3:18-cv-08036•SUGG v. VIRTUSA et al
Texte intégral
NOT FOR PUBLICATION UNITED STATES DISTRICT COURT DISTRICT OF NEW JERSEY GWYNNE SUGG, individually and on behalf of all those similarly situated, Plaintiff, v. VIRTUSA, Defendant. Civil Action No. 18-08036 (GC) (JTQ) OPINION CASTNER, District Judge THIS MATTER comes before the Court upon three motions: Plaintiff Gwynne Sugg’s Motion for Class Certification pursuant to Federal Rule of Civil Procedure (Rule) 23 (ECF Nos. 141, 142); Defendant Virtusa’s Motion to Exclude the Expert Reports of David Lang, Ph.D. (ECF No. 144); and Defendant’s Motion to Exclude the Expert Report of Ronil Hira, Ph.D. (ECF No. 145). The parties briefed each of the Motions. (See ECF Nos. 142, 144-145, 150-151, 153, 157- 158, 160.) The Court has carefully considered the parties’ submissions and decides the matter without oral argument pursuant to Rule 78(b) and Local Civil Rule 78.1(b). For the reasons set forth below, and other good cause shown, Defendant’s Motion to Exclude the Expert Reports of David Lang, Ph.D. is GRANTED in part and DENIED in part; Defendant’s Motion to Exclude the Expert Report of Ronil Hira, Ph.D. is DENIED; and Plaintiff’s Motion for Class Certification is DENIED. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 1 of 44 PageID: <pageID>
2 I. BACKGROUND A. Procedural Background On February 28, 2018, Plaintiff filed a putative class action in the Superior Court of New Jersey, Law Division, Monmouth County. (ECF No. 1-1.) On April 19, 2018, Defendant removed the case to this Court. (ECF No. 1.) After retaining new counsel, Plaintiff filed a First Amended Complaint (FAC) on July 11, 2018. (ECF No. 7.) Plaintiff’s FAC contains three counts: disparate treatment on the basis of race in violation of 42 U.S.C. § 1981 (Count I); an unenforceable and unconscionable non-competition agreement in violation of New Jersey state law (Count II); and breach of contract and unjust enrichment in violation of New Jersey state law (Count III). (Id.)1 The first two Counts are brought by Plaintiff and the putative class, and the third Count is brought only on behalf of Plaintiff. (Id.) On September 7, 2018, Defendant moved to dismiss the FAC. (ECF No. 11.) The Court denied the Motion to Dismiss with respect to Counts I and III. (ECF No. 24.) Plaintiff voluntarily withdrew Count II. (Id. at 2 n.2.)2 The parties then moved into discovery. As is relevant here, Plaintiff served its Expert Reports—authored by David Lang, Ph.D. and Ronil Hira, Ph.D., respectively—on July 29, 2024. (ECF No. 135-1 at 6; ECF No. 139 at 6.)3 Virtusa served a rebuttal to Dr. Lang’s Report—authored by Paul White, Ph.D.—on March 10, 2025. (ECF No. 135-1 at 6; ECF No. 139 at 10.) On May 5, 2025, Dr. Lang filed an Expert Reply Report. (ECF No. 135-1 at 6; ECF No. 139 at 11.) The 1 The Court has subject matter jurisdiction over this action pursuant to 28 U.S.C. § 1331. 2 Page numbers for record cites (i.e., “ECF Nos.”) refer to the page numbers stamped by the Court’s e-filing system and not the internal pagination of the parties. 3 On September 18, 2024, Leo Sugg passed away. (ECF No. 121.) Following a consent motion, this Court substituted Gwynne Sugg for her deceased husband. (ECF No. 129.) Accordingly, the parties jointly proposed an Amended Scheduling Order on February 14, 2025 (ECF No. 125), which the Court entered on March 5, 2025, (ECF No. 126.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 2 of 44 PageID: <pageID>
3 parties completed fact and expert discovery on May 23, 2025. (See ECF No. 126.) After the completion of discovery, the parties filed the instant Motions. On May 28, 2025, Defendant filed a motion to strike Dr. Lang’s Expert Reply Report. (ECF No. 135.) On June 25, 2025, Plaintiff filed a class certification motion. (ECF No. 141.) On that same date, Defendant filed motions to exclude the expert reports of Drs. Lang and Hira. (ECF Nos. 144, 145.) On December 23, 2025, the Honorable Justin T. Quinn, U.S.M.J., granted Defendant’s motion to strike Dr. Lang’s Expert Reply Report, and Plaintiff did not appeal that decision. (ECF No. 166.) B. Factual Background 1. Facts Relevant to the Motion for Class Certification a. Virtusa’s Business Model Virtusa provides information technology (IT) and consulting services to customers across the globe. (ECF No. 146 at 3; ECF No. 153-3 at 1.)4 The company was founded in Sri Lanka but maintains its corporate headquarters in Massachusetts. (ECF No. 142 at 7; ECF No. 146 at 3.) As of 2024, Virtusa employs over 27,000 individuals, (ECF No. 142-5), and nearly 70% of those individuals live in India. (ECF 142-4 at 27; 142-5 at 3.) Approximately 2,700 employees live in the United States. (ECF No. 142-6 at 7.) Virtusa employs an “offshore-onsite” business model to serve its United States clients. Under this model, most of the work performed for these clients is done “offshore” in India or Sri Lanka. (ECF No. 153-2 at 1.) Some of the work is done “on-site” at customer facilities in the United States. (Id.) Regardless, employees who work on these client 4 Because the Court finds that Dr. Hira’s Expert Report (ECF No. 146) is admissible under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), see supra Section III.A.2., the Court may rely on it for determining the facts for the purposes of class certification. See In re Blood Reagents Antitrust Lit., 783 F.3d 183, 187 (3d Cir. 2015) (“[A] plaintiff cannot rely on challenged expert testimony, when critical to class certification, to demonstrate conformity with Rule 23 unless the plaintiff also demonstrates, and the trial court finds, that the expert testimony satisfies the standard set out in Daubert.”). Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 3 of 44 PageID: <pageID>
4 projects—as opposed to “selling, general, and administrative” employees who work on internal Virtusa matters—are referred to as “billable employees.” (ECF No. 153-3 at 1-2.) These billable employees make up 90% of Virtusa’s workforce. (Id. at 2.) For all employees, Virtusa has a six- Tier job hierarchy based on employee roles and years of experience. Tier 4 (campus hires) is the lowest, followed by Tier 3 (two to five years’ experience), Tier 2 (five to eight years), Tier 1 (seven to twelve years), Tier 0 (ten to sixteen years), and Global Tier 0 (the highest Tier, comprised of Senior Directors and Vice Presidents, ranging in experience from fifteen to thirty years). (ECF No. 142-4 at 34-35.) Within this hierarchy, there are technology, delivery, and consulting tracks for billable employees, and sales and administration tracks for non-billable employees. (Id. at 31- 32, 34.) b. Virtusa’s Hiring Practices When a billable role becomes available,5 the company will either staff the role with a current employee or an external candidate. (Id. at 16, 22-23.) Virtusa generally first looks to availability among internal employees before hiring externally. (See, e.g., ECF No. 142-10 at 6- 8, 12.) When hiring externally, Virtusa’s Talent Acquisition Group employs various “modes of hiring” to attract candidates. (ECF No. 153-4 at 23-24.) It finds candidates through its own job portal postings; LinkedIn and other electronic job boards; external recruiters; external vendors who provide personnel; and internal recruiters. (Id. at 24-25.) Once candidates are sourced, they go through three rounds of interviews. First, technical recruiters conduct screening interviews to ensure applicants “will fit the profile.” (Id. at 25.) Second, “strategic business unit[] representatives”—similar to hiring managers—conduct interviews. (Id. at 26.) Third, practice 5 The hiring process for non-billable employees proceeds differently. See supra Section I.B.1.f (citing ECF No. 153-3 at 3-4). Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 4 of 44 PageID: <pageID>
5 managers—“high-end technology experts”—conduct interviews. (Id.) During the course of these interviews, applicants are asked whether they would be open to relocation and open to travel “to be sure” the applicant is aligned with the needs of the client. (Id. at 28.) If the applicant makes it through these three rounds, Virtusa presents the applicant to the client. (Id. at 28.) Between 80% and 90% of the time, the client then interviews the applicant. (Id. at 29.) The client and practice manager both play a “critical role” in deciding whether to extend an offer to an external applicant. (Id. at 32.) Virtusa also considers affordability when making its external hiring decisions. (Id.) During the application process, Virtusa does not collect documents from candidates other than their resumes, though it does inform candidates that they must have work authorization, and it does not ask candidates to self-identify their race. (Id. at 27-30; ECF No. 142-4 at 9-10.) c. Virtusa’s Promotion Practices Multiple players are involved in Virtusa’s promotion process. (ECF No. 153-3 at 3-4; ECF No. 153-4 at 53; see also ECF No. 142-16; ECF No. 142-18.) The process begins when an individual manager makes a promotion recommendation, and that recommendation is forwarded to the next level of management or to Human Resources personnel who make the ultimate decision. (ECF No. 153-3 at 3.) The promotion recommendation is based on a number of considerations: (1) employees may not be promoted if they lack required technical certifications, (2) an employee must have completed a required number of months of service in their current role, which varies depending on the tier an employee is in; (3) an employee must meet performance rating requirements; (4) an employee’s utilization levels and contribution to projects are considered; (5) an employee’s initiative, learning ability, and general growth potential are considered; (6) supervisor and client feedback; and (7) performance against expected goals[.] (Id. at 3-4.) But these factors change depending on the Tier in which the employee is seeking the promotion. (Id. at 4.) For junior level positions, promotions are based on tenure and performance Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 5 of 44 PageID: <pageID>
6 ratings, which turn on the employee’s potential. (ECF No. 153-4 at 53-54.) These employees— ones moving from Tier 4 to Tier 3—can be considered for promotion generally within 12 to 18 months. (Id. at 54; ECF No. 142-18 at 4.) As the employee moves up, the focus becomes less on potential and more on whether the individual is in fact contributing to the practice, and the employee can be considered for promotion generally after 24 to 36 months. (ECF No. 153-4 at 54; ECF No. 142-18 at 4.) To be promoted to the top positions—i.e., into Tiers 1 and 0—tenure and performance are key parameters, but as Virtusa’s Chief People Officer testified, affordability and a business need for the role are also “super” important. (ECF No. 153-4 at 54; ECF No. 142- 18 at 3.) Performance rating is a key factor throughout, but it applies differently depending on the type of job sought and the corresponding Tier. Billable employees are reviewed, at least in part, based on a qualitative scale. (ECF No. 142-18 at 3-4.) Employees who receive the top two scores on this scale—“Outstanding” and “Exceeds Expectation Consistently”—are eligible to be promoted at an expedited pace. (Id.) But billable employees seeking to be promoted within Tier 0 cannot do so unless they receive either of these two ratings. (Id. at 4.) By contrast, non-billable roles—like sales—do not employ the qualitative scale. (ECF No. 153-3 at 6.) Instead, these employees are primarily evaluated based on their “revenue impact” and “margin.” (Id.) Regardless of the type of role, “360-degree feedback”—that is feedback by not just managers but also peers and subordinates—is implemented when evaluating promotions into and within Tier 0, but not at more junior Tiers. (ECF No. 153-3 at 4; ECF No. 153-4 at 55-56.) Promotion into or within Tier 0 may also be accompanied by an interview with an internal or external review panel. (ECF No. 142-18 at 3; ECF No. 153-3 at 4.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 6 of 44 PageID: <pageID>
7 d. Virtusa’s Termination Practices Virtusa’s termination procedures are related to its “benching” practices. Being “on the bench” refers to a billable employee’s status when he or she is not currently staffed on a billable project. (ECF No. 153-4 at 35.) Non-billable employees cannot be on the bench because they are not staffed on client projects. (Id. at 49.) A billable employee may end up on the bench for three reasons. First, and most commonly, their previous project has come to an end. (Id. at 48.) Second, the client cancels a new project that the employee was already staffed on. (Id.) Third, the client requests the employee be removed from the project because of dissatisfaction with the employee’s performance. (Id.) If an employee is on the bench for too long, they run the risk of termination. (See, e.g., ECF No. 142-29 at 8-9.) e. Virtusa’s Visa Policies As of 2020, more than half of Virtusa’s United States-based employees are H-1B or L-1 visa holders. (ECF No. 146 at 5.) H-1B visas allow companies to sponsor employees in specialty occupations requiring application of highly specialized knowledge and attainment of at least a bachelor’s degree or equivalent. (Id. at 6-7.) See also 8 C.F.R. § 214.2(h)(1)(ii)(B). L-1 visas allow multinational corporations to transfer a worker from a foreign office to a United States office. (ECF No. 146 at 10.) See also 8 C.F.R. § 214.2(l). L-1B visas apply to transferees with “specialized knowledge” and L-1A apply to transferees who are executives or managers. (ECF No. 146 at 10.) It is Virtusa’s alleged practices pertaining to visa-holding employees, and the possibility of discriminatory treatment in implementation of those policies on the basis of race, that are primarily at issue here. Virtusa’s alleged practices largely fall within two buckets: visa preparation and visa utilization. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 7 of 44 PageID: <pageID>
