal.

Concurring, Judge Miller wrote that he joined the opinion in full. He wrote separately to note several reasons why testimony of the kind offered in this case will rarely, if

* This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader.

YU V. IDAHO STATE UNIVERSITY 3

ever, be admissible. First, expert testimony is not admissible simply to cast doubt on the credibility of other witnesses, but that is essentially what the expert did here. Second, before allowing scientific evidence to be presented to a jury, the district court must assess whether it rests on tested scientific principles. Here, though, the expert’s claimed ability to identify aversive racism did not appear to rest on the kind of scientific principles that the Supreme Court has demanded. Third, the expert relied on the use of race-neutral explanations as evidence of unconscious bias. The defendant in a Title VI case, however, is required to present a race-neutral explanation for its action once the plaintiff has presented a prima facie case of discrimination.

COUNSEL

Ronaldo A. Coulter (argued), Idaho Employment Law Solutions, Eagle, Idaho, for Plaintiff-Appellant.

Michael E. Kelly (argued), Special Deputy Attorney General; Shannon M. Graham, Attorney; Kelly Law PLLC, Garden City, Idaho, for Defendant-Appellee.

Olivia N. Sedwick, Murnaghan Appellate Advocacy Fellow, Public Justice Center, Baltimore, Maryland, for Amici Curiae Public Justice Center, Reed T. Korematsu Center for Law and Equality, Chinese American Progressive Action, Dr. Russell Jeung, LatinoJustice, and Chinese for Affirmative Action.

Lisa Hogan and Martha L. Fitzgerald, Brownstein Hyatt Farber Schreck LLP, Denver, Colorado, for Amici Curiae National Latinx Psychology Association, Society of Indian Psychologists, Association of Black Psychologists, and Asian-American Psychological Association.

Eva Paterson, Rau Mona Tawatao, and Christina Alvernaz, Equal Justice Society, Oakland, California; Kass Harstad and Erika Birch, Strindberg & Scholnick LLC, Boise, Idaho; for Amici Curiae Equal Justice Society, Legal Aid at Work, National Employment Lawyers Association, and Public Rights Project.

ORDER

The Opinion filed on August 31, 2021, is WITHDRAWN and replaced with a superseding Opinion filed concurrently with this Order. Future petitions for rehearing will be permitted under the deadlines outlined in Federal Rules of Appellate Procedure 35(c) and 40(a)(1).

IT IS SO ORDERED.

OPINION

GOULD, Circuit Judge:

Jun Yu, a Chinese international student, enrolled in Idaho State University’s (“ISU”) Doctoral Program in Clinical Psychology (the “Program”) in 2008. He completed the requisite four years of instruction and wrote and successfully defended his dissertation. However, he failed to complete the last requirement of the Program, satisfactory completion of a professional internship consisting of 2,000 clinical hours over the course of 11 months. After Yu was dismissed from the internship, ISU dismissed Yu from the Program altogether.

YU V. IDAHO STATE UNIVERSITY 5

Yu filed the present suit, alleging that ISU violated Title VI because it intentionally discriminated against him based on his race or national origin. At trial, Yu relied in part on the expert testimony of Dr. Leslie Wade Zorwick. Dr. Zorwick opined that Yu was a victim of “aversive racism,” a theory of prejudice that the district court, the parties, and Dr. Zorwick compare to “unconscious” or “implicit” bias. After a bench trial, the district court found that Yu had failed to show that ISU intentionally discriminated against him. Yu appealed the verdict. We have jurisdiction pursuant to 28 U.S.C. § 1291, and we affirm judgment in favor of ISU. The evidence in the record shows that the district court permissibly found that ISU did not intentionally discriminate against Yu.

I. FACTUAL AND PROCEDURAL BACKGROUND

Yu began taking courses in the Program in fall 2008. Although Yu’s academic progress was consistently evaluated as “satisfactory,” his professional progress dropped to “unsatisfactory” in his fall 2011 evaluation. A number of Yu’s supervisors commented on his limited English language fluency throughout his time at ISU. Supervisors also commented, in what can only be characterized as repeated criticisms, that Yu had trouble “form[ing] alliances” with clients and patients, “need[ed] more practice counseling patients,” and possessed limited “ability to adjust treatment.” In light of these comments, the Clinical Training Committee (the “Committee”), which was tasked with evaluating the training progress of Yu, expressed that it could only support Yu applying to professional internship sites where “his Chinese language is a strength, rather than a liability.”

