Sophia BALOW; Ava Boutrous; Julia Coffman; Kylie Goit; Emma Inch; Sheridan Phalen; Madeline Reilly; Olivia Starzomski; Sarah Zofchak; Taylor Arnold; Elise Turke, individually and on behalf of all those similarly situated, Plaintiffs-Appellants, v. MICHIGAN STATE UNIVERSITY; Michigan State University Board of Trustees; Samuel L. Stanley, Jr.; Bill Beekman, Defendants-Appellees.
Michigan State University (MSU) eliminated both its men's and women's swimming-and-diving teams. Members of the women's swimming-and-diving team (“student-athletes”) sued, arguing that MSU fails to provide women athletes with equal participation opportunities as required by Title IX. The district court denied the student-athletes’ request for a preliminary injunction. We VACATE the district court's order and REMAND for further proceedings consistent with this opinion.
A. MSU's Elimination of Its Swimming-and-Diving Teams
Before the end of the 2019–20 academic year, MSU had the following Division I sports teams: men's baseball, basketball, cross country, football, golf, ice hockey, soccer, swimming and diving, tennis, track and field, and wrestling; and women's basketball, cross country, field hockey, golf, gymnastics, rowing, soccer, softball, swimming and diving, tennis, track and field, and volleyball. R. 8-2 (Breske Decl. at 14) (Page ID #362). On October 22, 2020, MSU announced it would no longer sponsor the men's and women's swimming-and-diving teams after the 2020–21 school year. R. 1 (Compl. ¶ 130) (Page ID #39). During the 2019–20 school year, the teams had 29 men and 33 women. R. 8-2 (Breske Decl. at 14) (Page ID #362).
Eleven women student-athletes sought a preliminary injunction to prevent MSU from eliminating the women's swimming-and-diving team. R. 1 (Compl.) (Page ID #1–55). They argued that MSU failed to provide women with substantially proportionate athletic opportunities, as required by Title IX. The student-athletes and MSU agree on the gender breakdown of the undergraduate student body as a whole: in the 2018–19 school year, 48.8% of undergraduate students were male and 51.2% were female; and, in the 2019–20 school year, 49.1% were male and 50.9% were female. R. 2-14 (Lopiano Rep. at 20) (Page ID #217); R. 8-2 (Breske Decl. at 10, 14) (Page ID #358, 362). The parties disagree, however, about the number of male and female athletes at MSU.
The district court denied the student-athletes’ motion for a preliminary injunction, finding that they were not likely to succeed on the merits of their Title IX claim. The student-athletes timely appealed. R. 18 (Notice of Appeal) (Page ID #757).
B. Statutory and Regulatory Background
Title IX provides that “[n]o person in the United States shall, on the basis of sex, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any education program or activity receiving Federal financial assistance.” 20 U.S.C. § 1681(a). Agencies are “authorized and directed to effectuate the provisions of section 1681 ․ by issuing rules, regulations, or orders of general applicability.” Id. § 1682.
Regulations promulgated pursuant to Title IX extend its protections to athletics, 34 C.F.R. § 106.41(a); see also 45 C.F.R. § 86.41(a), and require that recipients “shall provide equal athletic opportunity for members of both sexes,” 34 C.F.R. § 106.41(c). The factors that determine whether equal opportunities are available include “[w]hether the selection of sports and levels of competition effectively accommodate the interests and abilities of members of both sexes.” 34 C.F.R. § 106.41(c)(1).
In 1979, the Secretary of the Department of Health, Education, and Welfare (HEW) 1 issued, after notice and comment, a Policy Interpretation that “clarifie[d] the meaning of ‘equal opportunity’ in intercollegiate athletics.” 44 Fed. Reg. 71,413, 71,414 (Dec. 11, 1979). This document established a three-part test to assess compliance:
(1) Whether intercollegiate level participation opportunities for male and female students are provided in numbers substantially proportionate to their respective enrollments; or
(2) Where the members of one sex have been and are underrepresented among intercollegiate athletes, whether the institution can show a history and continuing practice of program expansion which is demonstrably responsive to the developing interest and abilities of the members of that sex; or
(3) Where the members of one sex are underrepresented among intercollegiate athletes, and the institution cannot show a continuing practice of program expansion such as that cited above, whether it can be demonstrated that the interests and abilities of the members of that sex have been fully and effectively accommodated by the present program.
Id. at 71,418.
Only the first prong of this test is at issue. It defines participants as athletes:
a. Who are receiving the institutionally-sponsored support normally provided to athletes competing at the institution involved, e.g., coaching, equipment, medical and training room services, on a regular basis during a sport's season; and
b. Who are participating in organized practice sessions and other team meetings and activities on a regular basis during a sport's season; and
c. Who are listed on the eligibility or squad lists maintained for each sport, or
d. Who, because of injury, cannot meet a, b, or c above but continue to receive financial aid on the basis of athletic ability.
Id. at 71,415.
In 1996, the Department of Education issued a “Dear College” letter to clarify this three-part test. In addition to “confirm[ing] that institutions need to comply only with any one part of the three-part test in order to provide nondiscriminatory participation opportunities,” this letter clarified each of the test's three prongs. Office for Civil Rights, U.S. Dep't of Educ., Clarification of Intercollegiate Athletics Policy Guidance: The Three-Part Test (Jan. 16, 1996), https://www2.ed.gov/about/offices/list/ocr/docs/clarific.html (“1996 Letter”). It explained that substantial, not exact, proportionality is required “because in some circumstances it may be unreasonable to expect an institution to achieve exact proportionality—for instance, because of natural fluctuations in enrollment and participation rates or because it would be unreasonable to expect an institution to add athletic opportunities in light of the small number of students that would have to be accommodated to achieve exact proportionality.” Id. Substantial proportionality is determined “on a case-by-case basis, rather than through use of a statistical test.” Id.
The 1996 Letter further clarified:
OCR would also consider opportunities to be substantially proportionate when the number of opportunities that would be required to achieve proportionality would not be sufficient to sustain a viable team, i.e., a team for which there is a sufficient number of interested and able students and enough available competition to sustain an intercollegiate team. As a frame of reference in assessing this situation, OCR may consider the average size of teams offered for the underrepresented sex, a number which would vary by institution.
For instance, Institution A is a university with a total of 600 athletes. While women make up 52 percent of the university's enrollment, they only represent 47 percent of its athletes. If the university provided women with 52 percent of athletic opportunities, approximately 62 additional women would be able to participate. Because this is a significant number of unaccommodated women, it is likely that a viable sport could be added. If so, Institution A has not met part one.
