966 F.3d 1038

10th Cir.

2020

CHRISTINE FRAPPIED; CHRISTINE GALLEGOS; KATHLEEN GREENE; JOYCE HANSEN; KRISTINE JOHNSON; GEORGEAN LABUTE; JOHN ROBERTS; ANNETTE TRUJILLO; DEBBIE VIGIL, Plaintiffs - Appellants, and JENNIFER RYAN, Plaintiff, v. AFFINITY GAMING BLACK HAWK, LLC, Defendant - Appellee.

No. 19-1063

United States Court of Appeals for the Tenth Circuit

July 21, 2020

PUBLISH
Christopher M. Wolpert, Clerk of Court
EQUAL EMPLOYMENT OPPORTUNITY COMMISSION; NATIONAL EMPLOYMENT LAWYERS ASSOCIATION; THE EMPLOYEE RIGHTS ADVOCACY INSTITUTE FOR LAW & POLICY; AARP; AARP FOUNDATION, Amici Curiae.

Appeal from the United States District Court for the District of Colorado

(D.C. No. 1:17-CV-01294-RM-NYW)

Lisa R. Sahli, Lisa R. Sahli, Attorney at Law, LLC, Littleton, Colorado, and Barry D. Roseman, Roseman Law Offices, LLC, Denver, Colorado for Plaintiffs-Appellants.

Joshua B. Kirkpatrick (Jennifer S. Harpole and David C. Gartenberg, with him on the briefs), Littler Mendelson, P.C., Denver, Colorado for Defendant-Appellee.

Daniel B. Kohrman and William Alvarado Rivera, AARP Foundation, Washington, D.C., filed an Amici Curiae brief for AARP and AARP Foundation, in support of Appellants.

Darold W. Killmer and Liana Orshan, Killmer, Lane & Newman, LLP, Denver, Colorado, filed an Amici Curiae brief for the National Employment Lawyers Association and the Employee Rights Advocacy Institute for Law & Policy, in support of Appellants.

James L. Lee, Deputy General Counsel, Jennifer S. Goldstein, Associate General Counsel, Elizabeth E. Theran, Assistant General Counsel, and Susan R. Oxford, Attorney, Equal Employment Opportunity Commission, Office of General Counsel, Washington, D.C., filed an Amicus Curiae brief for the Equal Employment Opportunity Commission, in support of Appellants.

Before LUCERO, PHILLIPS, and MORITZ, Circuit Judges.

LUCERO, Circuit Judge.

This is an employment discrimination case. Plaintiffs allege defendant Affinity Gaming Black Hawk, LLC (“Affinity“) terminated them on the basis of age and sex. They brought disparate impact and disparate treatment claims under Title VII of the Civil Rights Act of 1964 and the Age Discrimination in Employment Act (“ADEA“). The district court dismissed (1) the Title VII disparate impact claim, (2) the Title VII disparate treatment claim, and (3) the ADEA disparate impact claim. It granted summary judgment in favor of Affinity on the ADEA disparate treatment claim. Exercising jurisdiction under 28 U.S.C. § 1291, we affirm the dismissal of the Title VII disparate treatment claim. With respect to the other claims, we reverse and remand to the district court.

I

Plaintiffs were employed at the Golden Mardi Gras Casino (“the Casino“). Affinity purchased the Casino in early 2012 and took over its operations in November 2012. In January 2013, Affinity laid off many of the Casino‘s employees. The terminations were not a reduction in force, and Affinity posted an advertisement on Craigslist listing 59 open positions.

Plaintiffs are nine Casino employees terminated by Affinity in January 2013.¹ Eight are women; one is a man. All were forty or older when they were terminated. The female plaintiffs brought “sex-plus-age” disparate impact and disparate treatment claims under Title VII, alleging they were terminated because Affinity discriminated against women over forty. All nine plaintiffs brought disparate impact and disparate treatment claims under the ADEA, alleging they were terminated because of their age.

The district court granted Affinity‘s motion to dismiss the Title VII sex-plus age claims and the ADEA disparate impact claim. It also denied plaintiffs’ motion for reconsideration of the order dismissing the disparate impact ADEA claim. Additionally, the court granted summary judgment in favor of Affinity on the remaining disparate treatment ADEA claim. This appeal followed.

II

We first address the district court‘s dismissal of plaintiffs’ Title VII sex-plus age disparate impact claim. We review de novo the dismissal of a claim under Federal Rule of Civil Procedure 12(b)(6). See Smith v. United States, 561 F.3d 1090, 1098 (10th Cir. 2009).

At the outset, we must determine whether sex-plus-age claims are cognizable under Title VII. “Title VII is a broad remedial measure, designed to assure equality of employment opportunities.” Pullman-Standard v. Swint, 456 U.S. 273, 276 (1982) (quotation omitted). The statute makes it unlawful for an employer to “discharge any individual . . . because of such individual‘s race, color, religion, sex, or national origin.” 42 U.S.C. § 2000e-2(a)(1). “[T]he ordinary meaning of ‘because of’ is ‘by reason of’ or ‘on account of,‘” so “Title VII‘s ‘because of’ test incorporates the simple and traditional standard of but-for causation.” Bostock v. Clayton Cty., Ga., 140 S. Ct. 1731, 1739 (2020) (quotations omitted). The statute also provides that a plaintiff may show discrimination by showing that his or her membership in a protected class was a “motivating factor” for the challenged employment practice. § 2000e-2(m). “An employer violates Title VII when it intentionally fires an individual employee based in part on sex.” Bostock, 140 S. Ct. at 1741.

Title VII also prohibits discrimination based on a combination of protected characteristics, such as “sex-plus-race” discrimination, i.e., discrimination targeted only at employees of a particular race and sex. See Connecticut v. Teal, 457 U.S. 440, 455 (1982) (“It is clear that Congress never intended to give an employer license to discriminate against some employees on the basis of race or sex merely because he favorably treats other members of the employees’ group.“); see also Hicks v. Gates Rubber Co., 833 F.2d 1406, 1416-17 (10th Cir. 1987) (same).

In this case, plaintiffs’ sex-plus-age claim is not based on a combination of protected characteristics enumerated in the statute; rather, the “plus-” characteristic is age, which is not a protected class under Title VII. Ample precedent holds that Title VII forbids “sex-plus” discrimination in cases in which the “plus-” characteristic is not itself protected under the statute. In Phillips v. Martin Marietta Corp., 400 U.S. 542 (1971) (per curiam), the employer refused to hire women with preschool-age children. Id. 543. The Supreme Court held that the employer violated Title VII by maintaining “one hiring policy for women and another for men—each having preschool-age children.” Id. 544. Even though Title VII does not prohibit discrimination against people with preschool-age children as a class, the Court recognized that discrimination against only women, not men, with preschool-age children is a form of sex discrimination cognizable under the statute. In Bostock, the Court acknowledged that the employer in Phillips “easily could have pointed to some other, nonprotected trait and insisted it was the more important factor in the adverse employment outcome.” 140 S. Ct. at 1744. But so long as sex plays a role in the employment action, it “has no significance” that a factor other than sex “might also be at work,” even if that other factor “play[s] a more important role [than sex] in the employer‘s decision.” Id.