During the fall 2011 semester, Dr. John Landers supervised Yu’s off-site clinical externship. Again, the results of that supervision were not favorable for Yu. Dr. Landers dismissed Yu from his externship before the externship was scheduled to end, testifying that Yu was never able “to grasp the communication nuances that are required” in a position at his site. Dr. Landers concluded that, based on the vulnerability of Dr. Landers’s patients, who were particularly high risk, Dr. Landers could not “afford to remediate or experiment and try to teach someone how to do things that they should know how to do with these particular patients.”

Before he was dismissed from the externship, Yu had applied for a professional internship through APPIC. 1 Unlike the externship, which can be completed during the first four years of study in the Program, the professional internship is completed during the fifth and final year. The professional internship requires 2,000 hours of clinical work over the course of eleven months, during which time students are expected to work independently with clients. After he was dismissed from the externship with Dr. Landers, Yu was informed that he had not been matched with any internships. In Yu’s spring 2012 evaluation, the Committee set out “[t]hree different internship paths” Yu could take. First, he could reapply the next November, but he would need to disclose in his applications that he had been dismissed from his externship. Second, Yu could propose his own, comparable internship subject to approval by the department. Third, he could pursue an internship in China, the option that the Committee recommended considering his language difficulties and his intent to return to China after completing the Program. Yu chose the second option: he

1 APPIC, the Association of Psychology Postdoctoral and Internship Centers, facilitates “matching” clinical psychologist doctoral students with internship placement sites.

YU V. IDAHO STATE UNIVERSITY 7

proposed working offsite at the Cleveland Clinic under the supervision of Dr. Leslie Speer and Dr. Thomas Frazier, and—separately—Dr. Cheryl Chase.

After only a few days, Dr. Speer told an ISU faculty member that Yu was not performing at the expected competency levels for a clinical psychology intern. At the faculty member’s request, Dr. Speer provided a formal evaluation, including a “joint written remedial plan” noting that Yu’s competency would be reevaluated in two months. Dr. Frazier concluded early in the internship that Yu was not prepared for the internship and ceased working with him.2 Dr. Speer then completed a second evaluation form on April 1, 2013, explaining that Yu “ha[d] not made progress” and expressing concerns that he “is unaware of [his] own limitations [and the] [c]ombination of [the] above factors put [him] at risk for causing harm to patients.” Yu was dismissed from the internship on April 3, 2013.

Yu was not then given the opportunity to pursue one of the other internship options originally presented to him, although he had already taken steps to arrange an internship in China. Instead, the Graduate Faculty of the Psychology Department voted to dismiss Yu from the Program as shown by its letter dated May 3, 2013. Drawing on the negative comments that it had received from the several clinical supervisors noted above and from ISU faculty members, the Committee concluded that Yu lacked “sufficient perspective-taking skills and conceptual abilities to become a clinical psychologist.” When the Graduate Faculty of the Psychology Department unanimously upheld its decision on

2 The record does not indicate that Dr. Frazier gave Yu notice, feedback, or a remedial plan. appeal, it expressed that Yu may pose a risk to clients and patients, even if he were to return to China.

After exhausting his administrative appeals, Yu filed a complaint against ISU in the United States District Court for the District of Idaho in September 2015. The district court denied ISU’s motion for summary judgment on Yu’s Title VI disparate treatment claim, and held a four-day bench trial.

During this trial, Dr. Zorwick testified as an expert witness regarding aversive racism. Dr. Zorwick described aversive racism as the “dominant theory” of modern race- based prejudice. According to Dr. Zorwick, the theory of aversive racism rests on the assumption that modern prejudice no longer manifests through explicit and overt expressions. Rather, Dr. Zorwick explained that individuals may simultaneously hold a generalized belief in egalitarianism while being profoundly influenced by negative racial stereotypes that they have absorbed through their socialization. The result of Dr. Zorwick’s aversive racism theory is that even people who expressly profess to be egalitarian may treat someone differently because of their race.