8 The clearest example of Virtusa’s visa preparation policy is its alleged implementation of “Project Catalyst.” Under this program, introduced in 2013, Virtusa sought to increase the percentage of employees in the United States that were H1-B or L-1 visa holders to decrease costs. (See generally ECF No. 142-21 (Virtusa presentation outlining Project Catalyst).) Accordingly, and following the lead of its competitors, Virtusa considerably increased its number of visa applications. (See ECF No. 142-25 at 3; ECF No. 146 at 21.) And it took a proactive approach— submitting more applications than it could use and filing them before specific positions opened up. (See ECF No. 146 at 19; ECF No. 142-50 at 6, ECF 142-52 at 2.) By 2014, more than 15% of Virtusa’s United States employees were H1-B or L-1 visa holders, by 2016, more than 50%, and by 2018 it reached 65%. (ECF No. 146 at 21.) Having secured a large number of visa-holding employees, Virtusa ensured its employees used those visas. Under the “Cost Per Consultant” initiative, Virtusa sought to bring visa-ready employees to the United States. (See ECF No. 142-31 (Virtusa presentation on Cost Per Consultant initiative).) Accordingly, it “[p]roactively train[ed] visa ready [offshore employees] for onsite roles.” (ECF No. 142-32 at 2.)6 And pursuant to Virtusa’s rotation policy, implemented in 2014, 6 Defendant argues that Virtusa never implemented the Cost Per Consultant initiative, and that it was only recommended by consulting firm McKinsey & Company, which was hired to identify “market strategy” and “profitable growth” opportunities. (ECF No. 153 at 20; ECF No. 153-4 at 46.) In support, Defendant cites to the deposition transcript of Virtusa’s former Global Head of Human Resources. (ECF No. 153 at 20.) In that deposition, when asked whether Virtusa implemented the program, the employee responded: “not really to the depth and the number” that McKinsey recommended, and to the extent Virtusa “might have implemented” the program, that implementation was “very thin.” (ECF No. 153-4 at 62-63.) “[T]he Court need not determine at [the class certification] stage whether [Virtusa] actually ha[d]” the Cost Per Consultant initiative. Palmer v. Cognizant Tech. Sols. Corp., No. 17-6848, 2022 WL 18214014, at *27 (C.D. Cal. Oct. 27, 2022). Instead, “[f]actual determinations supporting Rule 23 findings must be made by a preponderance of the evidence.” In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 307 (3d Cir. 2008). Plaintiff has satisfied that burden with respect to the notion that Virtusa implemented some version of the initiative. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 8 of 44 PageID: <pageID>
9 if an employee was staffed on a project for more than 18 months and operated at a margin of under 30%, then Virtusa would implement a plan to increase that margin or replace the employee with a “lower cost onsite hiring” or a “visa ready” offshore employee. (ECF No. 142-40 at 2.) This rotation policy appeared to apply to both visa-holding and non-visa-holding United States-based employees. (ECF No. 153-4 at 45.) Finally, Virtusa implemented a new bench policy following the influx of visa-holding employees. Under this policy, Virtusa helped visa-holding employees identify new opportunities after only one day on the bench but waited seven days to assist non- visa employees. (See ECF No. 142-29 at 8-9 (Virtusa presentation on bench policy).) If a visa- holding employee was still on the bench after 60 days, the employee would be “immediately sen[t] . . . back offshore.” (Id. at 9.) The implication being that a visa-holder is still valuable to Virtusa because Virtusa could pay that employee significantly less in Sri Lanka or India. (ECF No. 146 at 23-24.) By contrast, a non-visa holder that remained on the bench for 90 days would be subject to a discussion on “alternatives and options”—suggesting the possibility of termination. (ECF No. 142-29 at 8.) f. Mr. Sugg’s Experience Leo Sugg, a white male from the United States, worked for Virtusa from 2014 to 2017 as Vice President and the Global Account Executive for one of Virtusa’s clients, American International Group, Inc. (AIG). (ECF No. 142-11 at 6, 22; ECF No. 143-4 at 4, 7-8.) Virtusa labels employees like Mr. Sugg “Client Partners” because they are responsible for managing the relationship with one client. (See ECF No. 153-3 at 2.) Client Partners are not hired through the usual three-round interview process. (Id. at 3.) Instead, the process “proceeds on an as needed basis and involves the discretion of individual executives and personnel at Virtusa.” (Id.) Accordingly, Mr. Sugg was hired by Executive Vice President James Francis because of Sugg’s Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 9 of 44 PageID: <pageID>
10 prior experience working with AIG. (ECF No. 143-4 at 1, 3.) Mr. Sugg’s role was “principally relationship driven, business development driven, making sure [AIG] was happy.” (ECF No. 142- 11 at 7.) He was not a billable employee. (See ECF No. 153-3 at 2.) During his tenure, Mr. Sugg was not promoted, nor did he seek a promotion. (ECF No. 142-11 at 25; ECF No. 153 at 42; ECF No. 157 at 18.) Mr. Sugg could not name a position that existed at Virtusa to which he could have been promoted, though he testified that Virtusa “could have gotten creative” and “given [him] more responsibility.” (ECF No. 142-11 at 27.) Mr. Sugg also testified that during his tenure, he was told by Virtusa’s President, Raj Rajogpal, that Mr. Sugg did not “look like someone who sells technology” and should “find someone from India to work with [him].” (Id. at 22-23.) Francis terminated Sugg in 2017 because “Virtusa’s business with AIG fell dramatically.” (ECF No. 143-4 at 6.) Plaintiff contends that during this period Mr. Sugg “still performed well.” (ECF No. 142 at 19.) Mr. Sugg was replaced by Sameer Menon, but when Menon similarly failed to grow the AIG business, he was transitioned to a new role in banking rather than terminated. (ECF No. 142-11 at 26-27.) Mr. Sugg stated that he suffered “stress” and “anxiety associated with switching employers” and sought help from two mental health professionals. (Id. at 3-5.) 2. Facts Relevant to the Motions to Exclude Expert Reports Defendant challenges the admissibility of two of Plaintiff’s experts: (1) Dr. Lang (labor economics) and (2) Dr. Hira (business models of information technology services companies). (See ECF Nos. 144, 145.) Plaintiff relies on Dr. Lang’s Reports for statistical evidence of the disparate treatment non-South Asians faced as a result of Virtusa’s policies, and he relies on Dr. Hira’s Report to contextualize Virtusa’s actions. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 10 of 44 PageID: <pageID>
11 a. Dr. David Lang Plaintiff retained Dr. Lang7 to analyze employment data provided by Virtusa and opine on whether racial disparities existed in Virtusa’s workforce with respect to hiring, promotion, and termination, and if so, whether those disparities were statistically significant. (ECF No. 144-4 at 3.) Dr. Lang concluded that there was a “significant hiring disparity between White and Asian hiring rates.” (Id. at 15.)8 Specifically, Dr. Lang found that “[r]egardless of the number of applicants for a position, the probability that the hired applicant is Asian is very high (84.6% to 93.2%)” and that “[w]hen the number of applicants for a position is relatively high (11 or more applicants), the likelihood of the hired employee being Asian increases significantly” to 93.2%. (Id.) This disparity also existed within Virtusa’s job Tiers: “Regardless of job [T]ier, the probability that the hired applicant is Asian is very high (59.6% to 93.8%).” (Id.) Dr. Lang provided no data on or analysis of the applicant pool. (See generally id. at 5-6.) As for promotions, Dr. Lang found that “White employees are significantly less likely to receive employee ratings of ‘Exceeds Expectations’ (the second highest rating) compared to Asian employees.” (Id. at 15.) White employees were more likely to receive the highest rating, but the difference was not statistically significant. (Id. at 7.) Based on an analysis of less than 20 white employees, Dr. Lang also found that white employees who started at Tier 4 were significantly less likely to be promoted to Tiers 3, 2, and 1 than were Asian employees. (Id. at 9, 15.) However, 7 Dr. Lang holds a Master of Arts degree and Doctorate in economics from Washington University in St. Louis. (ECF No. 144-4 at 3.) Dr. Lang also holds a Bachelor of Arts degree in economics from Stanford University. (Id.) Dr. Lang is a Full Professor and Chair of the Economics Department at California State University, Sacramento. (Id.) 8 Dr. Lang excluded other races from his analyses because “the number of observations was too small.” (ECF No. 144-4 at 3.) As discussed supra, Dr. Lang focused on Asian employees rather than South-Asian employees—the reference group identified in the FAC, (ECF No. 7), and the proposed classes, (ECF No. 141 at 3.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 11 of 44 PageID: <pageID>
12 there was no overall significant difference in the percentage of white versus Asian employees who were never promoted. (Id. at 8.) Dr. Lang attributed non-significance to a small sample size: there were only approximately 17 white individuals not promoted. (See id.) Finally, with respect to terminations, Dr. Lang found that “White employees at Virtusa were 2.19 times more likely to be involuntarily terminated than Asian employees,” employees with visas were 1.68 times less likely to be involuntarily terminated than employees without a visa, and that “[a]fter accounting for gender, job [T]ier and age, being Asian (as opposed to being White) leads to an 8.2 percentage point reduction in the probability of being terminated involuntarily.” (Id. at 12-13, 15.) Dr. Lang drew from a sample of approximately 18 white individuals, and 101 non-visa holding individuals, who were involuntarily terminated. (See id. at 12.) Dr. Lang conducted his analyses based off of data provided by Virtusa. Sometimes the data included the Virtusa employee’s race and nationality, sometimes it included just race, sometimes just nationality, and sometimes neither, which is why sample sizes for different types of analyses varied. (Id. at 4.) Because the study—and the underlying 42 U.S.C. § 1981 claim—are based on race rather than nationality, Dr. Lang used nationality as a proxy for race in all of his analyses when data about race was unavailable but nationality was available. (Id.) Dr. Lang only implemented this proxy system for employees from India or Sri Lanka because Dr. Lang found that—where race and nationality data was available—99.6% of Indian and 100% Sri Lankan individuals also identified as Asian, whereas American employees were too racially diverse to implement the proxy. (Id.) Beyond this common thread, Dr. Lang employed slightly a different methodology for his hiring, promotion, and termination analyses. To opine on Virtusa’s hiring practices, Dr. Lang employed a “name-matching” method. (ECF No. 144-3 at 10-11.) Based on the data provided to him, 4,420 employees were hired by Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 12 of 44 PageID: <pageID>
13 Virtusa between October 31, 2018 and March 12, 2024. (ECF No. 144-4 at 5.) Dr. Lang wrote that for this dataset, “race and nationality [were] seldom reported,” so he employed name matching. (Id.) Under this method, Dr. Lang compared the first and last names of applicants with the names of employees—whose race was known to Virtusa—that were coded as non-Asian or Asian. (Id.) For example, Dr. Lang testified in his deposition that if an applicant were named “Rajesh Koothrappali,” and there were Asian employees with the first name “Rajesh” and the last name “Koothrappali,” then that applicant would be coded as Asian so long as there was not a non- Asian employee with the first name “Rajesh” or the last name “Koothrappali.” (ECF No. 144-3 at 12-13.) For the applicant to be coded as Asian, there would not need to be an employee named “Rajesh Koothrappali”—there would just need to be at least one employee with the first name “Rajesh” and another with the surname “Koothrappali.” (Id. at 13.) Conversely, Dr. Lang testified that if an applicant were named “Rajesh Smith” but there were non-Asian employees named “David Smith” or “Mary Smith,” then “Rajesh Smith” would not be coded as Asian. (Id.) Defendants raised significant concerns with Dr. Lang’s Report and deposition testimony. First, Dr. Lang testified that because of the incomplete race and nationality data related to hiring, he “was unable to perform” an “error analysis.” (ECF No. 150-3 at 6.)9 Second, Dr. Lang did not “distinguish between South Asian and other forms of Asian ethnicity,” but Dr. Lang stated that this concern was “theoretical[]” because “the vast majority of the data . . . happened to be of two particular [countries: India and Sri Lanka].” (ECF No. 144-3 at 13-14.) Third, Dr. Lang stated that the specific method he employed “does not exist out in the world” but noted that “[o]ther individuals have used name matching techniques.” (Id. at 15.) While Dr. Lang could not point to 9 In Plaintiff’s briefing, Plaintiff cites one study for the proposition that name-matching studies have an error rate of between 4% and 30%. (See ECF No. 150 at 13.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 13 of 44 PageID: <pageID>