As another example, at Institution B women also make up 52 percent of the university's enrollment and represent 47 percent of Institution B's athletes. Institution B's athletic program consists of only 60 participants. If the University provided women with 52 percent of athletic opportunities, approximately 6 additional women would be able to participate. Since 6 participants are unlikely to support a viable team, Institution B would meet part one.
“In reviewing a district court's decision to deny a preliminary injunction, we evaluate the same four factors that the district court does: ‘(1) whether the movant has a strong likelihood of success on the merits; (2) whether the movant would suffer irreparable injury without the injunction; (3) whether issuance of the injunction would cause substantial harm to others; and (4) whether the public interest would be served by issuance of the injunction.’ ” Kentucky v. U.S. ex rel. Hagel, 759 F.3d 588, 600 (6th Cir. 2014) (quoting City of Pontiac Retired Emps. Ass'n v. Schimmel, 751 F.3d 427, 430 (6th Cir. 2014) (en banc) (per curiam)). “These factors are to be balanced against one another and should not be considered prerequisites to the grant of a preliminary injunction.” Leary v. Daeschner, 228 F.3d 729, 736 (6th Cir. 2000).
“This court reviews the district court's decision for an abuse of discretion.” Id. We review the district court's factual findings for clear error, but “[i]f pure legal conclusions are involved in the district court's determination ․, those conclusions are subject to de novo review.” Id. at 736–37.
A. Genuine Participation Opportunities
When determining whether participation opportunities are substantially proportionate to enrollment, “OCR's analysis begins with a determination of the number of participation opportunities afforded to male and female athletes in the intercollegiate athletic program.” 1996 Letter. The student-athletes argue that MSU inflated its number of female athletes by failing to accord certain female athletes genuine participation opportunities, both on the rowing team and on the track-and-field and cross-country teams. The district court rejected this argument. Balow v. Mich. State Univ., No. 1:21-cv-44, 2021 WL 650712, at *6–7 (W.D. Mich. Feb. 19, 2021). We address each sport in turn.
First, the district court did not clearly err in finding that MSU did not inflate its count of women rowers by including “novice” rowers in its count of athletes and having a larger-than-average team. Although the number of novice rowers contributes to the large size of the women's rowing team, novice rowing is “an integral part of the sport,” and Big Ten meets include events that are reserved for only novice rowers. R. 8-3 (Chavers Decl. ¶¶ 3–11) (Page ID #367–69). The rowing coach submitted a declaration stating that novice rowers “are full-fledged members of the MSU rowing team” who “receive the same practice gear and competition gear and participate in the same training and conditioning activities as the rest of the team.” Id. ¶ 5 (Page ID #368). Thus, novice rowers meet the regulatory definition of participant. See 44 Fed. Reg. at 71,415.
Second, the district court did not clearly err in finding that MSU did not inflate the number of women's track-and-field and cross-country athletes. Although some athletes did not participate in any races, Title IX does not require that athletes participate in competitions to be counted. See 44 Fed. Reg. at 71,415; 1996 Letter (“In determining participation opportunities, OCR includes ․ those athletes who practice but may not compete.”); see also Biediger v. Quinnipiac Univ. (“Biediger III”), 691 F.3d 85, 93 (2d Cir. 2012) (“It is not necessary for an athlete to meet minimum criteria of playing time ․ to count as a participant.”); Anders v. Cal. State Univ., Fresno, No. 1:21-cv-179-AWI-BAM, 2021 WL 1564448, at *12 (E.D. Cal. Apr. 21, 2021) (“ ‘[B]ench warming’ is a fact of life in most sports.”).
Against these conclusions, the student-athletes point to cases in which athletes were not accorded genuine participation opportunities. But those cases are different from the current one. In this case, the university did not pressure teams to have larger or smaller rosters than the coach would prefer, see Biediger v. Quinnipiac Univ. (“Biediger I”), 616 F. Supp. 2d 277, 283–84 (D. Conn. 2009), nor are women's teams larger than average while men's teams are smaller than average, see Portz v. St. Cloud State Univ. (“Portz II”), 401 F. Supp. 3d 834, 863 (D. Minn. 2019). At MSU, the coach determines the size of the team based on “interest and the Big Ten's competition requirements.” R. 8-3 (Chavers Decl. ¶ 12) (Page ID #369). A coach's preference for a larger team does not mean that team members lack genuine participation opportunities.
Ultimately, the district court did not clearly err in finding that MSU did not inflate its number of women athletes.
B. Calculating the Participation Gap
After determining the number of participants, the district court considered the participation gap as a percentage of the size of the athletic program. 2021 WL 650712, at *11. This was improper. The correct inquiry focuses on the number of participation opportunities, not the gap as a percentage of the athletic program.
The text of the 1979 Policy Interpretation and the 1996 Letter prove this point. The language of the 1979 Policy Interpretation is clear: schools must provide participation opportunities for males and females “in numbers substantially proportionate to their respective enrollments.” 44 Fed. Reg. at 71,418 (emphasis added). The Dear College Letter likewise focused on the number, not percentage, of participation opportunities. See 1996 Letter.
The district court, however, justified its consideration of percentages based on language from the 1996 letter that provided that “this determination depends on the institution's specific circumstances and the size of its athletic program.” 1996 Letter. The district court reasoned that “[i]f the size of an athletic program is relevant, then the size of the participation gap in relation to the size of the athletic program should also be relevant.” 2021 WL 650712, at *11.
This logic ignores the clear text of the 1979 Policy interpretation and misinterprets the reasoning of the 1996 Letter. The 1979 Policy Interpretation never refers to percentages and discusses only the “numbers” of participation opportunities provided. 44 Fed. Reg. 71,413, 71,418 (Dec. 11, 1979). Although the 1996 Letter refers to both numbers and percentages, its central focus is on the numbers. It asks whether a school provides “participation opportunities for male and female students in numbers substantially proportionate to their respective full-time undergraduate enrollments,” calculates compliance based on “the number of participation opportunities,” and notes that opportunities are substantially proportionate “when the number of opportunities that would be required to achieve proportionality would not be sufficient to sustain a viable team.” 1996 Letter (emphasis added).
Importantly, the 1996 Letter never discussed the participation gap as a percentage. Although it refers to percentages in other contexts, it uses only numbers to refer to the participation gap. Percentages are helpful in comparing the gender ratio of the athletic program to the gender ratio of the undergraduate body. They are not, however, the correct tool for measuring the participation gap.