Dr. Zorwick testified that there are five characteristic hallmarks of aversive racism: (1) “the presence of . . . ambiguity surrounding decision-making,” (2) “race neutral explanations after the fact,” (3) “the expression of microaggressions,” which are small events “that communicate who is and [is not] valued,” (4) “challenging interracial interactions and relationships,” and (5) “the use of post hoc justification.” After reviewing the documents in this case, Dr. Zorwick purported to identify examples of all five hallmarks in the interactions between Yu and ISU. Dr. Zorwick stressed that ISU imposed ambiguous English fluency standards on Yu, “that [ISU] faculty framed [Yu’s]

YU V. IDAHO STATE UNIVERSITY 9

multilingualism as primarily a liability,” that ISU showed no “awareness of the structural barriers that . . . Asian international students face,” and that after ISU decided to dismiss Yu, it focused only on the negative evaluations he received. Dr. Zorwick opined that Yu’s race and international status impacted the way he was treated by the ISU faculty through a pattern of intentional, repeated choices made by ISU faculty.

The district court did not dismiss the theories of Dr. Zorwick summarily or with light treatment. Instead, the district court, in nearly 20 pages of thoughtful analysis, concluded that Dr. Zorwick’s testimony did not demonstrate that Yu was the victim of intentional discrimination. Similarly, the district court found the remaining testimony in favor of Yu to be unpersuasive. The district court determined that Yu did not show that (1) he was qualified to continue in the Program, or that (2) ISU engaged in “actions or inactions intended to discriminate against Yu on the basis of race or national origin.” The district court concluded that ISU was entitled to judgment in its favor. This appeal followed.

While this appeal was pending, three amicus briefs were filed in support of Yu, including a brief submitted collectively by The Equal Justice Society, Legal Aid at Work, The National Employment Lawyers Association, and Public Rights Project (“EJS”), contending that implicit bias may be probative or used as evidence of intentional discrimination under Title VI.

II. STANDARD OF REVIEW

“Following a bench trial, the judge’s findings of facts are reviewed for clear error.” Lentini v. Cal. Ctr. for the Arts, 370 F.3d 837, 843 (9th Cir. 2004) (citation omitted). Federal Rule of Civil Procedure 52(a), which governs our review, states: “Findings of fact . . . must not be set aside unless clearly erroneous, and the reviewing court must give due regard to the trial court’s opportunity to judge the witnesses’ credibility.” Fed. R. Civ. P. 52(a)(6). Under this “significantly deferential” standard, “we will accept the lower court’s findings of fact unless we are left with the definite and firm conviction that a mistake has been committed.” N. Queen Inc. v. Kinnear, 298 F.3d 1090, 1095 (9th Cir. 2002) (quoting Allen v. Iranon, 283 F.3d 1070, 1076 (9th Cir. 2002)). We will “reverse only if the district court’s findings are clearly erroneous to the point of being illogical, implausible, or without support in inferences from the record.” Oakland Bulk & Oversized Terminal, LLC v. City of Oakland, 960 F.3d 603, 613 (9th Cir. 2020). “The district court’s conclusions of law following a bench trial are reviewed de novo.” Lentini, 370 F.3d at 843 (citing Brown v. United States, 329 F.3d 664, 671 (9th Cir. 2003)).

The United States Supreme Court has held that “a finding of intentional discrimination is a finding of fact.” See Anderson v. Bessemer City, 470 U.S. 564, 573 (1985). However, “misunderstanding or applying an erroneous definition of intentional discrimination” may constitute legal error. See Pullman-Standard v. Swint, 456 U.S. 273, 287– 88 (1982).

Because the issue whether there was intentional discrimination is a factual one, the clear error standard is relevant to our appellate review. “[I]n ruling on a motion for summary judgment, ‘[t]he evidence of the nonmovant is to be believed, and all justifiable inferences are to be drawn in his favor.’” Tolan v. Cotton, 572 U.S. 650, 651 (2014) (per curiam) (citation omitted). But once there has been a trial, as in this case, the avenues for further review are narrowed.

YU V. IDAHO STATE UNIVERSITY 11

On appeal of a bench trial finding no intentional discrimination, we have said that “[t]he only issue before us . . . is whether we can definitely and firmly say that the trial court clearly erred in finding that [the appellant] failed to prove discrimination.” Casillas v. U.S. Navy, 735 F.2d 338, 343 (9th Cir. 1984). As the Supreme Court has expressed, “[t]his standard plainly does not entitle a reviewing court to reverse the finding of the trier of fact simply because it is convinced that it would have decided the case differently.” Anderson, 470 U.S. at 573.

III. DISCUSSION

A.