14 an authoritative treatise on name matching or specific peer-reviewed studies that have employed name matching, he stated that it was “almost certain” that he had come across peer-reviewed studies that employed name matching given that the articles he reads are almost always peer reviewed. (ECF No. 150 at 7-8.)10 Finally, Dr. Lang did not control for certain variables. (See ECF No. 144-3 at 17, 33.) For example, he did not consider the racial composition of the applicant pools or the racial composition of the job market in his Report. (See ECF No. 144-1 at 14; ECF No. 144-4 at 5-6; ECF No. 150-2 at 12.) As for promotions, employee ratings, and job flow, Dr. Lang drew on data for full-time employees where race information was available or ascertainable based on the proxy system. (ECF No. 144-4 at 7.) For each of these analyses, Dr. Lang used t-tests, “inferential statistic[s] used to determine if there is a statistically significant difference between the means of two variables,” to arrive at conclusions. (Id. at 5.) However, neither in his Report nor in his deposition does Dr. Lang mention controlling for any potentially relevant factors beyond race in conducting his analyses. (See ECF No. 144-3 at 34-36.) Finally, as for termination, Dr. Lang similarly used t-tests to find several statistically significant differences involving employees terminated after July 11, 2014. (ECF No. 144-4 at 11- 13.) Dr. Lang does not appear to control for any other factors in making most of these assessments, though he did conduct one logistic regression that controlled for gender, job Tier, and age in evaluating race-based discrimination. (Id. at 14.) 10 In Plaintiff’s briefing, Plaintiff cites several peer reviewed name-matching studies. (See ECF No. 150 at 12-13.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 14 of 44 PageID: <pageID>
15 b. Dr. Ronil Hira Plaintiff retained Dr. Hira11 to analyze Virtusa’s business model and discuss how that model impacts Virtusa’s employment practices. (ECF No. 146 at 3.) In preparation for his Report, Dr. Hira reviewed 60 documents produced by Virtusa, deposition transcripts, and publicly available data, including government visa data and Virtusa’s Securities and Exchange Commission (SEC) filings. (ECF No. 146 at 2, 26-28.) Dr. Hira began his Report by providing an overview of Virtusa. Among other things, he noted “[f]or its onsite workforce, Virtusa relies heavily on importing foreign employees from India and Sri Lanka who require a visa.” (Id. at 5.) Dr. Hira recounted that employers are required to pay those employees with H-1B visas as much as they pay similarly experienced non-visa workers in the United States. (Id. at 5-6.) Only 85,000 new H-1B visas may be issued each year, and demand has far exceeded that cap. (Id. at 6.) Thus, the United States has implemented a lottery system for those spots. (Id.) Dr. Hira opined, based on publicly available data and documents obtained during discovery, that Virtusa submits more visa applications than it needs, with the knowledge that only a portion of them will be successful. (Id.) Dr. Hira reported that Virtusa is one of a series of employers that “improperly attest[] on the H-1B visa petition that there is an 11 Dr. Hira holds a doctorate in Public Policy from George Mason University. (ECF No. 146 at 30.) He also holds a Master of Science from George Mason and a Bachelor of Science from Carnegie-Mellon University, both in Electrical Engineering. (Id.) Dr. Hira is a professor of political science at Howard University and a research associate at the Economic Policy Institute. (Id. at 2.) Dr. Hira has published two books, three peer reviewed articles, six book chapters, and testified seven times before the United States Congress and twice before the Maryland State House of Delegates on the topics of business models and employment practices of information technology service companies. (Id.) Dr. Hira has also published dozens of policy briefs on these topics. (Id.) Dr. Hira served as an expert in Palmer v. Cognizant Technology Solutions Corp., Civ. No. 17- 06848 (C.D. Cal.). (Id. at 2-3.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 15 of 44 PageID: <pageID>
16 actual open position in the U.S. for the sponsored employee to fill when, in reality, no such position exists.” (Id. at 9.) Dr. Hira then explained why companies like Virtusa employ these visa policies. Dr. Hira concluded that “from [his] experience, IT outsourcing firms like Virtusa have adopted workforce strategies designed to predominantly fill U.S. positions with visa workers from countries in South Asia, rather than hiring Americans locally, because of deep-seated racial and ethnic preferences, rather than any business necessity.” (Id. at 11.) Dr. Hira drew this conclusion based on several facts. First, he opined that management and senior members of IT outsourcing firms are generally comprised of South Asians, like Virtusa, and they prefer to hire employees who are racially and ethnically similar to them. (Id. at 12.) Second, Dr. Hira stated that management prefers to hire visa workers because of the bargaining power: H1-B employees will likely not leave Virtusa because the employees would have to return to 10% wages in India or Sri Lanka, as compared to United States wages, so management can exert control over their working condition in an “indenturing”-like manner. (Id. at 12-13.) Indeed, as Dr. Hira opined, Virtusa requires visa employees to sign an “HR bond” before processing their visas, and if the employee leaves the company within less than two years, the employee is required to re-pay the bond amount. (Id. at 13.) Thus, Dr. Hira opined that even though Virtusa is required to pay H1-B visa employees at equal rates to U.S. employees, Virtusa pays them less. (Id. at 14.) Dr. Hira noted that these cost- saving measures came from McKinsey & Company recommendations. (Id.) II. LEGAL STANDARD A. Expert Testimony Federal Rule of Evidence 702 governs the admissibility of expert testimony. The Rule reads as follows: Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 16 of 44 PageID: <pageID>
17 A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if the proponent demonstrates to the court that it is more likely than not that: (a) the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert’s opinion reflects a reliable application of the principles and methods to the facts of the case. Fed. R. Evid. 702. Under Rule 702, “[d]istrict courts are tasked with a ‘rigorous gatekeeping function,’” to ensure that expert testimony is both relevant and reliable. Cohen v. Cohen, 125 F.4th 454, 460 (3d Cir. 2025) (quoting Elcock v. Kmart Corp., 233 F.3d 734, 744 (3d Cir. 2000)). Notwithstanding, “Rule 702 . . . has a liberal policy of admissibility.” Pineda v. Ford Motor Co., 520 F.3d 237, 243 (3d Cir. 2008) (quoting Kannankeril v. Terminix Int’l, Inc., 128 F.3d 802, 806 (3d Cir. 1997)). Expert testimony will be admissible if: (1) “the expert is qualified”; (2) “the proposed testimony is reliable and concerns matters requiring scientific, technical, or specialized knowledge”; and (3) “the expert’s testimony is ‘sufficiently tied to the facts of the case,’ . . . so that it ‘fit[s]’ the dispute and ‘will assist the trier of fact[.]’” Cohen, 125 F.4th at 460 (quoting Daubert, v. Merrell Dow Pharms., Inc., 509 U.S. 579, 591 (1993)). The party proposing the expert testimony must show that each prong is satisfied by a preponderance of the evidence. Oddi v. Ford Motor Co., 234 F.3d 136, 144 (3d Cir. 2000). Importantly, the “gatekeeping function is necessarily ‘flexible,’ . . . granting district courts ‘latitude in deciding how’ these requirements are met.” Cohen, 125 F.4th at 460 (first quoting Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 17 of 44 PageID: <pageID>
18 Daubert, 509 U.S. at 594, and then quoting Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152 (1999)). “Such ‘discretionary authority’ permits a district court to ‘decide whether or when special briefing or other proceedings are needed to investigate reliability.’” Id. (quoting Kumho Tire Co., 526 U.S. at 152). 1. Prong One – Qualification “Qualification refers to the requirement that the witness possess specialized expertise.” Schneider ex rel. Est. of Schneider v. Fried, 320 F.3d 396, 404 (3d Cir. 2003). The basis for this specialized expertise “can be practical experience as well as academic training and credentials.” Elcock v. Kmart Corp., 233 F.3d 734, 741 (3d Cir. 2000) (quoting Waldorf v. Shuta, 142 F.3d 601, 628 (3d Cir. 1998)). While courts have interpreted the special expertise requirement “liberally,” “at a minimum, a proffered expert witness . . . must possess skill or knowledge greater than the average layman.” Elcock, 233 F.3d at 741 (quoting Waldorf, 142 F.3d at 625). Importantly, however, an expert does not have to “be the best qualified” or “have the specialization that the court considers most appropriate.” Pineda, 520 F.3d at 244. 2. Prong Two – Reliability The “reliability requirement ensures that an expert’s testimony is ‘based on the methods and procedures of science, not on subjective belief and unsupported speculation.’” Cohen, 125 F.4th at 461-62 (quoting Karlo v. Pittsburgh Glass Works, LLC, 849 F.3d 61, 80-81 (3d Cir. 2017)). “[A]dmissibility does not hinge on ‘whether a particular scientific opinion has the best foundation, or even whether the opinion is supported by the best methodology or unassailable research.’” Id. at 462 (quoting In re TMI Litig., 193 F.3d 613, 665 (3d Cir. 1999)). Rather, courts look to whether the expert’s testimony is supported by “good grounds.” Id. (quoting UGI Sunbury LLC v. A Permanent Easement for 1.7575 Acres, 949 F.3d 825, 834 (3d Cir. 2020)). In conducting this analysis, courts are to consider “all aspects of an expert’s testimony: the methodology, the Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 18 of 44 PageID: <pageID>
19 facts underlying the expert’s opinion, [and] the link between the facts and the conclusion.” ZF Meritor, LLC v. Eaton Corp., 696 F.3d 254, 291 (3d Cir. 2012) (quoting Heller v. Shaw Indus., Inc., 167 F.3d 146, 155 (3d Cir. 1999)). “The court’s focus must be on the principles and methods, rather than on the ultimate conclusions reached.” Milltown-Ford Ave. Redevelopment Agency v. SB Bldg. Assocs., L.P., Civ. No. 19-21494, 2024 WL 3690776, at *16 (D.N.J. Aug. 7, 2024) (citing In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 742 (3d Cir. 1994)). Because, however, “‘conclusions and methodology are not entirely distinct from one another,’ . . . the court may conduct ‘at least a limited review of the expert’s conclusions “in order to determine whether they could reliably flow from the facts known to the expert and the methodology used.”’” Id. (first quoting General Elec. Co. v. Joiner, 522 U.S. 136, 146 (1997), and then quoting In re Human Tissue Prods. Liab. Litig., 582 F. Supp. 2d 644, 656 (D.N.J. 2008)). As the Third Circuit recently reiterated, “there is no ‘definitive checklist or test,’” for courts to employ when analyzing what constitutes “good grounds.” Cohen, 125 F.4th at 462 (quoting Daubert, 509 U.S. at 593). Nevertheless, along with “any other[ factors] that are relevant,” courts consistently consider the following factors in determining the reliability of a proposed expert: (1) whether a method consists of a testable hypothesis; (2) whether the method has been subject to peer review; (3) the known or potential rate of error; (4) the existence and maintenance of standards controlling the technique’s operation; (5) whether the method is generally accepted; (6) the relationship of the technique to methods which have been established to be reliable; (7) the qualifications of the expert witness testifying based on the methodology; and (8) the non-judicial uses to which the method has been put. In re Paoli, 35 F.3d at 742 & n.8 (citing Daubert, 509 U.S. at 591-94; United States v. Downing, 753 F.2d 1224, 1238-39 (3d Cir. 1985); see also Cohen, 125 F.4th at 462 (noting that “some analysis of these factors is necessary” (quoting UGI Sunbury, 949 F.3d at 834)). Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 19 of 44 PageID: <pageID>
20 “Although these factors are neither exhaustive nor applicable in every case, they provide a convenient starting point for analyzing” reliability. Kannankeril, 128 F.3d at 806-07. But in non- scientific cases, the relevant reliability concerns “may focus upon personal knowledge or experience,” rather than “scientific foundations.” Kumho Tire Co., 526 U.S. at 150; see also United States v. Price, 458 F.3d 202, 212 (3d Cir. 2006) (“It is settled law that an expert may testify about common behavior patterns in a profession or subculture.”); Wolfe v. Allstate Prop. & Cas. Ins. Co., Civ. No. 10-800, 2012 WL 12929813, at *5 (M.D. Pa. July 23, 2012) (“While rate of error, publication,[] peer review[,] and testing are imperative in ascertaining the reliability of a new scientific methodology . . . their application to testimony regarding customs and practices in [a business] industry” may be “both illogical and irrelevant.”); S.Y. v. Roman Cath. Diocese of Paterson, Civ. No. 20-2605, 2024 WL 1231333, at *6 (D.N.J. Mar. 21, 2024) (“[I]n nonscientific cases, the relevant reliability concerns will focus upon personal knowledge and experience of the witness and the methodology used will be applying that experience to the facts of the case.”) (internal quotation marks and citation omitted). Importantly, “good grounds” do not require that an expert’s opinion be perfect. Matter of Soued, 689 F. Supp. 3d 1, 11 (D.N.J. 2023). “Indeed, ‘an expert opinion is not inadmissible because it may contain flaws, nor is it excludable because it provides testimony regarding only one facet or aspect of an action but does not prove the whole case; such vulnerabilities affect the weight of the testimony, not its admissibility.’” Id. (quoting N.J. Dep’t of Env’t Prot. v. Amerada Hess Corp., Civ. No. 15-6468, 2019 WL 4052431, at *6 (D.N.J. Aug. 28, 2019)). If good grounds are shown, an expert’s testimony “should be tested by the adversary process—competing expert testimony and active cross-examination—rather than excluded from jurors’ scrutiny for fear that they will not grasp its complexities or satisfactorily weigh its inadequacies.” United States v. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 20 of 44 PageID: <pageID>