We have also held that Title VII prohibits sex-plus discrimination even when the “plus-” characteristic is not itself protected. In Coleman v. B-G Maintenance Management of Colorado, Inc., 108 F.3d 1199 (10th Cir. 1997), we held that Title VII forbids “discrimination against subclasses of women.” Id. 1203. But when the “plus-” characteristic is not itself protected, sex-plus discrimination claims must be premised on sex alone. See id. We explained:

Title VII contemplates [sex]-plus claims because when one proceeds to cancel out the common characteristics of the two classes being compared (e.g., married men and married women), as one would do in solving an algebraic equation, the cancelled-out element proves to be that of married status, and sex remains the only operative factor in the equation. Thus, although the protected class need not include all women, the plaintiff must still prove that the subclass of women was unfavorably treated as compared to the corresponding subclass of men.

Id. (emphases in original) (alteration, citation, and quotation omitted); see also Chadwick v. WellPoint, Inc., 561 F.3d 38, 43 (1st Cir. 2009) (“The terminology may be a bit misleading[;] . . . the ‘plus’ does not mean that more than simple sex discrimination must be alleged; rather, it describes the case where not all members of a disfavored class are discriminated against.” (quotation omitted)); Back v. Hastings On Hudson Union Free Sch. Dist., 365 F.3d 107, 118 (2d Cir. 2004) (“The term ‘sex plus’ or ‘gender plus’ is simply a heuristic. It is, in other words, a judicial convenience developed in the context of Title VII to affirm that plaintiffs can, under certain circumstances, survive summary judgment even when not all members of a disfavored class are discriminated against.“).

In its recent decision in Bostock, the Supreme Court stated that when determining whether a person is subjected to discrimination under Title VII, “our focus should be on individuals, not groups.” 140 S. Ct. at 1740. In dicta, the Court examined a hypothetical situation in which an employer has “a policy of firing any woman he discovers to be a Yankees fan.” Id. 1742. It explained that a termination because of such a policy would constitute discrimination “‘because of sex’ if the employer would have tolerated the same allegiance in a male employee,” and that such discrimination would satisfy Title VII‘s but-for causation standard. Id.; see also id. 1739 (“[A] defendant cannot avoid liability just by citing some other factor that contributed to its challenged employment decision. So long as the plaintiff‘s sex was one but-for cause of that decision, that is enough to trigger the law.“).² If the hypothetical employer has a policy under which it fires all Yankees fans, a termination based solely on that policy is because of status as a Yankee fan, not because of sex. But if its policy is to fire only female Yankees fans, it engages in

prohibited discrimination because such terminations are based in part on sex. This affirms our ruling in Coleman that a female sex-plus plaintiff must show that her employer treated her unfavorably relative to a male employee who also shares the “plus-” characteristic.³

But Bostock‘s requirement that we focus on individual discrimination contradicts Coleman‘s holding that in order to prevail, a female sex-plus plaintiff must “prove that the subclass of women” to which she belongs “was unfavorably treated as compared to the corresponding subclass of men.” 108 F.3d at 1203. In light of Bostock, we conclude that a sex-plus plaintiff does not need to show discrimination against a subclass of men or women. Instead, if a female plaintiff shows that she would not have been terminated if she had been a man—in other words, if she would not have been terminated but for her sex—this showing is sufficient to establish liability under Title VII.⁴ She need not show her employer discriminated against her entire subclass. In short, we conclude that after Bostock, the class of sex-plus claims cognizable under Title VII is broader than we recognized in Coleman: we no longer require sex-plus plaintiffs to show discrimination against

an entire subclass.

Turning to plaintiffs’ claim in this case, no circuit court has yet addressed whether Title VII prohibits sex-plus-age discrimination.⁵ Several district courts, however, have accepted the viability of sex-plus-age claims under the statute.⁶ The Equal Employment Opportunity Commission has also recognized the validity of such claims. See 2 EEOC Compliance Manual § IIA, 2009 WL 2966754 (Aug. 6, 2009) (“The EEO statutes prohibit discrimination against an individual based on his/her membership in two or more protected classes. . . . [I]ntersectional discrimination can involve more than one EEO statute, e.g., discrimination based on age and disability, or based on sex and age.“).

We hold that sex-plus-age claims are cognizable under Title VII. There is no material distinction between a sex-plus-age claim and the other sex-plus claims we have previously recognized for which the “plus-” characteristic is not protected under Title VII. Like claims for which the “plus-” factor is marital status or having preschool-age children, a sex-plus-age claim alleges discrimination against an employee because of sex and some other characteristic. It is thus a sex discrimination claim, albeit one that alleges that the discrimination was based only in part on sex. See Chadwick, 561 F.3d at 43 (“[R]egardless of the label given to [a sex-plus] claim, the simple question posed by sex discrimination suits is whether the employer took an adverse employment action at least in part because of an employee‘s sex.” (emphasis omitted)). Like any other sex-plus plaintiff, a sex-plus age plaintiff must show unfavorable treatment relative to an employee of the opposite sex who also shares the “plus-” characteristic. For the female sex-plus-age plaintiffs in this case, the relevant comparator would be an older man.

Affinity argues that plaintiffs should not be able to bring Title VII sex-plus age claims because of the availability of relief under the ADEA. But ADEA claims and Title VII sex-plus-age claims address different harms. An ADEA claim addresses discrimination against an older worker because of his or her age, whereas a Title VII sex-plus-age claim brought by an older woman addresses discrimination against her because of her sex. In Bostock, the Court held, “if changing the employee‘s sex would have yielded a different choice by the employer—a statutory violation has occurred.” 140 S. Ct. at 1741. Thus, termination is “because of sex” if the employer would not have terminated a male employee with the same “plus-” characteristic. Although in some cases a plaintiff may be able to bring both a Title VII sex-plus-age claim and an ADEA age discrimination claim, the two claims would address two distinct kinds of discrimination—sex discrimination and age discrimination, respectively. Thus, allowing a plaintiff to bring a sex-plus-age claim under Title VII would not allow him or her to circumvent the requirements of the ADEA.

Affinity also contends that because the ADEA and Title VII are structured differently, involve different burdens of proof, and provide different remedies, Congress did not intend to allow sex-plus-age claims under Title VII. Though Affinity appears to argue that all discrimination claims with some age-related component must be brought only under the ADEA, the ADEA includes no such requirement. Nothing in the ADEA limits a plaintiff‘s ability to bring a claim under Title VII. To the contrary, by passing the ADEA, Congress intended to broaden protections against employment discrimination to cover older workers. See generally 29 U.S.C. § 621. It did not intend to limit existing protections provided under Title VII.