Although a few examples in the Letter speak in terms of percentages, none of these examples contemplates calculating the participation gap as a percentage. The dissent appears to rely on two examples that it claims “illustrat[e] the participation gap as a percentage.” Dissenting Op. at 1065. The first involves an institution with an enrollment that is 52% male and 48% female, in which 52% of athletes are male and 48% are female. If the enrollment shifts to 51% male and 49% female, the school need not “fine tune its program.” 1996 Letter. The second involves an institution that had a consistent enrollment rate of 50% for women, which spiked to 52% in a certain year. Neither example illustrates how to calculate the participation gap. They stand only for the principle that fluctuations in enrollment will not force a school out of compliance. Comparing these examples with the two that immediately follow shows that they do not support the claim that the participation gap is measured as a percentage. The next two examples involve schools of various sizes in which women make up 52% of the university's enrollment but only 47% of the institution's athletes. Unlike the prior examples, these examples offer instruction on how to calculate a participation gap, and they calculate it as a number. Id. They show that, although percentages are relevant, the ultimate focus should be on the numerical participation gap. 2
The district court further implies that participation gaps that are lower than two percent satisfy substantial proportionality. This bright line is inconsistent with the 1996 Letter. Substantial proportionality “depends on the institution's specific circumstances and the size of its athletic program” and is determined “on a case-by-case basis.” 1996 Letter; see, e.g., Lazor v. Univ. of Connecticut, ––– F. Supp. 3d ––––, ––––, 2021 WL 2138832, at *6 (D. Conn. May 26, 2021) (finding “the defense that a participation gap percentage of less than 2% satisfies the test for substantial proportionality” to be “unpersuasive”); Robb v. Lock Haven Univ., No. 4:17-CV-00964, 2019 WL 2005636, at *8 (M.D. Pa. May 7, 2019) (“While [a 3.35% gap] could be termed a ‘borderline case’ in terms of raw statistics, a glance at Lock Haven's long history of Prong One nonsatisfaction reveals that gap cannot be attributed to natural fluctuations in the student body, and the number of lost opportunities that gap represents—36—is not too small to support a new varsity team.” (footnotes omitted)); see also Equity in Athletics, Inc. v. Dep't of Educ., 639 F.3d 91, 110 (4th Cir. 2011) (“DOE has not specified a magic number at which substantial proportionality is achieved.”). “[W]e do not, in any event, understand the 1996 Clarification to create a statistical safe harbor at [two percent] or any other percentage.” Biediger III, 691 F.3d at 106. 3
While the percentage gap may be relevant, substantial proportionality should be determined by looking at the gap in numerical terms, not as a percentage. A school may fail to achieve substantial proportionality even if its participation gap is only a small percentage of the size of its athletic program.
C. The Participation Gap at MSU
The district court did not make any finding as to the size of the participation gap.4 2021 WL 650712, at *6, 10–11. MSU used internal Title IX data to calculate a participation gap of 12 before the elimination of the swimming-and-diving teams and 15 after the elimination of these teams. R. 8-8 (O'Brien Rep. at 28–29) (Page ID #443–44). The student-athletes relied on data reported pursuant to the Equity in Athletics Disclosure Act (EADA) and web-roster data to calculate a participation gap of 25 in 2018–19 and 35 in 2019–20. R. 2-14 (Lopiano Report at 20, 35) (Page ID #217, 232).
Title IX counts participants differently than EADA. Compare 44 Fed. Reg. at 71,415 with U.S. Dep't of Educ., Office of Postsecondary Educ., User's Guide for the Equity in Athletics Disclosure Act Web-Based Data Collection (2019), at 31, https://surveys.ope.ed.gov/athletics2k20/wwwroot/documents/2019_EADA_Users_Guide. pdf (“EADA User's Guide”). For Title IX purposes, athletes include those who (a) “receiv[e] the institutionally-sponsored support normally provided to athletes competing at the institution involved”; (b) “are participating in organized practice sessions and other team meetings and activities on a regular basis”; and (c) “are listed on the eligibility or squad lists maintained for each sport”; or (d) “because of injury, cannot meet a, b, or c above but continue to receive financial aid on the basis of athletic ability.” 44 Fed. Reg. at 71,415. For EADA purposes, participants are students who, as of the day of a varsity team's first scheduled contest “[a]re listed by the institution on the varsity team's roster”; “[r]eceive athletically related student aid”; or “[p]ractice with the varsity team and receive coaching from one or more varsity coaches.” EADA User's Guide.
Nevertheless, at the preliminary-injunction stage, it may be appropriate to rely on EADA data to calculate the size of the participation gap. See Univ. of Texas v. Camenisch, 451 U.S. 390, 395, 101 S.Ct. 1830, 68 L.Ed.2d 175 (1981) (“[A] preliminary injunction is customarily granted on the basis of procedures that are less formal and evidence that is less complete than in a trial on the merits.”). In the types of cases at issue, schools, not plaintiffs, are the only parties who have access to the underlying Title IX data. In Ohlensehlen v. University of Iowa, 509 F. Supp. 3d 1085, 1098 (S.D. Iowa 2020), the court credited EADA data in light of the university's refusal to disclose the raw data underlying its Title IX figures. The court explained that when “[d]efendants declined to produce the NCAA squad lists, time and hour limitation records, and competition results that make up the raw data for official Title IX counts that they say supports their figures, despite Plaintiffs’—and the Court's—requests for them to do so,” defendants’ position that the court should consider official Title IX, not EADA or web roster, data “is especially disingenuous.” 5 Id. at 1098, 1101; see also Biediger I, 616 F. Supp. 2d at 297 (relying on EADA data to support a preliminary injunction).
Although, at the preliminary-injunction stage, it may be possible to show a strong likelihood of success on the merits based on EADA data, as litigation progresses, the appropriate inquiry turns on Title IX data, which counts participation precisely for this purpose.
D. Substantial Proportionality
The district court found that both the student-athletes’ calculation of the participation gap and MSU's calculation of the participation gap meet the substantial-proportionality threshold because they are smaller than the average-size team at MSU. 2021 WL 650712, at *10. The district court erred when it compared the participation gap to the size of the average team at MSU, rather than the size of a viable team.
The language of the 1996 clarification is clear. Opportunities are substantially proportionate:
when the number of opportunities that would be required to achieve proportionality would not be sufficient to sustain a viable team, i.e., a team for which there is a sufficient number of interested and able students and enough available competition to sustain an intercollegiate team. As a frame of reference in assessing this situation, OCR may consider the average size of teams offered for the underrepresented sex, a number which would vary by institution.
1996 Letter. The text of the Letter provides a clear answer about how to define a viable team: it uses “i.e.,” to define a viable team as “a team for which there is a sufficient number of interested and able students and enough available competition to sustain an intercollegiate team.” Id.