21 Mitchell, 365 F.3d 215, 244 (3d Cir. 2004) (quoting Ruiz-Troche v. Pepsi Cola of P.R. Bottling Co., 161 F.3d 77, 85 (1st Cir. 1998)). 3. Prong 3 – Fit “The third prong of admissibility, fit, concerns whether the expert’s testimony will be helpful to the trier of fact.” In re Johnson & Johnson Talcum Powder Prods. Mktg., Sales Pracs. & Prods. Litig., 509 F. Supp. 3d 116, 132 (D.N.J. 2020) (citing Fed. R. Evid. 702(a)). The Third Circuit has explained that “admissibility depends in part on ‘the proffered connection between the scientific research or test result to be presented and particular disputed factual issues in the case.’” In re Paoli, 35 F.3d at 743 (quoting Downing, 753 F.2d at 1237). “Fit is an issue of relevance and simply means that scientific validity of the method or principles applies to the issues at hand.” Geiss v. Target Corp., Civ. No. 09-2208, 2013 WL 4675377, at *5 (D.N.J. Aug. 30, 2013) (citing United States v. Ford, 481 F.3d 215, 220 n.6 (3d Cir. 2007)). “Whether an expert’s testimony meets the fit requirement ‘is not always obvious, and scientific validity for one purpose is not necessarily scientific validity for other, unrelated purposes.’” In re Johnson & Johnson Talcum Powder Prods. Mktg., 509 F. Supp. 3d at 132 (quoting In re Paoli, 35 F.3d at 743). B. Class Certification “The class action is an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only.” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 348 (2011) (internal quotation marks and citation omitted). The “party proposing class-action certification bears the burden of affirmatively demonstrating by a preponderance of the evidence [its] compliance with the requirements of Rule 23.” Byrd v. Aaron’s Inc., 784 F.3d 154, 163 (3d Cir. 2015), as amended (Apr. 28, 2015) (citing Comcast Corp v. Behrend, 569 U.S. 27, 33 (2013)). “Rule 23 does not set forth a mere pleading standard. A party seeking class certification must affirmatively demonstrate his compliance with the Rule—that is, he must be prepared to Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 21 of 44 PageID: <pageID>
22 prove that there are in fact sufficiently numerous parties, common questions of law or fact, etc.” Dukes, 564 U.S. at 350. A court must engage in a “rigorous analysis” to determine whether all of Rule 23’s prerequisites have been satisfied. Id. (quoting Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 161 (1982)). This rigorous analysis has “three key aspects”: First, the court must find that the requirements of Rule 23 are met and any factual determinations supporting Rule 23 findings must be made by a preponderance of the evidence. Second, the court must resolve all factual or legal disputes relevant to class certification, even if they overlap with the merits. Third, the court must consider all relevant evidence and arguments, including expert testimony, whether offered by a party seeking class certification or by a party opposing it. In re Lamictal Direct Purchaser Antitrust Litig., 957 F.3d 184, 191 (3d Cir. 2020) (internal citations and quotations omitted) (cleaned up). “Merits questions may be considered to the extent—but only to the extent—that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied.” Williams v. Jani-King of Philadelphia Inc., 837 F.3d 314, 322 (3d Cir. 2016) (quoting Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455, 466 (2013)). III. DISCUSSION A. Motions to Exclude Plaintiff’s Experts 1. Dr. Lang Defendant raises two arguments in support of excluding Dr. Lang’s Expert Report and corresponding testimony.12 First, Defendant contends Dr. Lang’s methodology was unreliable. (ECF No. 144-1 at 7-20.) Second, Defendant argues Dr. Lang improperly offered a legal 12 Defendant also seeks to exclude Dr. Lang’s Expert Reply Report, but the Magistrate Judge has already granted Defendant’s motion to strike that Report on other grounds. (See ECF No. 166). Therefore, the Daubert motion is moot with respect to this Report. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 22 of 44 PageID: <pageID>
23 conclusion. (Id. at 20-22.) Plaintiff disagrees on both counts. (See ECF No. 150.) There is no dispute over whether Dr. Lang is qualified to opine or whether his opinions are relevant. (See generally ECF No. 144-1.) a. Reliability Defendant argues Dr. Lang’s methodologies are unreliable for two reasons. First, Defendant challenges Dr. Lang’s use of the “name-matching” method in support of hiring statistics. (ECF No. 144-1 at 7-11.) Second, Defendant argues that Dr. Lang failed to control for relevant factors in his hiring, promotion, and termination analyses. (Id. at 11-20.) The Court is unpersuaded that Dr. Lang’s Report should be excluded on these grounds.13 i Name Matching First, Defendant argues the name-matching method is unreliable because “[i]t has not been subjected to peer review, publication, or scrutiny by Dr. Lang’s peers. . . . Nor did Dr. Lang perform an analysis of this methodology to determine its error rate.” (Id. at 7.) Plaintiff argues “name- matching is a well-established methodology that has been subject to peer review, publication, and scientific scrutiny.” (ECF No. 150 at 12.) As for peer review, it appears that the specific type of 13 Defendant appears to raise a third argument: Dr. Lang’s proxy coding mechanism for race of Virtusa employees—separate from the name-matching method used for the hiring analysis— inflated the number of persons deemed to be “Non-Asians suffering from the employment practices complained of by Plaintiff, that is, non-South Asians not being hired, receiving poor review ratings, not receiving promotions, and being involuntarily terminated.” (ECF No. 144-1 at 9 (citing ECF No. 140-1).) The document Defendant cites to in support of this argument is a declaration from Defendant’s rebuttal expert, Dr. Paul White. In that declaration, Dr. White raises the inflation concern. (See ECF No. 140-1 at 3.) However, the concern only arises in response to Dr. Lang’s stricken Expert Reply Report, (id. at 2.), so the Court need not address this third argument. To the extent there are issues in the original Report that remain unaddressed because the Expert Reply Report is stricken, the “identification of such flaws in generally reliable scientific evidence is precisely the role of cross-examination” and should not serve as grounds for inadmissibility. Quiet Tech. DC-8, Inc. v. Hurel-Dubois UK Ltd., 326 F.3d 1333, 1345 (11th Cir. 2003) (collecting cases, including In re TMI Litig., 193 F.3d 613, 692 (3d Cir. 1999)). Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 23 of 44 PageID: <pageID>
24 name-matching Dr. Lang employed is not peer-reviewed, but that name-matching as a broader methodology is peer-reviewed. (See ECF No. 144-3 at 15.) While Dr. Lang could not point to specific peer-reviewed, name-matching articles in his deposition or a treatise in support of name matching, he testified that he was “almost certain” articles existed, (ECF No. 150-3 at 7-8), and Plaintiff’s brief highlights several instances, (ECF No. 150 at 12-13.) For example, Plaintiff cites a peer-reviewed article that itself reviews 13 articles that have employed name matching. (Id. at 12 (citing Pablo Mateos, A Review of Name-based Ethnicity Classification Methods and their Potential in Population Studies, 13 Population, Space and Place 243 (2007)). In that article, Dr. Mateos sought to “extract[] methodological commonalities” in name-matching studies. Mateos at 243. First, the 13 studies all included “a name reference list [that was] independently built . . . .” Id. at 249 (emphasis in original). “Generally, [the studies] all used some type of ‘ethnic origin information’ in the reference population, such as self-reported ethnicity, [or] country of birth or nationality, to classify individuals into ethnic groups” in the reference list. Id. at 250. Second, the studies included “a separate target population” and individuals from that target population were “classified into ethnic groups.” Id. at 249 (emphasis in original.) There were two common approaches for classifying such individuals: manually creating a “rich number of ‘fuzzy rules’” that the classifier applies “in order to decide the group into which an individual should be assigned”; or “apply[ing] an automated algorithm to search for the name of each individual in the target population against the reference list, and then assign . . . [an] ethnic group for that name to the individual.” Id. at 252. Third, “all of the studies measure the accuracy of the name-based classification by comparing it to a ‘gold standard’ for the ethnicity of the individuals in the target population, which had to be previously known through an independent source.” Id. In other words, the authors had access to information confirming the Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 24 of 44 PageID: <pageID>
25 ethnicities of the individuals in the target populations, so they could compare their results to that information and assess their approach’s accuracy. The study found that the 13 articles’ approaches had a positive predictive value—“the proportion of persons classified” into an ethnic group that “were correctly classified as such”—of between 70% and 96%. Id. at 254. There were “no substantial differences between the accuracy of the manual [versus] automatic classification methods.” Id. Dr. Mateos notes that one limitation of the studies was that “[n]ames usually only reflect patrilineal heritage” and thus name matching is “incapable of identifying mixed ethnicity or women’s ethnicity in mixed marriages.” Id. at 255. Nonetheless, the study concluded that while “this methodology cannot entirely replace self assigned ethnicity information, it provides a sufficient level of classification confidence to be used in the measurement of inequalities . . . .” Id. at 256. On the whole, Dr. Lang’s Expert Report survives Rule 702’s “liberal policy of admissibility.” Pineda, 520 F.3d at 243 (citation omitted). Many of the Paoli factors are sufficiently present. See 35 F.3d at 742 n.8. For example, Dr. Lang is qualified as an expert and employed hypothesis testing to arrive at his conclusions. Further, the method of name-matching has been subject to peer review, and Dr. Lang’s approach is substantially similar to those in Dr. Mateos’s article. Dr. Lang created a reference list based on the known ethnicities of employees and classified applicants, based on a “rich number” of rules, in reference to that list. (ECF No. 144-3 at 12-13; ECF No. 144-4 at 5.) Dr. Lang could not perform an error analysis for this particular study, (ECF No. 150-3 at 6), but Dr. Mateos identified an error rate between 4% and 30% in 13 name-matching articles, Mateos at 254. Cases in which courts have considered name matching in the context of Daubert motions are also instructive. In Palmer v. Cognizant Technology Solutions Corporation, the court Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 25 of 44 PageID: <pageID>
26 considered an expert’s analysis that employed name-matching to categorize employees as South Asian. See Civ. No. 17-6848, 2022 WL 18214014, at *11 (C.D. Cal. Oct. 27, 2022). Where available, the expert used birth country and passport information to code employees as South Asian. Id. To determine if other employees were South Asian, the expert analyzed their last names—i.e., employed name matching. Id. The court found the methodology permissible even if “somewhat imprecise.” Id. at *12 n.17. By contrast, the court deemed name-matching inadmissible in Koehler v. Infosys Ltd. Inc., 628 F. Supp. 3d 835, 862-63, 891 (E.D. Wis. 2022). There, the plaintiff’s expert employed name-matching as follows: [The expert] considered an Infosys employee to be Indian or South Asian if (a) the employee was born in India or South Asia, or (b) the employee’s last name matched the last name of another Infosys employee born in India or South Asia and was not one of the sixty- three “Western” names removed from [a list created by the expert]. [The expert] considered an applicant to be Indian or South Asian if the applicant’s last name matched the last name of an Infosys employee born in India or South Asia and was not one of the sixty- three “Western-sounding” names removed from the list. If an employee or applicant did not meet one of the preceding criteria, that employee or applicant was considered non-Indian or non-South Asian. Id. at 856-57 (emphases in original). The court concluded the methodology was not reliable because the record reflected that it was “created for this litigation, not tested, not peer-reviewed, [and had] no known history of error rates.” Id. at 891. But many of those potential issues are not apparent here. Dr. Lang testified that he employed a “conservative approach . . . to mitigate potential errors.” (ECF No. 150-3 at 6.) And unlike in Infosys, this Court has been presented with evidence that name matching is a peer-reviewed method with known error rates. While it is true that Dr. Lang created his specific name-matching method for purposes of this litigation, the method was designed in response to the unique data with which he was presented. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 26 of 44 PageID: <pageID>
27 ii Controlling for Other Variables Defendant’s argument that Dr. Lang’s Report is unreliable because it does not control for other variables is similarly unpersuasive. With respect to hiring, Defendant argues that Dr. Lang “simply counted the number of allegedly ‘Asian’ and allegedly ‘Non-Asian’ employees . . . hired and found the gap between the two significant.” (ECF No. 144-1 at 13-14.) But Dr. Lang did not account for, among other things, “the relative demographic proportions of qualified, interested, and available candidates who presented themselves for hiring consideration” nor “did he consider the factors that were considered by Virtusa’s hiring managers” such as skills the customer required in Virtusa employees or whether the applicant was open to travel or relocating. (Id. at 14.) As for performance and promotions, Defendant argues that Dr. Lang again employed a binary analysis— he tested “whether an individual was ‘Asian’ or ‘Non-Asian’ . . . and whether they were promoted.” (Id. at 17.) But, Defendant argues, Dr. Lang “made no effort to determine what factors were considered by Virtusa’s supervisors in performance reviews or promotions,” or “which employees were interested in or sought a promotion.” (Id. at 18.) Finally, Defendant contends Dr. Lang’s termination analyses failed to account for factors such as Virtusa supervisors’ considerations in making termination decisions, changes in the job market, and tenure and disciplinary records of terminated employees. (Id. at 19-20.) Plaintiff argues Dr. Lang was not required to control for other explanatory factors in his hiring, promotion, or termination analyses. (ECF No. 150 at 16- 30.) The Court agrees that Dr. Lang’s methods are sufficiently reliable under Daubert. As the Supreme Court stated, “[w]hile the omission of variables from a regression analysis may render the analysis less probative than it otherwise might be, it can hardly be said, absent some other infirmity, that an analysis which accounts for the major factors must be considered unacceptable Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 27 of 44 PageID: <pageID>