Our conclusion is consistent with Title VII‘s legislative purpose. “In forbidding employers to discriminate against individuals because of their sex, Congress intended to strike at the entire spectrum of disparate treatment of men and women resulting from sex stereotypes.” City of L.A., Dep‘t of Water & Power v. Manhart, 435 U.S. 702, 707 n.13 (1978) (quotation omitted); see also Price Waterhouse v. Hopkins, 490 U.S. 228, 251 (1989) (“As for the legal relevance of sex stereotyping, we are beyond the day when an employer could evaluate employees by assuming or insisting that they matched the stereotype associated with their group.“).

Research shows older women are subjected to unique discrimination resulting from sex stereotypes associated with their status as older women. See, e.g., Nicole Buonocore Porter, Sex Plus Age Discrimination: Protecting Older Women Workers, 81 Denv. U.L. Rev. 79, 94-101 (2003); Patti Buchman, Note, Title VII Limits on Discrimination Against Television Anchorwomen on the Basis of Age-Related Appearance, 85 Colum. L. Rev. 190 (1985) (discussing sex-plus-age discrimination in local television broadcasting). This discrimination is distinct from age discrimination standing alone. Jourdan Day, Closing the Loophole-Why Intersectional Claims Are Needed to Address Discrimination Against Older Women, 75 Ohio St. L.J. 447, 474 (2014) (“The intersectionality of two immutable characteristics is not the same as simply possessing two separate characteristics. While an individual can be both ‘old’ and be a ‘woman,’ being an ‘older woman’ is substantively different.“). One study, based on 40,000 job applications, found “much stronger and more robust evidence of age discrimination against older women than against older men.” Neumark et al., Is It Harder for Older Workers to Find Jobs? New and Improved Evidence from a Field Experiment, 127 J. Pol. Econ. 922, 966 (2019). As we have explained, if discrimination is targeted more at older women than at older men, that differential treatment is not merely a manifestation of “stronger” age discrimination—it is itself a form of sex discrimination aimed at older women.

Recognizing claims for “intersectional” discrimination best effectuates congressional intent to prohibit discrimination based on stereotypes. See Kimberlé Crenshaw, Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Politics, 1989 U. Chi. Legal F. 139, 150 (1989) (identifying mistaken assumption that “a discriminator treats all people within a race or sex category similarly“). A failure to recognize intersectional discrimination “obscures claims that cannot be understood as resulting from discrete sources of discrimination.” Id. 140; see also Jefferies, 615 F.2d at 1032-33 (“If both black men and white women are considered to be within the same protected class as black females . . . , no remedy will exist for discrimination which is directed only toward black females.“). Intersectional discrimination against older women is a form of discrimination based on sex stereotypes that Title VII was intended to prohibit. And discrimination against older women that does not target older men is a form of sex discrimination.

The district court dismissed plaintiffs’ Title VII disparate impact claim solely because it concluded plaintiffs could not bring sex-plus-age claims under Title VII. Affinity does not challenge the claim on any other grounds. Because we hold that sex-plus-age claims are cognizable under Title VII, we reverse the dismissal of this claim.

III

We turn to the district court‘s dismissal of plaintiffs’ Title VII disparate treatment claim, which we review de novo. See Smith, 561 F.3d at 1098. The district court dismissed this claim partly because it erroneously concluded sex-plus age claims are not cognizable under Title VII. It also set forth two additional grounds for dismissal: (1) plaintiffs did not specify the age at which one is an “older woman,” and (2) the Third Amended Complaint, which was the operative complaint in this case, contained insufficient allegations of sex discrimination.

A

We first address whether plaintiffs adequately alleged the “plus-” characteristic by specifying which employees are “older women.” In Coleman, we held that sex-plus plaintiffs must specify a corresponding subclass of members of the opposite sex. 108 F.3d at 1204. As we have explained, Bostock requires us to focus our analysis on individuals, not groups. Although we no longer require sex-plus plaintiffs to show discrimination targeting a particular subclass, sex-plus plaintiffs must still specify the “plus-” characteristic on which they premise their claims. Such specificity is necessary for a court to assess whether an employer discriminated against a sex-plus plaintiff relative to an employee of the opposite sex who shares the “plus-” characteristic.

Turning to the operative complaint in this case, we note that it sets forth each plaintiff‘s age. It also repeatedly and interchangeably refers to employees “age forty or older” and employees “in the protected age group.”⁷ In order to be protected under the ADEA, a worker must be at least forty years old. 29 U.S.C. § 631(a). For sex-plus-age claims under Title VII, however, there is no “protected age group“—as explained above, Title VII does not bar age discrimination; it merely bars discrimination based in part on sex. We acknowledge that the complaint does not explicitly state that the “plus-” characteristic is being forty or older, but it does state that the claim was brought on behalf of “older women” and repeatedly refers interchangeably to employees in the protected age group and employees at least forty years old. Although the allegations in the complaint are not of ideal clarity, we conclude that it sufficiently specifies that the “plus-” characteristic is being forty or older.

B

The district court also dismissed plaintiffs’ disparate treatment Title VII claim because it concluded that the complaint contained insufficient allegations of sex discrimination. In determining whether a Title VII disparate treatment claim is plausibly alleged, we do not require plaintiffs to establish a prima facie case.

Instead, we consider whether they have set forth a plausible claim in light of the elements of their claim. See Khalik v. United Air Lines, 671 F.3d 1188, 1192 (10th Cir. 2012) (“While the 12(b)(6) standard does not require that [a p]laintiff establish a prima facie case in her complaint, the elements of [the] alleged cause of action help to determine whether [a p]laintiff has set forth a plausible claim.“). In general, a Title VII plaintiff bringing a claim of employment discrimination in a termination decision must show four elements: “(1) he [or she] belongs to a protected class; (2) he [or she] was qualified for his [or her] job; (3) despite his [or her] qualifications, he [or she] was discharged; and (4) the job was not eliminated after his [or her] discharge.” Kendrick v. Penske Transp. Servs., Inc., 220 F.3d 1220, 1229 (10th Cir. 2000). A plaintiff pleading a sex-plus claim is not required to allege more than these elements. See Franchina v. City of Providence, 881 F.3d 32, 53 (1st Cir. 2018) (holding sex-plus plaintiffs need not “allege more than what is required for traditional sex discrimination claims“).

The complaint alleges the age and sex of each plaintiff. When these allegations are considered in conjunction with its references to employees “age forty or older” and employees “in the protected age group,” the complaint alleges that each plaintiff shares the “plus-” characteristic: being forty or older. It alleges facts related to each plaintiff‘s qualifications and states that each one was terminated. The complaint also includes statistics about the ages of the terminated and retained workers. Notably, the complaint does not include any specific factual allegations that would suggest that Affinity discriminated against any individual plaintiff because of sex.