It is true that the Letter states that the average size of team may be used “[a]s a frame of reference.” 1996 Letter. Yet, this language presents a clear contrast with the language in the previous sentence: “i.e.” defines viable to mean that there is sufficient interest, ability, and competition for a team, but the “average size of teams” is only “a frame of reference” in making this determination. 6 The Letter provides “no indication that, as long as the participation gap is less than the university's average women's team size, the university meets prong one and complies with Title IX.” Lazor, ––– F. Supp. 3d at ––––, 2021 WL 2138832, at *4.
This interpretation is buoyed by language elsewhere in the Letter. The Letter emphasizes that there are no “strict numerical formulas or ‘cookie cutter’ answers.” 1996 Letter. An interpretation that conflates “viable team” with “average team” creates a strict numerical formula. 7 The language about the lack of strict numerical formulas makes sense only when qualitative factors, such as interest and ability, impact the definition of a “viable team.” This also comports with another purpose of the Letter: the Letter consistently focuses on whether a school accommodates the interests and abilities of the underrepresented sex.
Based on the clear language of the guidance, a viable team is not an average one, but is instead one “for which there is a sufficient number of interested and able students and enough available competition to sustain an intercollegiate team.” 1996 Letter.
If, on remand, the district court determines that MSU is not in compliance with Title IX, there is a question of what the appropriate remedy should be. At the preliminary injunction stage, the appropriate remedy when a school seeks to eliminate a women's team in violation of Title IX is typically an injunction that prevents them from doing so. See, e.g., Biediger v. Quinnipiac Univ. (“Biediger II”), 728 F. Supp. 2d 62, 113–14 (D. Conn. 2010); Cohen v Brown Univ., 809 F. Supp. 978, 981, 1001 (D.R.I. 1992), aff'd 991 F.2d 888 (1st Cir. 1993) (granting preliminary injunction after school demoted two men's and two women's varsity teams, affecting between 34 and 37 men and between 22 and 23 women). This is true even though, as a more permanent matter, a school may be “entitled to determine its own method for achieving statutory compliance.” Biediger II, 728 F. Supp. 2d at 113–14 (granting preliminary injunction after school announced it would cut two men's teams and one women's team); see Cohen, 991 F.2d at 906 (“[R]equiring Brown to maintain the women's volleyball and gymnastics teams in varsity status for the time being is a remedial choice within the district court's discretion” but “[t]hat is not to say ․ that the same remedy will be suitable at trial's end if the Title IX charges prove out against Brown.”). In this case, whether a preliminary injunction is appropriate depends on both the district court's finding of the size of the participation gap and its weighing of the preliminary-injunction factors. 8 This issue should be decided in the first instance by the district court, with the benefit of our clarification on how to determine substantial proportionality.
We VACATE the district court's order and REMAND for further proceedings consistent with this opinion.
Due to the athletic department's projected budget deficit of “$35–40 million” and “major upgrades and repairs” needed for the swimming and diving facilities, MSU announced in October 2020 that it would “discontinue the men's and women's swimming and diving team[s] after the conclusion of the 2020-21 season” (affecting 29 men and 33 women). (R. 8-6, ¶ 5 (emphasis added); R. 8-7; R. 8-2, PgID 353, 365). Although that is a neutral decision, members of the women's team sued MSU and contemporaneously sought a preliminary injunction to require that MSU continue only “its women's varsity swimming and diving team.” (R. 2, PgID 57-58 (emphasis added); R. 1, PgID 54).
The decision to eliminate the teams resulted in a female participation gap of 15 or 0.87%—as shown in the detailed spreadsheets maintained (and provided to plaintiffs) by MSU's Title IX compliance officer. (R. 8-2, PgID 353, 362; R. 8-8, PgID 443-44). Considering MSU's average team size for females (35) and the participation gap as both a number and a percentage, the district court concluded that plaintiffs “have not shown a substantial likelihood of success” on the merits because “[b]ased on MSU's numbers,” “MSU's evidence indicates its participation numbers are substantially proportionate.” Balow v. Michigan State Univ., No. 1:21-cv-44, 2021 WL 650712, at *9, *11-12 (W.D. Mich. Feb. 19, 2021). After analyzing and balancing all four preliminary injunction factors, the district court denied plaintiffs’ requested injunction.
Yet the majority finds fault in the district court's decision and remands. In doing so, the majority announces legal standards that no other federal circuit court has adopted—and for good reason—because the standards blatantly contradict Title IX and agency guidance. Today's decision now means: (1) courts may rely on EADA data to grant a preliminary injunction, even though the EADA does not count “participants” in the same way as Title IX; (2) courts cannot consider the participation gap as a percentage; (3) courts cannot consider a school in compliance when the participation gap is less than the average size of the school's teams for the underrepresented sex; and (4) courts may grant an injunction and require a school to reinstate a particular sports team pending a final judgment perhaps years in the future. Any short-lived victory plaintiffs may have won today will hamstring schools and come full circle to harm all athletes in the future. “After all, in the law,” there must be evenhandedness, for “what is sauce for the goose is normally sauce for the gander.” Heffernan v. City of Paterson, 578 U.S. 266, 136 S. Ct. 1412, 1418, 194 L.Ed.2d 508 (2016). I would affirm the district court's decision to deny plaintiffs’ motion for a preliminary injunction.
All involved agree that the general legal framework is outlined in a 1979 Policy Interpretation 1 and a 1996 Letter, 2 interpreting Title IX's implementing regulations. (Maj. Op. 1053–55; Appellant Br. 14-18; Appellee Br. 23-27; Balow, 2021 WL 650712, at *2-3). As to plaintiffs’ likelihood of success on the merits, the question here is: “Whether intercollegiate level participation opportunities for male and female students are provided in numbers substantially proportionate to their respective enrollments.” 44 Fed. Reg. at 71,418 (emphasis added). “[E]xact proportionality” is not the test. (1996 Letter, PgID 489). Plaintiffs have the burden to show a sufficient “statistical disparity.” Horner ex rel. Horner v. Kentucky High Sch. Athletic Ass'n, 206 F.3d 685, 695-96 (6th Cir. 2000) (citation omitted).
Plaintiffs have not met their burden. The parties offer the following female participation gaps at MSU. 3
According to MSU (based on Title IX data):
• In 2018–2019, MSU's participation gap was 27 or 1.4%, and the average size of MSU's female teams was 35.
• In 2019–2020, the participation gap was 12 or 0.65%, and the average size of MSU's female teams was again 35. After deducting the 29 men and 33 women displaced by MSU's decision to eliminate the swimming and diving teams—and assuming all else remains the same—the participation gap is 15 or 0.87%.