28 as evidence of discrimination . . . . Normally, failure to include variables will affect the analysis’ probativeness, not its admissibility.” Bazemore v. Friday, 478 U.S. 385, 400 (1986) (internal citation omitted). Following Bazemore, courts in and out of this Circuit have continued to admit simple statistical analyses under Daubert that failed to control for other variables. See Adams v. Ameritech Servs., Inc., 231 F.3d 414, 425, 428 (7th Cir. 2000) (finding report that examined “correlations that existed between the ages of employees and the companies’ decisions to terminate” admissible even when expert did not “run a multiple-regression analysis” because disputes over inclusion of variables “went to the weight of the evidence . . . not to its admissibility”); Obrey v. Johnson, 400 F.3d 691, 695-97 (9th Cir. 2005) (finding study “based entirely on statistical disparities” between “the race of managers selected. . . compared to the race of those who applied for managerial positions” admissible because “objections to a study’s completeness generally go to the weight, not the admissibility of the statistical evidence . . . and should be addressed by rebuttal, not exclusion.” ) (internal quotation marks and citation omitted); Gutierrez v. Johnson & Johnson, Civ. No. 01-5302, 2006 WL 3246605, *5 (D.N.J. Nov. 6, 2006) (“Generally, decisions regarding what control variables to include in an expert report go to weight, not admissibility.”)14 While Defendant may have raised serious concerns with Dr. Lang’s statistical analyses of hiring, promotion, and termination data, those potential weaknesses go to the weight of Dr. Lang’s 14 The cases cited by Defendant are inapposite because they concern the type of statistical evidence needed to prove causation rather than the variables needed to be admissible under Daubert. See Bruno v. W.B. Saunders Co., 882 F.2d 760, 767 (3d Cir. 1989); Bell v. Lockheed Martin Corp., Civ. No. 08-6292, 2010 WL 3724271, at *7 (D.N.J. Sept. 15, 2010). The question here is “not whether the report[] proffered by the plaintiff[] prove[s] the entire case; it is whether [it was] prepared in a reliable and statistically sound way.” Adams, 231 F.3d at 425 (7th Cir. 2000); see also Obrey, 400 F.3d at 694. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 28 of 44 PageID: <pageID>
29 opinions rather than admissibility.15 Dr. Lang’s report and corresponding testimony “should be tested by the adversary process . . . rather than [be] excluded from jurors’ scrutiny for fear that they will not grasp its complexities or satisfactorily weigh its inadequacies.” Mitchell, 365 F.3d at 244 (citation omitted.) b. Opinion on the Ultimate Question Defendant argues that Dr. Lang’s conclusion should be excluded because it “invad[es] the exclusive province [of] the jury in applying the law to the facts” and “is an improper legal conclusion” that Dr. Lang is unqualified to make. (ECF No. 144-1 at 20.) Dr. Lang’s challenged conclusion is as follows: Based on these findings, it is my expert opinion that there is significant evidence that non-Asians employees are being discriminated against at Virtusa in terms of hiring, employee ratings and promotions, and through terminations. Further, there is significant evidence that visa employees are being favored at Virtusa with lower involuntary termination rates. 15 It is true that “[t]here may . . . be some regressions so incomplete as to be inadmissible as irrelevant.” Bazemore, 478 U.S. at 400 n.10. These analyses would face challenges under Daubert’s “fit” rather than “reliability” prong. Indeed, Defendant appears to make such an argument with respect to the hiring analysis. Defendant argues it is “egregious” that Dr. Lang failed to control for “the racial composition of the applicant pools.” (See ECF No. 144-1 at 14.) If Dr. Lang found the applicant pool was predominately Asian, Defendant argues, then it would not be surprising that the hires were predominately Asian, so any analysis that does not factor in this information would be irrelevant for finding disparate treatment. Id. But a statistical showing of discrimination need not be “based on [an] analysis of the characteristics of actual applicants” because the “application process might itself not adequately reflect the actual potential applicant pool, since otherwise qualified people might be discouraged from applying because of a self- recognized inability to meet the very standards challenged as being discriminatory.” Dothard v. Rawlinson, 433 U.S. 321, 330 (1977) (citing Int’l Brotherhood of Teamsters v. United States, 431 U.S. 324, 365-367 (1977); see also Buchanan v. Tata Consultancy Servs., Ltd., Civ. No. 15-01696, 2017 WL 6611653, at *11 (N.D. Cal. Dec. 27, 2017) (finding expert’s failure to consider applicant pool was 40% South Asian went to weight, not admissibility). Therefore, this is an issue more appropriately directed to the weight of Dr. Lang’s opinions as opposed to admissibility. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 29 of 44 PageID: <pageID>
30 (Id. (quoting ECF No. 144-4 at 16.) Plaintiff argues that this conclusion is permissible because Dr. Lang “does not offer the opinion that Virtusa engaged in a pattern or practice of race discrimination, only that his statistical analyses offer evidence of discrimination.” (ECF No. 150 at 31 (emphasis in original) (citing ECF No. 144-4 at 16).) And in any event, Plaintiff submits, Dr. Lang, “will not offer the legal conclusion that Virtusa engaged in intentional race discrimination” but instead will testify that “his statistical analyses show disparities that are not explained by chance and not explained by factors other than race and are therefore suggestive of discrimination.” (Id. at 32.) “An opinion is not objectionable just because it embraces an ultimate issue.” Fed. R. Evid. 704(a). But the “use of legal terms of art or legal conclusions by an expert” are inadmissible because they “could lead the jury to be prejudicial against the opposing party.” O’Bryant v. Johnson & Johnson, Civ. No. 20-2361, 2022 WL 7670296, at *6 (D.N.J. Oct. 13, 2022) (internal quotation marks and citation omitted). The “term ‘discrimination’ has a specialized legal meaning that is more precise than the lay understanding of the term.” United States v. Xue, 597 F. Supp. 3d 759, 774 (E.D. Pa. 2022) (quoting Burkhart v. Washington Metro. Area Transit Auth., 112 F.3d 1207, 1212 (D.C. Cir. 1997)). Accordingly, courts have held that statistical experts may not draw conclusions about whether a party acted in a discriminatory manner. See, e.g., Chen-Oster v. Goldman, Sachs & Co., Civ. No. 10-6950, 2022 WL 814074, at *10 (S.D.N.Y. Mar. 17, 2022) (“[T]o the extent that [the expert] uses the term ‘sex discrimination,’ which is a legal term of art, he is making a legal conclusion that is improper as it usurps the role of the jury. . . . [T]he expert may opine . . . so long as he does not use the term ‘sex discrimination.’”) (citation omitted); Simpson v. Dart, Civ. No. 18-0553, 2021 WL 1906469, at *2 (N.D. Ill. May 12, 2021) (finding expert could not opine on “any . . . legal conclusion about ‘discrimination’ as understood as a legal Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 30 of 44 PageID: <pageID>
31 term of art”); Chicago Tchrs. Union, Loc. v. Bd. of Ed. of City of Chicago, Civ. No. 12-10311, 2020 WL 914881, at *13 (N.D. Ill. Feb. 25, 2020) (“[Expert] cannot opine on whether the Board discriminated against employees nor whether there is statistically significant evidence of racial discrimination.”). Dr. Lang’s conclusion must be excluded. Dr. Lang’s Report opines on the legal term of art—“discrimination.” It is no matter that Dr. Lang caveats his written conclusion by noting there is “evidence” of discrimination rather than that there plainly is discrimination. See Chicago Tchrs. Union, at *13. The same goes for his anticipated testimony that his analyses “are suggestive of discrimination.” Therefore, Dr. Lang’s conclusion must be excluded to the extent it uses the word “discrimination,” and should Dr. Lang testify, he may not testify that his findings are “suggestive of discrimination.” Rather, Dr. Lang’s conclusions must be limited to his statistical findings. Defendant’s Motion is therefore granted in part and denied in part. 2. Dr. Hira Defendant argues Dr. Hira’s Expert Report and corresponding testimony should be excluded on three Rule 702 grounds. First, Defendant argues Dr. Hira is not qualified to opine on the topics in his Report. (ECF No. 145-1 at 10-11; ECF No. 160 at 9-12.) Second, Defendant contends Dr. Hira’s report is not reliable because Dr. Hira did not employ a proper methodology and of the 36,000 documents produced in discovery, Dr. Hira only reviewed 60 that were picked by Plaintiff. (ECF No. 145-1 at 11; ECF No. 160 at 5-9.) Third, Defendant argues Dr. Hira’s report is irrelevant because it would not assist the Court in deciding whether to certify a class. (ECF No. 145-1 at 8-9.) In addition to these Rule 702 considerations, Defendant argues Dr. Hira made improper legal conclusions that should be excluded. (ECF No. 145-1 at 9; ECF No. 160 at 12-18). The Court finds these arguments unpersuasive. In making this determination, the Court Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 31 of 44 PageID: <pageID>
32 finds particularly instructive the decision in Cognizant, 2022 WL 18214014, at *15-17 (denying motion to exclude Dr. Hira’s testimony concerning business model of Virtusa’s competitor). a. Qualification Defendant’s initial argument goes to the first prong of the 702 inquiry—qualification. To be sufficiently qualified, the individual must have “specialized expertise” on the topic of their testimony, and the basis for this expertise can be “academic training and credentials.” Elcock v. Kmart Corp., 233 F.3d 734, 741 (3d Cir. 2000) (internal quotation marks and citation omitted). This is a “liberal” standard. Id. (“[A]t a minimum, a proffered expert witness . . . must possess skill or knowledge greater than the average layman.”) (internal quotation marks and citation omitted). Here, Dr. Hira’s report states that he is opining on “business models of information technology services companies.” (ECF No. 146 at 3.) With a Ph.D. in political science, Dr. Hira has “published two books, three peer reviewed articles, six book chapters, and testified seven times before the U.S. Congress and twice before the Maryland State House of Delegates on” the topic of business models and employment practices of information technology services companies. (ECF No. 146 at 2; see also id. at 32-47.) This topic is therefore “well within his area of expertise” and “he is qualified to opine.” Cognizant, 2022 WL 18214014, at *17. b. Reliability Defendant’s next argument goes to the second prong of the 702 inquiry—reliability. Defendant raises two concerns. First, Defendant contends Dr. Hira did not employ a “legitimate, structured analytical method” and instead relayed a factual narrative as opposed to a legitimate expert opinion. (ECF No. 145-1 at 5, 9-11, 14.) Second, Defendant argues that Dr. Hira’s opinions are not reliable because he only reviewed a small portion of allegedly cherry-picked documents. (Id. at 11; ECF No. 160 at 6.) As to the first point, Defendant’s argument is unpersuasive because Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 32 of 44 PageID: <pageID>
33 in cases involving “specialized,” rather than scientific knowledge, the relevant reliability concerns “may focus upon personal knowledge or experience,” rather than “scientific foundations.” Kumho Tire Co., 526 U.S. at 150. Indeed, standard reliability considerations such as rate of error, publication, and peer review are “illogical and irrelevant” when it comes to opinions “regarding customs and practices” in a particular business industry. Wolfe, 2012 WL 12929813, at *5. So long as Dr. Hira applied his “experience to the facts of the case,” then his conclusions are sufficiently reliable to be admissible. S.Y., 2024 WL 1231333, at *6 (internal quotation marks and citation omitted). Here, Dr. Hira drew on his experience studying offshore business models in providing his analysis. This was sufficient. See Cognizant, 2022 WL 18214014, at *16 (stating that Dr. Hira “appears to rely primarily on his own expertise and understanding in opining on the industry generally”). And while courts have held that “[a]cting simply as a narrator of the facts” does not pass muster under Daubert, see SEC v. Tourre, 950 F. Supp. 2d 666, 675 (S.D.N.Y. 2013), Dr. Hira does more than that. In paragraphs 16 to 20 of his Report—the section challenged by Defendant (ECF No. 145-1 at 14)—Dr. Hira draws on his experience studying offshore business models to explain why, in his opinion, other companies predominately fill their positions with South Asian visa workers and the ways in which those companies are similar to Virtusa. (ECF No. 146 at 11- 14.)16 16 Defendant argues this same section of Dr. Hira’s Report is also improper because it “focuses singularly on the question of Virtusa’s motivation.” (ECF No. 145-1 at 16.) It is true that “experts may not provide testimony concerning . . . . the intent, motives or states of mind of [defendant] corporations . . . .” Krys v. Aaron, 112 F. Supp. 3d 181, 203 (D.N.J. 2015) (internal quotation marks and citation omitted). This is because such testimony “offers inappropriate legal conclusions and invades the province of the jury.” TAKTL, LLC v. IWR, N. Am., LLC, Civ. No. 18-1546, 2024 WL 4343141, at *11 (W.D. Pa. Sept. 30, 2024). But an “expert may testify about common behavior patterns in a profession or subculture.” United States v. Price, 458 F.3d 202, 212 (3d Cir. 2006). And “it is permissible to testify that facts and circumstances surrounding that Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 33 of 44 PageID: <pageID>