We are mindful that in analyzing plaintiffs’ sex-plus-age claim, “our focus should be on individuals, not groups.” Bostock, 140 S. Ct. at 1740. But the complaint in this case is devoid of allegations that any individual plaintiff was terminated because of sex. Accordingly, we must evaluate whether the allegations about how plaintiffs were treated as a group are sufficient to give rise to a plausible inference of sex discrimination against the individual plaintiffs. As we have explained, female sex-plus plaintiffs must show discrimination compared to men who share the same “plus-” characteristic. Consider a hypothetical case in which an employer fires all employees over forty, and the older women who are fired file a complaint of sex-plus-age discrimination. If that complaint does not plausibly allege that any of the terminations were based on sex, it would not state a sex-plus-age claim. Rather, the allegation that all employees over forty were fired would only give rise to an inference that the employees were fired because of their age.⁸ These hypothetical employees would need to rely solely on an age-discrimination theory.

In this case, because there are no allegations that would permit us to focus our analysis on whether Affinity discriminated against individual older women plaintiffs because of sex, we must instead evaluate whether the allegations in the complaint

about plaintiffs as a group give rise to the inference of discrimination against them as individuals because of sex. We conclude that they do not. Other than the statistical allegations in the complaint, the only allegations concerning discrimination against older women as compared to older men are the statements that Affinity‘s actions “had a discriminatory impact on older workers, and older females in particular” and that Affinity “viewed older females unfavorably and/or graded older females more harshly than . . . older males.” These allegations are conclusory. The only non conclusory allegations in the complaint involve statistical data.

Statistics may be probative in Title VII disparate treatment cases, particularly ones involving mass terminations. See Stone v. Autoliv ASP, Inc., 210 F.3d 1132, 1139 (10th Cir. 2000) (“While a balanced workforce cannot immunize an employer from liability for specific acts of discrimination, statistics concerning employees terminated in a [reduction in force] are probative to the extent they suggest that older employees were not treated less favorably than younger employees.” (citations and quotation omitted)); see also Bauer v. Bailar, 647 F.2d 1037, 1045 (10th Cir. 1981) (“Statistical evidence is, of course, relevant to a claim of disparate treatment and should be given proper effect by the courts.“). Even though the complaint in this case addresses only treatment of older women as a group, the statistics therein may give rise to a plausible inference of discrimination against plaintiffs as individuals.

Much of plaintiffs’ statistical data compares the treatment of older women to younger women, and the overall treatment of men to the overall treatment of women. As we have explained, neither comparison is relevant to a sex-plus-age claim. We therefore limit our focus to the allegations comparing the proportion of terminated women forty or older to the proportion of terminated men forty or older.

The complaint includes data about the proportion of older men and older women Affinity fired. It alleges that the termination rate of older women is statistically significant under a “chi-square test (one-tailed Fisher Exact test).” “Fisher‘s Exact test is used to calculate the probability that the number of individuals of a particular race-selected or gender-selected classification would be the same as the number actually selected, if the selection were independent of race or gender. . . . It is designed to examine statistical significance in small sample sizes.” United States v. Hernandez-Estrada, 749 F.3d 1154, 1164 (9th Cir. 2014). The probability value, or p-value, that results “represents the likelihood that an apparent association observed in a data set is the product of random chance rather than a true relationship.” In re Lipitor (Atorvastatin Calcium) Mktg., Sales Practices & Prod. Liab. Litig. (No II) MDL 2502, 892 F.3d 624, 634 (4th Cir. 2018).

The complaint alleges that applying the Fisher‘s Exact test results in a p-value of 0.01231.⁹ This means that there is a 1.231% chance that the number of older women terminated by Affinity is the same as it would have been had Affinity terminated employees on a random (i.e., non-discriminatory) basis. In other words,

there is a 98.769% chance that Affinity‘s basis for terminating older women was not random. A p-value of 0.01231 rises to the level of statistical significance. See Apsley v. Boeing Co., 691 F.3d 1184, 1198 (10th Cir. 2012) (observing that a p value of 0.05 (5%) is generally considered statistically significant).¹⁰

In order to determine whether the statistical data in the complaint suffices to raise a plausible inference of sex discrimination, we must determine what inferences may be drawn from the statistically significant p-value. Cf. Turner v. Pub. Serv. Co. of Colo., 563 F.3d 1136, 1148 (10th Cir. 2009) (“[In order for] statistics to be probative of discrimination, they must relate to the proper population.” (quotation omitted)). The p-value plaintiffs advance represents the likelihood that the number of older women terminated by Affinity is the same as it would have been had Affinity terminated all of its employees at random, regardless of sex or age. Notably, plaintiffs did not exclude younger workers when applying Fisher‘s Exact test, so their proffered p-value does not reflect the likelihood that the number of older women terminated by Affinity is the same as if Affinity had terminated its older workers (men and women) at random. Thus, at best, plaintiffs’ have raised a plausible inference of discrimination based on age, sex, or some combination of the two.

Of course, under Bostock, it is insignificant if a factor other than sex, such as age, “play[s] a more important role in the employer‘s decision.” 140 S. Ct. at 1744. Accordingly, plaintiffs could survive a motion to dismiss if they raise a plausible inference of discrimination based on sex alone or on some combination of sex and age. But a sex-plus-age plaintiff must plead facts sufficient to raise an inference that alleged discrimination was not based solely on age—the “plus” factor. Otherwise, he or she has not alleged that the discrimination was at least in part because of sex.

In this case, the statistics plaintiffs allege and the p-value calculated therefrom reflect at least three possibilities: discrimination based on (1) sex alone, (2) a combination of sex and age, or (3) age alone. Although the p-value is probative of whether Affinity discriminated against older women, because the plaintiffs did not compare older women to only older men in calculating it, the p-value does not itself give rise to a plausible inference of discrimination because of sex.

Turning to plaintiffs’ remaining statistical allegations, we ask whether these “nudge[] their claims across the line from conceivable to plausible.” Bell Atl. Corp. v. Twombly, 550 U.S. 544, 570 (2007). These statistics are presented in a confusing and convoluted manner, and they appear contradictory. In paragraph 33, the complaint alleges Affinity terminated 10 of its 18 male employees forty or older (56%) and 16 of its 22 female employees forty or older (73%). But in the very next paragraph, it alleges Affinity terminated “at least” 19 women forty or older. This paragraph also states both that 33 women were terminated and that 30 women were terminated. In fact, it is unclear which of the contradictory statistical allegations in the complaint plaintiffs used in applying Fisher‘s Exact test. In its briefing before this court, plaintiffs blame these discrepancies on “incomplete” data and differing data sets. The issue, however, is not insufficient or incomplete data; it is irreconcilable, self-contradictory data. Plaintiffs’ contradictory allegations do not permit us to infer that Affinity‘s discrimination was not based on age alone.¹¹

In sum, plaintiffs’ statistics suffice to allege only that age and sex are merely possible causes of Affinity‘s termination decisions. Because they allege no other facts that would give rise to an inference of disparate treatment of women over forty as compared to men over forty, we conclude that plaintiffs have failed to state a plausible Title VII sex-plus-age disparate treatment claim.