To prove Title VI discrimination, a plaintiff must demonstrate that he was “subjected to discrimination” due to “race, color, or national origin,” by a “program or activity receiving Federal financial assistance.” 42 U.S.C. § 2000d. Private parties seeking judicial enforcement of Title VI’s nondiscrimination protections must prove intentional discrimination. Alexander v. Sandoval, 532 U.S. 275, 280– 81 (2001). In other words, “Title VI . . . proscribe[s] only those racial classifications that would violate the Equal Protection Clause or the Fifth Amendment.” Grutter v. Bollinger, 539 U.S. 306, 343 (2003) (quoting Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 287 (1978) (opinion of Powell, J.)).

At the trial stage, “the ultimate factual issue in the case” is “whether the defendant intentionally discriminated against the plaintiff.” U.S. Postal Serv. Bd. of Governors v. Aikens, 460 U.S. 711, 715 (1983) (citation omitted). The finder of fact should “consider all the evidence,” id. at 714 n.3, and look to the “totality of the relevant facts” to determine whether the defendant has engaged in intentional discrimination, see Washington v. Davis, 426 U.S. 229, 242 (1976).

B.

On appeal, Yu and EJS challenge the following language employed by the district court: “Dr. Zorwick apparently would suggest that even the most egalitarian individuals, of whatever race, can be unaware of their unconscious bias (as the name describes) but still be intentionally racist. That simply makes no sense.” But, regardless of what the district court said regarding unconscious bias, it committed no legal error even under Yu’s view of the law because it admitted the testimony of Dr. Zorwick, deemed her qualified on the topic of aversive racism, and considered her testimony at great length in its decision. The district court offered alternative reasoning to give only limited weight to Dr. Zorwick’s testimony and to support the district court’s ultimate factual finding that ISU did not intentionally discriminate against Yu. Cf. Swint, 456 U.S. at 287 (noting that a district court’s findings of fact “may be set aside” if they “rest on an erroneous view of the law”) (emphasis added); see also Gonzales v. Police Dep’t, City of San Jose, 901 F.2d 758, 759 (9th Cir. 1990) (concluding that the combined effect of two “material legal errors” was “sufficiently serious to warrant a remand”).

After dedicating nearly 20 pages of its findings of fact to analyzing Dr. Zorwick’s testimony, the district court found that—contrary to Dr. Zorwick’s conclusion—Yu was not a victim of aversive racism. 3 The district court found that the

3 Yu and EJS both contend that the district court erred in failing to consider Dr. Zorwick’s testimony as probative on the issue of intentional discrimination. Although we have held that failure to consider highly

YU V. IDAHO STATE UNIVERSITY 13

circumstances of the instant case did not exhibit all the hallmarks of aversive racism that Dr. Zorwick set out in her testimony. According to the district court, Dr. Zorwick’s “testimony about supposed specific examples of [the] ‘hallmarks’ [of aversive racism] simply are not persuasive against the testimony of those who were involved in the events and within the context of a doctoral program in clinical psychology.” The district court was well within its discretion to disagree with Dr. Zorwick’s conclusions as they applied to the specific facts of this case. See Martin v. F.E. Moran, Inc., No. 13 C 3526, 2018 WL 1565597, at *29 (N.D. Ill. Mar. 30, 2018) (considering testimony on implicit bias at a bench trial but determining that “Plaintiffs failed to tie [the expert’s] opinions regarding implicit bias to the individual circumstances surrounding their employment . . . as required to prove intentional discrimination actually occurred”). The district court was simply not persuaded by Dr. Zorwick’s testimony, and that decision was not “illogical, implausible, or without support in inferences that may be drawn from the record.” United States v. Hinkson, 585 F.3d 1247, 1263 (9th Cir. 2009) (en banc).

In view of the “totality of the relevant facts,” Washington, 426 U.S. at 242, we hold that the district court did not clearly err in finding that Yu failed to show that ISU intentionally discriminated against him. In other words, the district court’s finding that ISU did not intentionally discriminate against Yu is “plausible in light of the record