34 As to the second point, while it is true that Dr. Hira reviewed a hand-picked, small portion of the documents received during discovery, his Report “is based on decades of research about the industry and the visa regime under which [Virtusa] operates, in addition to the non-public, [Virtusa]-specific documents produced in this action. Dr. Hira also relied on public information about [Virtusa], including [Virtusa]’s own 10-K and USCIS data about H-1B visa approvals.” Cognizant, 2022 WL 18214014, at *16. (See also ECF No. 146 at 27-28.) To the extent it is noteworthy that Dr. Hira has not reviewed the vast majority of the documents produced in discovery such that it is possible that those documents therefore contradict his findings, this fact goes to the weight of Dr. Hira’s testimony rather than its admissibility. See S.Y., 2024 WL 1231333, at *7; Cognizant, 2022 WL 18214014, at *16. c. Fit Defendant’s subsequent argument goes to the third prong of the 702 inquiry—fit or relevance. “The third prong of admissibility, fit, concerns whether the expert’s testimony will be helpful to the trier of fact.” In re Johnson & Johnson Talcum Powder Prods. Mktg., Sales Pracs. & Prods. Litig., 509 F. Supp. 3d 116, 132 (D.N.J. 2020) (citing Fed. R. Evid. 702(a)). Defendant argues that Dr. Hira’s Report will not be helpful for class certification because it does “nothing to behavior may be ‘indicative’ of a certain intent.” Eddystone Rail Co., LLC v. Bridger Logistics, LLC, Civ. No. 17-00495, 2022 WL 22885376, at *2 (citing United States v. Farrish, 297 F. App’x 162, 166 (3d Cir. 2008). Nowhere in the challenged section of Dr. Hira’s Report does Dr. Hira usurp the role of the factfinder by opining on Defendant’s intent. Instead, Dr. Hira opines that other companies with similar practices to Defendant have engaged in outsourcing “because of deep-seated racial and ethnic preferences, rather than any business necessity.” (ECF No. 146 at 11.) Because the fact finder is left with the ultimate task of deciding whether Virtusa’s intent was discriminatory, the section of the Report is proper. Cf. Palmer v. Cognizant Tech. Sols. Corp., Civ. No. 17-6848, 2023 WL 4155403, at *12 (C.D. Cal. June 1, 2023) (declining to exclude analysis when “Dr. Hira provide[d] context from which the jury could properly make informed inferences about [defendant]’s intent or motive.”) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 34 of 44 PageID: <pageID>
35 answer any common question concerning hiring, termination, or promotions; do[es] not establish any policy or practice; and do[es] not help Plaintiff avoid the obvious fact that hiring, termination, and promotion decisions at Virtusa are highly individualized and predicated upon the discretion of hundreds or thousands of Virtusa employees and clients.” (ECF No. 145-1 at 9 (emphasis in original).) But the question on a Daubert motion is not whether the expert will make or break a subsequent decision but rather whether it will be “helpful” in making those decisions. In re Johnson & Johnson, 509 F. Supp. 3d at 132. Here, “Dr. Hira’s conclusions about how a business model reliant on H-1B visa holders would help maximize profits lend support to [Plaintiff’s] case for commonality because they help to explain why a company might pursue” certain visa policies and “thus substantiate [Plaintiff’s] contention that [Virtusa] had such a policy.” Cognizant, 2022 WL 18214014, at *15. Additionally, “Dr. Hira’s opinion that cultural preferences—not only unconscious bias, but also supervisors’ preferences relating to the relationship between supervisors and supervisees—might lead managers to develop a preference for [South Asian] workers even in local hiring is obviously relevant to [Plaintiff’s] claims.” Id. at *16. And while Dr. Hira’s conclusions about Virtusa are more limited than his opinions about the industry as a whole given the dataset Dr. Hira reviewed, the Court will not exclude his opinions on that ground but acknowledges that “these limitations may limit their weight.” Id. d. Opinions on the Ultimate Question Finally, Defendant challenges certain statements on the grounds that they are legal conclusions. (ECF No. 145-1 at 9 (quoting ECF No. 146 at 9, 12-14, 24).)17 These statements 17 Defendant introduces additional statements in its reply brief, (see ECF No. 160 at 12-13), but it “is axiomatic in federal practice that arguments raised for the first time in a reply brief should Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 35 of 44 PageID: <pageID>
36 are: (1) “some IT outsourcing firms have wrongfully used the H-1B visa program to secure visas for future or fictitious work, improperly attesting on the H-1B visa petition that there is an actual open position in the U.S. for the sponsored employee to fill when, in reality, no such position exists,” (ECF 146 at 9); (2) the “indenturing nature” of the H-1B visa program, (id. at 12-13); and (3) that underpayment “constitutes wage theft.” (Id. at 14.)18 The first statement is proper. Courts have upheld similar statements concerning compliance with laws or regulations when the statements concerned business customs. See, e.g., United States v. Leo, 941 F.2d 181, 196-97 (3d Cir. 1991); Bartoli v. Novartis Pharms. Corp., Civ. No. 13-0724, 2014 WL 1515870, at *7 (M.D. Pa. Apr. 17, 2014). And the use of the word “wrongfully” does not make the statement excludable. Cf. Gov’t Emps. Ins. Co. v. Adams Chiropractic Ctr., Civ. No. 19-20633, 2023 WL 4248817, at *7 (D.N.J. June 29, 2023) (finding expert “report d[id] not improperly instruct the jury on what result to reach simply by employing the word ‘legitimate.’”) The second statement is proper because it concerns industry customs rather than legal conclusions. See Price, 458 F.3d at 212 (3d Cir. 2006) (“It is settled law that an expert may testify about common behavior patterns in a profession or subculture.”). The third statement is arguably improper. See Cognizant, 2022 WL 18214014, at *16 n. 25 (“Dr. Hira’s specific conclusion that Cognizant engages in ‘wage theft’ does appear to be beyond the scope of his expertise.”). However, Dr. Hira has agreed to withdraw the phrase, so the point is moot. (See ECF No. 151 at 14 n.3) be disregarded.” Wilson v. Parker, Civ. No. 18-2954, 2018 WL 6696783, at *5 (D.N.J. Dec. 20, 2018) (internal quotation marks and citation omitted). 18 Defendant also challenges Dr. Hira’s assertion that Virtusa’s visa policies are not the result of “business necessity.” (ECF No. 145-1 at 9 (citing ECF No. 146 at 24).) This statement is admissible for the same reason Dr. Hira’s similar statement was proper. See supra n.16 (citing ECF No. 146 at 11.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 36 of 44 PageID: <pageID>
37 Therefore, Defendant’s motion to exclude Dr. Hira’s Expert Report is denied. B. Motion for Class Certification Under Rule 23(a)(1)-(4), a class may be certified only if: (1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class. Fed. R. Civ. P. 23(a). These requirements are respectively referred to as the numerosity, commonality, typicality, and adequacy requirements. See id. “[E]very putative class action must satisfy the four requirements of Rule 23(a) and the requirements of either Rule 23(b)(1), (2), or (3).” Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 590 (3d Cir. 2012). Here, Plaintiff seeks certification under Rule 23(b)(3). (ECF No. 142 at 37.) Where a party seeks to certify a class under Rule 23(b)(3), the moving party must satisfy three additional requirements. See Byrd, 784 F.3d at 163. First, the party “must prove by a preponderance of the evidence that the class is ascertainable.” Id. Second, the party seeking certification must show that “questions of law or fact common to class members predominate over any questions affecting only individual members.” Fed. R. Civ. P. 23(b)(3). Third, the party must show “that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” Id. These additional requirements are, respectively, referred to as the ascertainability, predominance, and superiority requirements. See, e.g., Byrd, 784 F.3d at 161 n.4, 162-64; Johnson v. Comodo Grp., Inc., Civ. No. 16-04469, 2024 WL 2933195, at *3 (D.N.J. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 37 of 44 PageID: <pageID>
38 Jun. 10, 2024). A district court “must first determine whether the putative class satisfies the . . . provisions of Rule 23(a), and then proceed[] to analyze the . . . provisions of Rule 23(b)(3).” In re Citizens Bank, N.A., 15 F.4th 607, 612 (3d Cir. 2021). Plaintiff argues that pursuant to Rule 23(b)(3), common questions of law and fact predominate over individualized questions in this matter, and that a class action is superior to other available methods of adjudicating the claims. (ECF No. 142 at 37-46.) Plaintiff proposes the following three classes: Hiring Class: All non-South Asian applicants who applied for a position with Virtusa in the United States between February 28, 2014 and the date of class certification and who were not hired. Promotion Class: All non-South Asian employees who worked for Virtusa in the United States between February 28, 2014 and the date of class certification and who were not promoted. Terminations Class: All non-South Asian employees who worked for Virtusa in the United States between February 28, 2014 and the date of class certification and who were involuntarily terminated. (ECF No. 141 at 3.) The Court addresses whether class certification is appropriate for each of the three classes. Citizens Bank, 15 F.4th at 612. 1. Rule 23(a) a. Numerosity The numerosity requirement is disputed by the parties. (ECF No. 142 at 23-24; ECF No. 153 at 22-27; ECF No. 157 at 5-8.) Numerosity is met where “the class is so numerous that joinder of all members is impracticable.” See Fed. R. Civ. P. 23(a)(1). Although there is no minimum number of members needed to satisfy numerosity, courts hold as a general rule that if the proposed members exceed forty people, the first prong of Rule 23(a) has been met. See Stewart v. Abraham, 275 F.3d 220, 226-27 (3d Cir. 2001). Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 38 of 44 PageID: <pageID>
39 i Hiring Class Plaintiff estimates—based on Dr. Lang’s declaration—that the Hiring Class is in the tens of thousands with a cap of 23,589. (ECF No. 142 at 24 (citing ECF No. 142-1 ¶¶ 2-4).) Dr. Lang reviewed data provided by Virtusa which revealed 34,806 individuals who applied to and were rejected from Virtusa during the class period. (ECF No. 142-1 ¶ 2.) Dr. Lang subtracted instances in which individuals applied multiple times and were rejected multiple times and arrived at 23,589. (Id. ¶ 3). Dr. Lang stated that race and national origin data were seldom available for application data but would be able to “ascertain the race of Virtusa’s applicants by utilizing a names-based analysis similar to what” he employed in his Expert Report. (Id. ¶ 4-5.) Defendant argues that Dr. Lang, on whom Plaintiff solely relies, “presents no actual evidence of [the class size number]. Rather, he simply counts the total number of applicants, regardless of race.” (ECF No. 153 at 25 (emphasis in original).) In other words, the only information Dr. Lang provided is that 23,589 individuals were rejected and an allusion to future name matching is insufficient. Plaintiff replies that “common sense” dictates that the numerosity requirement is met. (ECF No. 157 at 7 (citing Marcus, 687 F.3d at 596-97.) The question therefore boils down to whether the fact that 23,589 individuals were rejected by Virtusa is enough to establish, by a preponderance of the evidence, that at least 40 non-South Asian individuals were rejected. The Court finds that this is not enough. The Third Circuit has given the numerosity requirement “real teeth” in recent years. Mielo v. Steak‘n Shake Operations, Inc., 897 F.3d 467, 484 (3d Cir. 2018) (citation omitted). “[W]here a putative class is some subset of a larger pool, the trial court may not infer numerosity from the number in the larger pool alone.” Hayes v. Wal-Mart Stores, Inc., 725 F.3d 349, 358 (3d Cir. 2018.) It is true that when a plaintiff “show[s] sufficient circumstantial evidence specific to the products, problems, parties, and Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 39 of 44 PageID: <pageID>
40 geographic areas actually covered by the class definition . . . . the court [may] rely on ‘common sense’ to forgo precise calculations and exact numbers.” Marcus, 687 F.3d at 596-97. But the Third Circuit has repeatedly found plaintiffs have not satisfied numerosity when applying this principle. See Marcus, 687 F.3d at 595-97 (numerosity unsatisfied when 740,102 vehicles leased nationwide but no evidence in the record about vehicles leased in New Jersey—the geographic limitation of the class); Hayes, 725 F.3d at 357-58 (numerosity unsatisfied when “the only conclusion that can be drawn from the evidence presented to the trial court is that the number of class members would be equal-to-or-less-than 3,500 and equal-to-or-greater-than zero”); Mielo, 897 F.3d at 486 (numerosity unsatisfied when evidence of “between 14.9 million to 20.9 million persons with mobility disabilities who live in the United States” but “no evidence that would permit [the Third Circuit] to use ‘common sense’ to determine—rather than speculate about—the portion of those disabled individuals who have actually patronized a relevant Steak ‘n Shake restaurant, let alone the portion who have experienced or will experience an ADA violation at one of those restaurants.”). In these type of cases, the “high[] likel[ihood] that at least 40 individuals” from the larger pool would be in the sub-class is insufficient. Mielo, 897 F.3d at 486. And even though it may be “tempting to assume [that the] class meets” the numerosity requirement based on the larger pool, the Court must not “giv[e] in to such temptation” and “excuse speculation.” Id. at 485 (internal quotation marks and citation omitted). Ultimately, it is a “plaintiff's burden to demonstrate numerosity by a preponderance of the evidence.” Hayes, 725 F. 3d at 358. And future calculations are of no import because courts “cannot take a wait-and-see approach to numerosity or any other requirement of Rule 23.” Id. Here, Plaintiff has not established numerosity. The only evidence Plaintiff cites is the size of the larger pool of rejected applicants. Even though it is likely at least 40 members of this pool Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 40 of 44 PageID: <pageID>