IV

We next review de novo the district court‘s dismissal of plaintiffs’ disparate impact claim under the ADEA. See Smith, 561 F.3d at 1098. The ADEA makes it “unlawful for an employer to . . . discharge any individual . . . because of such individual‘s age.” 29 U.S.C. § 623(a)(1). Although plaintiffs do not need to establish a prima facie case in the complaint, we consider the elements of an ADEA disparate impact claim to help determine whether plaintiffs’ claim is plausible. Cf. Khalik, 671 F.3d at 1192 (same in Title VII context).

To establish a prima facie case of disparate impact discrimination under the ADEA, “plaintiffs must show that a specific identifiable employment practice or policy caused a significant disparate impact on a protected group.” Pippin v. Burlington Res. Oil & Gas Co., 440 F.3d 1186, 1200 (10th Cir. 2006) (quotation omitted). “[A]n employee must point to both a significant disparate impact and to a particular policy or practice that caused the disparity.” Id. (emphasis omitted). In the context of disparate impact claims under the Fair Housing Act, the Supreme Court has held that “[a] plaintiff who fails to allege facts at the pleading stage or produce statistical evidence demonstrating a causal connection cannot make out a prima facie case of disparate impact.” Tex. Dep‘t of Hous. & Cmty. Affairs v. Inclusive Cmtys. Project, Inc., 135 S. Ct. 2507, 2523 (2015).

The district court concluded that plaintiffs identified specific age-neutral employment policies.¹² Nevertheless, it dismissed plaintiffs’ ADEA disparate impact claim because it concluded plaintiffs did not offer statistics or data regarding the alleged disparate impact of the policies.¹³ Plaintiffs counter that the complaint alleges sufficient data to raise an inference of a disparate impact¹⁴ because paragraphs 34 and 35 of the complaint contain the following data:

	Under Forty	Forty or Older
Laid Off	31	29
Retained	32	14

Image in original document— termination data chart

Affinity responds that those paragraphs of the complaint do not contain this information. Affinity is incorrect.

Paragraph 34 states that Affinity laid off 19 women and 10 men forty or older—a total of 29, as shown in the chart. It also states that 60 employees in total were laid off. Because these two numbers allow us to calculate that the remaining 31 of the 60 employees were younger than forty, as shown above, we conclude that the complaint did include that allegation.

Paragraph 35 states that Affinity retained eight men and six women forty or older—totaling 14, as shown in the table. Paragraph 34 states 60 of 106 employees were laid off, meaning 46 were retained. If 14 of these retained employees were forty or older, 32 were younger than forty, as shown in the chart. Paragraph 35, however, states that 47 employees, not 46, were retained. Using that datum, if 14 retained employees were forty or older, 33 retained employees were under forty.¹⁵ Using these data, we calculate that Affinity laid off 67% (29 out of 43) of its workers forty or older. It terminated 48% (31 of 64) or 49% (31 of 63) of its workers under forty, depending on whether we rely on the statistics in Paragraph 34 or 35.

We must determine whether these statistics suffice to make plausible plaintiffs’ claim that Affinity‘s termination policies had a disparate impact on employees forty or older. In their response to Affinity‘s motion to dismiss, plaintiffs offered guidance about how to interpret the statistical data in the complaint. They stated that when Fisher‘s Exact test is applied to the data in the complaint, the resulting p-value is 0.04784.¹⁶ Affinity did not challenge this calculation below, nor does it do so on appeal. A p-value of 0.04784 means that there is a 4.784% chance that the correlation between age and termination “is the product of random chance rather than a true relationship.” In re Lipitor, 892 F.3d at 634. In Apsley, we explained that courts generally find a relationship statistically significant if the p-value is less than 5%. 691 F.3d at 1198. Thus, the statistics weigh in favor of finding that plaintiffs plausibly alleged a significant disparate impact.

Plaintiffs’ disparate impact claim is further buttressed by their allegations regarding the ages of the new hires who replaced plaintiffs and the other terminated employees. In Pippin, we affirmed the dismissal of an ADEA plaintiff‘s disparate impact claim at summary judgment because he alleged only that his employer terminated “more over-forty workers than under-forty employees.” 440 F.3d at 1201. We explained that “this statistic ha[d] little significance” absent relevant “comparables“—i.e. allegations about the ages of the other employees. Id.; see also Stone, 210 F.3d at 1138 (“[I]n order for [the plaintiff] to establish a prima facie case of age discrimination, he must establish that [his employer] retained or placed younger employees in similar positions.“). In this case, plaintiffs allege that of the 24 workers hired to replace the terminated workers in January and February 2013, three men and no women were forty or older, whereas “a substantial percentage (fifteen or seventy-one percent) were in their twenties at the time of hire.” As with many of the statistical allegations in the complaint, this allegation is confusingly phrased, but it appears to state that fifteen of the new hires—71% of the 24 new hires—were in their twenties. So construed, this allegation that a significant proportion of the new hires were in their twenties lends further support to plaintiffs’ allegation of discrimination.

Accepting plaintiffs’ non-conclusory allegations as true, we conclude it is plausible that Affinity‘s termination policies resulted in a significant disparate impact on workers forty or older. Plaintiffs’ factual allegations with respect to their ADEA disparate impact claim were therefore sufficient to survive a motion to dismiss.¹⁷

V

Finally, we address plaintiffs’ disparate treatment claim under the ADEA. We review de novo the district court‘s grant of summary judgment in favor of Affinity on this claim. See Pippin, 440 F.3d at 1191. “Summary judgment is proper if the evidence, viewed in the light most favorable to the non-moving party, presents no genuine issue of material fact and the court finds the moving party is entitled to judgment as a matter of law.” Id.

“To establish a disparate-treatment claim under the plain language of the ADEA, . . . a plaintiff must prove that age was the ‘but-for’ cause of the employer‘s adverse decision.” Gross v. FBL Fin. Servs., Inc., 557 U.S. 167, 176 (2009). In the absence of direct evidence of age discrimination, we apply the three-step burden-shifting analysis set forth in McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973). See Jones v. Okla. City Pub. Sch., 617 F.3d 1273, 1278 (10th Cir. 2010). At the first step of the analysis, a plaintiff must establish a prima facie case of age discrimination. See id. If the plaintiff succeeds, “the burden of production then shifts to the employer to identify a legitimate, nondiscriminatory reason for the adverse employment action. Once the employer advances such a reason, the burden shifts back to the plaintiff to prove the employer‘s proffered reason was pretextual.” Id. (citation omitted).