(R. 8-2, PgID 353, 357, 362; R. 8-8, PgID 443-44; Appellee Br. 28, 35).
According to plaintiffs’ expert (Lopiano):
• In 2018–2019, the participation gap was 25 or 1.3% (based on EADA data).
• In 2019–2020, the participation gap was 35 or 2% (based on website rosters). Plaintiffs do not contest that the average size of MSU's female teams was 35. (Appellant Br. 48-52).
(Appellant Br. 28-29; R. 2-14, PgID 217, 232).
The district court identified those figures, as does the majority. See Balow, 2021 WL 650712, at *6, *8-9; (Maj. Op. 1059).
The district court concluded—and the majority also acknowledges—that plaintiffs’ numbers are based “upon imperfect data, flawed assumptions, contradictory reasoning, and a skewed analysis.” Balow, 2021 WL 650712, at *9; (Maj. Op. 1056 – 1059–60, –––– – ––––). Indeed, the majority holds that “the district court did not clearly err in finding that MSU did not inflate its number of women athletes.” (Maj. Op. 1056–57); Balow, 2021 WL 650712, at *6-7. The majority also agrees with the district court that “Title IX counts participants differently than [the] EADA.” (Maj. Op. 1059); Balow, 2021 WL 650712, at *6; compare 44 Fed. Reg. at 71,415 (Title IX), with 34 C.F.R. § 668.47(b)(3) (EADA). Even plaintiffs’ expert admits that “EADA reports overcount female participation compared to Title IX participation counts because they use different metrics.” (R. 2-14, PgID 215 (emphasis added)). Yet the majority orders a remand.
There are five fundamental problems with the majority's reasoning.
Despite recognizing that Title IX counts participants differently than the EADA, the majority crafts a new rule by concluding that “at the preliminary-injunction stage, it may be possible to show a strong likelihood of success on the merits based on EADA data.” (Maj. Op. 1060). No federal appellate court has adopted such a rule, nor is it permissible to do so. Courts cannot simply say that a legal standard may “change its stripes” in the early stages of litigation. Comcast Corp. v. Nat'l Ass'n of African American-Owned Media, ––– U.S. ––––, 140 S. Ct. 1009, 1014, 206 L.Ed.2d 356 (2020). There is nothing in Title IX or its regulations to signal that Title IX's test for what counts as a participant “might be overlooked or modified in the early stages of a case.” See id. at 1016. MSU's Title IX compliance officer provided plaintiffs with the detailed Title IX spreadsheets maintained over the relevant years. (R. 8-2, PgID 353, 362; R. 8-8, PgID 443-44). Because this court has no authority to bend the definition of athletic “participant” under Title IX at any stage in a case, plaintiffs cannot substitute EADA data and website rosters for Title IX data. Thus, as the district court concluded, we are left with MSU's calculations—the only calculations that use Title IX participation counts.
The majority cites one district court case that credited EADA data because the defendant school “declined to produce” the underlying “raw data for official Title IX counts” that it offered, “despite Plaintiffs’—and the Court's—requests for [the school] to do so.” Ohlensehlen v. Univ. of Iowa, 509 F. Supp. 3d 1085, 1098, 1101 (S.D. Iowa 2020); (Maj. Op. 1059–60). But as the majority concedes, “this case is different from Ohlensehlen in two respects”: (1) the plaintiffs here moved for an injunction and never sought expedited discovery or filed a motion to compel MSU to produce its underlying raw data, see Fed. R. Civ. P. 26(d) and 37(a)(1)-(3); and (2) the district court never ordered MSU to disclose its underlying data. (Maj. Op. 1060 n.5). 4 The majority's notion that Title IX plaintiffs may rely on EADA data early in the litigation exceeds judicial authority and, at best, is dicta.
The majority erroneously represents that the district court “did not make any finding as to the size of the participation gap.” (Maj. Op. 1059). On the contrary, the district court rejected the logic and substitute data plaintiffs’ expert used to calculate a greater participation gap. Balow, 2021 WL 650712, at *6-8. As stated, the majority agrees with that much. (Maj. Op. 1056–57 – 1059–60, –––– – ––––). The district court then assessed proportionality “[b]ased on MSU's numbers.” 2021 WL 650712 at *9. “[U]sing one of the OCR's stated criteria for proportionality,” the court concluded that because MSU's participation gap “numbers are smaller than the average size of a women's team at MSU, which is 35 athletes,” “MSU's evidence indicates its participation numbers are substantially proportionate.” Id. (emphasis added); see also id. at *11 (finding that “MSU's participation gap appears to be lower than 2%”—the gap calculated by plaintiffs’ expert, Lopiano). The court underscored its reasoning when it explained:
Plaintiffs offer a much higher calculation of the participation gap, but as discussed above, their calculation depends upon imperfect data, flawed assumptions, contradictory reasoning, and a skewed analysis. Thus, Plaintiffs have provided a shaky foundation on which to argue that MSU's participation gap is statistically significant. The cracks in that foundation become even more apparent after considering the evidence offered by MSU.
In short, on the present record, the Court is not persuaded that MSU has improperly inflated its participation opportunities for women, or that Plaintiffs have shown a likelihood of success on their claim to the extent it requires them to demonstrate such inflation.
Id. at *9-10. It is clear that the district court adopted MSU's calculation of the participation gap. The district court was not required to incant magic words to say so. When the court went on later to say that “[f]urthermore,” MSU would likely satisfy substantial proportionality with a gap of 12, 25, or 35, the court was merely offering alternative reasoning. Id. at *10. There is no need to remand for the court to spell it out more clearly.
In assessing “substantial proportionality,” the majority concludes the district court erred when it “considered the participation gap as a percentage.” (Maj. Op. 1057). But courts may consider the participation gap as a percentage or a number.
First, under Title IX, Congress authorized courts to consider “statistical evidence of an imbalance” in terms of a “number or percentage.” 20 U.S.C. § 1681(b); see also Horner, 206 F.3d at 697.