probative evidence of intentional discrimination may constitute reversible legal error, we conclude that the district court did consider Dr. Zorwick’s testimony. See Gonzales, 901 F.2d at 761 (finding legal error where “there [wa]s no mention of the extensive testimony on the [defendant]’s violation of the San Jose Affirmative Action Plan” although such evidence was “highly relevant and probative”). viewed in its entirety.” Husain v. Olympic Airways, 316 F.3d 829, 835 (9th Cir. 2002), aff’d, 540 U.S. 644 (2004). Relying on the well-documented feedback from outside supervisors and from ISU faculty stating that Yu’s professional progress was unsatisfactory, Yu’s testimony and general affect in trial, 4 and the credibility of ISU faculty and supervisors who testified at trial, the district court found that “[t]he evidence presented in this case failed to persuade that Yu was dismissed for any reason other than his inability to gain, or to demonstrate, the degree of clinical competence expected of a fifth-year clinical psychology doctoral student.” When substantial testimony or other evidence supports the factual determination, as it does here with substantial evidence of supervisor criticism and adverse faculty evaluation, we cannot reverse the court’s judgment based on that factual determination in the absence of some supervening legal error or clear error. See Oakland Bulk, 960 F.3d at 613. On appeal, Yu points to no evidence in the record that would render this finding clearly erroneous.

In support of his discrimination claim, Yu emphasizes supervisors’ comments on his English fluency. He contends that such comments constitute direct evidence of linguistic discrimination, which is a proxy for unlawful national origin discrimination. See, e.g., In re Rodriguez, 487 F.3d 1001, 1008–09 (6th Cir. 2007). But the district court found that “[e]ach reference to [Yu’s] spoken English related in some manner, either expressly or implicitly, to his difficulties demonstrating clinical competence, including a difficulty in establishing rapport and forming therapeutic relationships with English-speaking clients.” Yu offers—and the record

4 The district court found that “Yu said nothing at trial to suggest he held himself accountable in any way for his well-documented poor clinical performance and numerous negative reviews.”

YU V. IDAHO STATE UNIVERSITY 15

reveals—no evidence or argument that leaves us with a “definite and firm conviction that a mistake has been committed.” See Easley v. Cromartie, 532 U.S. 234, 242 (2001) (quoting United States v. U.S. Gypsum Co., 333 U.S. 364, 395 (1948)). We stress that, in the field of clinical psychology, the ability to communicate effectively is imperative, as reflected in standard criteria under which Yu and his classmates were evaluated, including being sensitive to subtle patient signals, building rapport, and forming alliances with patients.

Yu also challenges the district court’s finding that he did not meet ISU’s legitimate educational expectations. Again, we see no reason to disturb the district court’s factual finding. Yu asserts that he would have successfully graduated had ISU (1) complied with ethical norms, 5 (2) complied with cultural competence standards, and (3) not engaged in aversive racism. But Yu only summarizes the evidence and arguments presented to the district court at trial, perhaps seeking an opportunity to re-try his case before this court on appeal. Given the highly deferential standard afforded to a district court’s factual findings at a bench trial,

5 Yu contends that ISU “substantial[l]y depart[ed] from accepted academic norms,” a standard employed in due process jurisprudence. See Regents of Univ. of Mich. v. Ewing, 474 U.S. 214, 225 (1985). The district court dismissed Yu’s due process claims on summary judgment, holding that the Eleventh Amendment precluded jurisdiction over them. The district court assumed, without deciding, that the standard was relevant and determined that ISU did not substantially depart from academic norms, and we discern no clear error in that determination. We thus decline to reach the legal question of whether the “substantially departed” due process standard is relevant to Title VI intentional discrimination. we decline to extend such an opportunity. See Anderson, 470 U.S. at 573.

C.

We decline to address whether implicit bias may be probative or used as evidence of intentional discrimination under Title VI because resolution of this issue is not necessary to the disposition of this appeal, and we see no benefit that would be served by commenting on it. 6

AFFIRMED.

MILLER, Circuit Judge, concurring:

This case turns on a factual question of a kind that is common in discrimination cases: Did Idaho State University dismiss Jun Yu from its doctoral program in clinical psychology because of his poor performance (as ISU maintained) or because of his race or national origin (as Yu alleged)? The district court held a four-day bench trial during which it examined documentary evidence and heard live testimony, including from ISU decision makers, Yu himself, and various expert witnesses. One of those experts was Dr. Leslie Wade Zorwick, a psychologist who opined that ISU’s decision making was affected by what she called “aversive racism”—that is, “unconscious” or “implicit” racial bias. After hearing all of the evidence, the district court found that ISU had not discriminated against Yu because of his race or national origin.

6 The Court nevertheless appreciates the work of all amici.

YU V. IDAHO STATE UNIVERSITY 17

The court affirms the district court’s judgment, and I join its opinion in full. I write separately to note several reasons why expert testimony of the kind offered in this case will rarely, if ever, be admissible.