A

The district court concluded that plaintiffs did not meet their burden of establishing a prima facie case of disparate treatment under the ADEA. In termination cases, the elements of a prima facie age discrimination case are typically that the plaintiff was “(1) within the protected class of individuals 40 or older; (2) performing satisfactory work; (3) terminated from employment; and (4) replaced by a younger person, although not necessarily one less than 40 years of age.” Adamson v. Multi Cmty. Diversified Servs., Inc., 514 F.3d 1136, 1146 (10th Cir. 2008). These “elements of a prima facie case under the McDonnell Douglas framework are neither rigid nor mechanistic, [and] their purpose is the establishment of an initial inference of unlawful discrimination warranting a presumption of liability in plaintiff‘s favor.” Id. This is particularly true in an age discrimination case. See Stone, 210 F.3d at 1139.

Affinity does not contest that plaintiffs established the first three elements of their prima facie case. At issue is the fourth element: whether plaintiffs have shown they were replaced by younger hires.¹⁸ Because the employees are grouped by position at the Casino and Affinity terminated fifty employees, there is no one-to-one correspondence between the employees who were terminated and those who replaced them.

Although this case does not involve a reduction in force, our precedents on that subject are instructive. In that context, we have held that when a terminated employee is “not replaced by someone so that the fourth factor can be analytically applied,” we adapt the McDonnell Douglas framework “to the particular type of adverse employment decision in question.” Greene v. Safeway Stores, Inc., 98 F.3d 554, 560 (10th Cir. 1996). With that adaptation, plaintiffs in reduction-in-force cases only need to “show that older employees were fired while younger ones in similar positions were retained.” Id.

Similarly, we adapt the McDonnell Douglas framework to the situation presented in this case. Because there is no clear one-to-one correspondence between the terminated employees and the new hires, we do not require each plaintiff to show he or she was individually replaced by a materially younger employee. Rather, plaintiffs need only demonstrate that for any given position, the new hires were materially younger than the terminated employees.

Plaintiffs offered the following evidence that the median age of the newly hired employees was significantly lower than that of the terminated employees:

Job Title	Median Age of Discharged Employees	Median Age of New Hires
Table Games Dealer	51.0	36.8
Cage Cashier	59.7	30.3
Casino Host	59.7	44.1
F&B Cashier	41.8	29.8

Image in original document— median age comparison table

Affinity argues the median-age analysis is inadmissible because it was created by plaintiffs’ counsel. It contends the analysis is therefore an unauthenticated out-of-court statement by a non-expert that cannot be considered at the summary judgment stage. We disagree.

“[W]e can consider only admissible evidence in reviewing an order granting summary judgment.” Johnson v. Weld Cty., Colo., 594 F.3d 1202, 1209 (10th Cir. 2010) (quotation omitted). Although “the form of evidence produced by a nonmoving party at summary judgment may not need to be admissible at trial, the content or substance of the evidence must be admissible.” Id. at 1210 (emphases and quotation omitted). The record reflects that plaintiffs’ counsel prepared the median-age analysis from spreadsheets Affinity provided in discovery that listed information about employees’ dates of birth, hire, and termination. Affinity does not argue that the underlying data is inadmissible. Rather, it appears to contend that the median-age analysis is inadmissible because it can only be admitted as expert testimony, and the expert‘s report submitted by plaintiffs did not contain such an analysis.

But the simple calculation of a median constitutes lay testimony under Federal Rule of Evidence 701 because it involves only basic arithmetic—one merely arranges numbers in order and picks the middle one (or averages the middle two). Cf. Ryan Dev. Co., L.C. v. Ind. Lumbermens Mut. Ins. Co., 711 F.3d 1165, 1170 (10th Cir. 2013) (no expert required where witnesses “used basic arithmetic, personal experience, and no outside expert reports” in performing calculations). The same is true of calculating age at the time of hiring or termination when given a birthdate and a date of hire or termination. Accordingly, we conclude that the district court properly considered the median-age analysis.

Affinity also contends that the data in the table does not support plaintiffs’ prima facie case because it includes terminated employees who are not plaintiffs in this litigation. Plaintiffs assert that because many of the job positions were interchangeable and many of the terminated employees were fired on the same date, it is impossible to ascertain who replaced whom. We agree. In such a situation, under “the flexible McDonnell Douglas approach,” Greene, 98 F.3d at 560, we may consider data including non-plaintiffs as well as plaintiffs.

Having determined that the median-age analysis presented above is relevant and admissible, we turn to whether the age differences adduced are sufficiently substantial to raise an inference of age discrimination. As shown in the table above, for each job title, the median age of the new hires was between 12 and 29 years younger than the median age of the discharged employees. Our sibling circuits have generally held that an age difference of ten or more years is sufficiently substantial, but an age difference of less than ten years is not. See Grosjean v. First Energy Corp., 349 F.3d 332, 336, 338 (6th Cir. 2003) (collecting cases from various circuits); France v. Johnson, 795 F.3d 1170, 1174 (9th Cir. 2015); Hartley v. Wis. Bell, Inc., 124 F.3d 887, 893 (7th Cir. 1997); cf. Munoz v. St. Mary-Corwin Hosp., 221 F.3d 1160, 1166 (10th Cir. 2000) (“[B]ecause plaintiff‘s replacement was only two years his junior—an obviously insignificant difference—the necessary inference of discrimination was precluded.“).

We agree with these decisions. In this case, the difference between the median ages of new hires and discharged employees ranged between 12 years and 29 years, depending on the job title. These disparities are sufficient to give rise to an inference of age discrimination. We thus conclude that plaintiffs established their prima facie case.

B

At the second step of the burden-shifting analysis, the employer bears the burden of production to identify a legitimate, nondiscriminatory reason for the adverse employment action. See Jones, 617 F.3d at 1278. In the Title VII context, we have explained that “the defendant does not at this stage of the proceedings need to litigate the merits of the reasoning, nor does it need to prove that the reason relied upon was bona fide, nor does it need to prove that the reasoning was applied in a nondiscriminatory fashion.” E.E.O.C. v. Flasher Co., 986 F.2d 1312, 1316 (10th Cir. 1992). “[T]his stage of the analysis only requires the defendant to articulate a reason for the discipline that is not, on its face, prohibited” and that is “reasonably specific and clear.” Id. at 1316 & n.4.