Second, the 1996 Letter also expressly includes both ways of looking at the participation gap: (1) the 1996 Letter considers whether the gap in terms of a number could “sustain a viable team,” and the letter gives two examples; and (2) the 1996 Letter also explains that “the  Policy Interpretation examines whether participation opportunities are ‘substantially’ proportionate to enrollment rates,” and the letter gives two examples illustrating the participation gap as a percentage. (1996 Letter, PgID 489-90 (participation gap of 1% and 2% “would satisfy” proportionality)). The majority, however, adopts a bright-line rule that courts may only consider “the gap in numerical terms, not as a percentage.” (Maj. Op. 1056–57, ––––). But we cannot “cherry pick” from the 1996 Letter. See Cyan, Inc. v. Beaver Cty. Emps. Ret. Fund, ––– U.S. ––––, 138 S. Ct. 1061, 1070, 200 L.Ed.2d 332 (2018). “Our license to interpret [agency guidance] does not include the power to engage in such freewheeling judicial policymaking.” See Pereida v. Wilkinson, ––– U.S. ––––, 141 S. Ct. 754, 766-67, 209 L.Ed.2d 47 (2021); Kisor v. Wilkie, ––– U.S. ––––, 139 S. Ct. 2400, 2413, 204 L.Ed.2d 841 (2019).
Third, all but one federal appellate court to have considered the matter has viewed the participation gap as a percentage. See, e.g., Equity in Athletics, Inc. v. Dep't of Educ., 639 F.3d 91, 97, 110 (4th Cir. 2011) (affirming summary judgment for university because plaintiff “provide[d] no support for its contention that a disparity as low as 2% [for men] (and, according to the record, not much above 1%) is substantially disproportionate as a matter of law”). 5
Here, the district court evaluated the participation gap in terms of both a percentage and a number. Balow, 2021 WL 650712, at *9 (concluding that “MSU's evidence indicates its participation numbers are substantially proportionate” because MSU's participation gap “numbers are smaller than the average size of a women's team at MSU, which is 35 athletes”); id. at *11 (concluding that “MSU's participation gap appears to be lower than 2%,” and plaintiffs “have not cited, and the Court is not aware, of any case where a gap lower than 2% failed to satisfy the test for substantial proportionality”). The district court did not err. 6
To be sure, the 1996 Letter explains that compliance “depends on the institution's specific circumstances and the size of its athletic program.” (1996 Letter, PgID 489); accord Balow, 2021 WL 650712, at *11. In that vein, the district court reasoned:
Schools with larger athletic programs are likely to see larger fluctuations in participation numbers from year to year. Ignoring the size of the participation gap in relation to the size of the athletics program would significantly hinder the ability of schools with larger programs to maintain compliance. They would be more likely to fall outside the safe harbor due to “natural fluctuations in enrollment and participation rates” that may be somewhat large in absolute numbers but are relatively small in relation to the size of their programs.
Balow, 2021 WL 650712, at *11 (quoting 1996 Letter, PgID 489). That is sound logic grounded in the 1996 Letter and the 1979 Policy Interpretation.
The district court rejected plaintiffs’ numbers and accepted MSU's, which showed a participation gap of 0.87% or 15 after the elimination of the men's and women's swimming and diving teams. No court has gone so far as to enjoin a school for such a minimal disparity. Plaintiffs have not shown a slight, much less substantial, likelihood of success on the merits.
The majority concludes that “[t]he district court erred when it compared the participation gap” (as a number) “to the size of the average team” for females at MSU. (Maj. Op. 1060). But that conclusion conflicts with the 1996 Letter and how the OCR has applied it. After explaining proportionality using a percentage, the 1996 Letter explicitly states:
OCR would also consider opportunities to be substantially proportionate when the number of opportunities that would be required to achieve proportionality would not be sufficient to sustain a viable team, i.e., a team for which there is a sufficient number of interested and able students and enough available competition to sustain an intercollegiate team. As a frame of reference in assessing this situation, OCR may consider the average size of teams offered for the underrepresented sex, a number which would vary by institution.
(1996 Letter, PgID 490 (emphasis added)). Yet the majority reasons that “a viable team is not an average one, but is instead one ‘for which there is a sufficient number of interested and able students and enough available competition to sustain an intercollegiate team.’ ” (Maj. Op. 1061 (citation omitted)). Again, however, we cannot “cherry pick” from the 1996 Letter. See Cyan, 138 S. Ct. at 1070. The majority's rule stands alone among federal circuit courts, and the majority cites one district court case to support its selective interpretation. (Maj. Op. 1060–61 – ––––) (citing Lazor v. Univ. of Connecticut, ––– F. Supp. 3d ––––, ––––, 2021 WL 2138832, at *4 (D. Conn. May 26, 2021)).
Moreover, the majority's interpretation is contrary to how the OCR has applied its 1996 Letter. MSU provided six specific examples where the OCR assessed compliance based upon a school's average team size for the underrepresented sex. The OCR, in fact, has found schools in compliance specifically because the participation gap was less than the average size of the school's female teams. See Letter from OCR to Univ. of Minnesota-Twin Cities, at 6 (Sept. 27, 2018) (approving participation gap of 28 where “the average size of teams offered for the underrepresented sex ․ was 35.85 female athletes”), https://www2.ed.gov/about/offices/list/ocr/docs/investigations/more/05152038-a.pdf; Letter to Innovative Horizons Charter School, at 5 (Jul. 15, 2016) (finding substantial proportionality achieved because the school had an “average female team size of 15” and “[f]emale under representation [sic] of 5 athletes is not enough to sustain a viable team, and is less than the average team size of 15”), https://www2.ed.gov/about/offices/list/ocr/docs/investigations/more/09151075-a.pdf; (Appellee Br. 30-31). If the 1996 Letter expressly allows an adjudicator to use “the average size of teams for the underrepresented sex” at a school as a benchmark and the OCR has applied the rule in that way, we cannot fault the district court for doing so and summarily adopt a conflicting interpretation. See Kisor, 139 S. Ct. at 2412, 2418.
Practically speaking, today's decision means that if the participation gap is greater than any team for which there is interest, ability, and available competition (i.e., a 4-person tennis team), a school must always add that team to comply with Title IX. (Maj. Op. 1056, 1060–61; see Appellant Br. 50-51). That is tantamount to requiring perfection, not substantial proportionality. (See 1996 Letter, PgID 489). Indeed, under the majority's reasoning, when a school eliminates an athletic program and there is a participation gap, student-athletes (male and female alike) may establish a Title IX violation by simply relying on the prior existence of their team to show that there is enough interest, ability, and competition for their team. That cannot be right.
Because the 1996 Letter and the OCR's decisions recognize that the average size of MSU's female teams is a valid benchmark for the participation gap, a remand is wholly unwarranted.