First, to be admissible, expert testimony must be helpful to the trier of fact. Fed. R. Evid. 702(a). Because it is the role of the trier of fact to assess credibility, expert testimony is not admissible simply to tell the jury that one party’s witnesses should be believed and the other party’s should not. See, e.g., United States v. Dorsey, 45 F.3d 809, 815 (4th Cir. 1995) (“[E]xpert testimony can be properly excluded if it is introduced merely to cast doubt on the credibility of other eyewitnesses, since the evaluation of a witness’s credibility is a determination usually within the jury’s exclusive purview.”); see also United States v. Hill, 749 F.3d 1250, 1263 (10th Cir. 2014); United States v. Rincon, 28 F.3d 921, 926 (9th Cir. 1994).

That is essentially what Dr. Zorwick did here. She explained that because of modern “social norms” against “the explicit and overt expression of prejudice,” people may “know negative stereotypes and be profoundly influenced by those negative stereotypes they’ve learned about race” without admitting it. She added that people “tend to come up with an explanation for our motives that makes us look good,” and that “that’s a reason why . . . we really need to pay attention to behaviors to decide if those motives that are explicitly expressed are maybe a little bit suspect.” In other contexts, we have held that an expert’s observations on a psychological phenomenon explaining a class of behavior can be admitted to assist the jury. See, e.g., United States v. Halamek, 5 F.4th 1081, 1087–89 (9th Cir. 2021); United States v. Vallejo, 237 F.3d 1008, 1020 (9th Cir. 2001); United States v. Bighead, 128 F.3d 1329, 1330–31 (9th Cir. 1997) (per curiam); United States v. Antone, 981 F.2d 1059, 1062 (9th Cir. 1992). Testimony about aversive racism or unconscious bias could, in principle, serve a similar function—at least to the extent that it tells the jury something beyond the obvious, commonsense fact that people’s stated motives are not always their true motives. See United States v. Fuentes-Cariaga, 209 F.3d 1140, 1142 n.3 (9th Cir. 2000) (district court may exclude expert testimony “about an issue within the ken of the jury’s knowledge”).

But Dr. Zorwick did not stop there. She went on to opine that an observer can identify behavior that is influenced by unacknowledged stereotypes—or “aversive racism”—by looking for the “five characteristic hallmarks that tend to be present if aversive racism is at play”: (1) “the presence of a lot of ambiguity surrounding decision-making,” (2) “the use of race neutral explanations after the fact,” (3) “the expression[] of microaggressions,” (4) “really challenging interracial interactions and relationships,” and (5) “the use of post hoc justification.” Because “each of these hallmarks” was present in this case, Dr. Zorwick concluded that “Yu’s race and international status impacted the way he was treated by the faculty in the Idaho State University Clinical Psychology Program,” or, in other words, that the claims of ISU’s witnesses that they acted for race-neutral reasons should not be believed.

That testimony is similar to purported expert assessments of credibility that courts routinely exclude. For example, in Nichols v. American National Insurance Co., 154 F.3d 875 (8th Cir. 1998), the Eighth Circuit held that a district court erred in allowing a psychiatric expert to testify about the concepts of “psychiatric credibility, malingering, recall bias, and secondary gain” and to use those concepts “to indicate that [the plaintiff’s] version of the facts was

YU V. IDAHO STATE UNIVERSITY 19

inconsistent and changed over time and that it was tainted by bias and desire for financial gain.” Id. at 883–84. The court explained that those “were inferences for the jury to draw from the admissible evidence before it,” and that the expert had “impermissibly instructed the jury on how to weigh that evidence.” Id. at 884. Similarly, we have held that a district court erred in permitting a psychological expert to testify “that the complaining witnesses were able to distinguish reality from fantasy and truth from falsehood” because the effect of the testimony was “to usurp the jury’s fact-finding function.” United States v. Binder, 769 F.2d 595, 602 (9th Cir. 1985), overruled on other grounds by United States v. Morales, 108 F.3d 1031, 1035 n.1 (9th Cir. 1997) (en banc); accord Nimely v. City of New York, 414 F.3d 381, 398 (2d Cir. 2005) (holding that the district court erred in admitting testimony of an expert who “stated that he ‘rejected’ the possibility that [law-enforcement witnesses] had lied, and explained various reasons why police officers have no incentive to give false statements in excessive force cases”).