Plaintiffs argue Affinity did not meet its burden of production. They note that during their depositions, various managers could not remember the precise reasons for firing particular plaintiffs. But in its motion for summary judgment, Affinity set forth specific and detailed legitimate, nondiscriminatory reasons for its termination of each plaintiff, including attendance issues, violations of company policy, rudeness to customers, various performance mistakes, and attitude issues. In addition to deposition testimony, Affinity cited to performance reviews, evaluations, written warnings, and “Corrective Counseling Notices” reflecting the reasons why plaintiffs were terminated.¹⁹ We conclude that Affinity has met its burden of production.

C

At the third step of McDonnell Douglas, a plaintiff must show the employer‘s proffered reason for the adverse employment decision was pretextual. See Jones, 617 F.3d at 1278. “A plaintiff can show pretext by revealing weaknesses, implausibilities, inconsistencies, incoherencies, or contradictions in the employer‘s proffered legitimate reasons for its action such that a reasonable factfinder could rationally find them unworthy of credence and hence infer that the employer did not act for the asserted non-discriminatory reason.” Garrett v. Hewlett-Packard Co., 305 F.3d 1210, 1217 (10th Cir. 2002) (quotation and alteration omitted). “[T]he evidence which a plaintiff can present in an attempt to establish that a defendant‘s stated reasons are pretextual may take a variety of forms,” and “[a] plaintiff may not be forced to pursue any particular means of demonstrating that a defendant‘s stated reasons are pretextual.” Kendrick, 220 F.3d at 1230 (quotation and alterations omitted). When reviewing for pretext, “[w]e are mindful we must not sit as a super-personnel department that second-guesses the company‘s business decisions, with the benefit of twenty-twenty hindsight.” Tyler v. RE/MAX Mountain States, Inc., 232 F.3d 808, 813-14 (10th Cir. 2000).

The district court concluded that plaintiffs failed to present sufficient evidence to raise a genuine issue of material fact that Affinity‘s proffered legitimate, nondiscriminatory reasons are pretextual. We disagree. Plaintiffs contend that the proffered reasons are pretextual because they are post-hoc justifications not given when they were laid off. Post-hoc justifications for termination constitute evidence of pretext. See Plotke v. White, 405 F.3d 1092, 1103 (10th Cir. 2005); see also Appelbaum v. Milwaukee Metro. Sewerage Dist., 340 F.3d 573, 579 (7th Cir. 2003) (“One can reasonably infer pretext from an employer‘s shifting or inconsistent explanations for the challenged employment decision.“); Santiago-Ramos v. Centennial P.R. Wireless Corp., 217 F.3d 46, 56 (1st Cir. 2000) (“[A] method of establishing pretext is to show that [the employer]‘s nondiscriminatory reasons were after-the-fact justifications, provided subsequent to the beginning of legal action.“).

When each plaintiff was terminated, he or she received a “Personnel Action Form” stating either “Failed to pass Introductory Period” or simply “Introductory Period.” None of the Personnel Action Forms includes any other reason for termination. The Affinity Employee Handbook defines “Introductory Period” as “[t]he first 90 days of continuous employment with [Affinity]” and provides that employee performance “may be reviewed” after the end of the period. This period began in November 2012, after Affinity took over operations at the Casino.

For plaintiffs Christine Frappied,²⁰ Kathleen Greene, Georgean LaBute, John Roberts, Annette Trujillo, and Debbie Vigil, none of the nondiscriminatory reasons Affinity proffered pertain to actions they took during the Introductory Period. For plaintiffs Christine Gallegos, Joyce Hansen, and Kristine Johnson, Affinity does cite issues that arose during that time. Gallegos received a Corrective Counseling Notice on December 9, 2012 for “failing to use the proper forms” for a currency transaction. As for Hansen, Affinity stated that on January 12, 2013, it received a written statement from another employee that Hansen used profanity in front of a customer. Hansen also received written warnings for missing information and a signature on a tip count sheet on December 25, 2012, and for missing a signature on January 2, 2013. And Johnson failed to write a return date and time on her “count” on December 12, 2012.

Nevertheless, most of the conduct that Affinity cites as its basis for terminating Gallegos, Hansen, and Johnson occurred before the Introductory Period—in some instances several years before. Further, these three plaintiffs worked in the Cage Department, and the Cage Manager testified at a deposition that she did not recall evaluating her employees’ performance during the Introductory Period. She also testified that she was never told how employees would be evaluated during that period. The Controller—the Cage Manager‘s supervisor—confirmed that the Cage Manager was not asked to evaluate employees during the Introductory Period.

A jury considering this evidence could reasonably believe that Affinity lacks credibility. When Affinity fired plaintiffs, it gave only the vague explanation that they had failed to pass the ninety-day Introductory Period. For six of the nine plaintiffs, it could not point to a single instance of poor performance or infraction of the rules that occurred during that time. For the other three plaintiffs—Gallegos, Hansen, and Johnson—Affinity primarily relies on evidence from before the Introductory Period. Further, Gallegos, Hansen, and Johnson‘s manager testified that she could not recall evaluating her employees’ performance during that period. The inconsistencies between Affinity‘s contemporaneous stated reasons and its detailed post-hoc explanations for terminating plaintiffs could support a jury‘s finding that Affinity lacks credibility. See Tyler, 232 F.3d at 814 (“[W]hen the plaintiff casts substantial doubt on many of the employer‘s multiple reasons, the jury could reasonably find the employer lacks credibility.“).

Affinity responds that it chose not to share with plaintiffs its detailed reasons for termination at the time it made those decisions. It provides a spreadsheet maintained by Scott Nelson, the Casino‘s general manager, that lists specific reasons for terminating each employee. The spreadsheet was purportedly prepared contemporaneously with Affinity‘s termination decisions. It states that Gallegos, Johnson, Labute, and Vigil were terminated because of “Attitude / C[ustomer ]S[ervice] Complaints“; Hansen, Roberts, and Trujillo were terminated because of “Attitude / Morale / Perf[ormance]“; and Greene was terminated because of “Customer Serv[ice] Complaints.” It does not list Frappied.

The information listed on the spreadsheet is inconsistent with Affinity‘s proffered reasons for termination. For example, Affinity does not list any facts relating to customer service complaints among its reasons for terminating Greene and Johnson. Thus, the spreadsheet does not undercut our conclusion that there are genuine issues of material fact as to pretext.

Further, plaintiffs point to evidence that Affinity‘s evaluations were subjective. “Courts view with skepticism subjective evaluation methods.” Garrett, 305 F.3d at 1218. Although Affinity lists a litany of complaints and infractions relating to each plaintiff, it offers no evidence that it used objective criteria to evaluate its employees. For example, the manager of the Food and Beverage Department—where plaintiffs Frappied, Labute, and Vigil were employed—testified at a deposition that he could not recall consulting their files or looking at customer comments. He was also not given guidelines for evaluating employees during the Introductory Period; rather, he based his decisions on his own evaluations. Although “actual performance may constitute a legitimate basis for different treatment,” id., and “the use of subjective criteria does not suffice to prove intentional age discrimination,” Furr v. Seagate Tech., Inc., 82 F.3d 980, 987 (10th Cir. 1996) (emphasis added), we conclude that plaintiffs’ evidence is at least sufficient to raise an issue of fact regarding whether their performance was the true basis for their terminations.²¹

In sum, we conclude that when the evidence is construed in the light most favorable to plaintiffs, it suffices to create a genuine issue of material fact that Affinity‘s stated reasons for termination are pretextual.