If all that were not enough, there is a problem with the relief plaintiffs seek. As the parties agree, the district court would need to “order MSU to reinstate the women's swimming and diving team” because it no longer exists. (Appellant Br. 12, 18; Reply Br. 16; Appellee Br. 51-56). 7 But even if plaintiffs were entitled to an injunction (and they are not), the most they may obtain is an order “that the university prepare a Title IX-compliant proposal for the ․ school year next fall.” Mayerova v. E. Michigan Univ., No. 19-1177, 2019 U.S. App. LEXIS 9373 *4 (6th Cir. Mar. 28, 2019) (order) (Norris, Sutton, and Cook, JJ.) (staying a district court injunction that required the school to reinstate the women's tennis and softball teams). For two reasons, that is true regardless of the participation gap.
First, Title IX does not require that MSU reinstate or continue to support a particular team. “Every court, in construing the  Policy Interpretation and the text of Title IX, has held that a university may bring itself into Title IX compliance by increasing athletic opportunities for the underrepresented gender (women in this case) or by decreasing athletic opportunities for the overrepresented gender.” Equity in Athletics, 639 F.3d at 103 (quoting Neal v. Bd. of Trs., 198 F.3d 763, 769-70 (9th Cir. 1999)); see also Kelley v. Bd. of Trs., 35 F.3d 265, 269 (7th Cir. 1994); Roberts, 998 F.2d at 830; Cohen v. Brown Univ., 991 F.2d 888, 898 n.15 (1st Cir. 1993). Indeed, that is the rule in this circuit: “An institution need not pour ever-increasing sums into its athletic programs in order to bring itself into compliance, but has the option of reducing opportunities for the overrepresented gender while keeping opportunities for the underrepresented gender stable.” Horner v. Kentucky High Sch. Athletic Ass'n, 43 F.3d 265, 275 (6th Cir. 1994); see also Miami Univ. Wrestling Club v. Miami Univ., 302 F.3d 608, 611 (6th Cir. 2002) (upholding decision to eliminate “men's soccer, tennis and wrestling teams” to decrease participation disparity). If reinstating the women's team is not an available remedy at the final judgment, why would a court be permitted to grant such relief while the litigation continues for years? By the end, plaintiffs will have graduated and obtained all they desired: to compete in their sport at MSU.
Second, plaintiffs’ requested injunction does not maintain the status quo. The “purpose of a preliminary injunction is merely to preserve the relative positions of the parties until a trial on the merits can be held.” Benisek v. Lamone, ––– U.S. ––––, 138 S. Ct. 1942, 1945, 201 L.Ed.2d 398 (2018) (quoting Univ. of Tex. v. Camenisch, 451 U.S. 390, 395, 101 S.Ct. 1830, 68 L.Ed.2d 175 (1981)). Where the participation gap was previously 12 or 0.65%, requiring MSU to reinstate only the women's swimming and diving team of 33 women would swing the participation gap further the other way—creating a participation gap of 21 for men. That hardly represents the status quo and is yet another reason why a remand is a fruitless exercise.
* * *
“[A] preliminary injunction is ‘an extraordinary remedy never awarded as of right.’ ” Benisek, 138 S. Ct. at 1944 (quoting Winter v. NRDC, Inc., 555 U.S. 7, 24, 129 S.Ct. 365, 172 L.Ed.2d 249 (2008)). For that reason, “the proof required for the plaintiff to obtain a preliminary injunction is much more stringent than the proof required to survive a summary judgment motion.” McNeilly v. Terri Lynn Land, 684 F.3d 611, 615 (6th Cir. 2012) (citation omitted). Plaintiffs have not carried their heavy burden. That should be enough to give pause before announcing binding principles of law that are contrary to Title IX, agency guidance, and the consensus among the other federal circuit courts. This court should affirm the district court's decision denying plaintiffs a preliminary injunction.
1. HEW was the Department of Education's predecessor.
2. The dissent argues that the term “substantial proportionality” “inherently requires reference to a ratio or percentage.” Dissent at 1066 n.6. As the dissent acknowledges, however, the relevant ratio comes from comparing the athletic opportunities to the gender breakdown of the undergraduate student body. This is the relevant ratio, not the percentage of the athletic opportunities relative to the size of the athletic program. This ratio is a variable in the equation that is used to calculate the participation-gap number. The fact that one ratio is used in evaluating substantial proportionality does not mean that every part of the compliance determination requires the use of a ratio.
3. Many cases (none of which are binding on this court) have drawn a bright line around two percent. See, e.g., Equity in Athletics, 639 F.3d at 110 (“EIA provides no support for its contention that a disparity as low as 2% (and, according to the record, not much above 1%) is substantially disproportionate as a matter of law.”); Boulahanis v. Bd. of Regents, 198 F.3d 633, 639 (7th Cir. 1999), abrogated on other grounds by Trentadue v. Redmon, 619 F.3d 648 (7th Cir. 2010) (explaining that “the University has achieved substantial proportionality” when “the athletic participation of men remained within three percentage points of enrollment”); Anders, 2021 WL 1564448, at *5 (“[C]ourts have held that a disparity of 2% or less between the underrepresented sex's percentage of participation opportunities and the underrepresented sex's percentage of enrollment is proof that an educational institution falls within the substantial proportionality safe harbor.” (internal citations and quotation marks omitted)); Portz v. St. Cloud State Univ. (“Portz I”), 196 F. Supp. 3d 963, 975 (D. Minn. 2016) (“[A] deviation of less than 3.5 percentage points typically keeps the ratios substantially proportionate.”). We do not find this reasoning persuasive in light of the clear language of the 1979 Policy Interpretation and the 1996 Letter.
4. Our dissenting colleague reads the district court's opinion as finding that MSU's numbers are accurate. Dissenting Op. at 1064–65. We do not read the district court's opinion in the same way. The district court found that, regardless of whether the gap was 25, 36, or 12, MSU complied with the substantial-proportionality requirement. 2021 WL 650712, at *10. We do not require the district court “to incant magic words” to make a finding regarding the size of the participation gap. Dissenting Op. at 1064–65. It is not clear to us, however, that the district court made any finding on this issue at all. The language quoted by the dissent is a rejection of the student-athletes’ argument that MSU inflated its participation numbers. Id. It does not bear on the parties’ dispute about the data source, which is wholly separate from whether MSU improperly inflated participation opportunities.
5. Admittedly, this case is different from Ohlensehlen in two respects. First, the district court did not ask MSU to disclose its underlying data. Second, although the student-athletes requested the underlying data from MSU, R. 13-7 (FOIA request) (Page ID #708); R. 13-10 (Limited Discovery Request) (Page ID #711–13), they did not pursue either avenue after MSU claimed that neither mechanism gave the student-athletes the right to access this information, R. 13-9 (FOIA Response) (Page ID #710); R. 13-11 (Email from MSU Attorney) (Page ID #714). Nevertheless, Ohlensehlen shows that, at the preliminary-injunction stage, there may be a need to rely on data other than official Title IX counts.