Second, Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), requires that before allowing scientific evidence to be presented to a jury, the district court must assess “the scientific validity . . . of the principles that underlie a proposed submission.” Id. at 594–95. “[A] key question to be answered in determining whether a theory or technique is scientific knowledge that will assist the trier of fact will be whether it can be (and has been) tested.” Id. at 593. “Additionally, in the case of a particular scientific technique, the court ordinarily should consider the known or potential rate of error.” Id. at 594. The same framework also applies in evaluating the reliability of “technical” and “other specialized knowledge.” Kumho Tire Co. v. Carmichael, 526 U.S. 137, 147 (1999) (quoting Fed. R. Evid. 702(a)). Dr. Zorwick’s claimed ability to identify aversive racism does not appear to rest on the kind of tested scientific principles that the Supreme Court has demanded. Dr. Zorwick’s only explanation of the basis for her methodology was the vague assertion that “[r]esearch finds that there are five characteristic hallmarks that tend to be present if aversive racism is at play.” She did not say what empirical testing had been conducted to demonstrate that applying those five factors allows for the accurate identification of aversive racism in a particular case. Dr. Zorwick noted that other scholars of aversive racism “sometimes use[] the language of race-based attitudes as being unconscious or implicit.” There is an extensive literature on implicit associations related to race, but studies of the phenomenon have yielded little evidence that such associations can be measured in a way that is useful for predicting biased behavior in individual cases. See, e.g., Frederick L. Oswald et al., Predicting Ethnic and Racial Discrimination: A Meta-Analysis of IAT Criterion Studies, 105 J. Personality & Soc. Psych. 171 (2013); Michael R. Andreychik & Michael J. Gill, Do Negative Implicit Associations Indicate Negative Attitudes?, 48 J. Experimental Soc. Psych. 1082 (2012). Indeed, even the leading advocates of implicit-association testing concede that “attempts to diagnostically use such measures for individuals risk undesirably high rates of erroneous classifications.” Anthony G. Greenwald et al., Statistically Small Effects of the Implicit Association Test Can Have Societally Large Effects, 108 J. Personality & Soc. Psych. 553, 557 (2015). But Dr. Zorwick did not say anything about what testing had shown the potential rate of error of her approach to be. Instead, she refused to acknowledge any possibility of error: When asked “how certain are you of your conclusion?,” she answered simply, “I am certain.”

YU V. IDAHO STATE UNIVERSITY 21

Third, the testimony offered here is objectionable because two of the factors set out by Dr. Zorwick in her five- factor test were “the use of race neutral explanations after the fact” and “the use of post hoc justification.” The district court correctly noted that those two factors are duplicative, but there is a more serious problem. Once a plaintiff has presented a prima facie case of discrimination, the defendant is required to present a race-neutral explanation for its action; if it fails to do so, it will effectively concede liability. McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802–03 (1973). It cannot be that a defendant’s effort to defend itself in litigation should be considered as evidence of its liability. Treating it as such would redefine the governing legal standard under the guise of an expert opinion.

Although Dr. Zorwick’s testimony on this point was not entirely clear, it is possible that she meant to refer only to pretextual—that is, false—explanations: “people pointing to things that, ‘Oh, that’s why we did it,’ but it’s not how they felt at the time.” But if that is what Dr. Zorwick meant, then her test is question-begging. Obviously, the fact that a defendant’s explanation is pretextual constitutes circumstantial evidence supporting an inference that the true reason was a discriminatory one. Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 147 (2000). The point of the trial was to determine whether ISU’s proffered race- neutral reason for Yu’s dismissal was or was not pretextual. An expert should not be permitted to opine on that question using a methodology that requires her to assume its answer. Stephens v. Union Pac. R.R. Co., 935 F.3d 852, 856–57 (9th Cir. 2019).

All that said, I do not fault the district court for admitting Dr. Zorwick’s testimony in this case. ISU did not object to the testimony, no doubt because “the Daubert gatekeeping obligation is less pressing in connection with a bench trial,” where the potential for prejudice or confusion is reduced. Volk v. United States, 57 F. Supp. 2d 888, 896 n.5 (N.D. Cal. 1999). And the district court correctly recognized that the testimony had minimal persuasive value. But when the opposing party objects, a court should not permit testimony of this kind to be presented to a jury.