VI

For the foregoing reasons, we AFFIRM the district court‘s dismissal of plaintiffs’ Title VII disparate treatment claim. With respect to its dismissal of plaintiffs’ Title VII and ADEA disparate impact claims, and its grant of summary judgment on the ADEA disparate treatment claim, we REVERSE the decision of the district court and REMAND for proceedings consistent with this decision. Plaintiffs’ Unopposed Motion to Seal Vol. IX of Plaintiffs-Appellants’ Appendix is GRANTED.

Notes

When plaintiffs filed suit at the district court, there were eleven plaintiffs. Plaintiff Judy Huck was voluntarily dismissed by the district court before it issued the orders now on appeal. Plaintiff Jennifer Ryan was voluntarily dismissed from this suit after entry of final judgment.

The Court acknowledged that Title VII also bars discrimination in cases in which sex is a “motivating factor,” but states that its analysis does not depend on the motivating-factor test. . Similarly, our analysis focuses only on the but-for causation standard.

We acknowledge that this framework requiring a comparison between male and female employees assumes that sex is binary. This case does not raise, and we do not address, sex discrimination involving intersex or gender non-binary individuals. Cf. (recognizing “inevitable inaccuracies of a binary sex policy” for intersex individuals).

Of course, a male plaintiff may also establish liability by showing he would not have been terminated but for his sex.

The Second and Sixth Circuits have acknowledged the issue but have not resolved it. See (“Having determined that Gorzynski has provided sufficient evidence of age discrimination to reach a jury, there is no need for us to create an age-plus-sex claim independent from Gorzynski‘s viable ADEA claim.“); (unpublished) (“[W]e decline the invitation to decide the ‘sex plus [age]’ charge partly because it is unnecessary for us to do so.“); .

See, e.g., ; ; ; ; ; , aff‘d on other grounds, . But see, e.g., .

Ryan, who was 39 years old when she was terminated, was one of the plaintiffs who brought the Title VII claim in the complaint. For the other plaintiffs, the “plus-” characteristic is being at least forty years old. But for Ryan, although the complaint alleges that she was the oldest beverage server employed by the Casino, it is unclear what the “plus-” characteristic is or how to compare her to a male employee who also shares the same “plus-” characteristic. In any event, Ryan was voluntarily dismissed from the suit before this appeal was filed, and plaintiffs do not challenge the dismissal of the Title VII claim as to her. We therefore do not address this issue.

Of course, if the complaint did include non-conclusory allegations that the older women in the group of terminated employees were fired because of age and sex, the older women could successfully state a claim for sex-plus-age discrimination.

In their response to Affinity‘s motion to dismiss, plaintiffs state that the “statistical disparity for females in the protected age group [is] 0.02074“—not 0.01231. They explain the discrepancy with the complaint by stating that “counsel was unable to access the online tool when analyzing the data” in the complaint.

Plaintiffs argued in their response to Affinity‘s motion to dismiss that “0.05%” is the “threshold” for statistical significance “for prima facie evidence of discrimination.” They cited no authority for this proposition and appear to have confused a p-value of 0.05, or 5%, with a p-value of 0.0005, or 0.05%. But they stated in a footnote in the same response that “[a] significance level of 0.05 establishes that the disparity would occur by chance less than 5% of the time.” It thus appears that plaintiffs argued that the threshold p-value for statistical significance is 0.05, or 5%, and any reference to “0.05%” is a clerical error.

The complaint includes one more allegation relating to the disparity between older men and older women. It states that of 24 new hires, three men and no women were forty or older. But absent an allegation about the number of people who applied, we do not know whether the disparity in hiring gives rise to an inference of discrimination.

Affinity does not challenge this conclusion.

The district court also dismissed because it concluded that plaintiffs did not specify which workers were in the protected class. But the ADEA protects only workers who are at least forty years old, § 631(a), and plaintiffs only bring their ADEA claims on behalf of employees who were at least forty when they were terminated.

Plaintiffs ask us to hold that the ADEA does not require plaintiffs to plead statistical data to establish a disparate impact claim. We do not address this issue because the statistical data plaintiffs allege is sufficient to raise a plausible inference of disparate impact under the ADEA.

Plaintiffs’ reply brief includes an amended table stating there were 33 retained employees under forty, implicitly acknowledging the discrepancy.

We acknowledge that this allegation is not contained in the complaint, but analysis of statistical data is not necessary to survive a motion to dismiss a disparate impact claim.

The complaint itself expresses skepticism about the disparate impact claim, stating that “[w]ithout complete data, which is discoverable in this action, Plaintiffs are unable to positively rule out the disparate impact of [Affinity‘s] facially neutral selection criteria on older workers as a class.” Affinity argues that because of this statement, the complaint does not plausibly state a disparate impact claim. But this statement hypothesizing about what discovery may show is not a specific factual allegation. Therefore, we disregard it. See (“[I]n ruling on a motion to dismiss, a court should disregard all conclusory statements of law and consider whether the remaining specific factual allegations, if assumed to be true, plausibly suggest the defendant is liable.“).

Plaintiffs urge us to hold that the fourth prong of the prima facie test can be satisfied if an ADEA plaintiff establishes that “the position remained open and the employer continued to seek applicants from persons of plaintiff[s‘] qualifications.” (alteration omitted) (quoting ). Because we ultimately conclude that plaintiffs established evidence of the fourth prong, we do not address this issue.

Plaintiffs contest the admissibility of one exhibit on which Affinity relies. But Affinity cited multiple other exhibits to support its proffered nondiscriminatory reasons for termination of each employee. We therefore do not address the admissibility of the exhibit.

Affinity states that Frappied was terminated in part because it learned that she intended to resign. Frappied confirmed that Affinity offered that reason when it fired her. But she also testified that just prior to her termination, she told her manager about medically ordered work restrictions due to a foot injury. Plaintiffs have thus established a genuine factual issue as to pretext for Frappied.

Additionally, for plaintiff Roberts, plaintiffs proffer evidence that Affinity‘s nondiscriminatory reasons for termination were not actually considered in his termination decision. Roberts was a Table Games employee, and the Table Games Director testified that in making his decision to terminate Roberts, he did not consider disciplinary actions. Affinity proffered these among its nondiscriminatory reasons for termination.