6. The dissent points to OCR letters that examine the average team size at institutions. Each of these letters involves circumstances in which the parties offered no evidence of whether there is sufficient interest, ability, and competition to field a viable team. In circumstances in which there is no information about interest, ability, and competition, it may be more appropriate to look at the average team as the primary point of reference.
7. Unlike the size of a viable team, the size of an average team can be calculated based on only quantitative, not qualitative, factors. Because the participation gap also involves a purely quantitative determination, comparing the participation gap to the average team becomes a purely mathematical calculation, in conflict with the 1996 Letter.
8. The dissent argues that an injunction cannot be appropriate because it does not maintain the status quo. Certainly, the costs of reinstating a team may impact the district court's valuation of the second and third preliminary injunction factors. That does not, however, mean that if a district court denies a preliminary injunction based on a misreading of the law, courts are without the ability subsequently to rectify that error. See Porter v. Lee, 328 U.S. 246, 251, 66 S.Ct. 1096, 90 L.Ed. 1199 (1946) (“It has long been established that where a defendant with notice in an injunction proceeding completes the acts sought to be enjoined the court may by mandatory injunction restore the status quo.”); Di Biase v. SPX Corp., 872 F.3d 224, 232 (4th Cir. 2017) (“[A] motion for preliminary injunction filed before the act to be enjoined has occurred, and subsequently intended to restore the status quo once it has been disturbed, is not moot.”). To hold otherwise would effectively render the denial of a preliminary injunction in such circumstances unreviewable.
1. Policy Interpretation: Title IX and Intercollegiate Athletics, 44 Fed. Reg. 71,413 (Dec. 11, 1979).
2. Office for Civil Rights (OCR), U.S. Dep't of Educ., Clarification of Intercollegiate Athletics Policy Guidance: The Three-Part Test (Jan. 16, 1996), https://www2.ed.gov/about/offices/list/ocr/docs/clarific.html (R. 8-10, PgID 484-94).
3. As explained below, the OCR and courts consider the participation gap as a percentage and a number. For example, if student enrollment is 48% females and female athletes comprise 46% of the athletic department, the participation gap is 2%. To calculate the participation gap for females as a number, the following formula is used: (total male athletes ÷ percentage of males in the student body) – total number of athletes = the female participation gap. (R. 2-14, PgID 211-12). It appears MSU has rounded the participation gap up to a whole number (i.e., 11.31 is rounded up to 12).
4. The majority cites one other district court decision that relied on EADA data because, due to rosters that were greater than what the court thought was necessary, the court believed that some students were “not receiving genuine opportunities to participate,” and the court also found that there was credible evidence of unreported male players. Biediger v. Quinnipiac Univ., 616 F. Supp. 2d 277, 297-98 (D. Conn. 2009); (Maj. Op. 1059). But again, the majority notes that this case is different. There is no evidence of unreported male players, and the majority agrees that “Title IX does not require that athletes participate in competitions to be counted.” (Maj. Op. 1056 – ––––).
5. See also Boulahanis v. Bd. of Regents, 198 F.3d 633, 636, 638-39 (7th Cir. 1999) (affirming summary judgment for university because, despite “the elimination of men's soccer and men's wrestling at the University, the athletic participation of men remained within three percentage points of enrollment”); Biediger v. Quinnipiac Univ., 691 F.3d 85, 106-07 (2d Cir. 2012) (upholding injunction where there was a “3.62% disparity” or 38 participation opportunities); Roberts v. Colorado State Bd. of Agric., 998 F.2d 824, 830 (10th Cir. 1993) (“[A] 10.5% disparity between female athletic participation and female undergraduate enrollment is not substantially proportionate.”); Pederson v. Louisiana State Univ., 213 F.3d 858, 878 (5th Cir. 2000). But see Ollier v. Sweetwater Union High Sch. Dist., 768 F.3d 843, 852, 856-57 (9th Cir. 2014) (affirming summary judgment for plaintiffs where the “6.7 percent disparity ․ was equivalent to 47 girls” and “47 girls can sustain at least one viable competitive team”). Notably, the majority cites Biediger to reject “a statistical safe harbor at [2%] or any ․ percentage,” (Maj. Op. 1058), but Biediger still viewed the proportionality gap in terms of a percentage and a number. Biediger, 691 F.3d at 106-08 (“3.62% disparity” or 38 roster positions).
6. It makes sense to view the participation gap as a number when assessing the number of athletes needed to field a viable, average size team. It also makes sense, however, to view the participation gap as a percentage when comparing the percentage of females in the student body and the percentage of females in the athletic department. The number of female students enrolled at MSU (18,192) cannot be compared to the number of female athletic participants at MSU (417). (R. 8-8, PgID 444; see R. 8-2, PgID 362). Because the two groups have different quantities, a percentage or ratio must be used to perform a “proportionality” comparison. After all, the lodestar is “substantial proportionality” between enrollment for a gender and that respective gender's athletic opportunities. See 44 Fed. Reg. at 71,418. The plain meaning of that phrase inherently requires reference to a ratio or percentage. See Proportionality, Oxford English Dictionary, OED (Oxford Univ. Press 2021) (“A formula or expression stating the proportionality of two or more quantities”), https://www.oed.com/view/Entry/152773?redirectedFrom=proportionality#eid; see also Proportional, Oxford English Dictionary (“[H]aving a constant ratio to another variable quantity.”); Proportional, Merriam-Webster's Unabridged Dictionary (2021) (“[H]aving the same or a constant ratio”), https://unabridgedmerriam-webster.com/unabridged/proportional. Because females make up 50.93% of MSU's enrollment and 50.06% of its athletic participants, (R. 8-8, PgID 443-44; see R.8-2, PgID 362), it's fair to say that MSU's “numbers [are] substantially proportionate.” 44 Fed. Reg. at 71,418; (see also 1996 Letter, PgID 489-90).
7. See also Michigan State University, 2020-21 Swimming and Diving Schedule, https://msuspartans.com/sports/swimming-and-diving/schedule; Jared Ramsey, MSU Swimming and Diving members struggle with the shift to student life, The State News (Nov. 3, 2021), https://statenews.com/article/2021/11/the-question-throughout-all-of-this-has-been-why-msu-swimming-and-divingmembers-struggle-with-the-shift-to-student-life?ct=content_open&cv=cbox_featured.
KAREN NELSON MOORE, Circuit Judge.
MOORE, J., delivered the opinion of the court in which GIBBONS, J., joined. GUY, J. (pp. 1062–69), delivered a separate dissenting opinion.