41 are not South Asian, Plaintiff provides no evidence for the Court to make this inference. Indeed, Dr. Lang’s Expert Report—while admissible under Daubert and not cited by Plaintiff in support of numerosity—only provides race estimates for the applicants who were hired, (ECF No. 144-4 at 5), so the Court has no way of determining, beyond speculation, the number of non-South Asian individuals who were not hired. And even if the numerosity issue would be resolved if Dr. Lang applied name matching to the applicant data, the Court cannot accept a “wait-and-see approach.” Hayes, 725 F.3d at 358. “[T]he only conclusion that can be drawn from the evidence presented to the trial court is that the number of class members would be equal-to-or-less-than [23,589] and equal-to-or-greater-than zero.” Id. at 357-58. The Court must therefore deny certification with respect to the Hiring Class. See Fuentes v. Super Bread II Corp., Civ. No. 18-6736, 2020 WL 7237942, at *6 (D.N.J. Dec. 9, 2020) (“Because a failure to meet the numerosity requirement is dispositive of [plaintiff’]s motion to certify a class, the [c]ourt will not address the other three requirements under Rule 23(a)—namely, commonality, typicality, and adequacy of representation—or the requirement[s] under Rule 23(b).”19 19 Even if Plaintiff provided enough information such that the Court could apply “common sense” and find numerosity satisfied, Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 596-97 (3d Cir. 2012), class certification would still be denied with respect to the Hiring Class because Mr. Sugg was in fact hired by Virtusa, (see ECF No. 143-4 at 4). It is axiomatic that a class representative must be a member of the proposed class. See Amchem Prods., v. Windsor, 521 U.S. 591, 594-95 (1997) (“Representatives must be part of the class . . . .”); In re Pett Food Prods. Liab. Lit., 629 F.3d 333, 343 (3d Cir. 2010) (“Class representatives must be part of the class . . . .”) (internal quotation marks and citation omitted). Courts have discussed this principle in relation to both the typicality and adequacy prongs. See E. Tex. Motor Freight Sys., Inc. v. Rodriquez, 431 U.S. 395, 403-04 (1977) (adequacy and typicality); Amchem Prods., v. Windsor, 521 U.S. at 594- 95 (adequacy); In re Pet Food Prods. Liab. Litig., 629 F.3d at 343 (adequacy); Martin v. Ford Motor Co., 292 F.R.D. 252, 269-70 (E.D. Pa. 2013) (adequacy and typicality). Here, it is uncontested that Mr. Sugg was hired, (see ECF No. 143-4 at 4; ECF No. 157 at 18), so Plaintiff fails the adequacy and typicality requirements. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 41 of 44 PageID: <pageID>
42 ii Promotion Class Plaintiff estimates that the Promotion Class is approximately 269 members. (ECF No. 142 at 24 (citing ECF No. 142-19 at 16).) Defendant argues that this number was derived from a faulty analysis in Dr. Lang’s Expert Reply Report—an analysis which resulted in dozens of individuals allegedly in the class who appeared to be South Asian and several individuals who were promoted or held a position from which they could not be promoted. (ECF No. 153 at 26-27.) Plaintiff responds that Dr. White’s opinion—a declaration filed with Defendant’s opposition brief—cannot be used because it was not disclosed as part of his expert report and is untimely. (ECF No. 157 at 6-7 (citing Fed. R. Civ. P. 37(c)(1)).) In any event, Plaintiff argues other evidence in the record allegedly identifies the requisite number of class members. (Id. (citing ECF No. 153-3 ¶ 22; ECF No. 157-1 ¶ 12).) Plaintiff has failed to demonstrate numerosity by a preponderance of evidence. Plaintiff’s chief hurdle is that Dr. Lang’s Expert Reply Report has been stricken by the Court. (See ECF No. 166.) And the only source Plaintiff relies on in his brief is that Report. (ECF No. 142 at 24 (citing ECF No. 142-19 at 16).) On reply, Plaintiff cites a declaration by the Vice President of Human Resources for Virtusa, (ECF No. 153-3 ¶ 22), along with Dr. Lang’s declaration filed with the reply brief, (ECF No. 157-1 ¶ 12.) But the employee’s declaration is an analysis of the Expert Reply Report and Dr. Lang’s declaration is a response to that analysis. An analysis of evidence struck from the record cannot sustain Plaintiff’s burden. Plaintiff is left with Dr. Lang’s original Expert Report, which reveals approximately 17 non-South Asian individuals who were not promoted. (See ECF No. 144-4 at 8 (Table 6 depicting that 23.3% of 73 white employees not promoted).) The Court has not been presented with any reason why joinder of these individuals would be impracticable, so the Court finds numerosity is unsatisfied. See Zangara v. Zager Fuchs, Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 42 of 44 PageID: <pageID>
43 P.C., Civ. No. 17-6755, 2019 WL 6310056, *2-3 (D.N.J. Nov. 25, 2019) (stating that as a “general rule[] . . . a class of twenty or fewer members is usually insufficiently numerous,” but “a class of twenty-one to forty members may or ‘may not meet the numerosity requirement depending on the circumstances,’” and finding class of 26 insufficient) (quoting In re Modafinil Antitrust Litig., 837 F.3d 238, 250 (3d Cir. 2016)). The Court must therefore deny certification with respect to the Promotion Class. See Fuentes, 2020 WL 7237942, at *6. iii Termination Class Plaintiff estimates that the Termination Class is approximately 249 members based on Dr. Lang’s Expert Reply Report. (ECF No. 142 at 24 (citing ECF No. 142-19 at 19).) Defendant argues Plaintiff cannot rely on the Reply Report, and the original Report only records significantly less than 40 individuals in the Class. (ECF No. 153 at 23-24.) Plaintiff fails to establish numerosity of the Termination Class for the same reason they fail in the Promotion context. Without the Reply Report, which the Court has stricken, the only evidence is Dr. Lang’s original Expert Report which puts the number of potential class members at approximately 18. (See ECF No. 144-4 at 12 (Table 9 depicting that of 103 white employees terminated, 17.7% were terminated involuntarily).) The Court has been provided with no explanation for why joinder of these individuals would be impracticable, and the Court therefore finds this number insufficient. See Zangara, 2019 WL 6310056, at *2-3. The Court accordingly denies certification with respect to the Termination Class. See Fuentes, 2020 WL 7237942, at *6. Therefore, class certification must be denied for each of the three proposed classes.20 20 Plaintiff requests that Plaintiff’s counsel, Kotchen & Low LLP and Jonathan Rudnick, be appointed class counsel. (ECF No. 141 at 4.) Plaintiff’s request is moot because the Court denies Plaintiff’s Motion for Class Certification. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 43 of 44 PageID: <pageID>
Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 44 of 44 PageID: <pageID>
OCR transcription for search; preserved printed pages are authoritative
Case 3:18-cv-08036-GC-JTQ Document167_ Filed 01/13/26 Page 1 of 44 PagelD: <pagelD> NOT FOR PUBLICATION UNITED STATES DISTRICT COURT DISTRICT OF NEW JERSEY GWYNNE SUGG, individually and on behalf of all those similarly situated, Plaintiff, Civil Action No. 18-08036 (GC) (JTQ) v. OPINION VIRTUSA, Defendant. CASTNER, District Judge THIS MATTER comes before the Court upon three motions: Plaintiff Gwynne Sugg’s Motion for Class Certification pursuant to Federal Rule of Civil Procedure (Rule) 23 (ECF Nos. 141, 142); Defendant Virtusa’s Motion to Exclude the Expert Reports of David Lang, Ph.D. (ECF No. 144); and Defendant’s Motion to Exclude the Expert Report of Ronil Hira, Ph.D. (ECF No. 145). The parties briefed each of the Motions. (See ECF Nos. 142, 144-145, 150-151, 153, 157- 158, 160.) The Court has carefully considered the parties’ submissions and decides the matter without oral argument pursuant to Rule 78(b) and Local Civil Rule 78.1(b). For the reasons set forth below, and other good cause shown, Defendant’s Motion to Exclude the Expert Reports of David Lang, Ph.D. is GRANTED in part and DENIED in part; Defendant’s Motion to Exclude the Expert Report of Ronil Hira, Ph.D. is DENIED; and Plaintiff’s Motion for Class Certification is DENIED. AUTHENTICATED U.S. GOVERNMENT INFORMATION GPO 2 I. BACKGROUND A. Procedural Background On February 28, 2018, Plaintiff filed a putative class action in the Superior Court of New Jersey, Law Division, Monmouth County. (ECF No. 1-1.) On April 19, 2018, Defendant removed the case to this Court. (ECF No. 1.) After retaining new counsel, Plaintiff filed a First Amended Complaint (FAC) on July 11, 2018. (ECF No. 7.) Plaintiff’s FAC contains three counts: disparate treatment on the basis of race in violation of 42 U.S.C. § 1981 (Count I); an unenforceable and unconscionable non-competition agreement in violation of New Jersey state law (Count II); and breach of contract and unjust enrichment in violation of New Jersey state law (Count III). (Id.)1 The first two Counts are brought by Plaintiff and the putative class, and the third Count is brought only on behalf of Plaintiff. (Id.) On September 7, 2018, Defendant moved to dismiss the FAC. (ECF No. 11.) The Court denied the Motion to Dismiss with respect to Counts I and III. (ECF No. 24.) Plaintiff voluntarily withdrew Count II. (Id. at 2 n.2.)2 The parties then moved into discovery. As is relevant here, Plaintiff served its Expert Reports—authored by David Lang, Ph.D. and Ronil Hira, Ph.D., respectively—on July 29, 2024. (ECF No. 135-1 at 6; ECF No. 139 at 6.)3 Virtusa served a rebuttal to Dr. Lang’s Report—authored by Paul White, Ph.D.—on March 10, 2025. (ECF No. 135-1 at 6; ECF No. 139 at 10.) On May 5, 2025, Dr. Lang filed an Expert Reply Report. (ECF No. 135-1 at 6; ECF No. 139 at 11.) The 1 The Court has subject matter jurisdiction over this action pursuant to 28 U.S.C. § 1331. 2 Page numbers for record cites (i.e., “ECF Nos.”) refer to the page numbers stamped by the Court’s e-filing system and not the internal pagination of the parties. 3 On September 18, 2024, Leo Sugg passed away. (ECF No. 121.) Following a consent motion, this Court substituted Gwynne Sugg for her deceased husband. (ECF No. 129.) Accordingly, the parties jointly proposed an Amended Scheduling Order on February 14, 2025 (ECF No. 125), which the Court entered on March 5, 2025, (ECF No. 126.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 2 of 44 PageID: <pageID> 3 parties completed fact and expert discovery on May 23, 2025. (See ECF No. 126.) After the completion of discovery, the parties filed the instant Motions. On May 28, 2025, Defendant filed a motion to strike Dr. Lang’s Expert Reply Report. (ECF No. 135.) On June 25, 2025, Plaintiff filed a class certification motion. (ECF No. 141.) On that same date, Defendant filed motions to exclude the expert reports of Drs. Lang and Hira. (ECF Nos. 144, 145.) On December 23, 2025, the Honorable Justin T. Quinn, U.S.M.J., granted Defendant’s motion to strike Dr. Lang’s Expert Reply Report, and Plaintiff did not appeal that decision. (ECF No. 166.) B. Factual Background 1. Facts Relevant to the Motion for Class Certification a. Virtusa’s Business Model Virtusa provides information technology (IT) and consulting services to customers across the globe. (ECF No. 146 at 3; ECF No. 153-3 at 1.)4 The company was founded in Sri Lanka but maintains its corporate headquarters in Massachusetts. (ECF No. 142 at 7; ECF No. 146 at 3.) As of 2024, Virtusa employs over 27,000 individuals, (ECF No. 142-5), and nearly 70% of those individuals live in India. (ECF 142-4 at 27; 142-5 at 3.) Approximately 2,700 employees live in the United States. (ECF No. 142-6 at 7.) Virtusa employs an “offshore-onsite” business model to serve its United States clients. Under this model, most of the work performed for these clients is done “offshore” in India or Sri Lanka. (ECF No. 153-2 at 1.) Some of the work is done “on-site” at customer facilities in the United States. (Id.) Regardless, employees who work on these client 4 Because the Court finds that Dr. Hira’s Expert Report (ECF No. 146) is admissible under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), see supra Section III.A.2., the Court may rely on it for determining the facts for the purposes of class certification. See In re Blood Reagents Antitrust Lit., 783 F.3d 183, 187 (3d Cir. 2015) (“[A] plaintiff cannot rely on challenged expert testimony, when critical to class certification, to demonstrate conformity with Rule 23 unless the plaintiff also demonstrates, and the trial court finds, that the expert testimony satisfies the standard set out in Daubert.”). Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 3 of 44 PageID: <pageID> 4 projects—as opposed to “selling, general, and administrative” employees who work on internal Virtusa matters—are referred to as “billable employees.” (ECF No. 153-3 at 1-2.) These billable employees make up 90% of Virtusa’s workforce. (Id. at 2.) For all employees, Virtusa has a six- Tier job hierarchy based on employee roles and years of experience. Tier 4 (campus hires) is the lowest, followed by Tier 3 (two to five years’ experience), Tier 2 (five to eight years), Tier 1 (seven to twelve years), Tier 0 (ten to sixteen years), and Global Tier 0 (the highest Tier, comprised of Senior Directors and Vice Presidents, ranging in experience from fifteen to thirty years). (ECF No. 142-4 at 34-35.) Within this hierarchy, there are technology, delivery, and consulting tracks for billable employees, and sales and administration tracks for non-billable employees. (Id. at 31- 32, 34.) b. Virtusa’s Hiring Practices When a billable role becomes available,5 the company will either staff the role with a current employee or an external candidate. (Id. at 16, 22-23.) Virtusa generally first looks to availability among internal employees before hiring externally. (See, e.g., ECF No. 142-10 at 6- 8, 12.) When hiring externally, Virtusa’s Talent Acquisition Group employs various “modes of hiring” to attract candidates. (ECF No. 153-4 at 23-24.) It finds candidates through its own job portal postings; LinkedIn and other electronic job boards; external recruiters; external vendors who provide personnel; and internal recruiters. (Id. at 24-25.) Once candidates are sourced, they go through three rounds of interviews. First, technical recruiters conduct screening interviews to ensure applicants “will fit the profile.” (Id. at 25.) Second, “strategic business unit[] representatives”—similar to hiring managers—conduct interviews. (Id. at 26.) Third, practice 5 The hiring process for non-billable employees proceeds differently. See supra Section I.B.1.f (citing ECF No. 153-3 at 3-4). Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 4 of 44 PageID: <pageID> 5 managers—“high-end technology experts”—conduct interviews. (Id.) During the course of these interviews, applicants are asked whether they would be open to relocation and open to travel “to be sure” the applicant is aligned with the needs of the client. (Id. at 28.) If the applicant makes it through these three rounds, Virtusa presents the applicant to the client. (Id. at 28.) Between 80% and 90% of the time, the client then interviews the applicant. (Id. at 29.) The client and practice manager both play a “critical role” in deciding whether to extend an offer to an external applicant. (Id. at 32.) Virtusa also considers affordability when making its external hiring decisions. (Id.) During the application process, Virtusa does not collect documents from candidates other than their resumes, though it does inform candidates that they must have work authorization, and it does not ask candidates to self-identify their race. (Id. at 27-30; ECF No. 142-4 at 9-10.) c. Virtusa’s Promotion Practices Multiple players are involved in Virtusa’s promotion process. (ECF No. 153-3 at 3-4; ECF No. 153-4 at 53; see also ECF No. 142-16; ECF No. 142-18.) The process begins when an individual manager makes a promotion recommendation, and that recommendation is forwarded to the next level of management or to Human Resources personnel who make the ultimate decision. (ECF No. 153-3 at 3.) The promotion recommendation is based on a number of considerations: (1) employees may not be promoted if they lack required technical certifications, (2) an employee must have completed a required number of months of service in their current role, which varies depending on the tier an employee is in; (3) an employee must meet performance rating requirements; (4) an employee’s utilization levels and contribution to projects are considered; (5) an employee’s initiative, learning ability, and general growth potential are considered; (6) supervisor and client feedback; and (7) performance against expected goals[.] (Id. at 3-4.) But these factors change depending on the Tier in which the employee is seeking the promotion. (Id. at 4.) For junior level positions, promotions are based on tenure and performance Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 5 of 44 PageID: <pageID> 6 ratings, which turn on the employee’s potential. (ECF No. 153-4 at 53-54.) These employees— ones moving from Tier 4 to Tier 3—can be considered for promotion generally within 12 to 18 months. (Id. at 54; ECF No. 142-18 at 4.) As the employee moves up, the focus becomes less on potential and more on whether the individual is in fact contributing to the practice, and the employee can be considered for promotion generally after 24 to 36 months. (ECF No. 153-4 at 54; ECF No. 142-18 at 4.) To be promoted to the top positions—i.e., into Tiers 1 and 0—tenure and performance are key parameters, but as Virtusa’s Chief People Officer testified, affordability and a business need for the role are also “super” important. (ECF No. 153-4 at 54; ECF No. 142- 18 at 3.) Performance rating is a key factor throughout, but it applies differently depending on the type of job sought and the corresponding Tier. Billable employees are reviewed, at least in part, based on a qualitative scale. (ECF No. 142-18 at 3-4.) Employees who receive the top two scores on this scale—“Outstanding” and “Exceeds Expectation Consistently”—are eligible to be promoted at an expedited pace. (Id.) But billable employees seeking to be promoted within Tier 0 cannot do so unless they receive either of these two ratings. (Id. at 4.) By contrast, non-billable roles—like sales—do not employ the qualitative scale. (ECF No. 153-3 at 6.) Instead, these employees are primarily evaluated based on their “revenue impact” and “margin.” (Id.) Regardless of the type of role, “360-degree feedback”—that is feedback by not just managers but also peers and subordinates—is implemented when evaluating promotions into and within Tier 0, but not at more junior Tiers. (ECF No. 153-3 at 4; ECF No. 153-4 at 55-56.) Promotion into or within Tier 0 may also be accompanied by an interview with an internal or external review panel. (ECF No. 142-18 at 3; ECF No. 153-3 at 4.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 6 of 44 PageID: <pageID> 7 d. Virtusa’s Termination Practices Virtusa’s termination procedures are related to its “benching” practices. Being “on the bench” refers to a billable employee’s status when he or she is not currently staffed on a billable project. (ECF No. 153-4 at 35.) Non-billable employees cannot be on the bench because they are not staffed on client projects. (Id. at 49.) A billable employee may end up on the bench for three reasons. First, and most commonly, their previous project has come to an end. (Id. at 48.) Second, the client cancels a new project that the employee was already staffed on. (Id.) Third, the client requests the employee be removed from the project because of dissatisfaction with the employee’s performance. (Id.) If an employee is on the bench for too long, they run the risk of termination. (See, e.g., ECF No. 142-29 at 8-9.) e. Virtusa’s Visa Policies As of 2020, more than half of Virtusa’s United States-based employees are H-1B or L-1 visa holders. (ECF No. 146 at 5.) H-1B visas allow companies to sponsor employees in specialty occupations requiring application of highly specialized knowledge and attainment of at least a bachelor’s degree or equivalent. (Id. at 6-7.) See also 8 C.F.R. § 214.2(h)(1)(ii)(B). L-1 visas allow multinational corporations to transfer a worker from a foreign office to a United States office. (ECF No. 146 at 10.) See also 8 C.F.R. § 214.2(l). L-1B visas apply to transferees with “specialized knowledge” and L-1A apply to transferees who are executives or managers. (ECF No. 146 at 10.) It is Virtusa’s alleged practices pertaining to visa-holding employees, and the possibility of discriminatory treatment in implementation of those policies on the basis of race, that are primarily at issue here. Virtusa’s alleged practices largely fall within two buckets: visa preparation and visa utilization. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 7 of 44 PageID: <pageID> 8 The clearest example of Virtusa’s visa preparation policy is its alleged implementation of “Project Catalyst.” Under this program, introduced in 2013, Virtusa sought to increase the percentage of employees in the United States that were H1-B or L-1 visa holders to decrease costs. (See generally ECF No. 142-21 (Virtusa presentation outlining Project Catalyst).) Accordingly, and following the lead of its competitors, Virtusa considerably increased its number of visa applications. (See ECF No. 142-25 at 3; ECF No. 146 at 21.) And it took a proactive approach— submitting more applications than it could use and filing them before specific positions opened up. (See ECF No. 146 at 19; ECF No. 142-50 at 6, ECF 142-52 at 2.) By 2014, more than 15% of Virtusa’s United States employees were H1-B or L-1 visa holders, by 2016, more than 50%, and by 2018 it reached 65%. (ECF No. 146 at 21.) Having secured a large number of visa-holding employees, Virtusa ensured its employees used those visas. Under the “Cost Per Consultant” initiative, Virtusa sought to bring visa-ready employees to the United States. (See ECF No. 142-31 (Virtusa presentation on Cost Per Consultant initiative).) Accordingly, it “[p]roactively train[ed] visa ready [offshore employees] for onsite roles.” (ECF No. 142-32 at 2.)6 And pursuant to Virtusa’s rotation policy, implemented in 2014, 6 Defendant argues that Virtusa never implemented the Cost Per Consultant initiative, and that it was only recommended by consulting firm McKinsey & Company, which was hired to identify “market strategy” and “profitable growth” opportunities. (ECF No. 153 at 20; ECF No. 153-4 at 46.) In support, Defendant cites to the deposition transcript of Virtusa’s former Global Head of Human Resources. (ECF No. 153 at 20.) In that deposition, when asked whether Virtusa implemented the program, the employee responded: “not really to the depth and the number” that McKinsey recommended, and to the extent Virtusa “might have implemented” the program, that implementation was “very thin.” (ECF No. 153-4 at 62-63.) “[T]he Court need not determine at [the class certification] stage whether [Virtusa] actually ha[d]” the Cost Per Consultant initiative. Palmer v. Cognizant Tech. Sols. Corp., No. 17-6848, 2022 WL 18214014, at *27 (C.D. Cal. Oct. 27, 2022). Instead, “[f]actual determinations supporting Rule 23 findings must be made by a preponderance of the evidence.” In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 307 (3d Cir. 2008). Plaintiff has satisfied that burden with respect to the notion that Virtusa implemented some version of the initiative. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 8 of 44 PageID: <pageID> 9 if an employee was staffed on a project for more than 18 months and operated at a margin of under 30%, then Virtusa would implement a plan to increase that margin or replace the employee with a “lower cost onsite hiring” or a “visa ready” offshore employee. (ECF No. 142-40 at 2.) This rotation policy appeared to apply to both visa-holding and non-visa-holding United States-based employees. (ECF No. 153-4 at 45.) Finally, Virtusa implemented a new bench policy following the influx of visa-holding employees. Under this policy, Virtusa helped visa-holding employees identify new opportunities after only one day on the bench but waited seven days to assist non- visa employees. (See ECF No. 142-29 at 8-9 (Virtusa presentation on bench policy).) If a visa- holding employee was still on the bench after 60 days, the employee would be “immediately sen[t] . . . back offshore.” (Id. at 9.) The implication being that a visa-holder is still valuable to Virtusa because Virtusa could pay that employee significantly less in Sri Lanka or India. (ECF No. 146 at 23-24.) By contrast, a non-visa holder that remained on the bench for 90 days would be subject to a discussion on “alternatives and options”—suggesting the possibility of termination. (ECF No. 142-29 at 8.) f. Mr. Sugg’s Experience Leo Sugg, a white male from the United States, worked for Virtusa from 2014 to 2017 as Vice President and the Global Account Executive for one of Virtusa’s clients, American International Group, Inc. (AIG). (ECF No. 142-11 at 6, 22; ECF No. 143-4 at 4, 7-8.) Virtusa labels employees like Mr. Sugg “Client Partners” because they are responsible for managing the relationship with one client. (See ECF No. 153-3 at 2.) Client Partners are not hired through the usual three-round interview process. (Id. at 3.) Instead, the process “proceeds on an as needed basis and involves the discretion of individual executives and personnel at Virtusa.” (Id.) Accordingly, Mr. Sugg was hired by Executive Vice President James Francis because of Sugg’s Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 9 of 44 PageID: <pageID> 10 prior experience working with AIG. (ECF No. 143-4 at 1, 3.) Mr. Sugg’s role was “principally relationship driven, business development driven, making sure [AIG] was happy.” (ECF No. 142- 11 at 7.) He was not a billable employee. (See ECF No. 153-3 at 2.) During his tenure, Mr. Sugg was not promoted, nor did he seek a promotion. (ECF No. 142-11 at 25; ECF No. 153 at 42; ECF No. 157 at 18.) Mr. Sugg could not name a position that existed at Virtusa to which he could have been promoted, though he testified that Virtusa “could have gotten creative” and “given [him] more responsibility.” (ECF No. 142-11 at 27.) Mr. Sugg also testified that during his tenure, he was told by Virtusa’s President, Raj Rajogpal, that Mr. Sugg did not “look like someone who sells technology” and should “find someone from India to work with [him].” (Id. at 22-23.) Francis terminated Sugg in 2017 because “Virtusa’s business with AIG fell dramatically.” (ECF No. 143-4 at 6.) Plaintiff contends that during this period Mr. Sugg “still performed well.” (ECF No. 142 at 19.) Mr. Sugg was replaced by Sameer Menon, but when Menon similarly failed to grow the AIG business, he was transitioned to a new role in banking rather than terminated. (ECF No. 142-11 at 26-27.) Mr. Sugg stated that he suffered “stress” and “anxiety associated with switching employers” and sought help from two mental health professionals. (Id. at 3-5.) 2. Facts Relevant to the Motions to Exclude Expert Reports Defendant challenges the admissibility of two of Plaintiff’s experts: (1) Dr. Lang (labor economics) and (2) Dr. Hira (business models of information technology services companies). (See ECF Nos. 144, 145.) Plaintiff relies on Dr. Lang’s Reports for statistical evidence of the disparate treatment non-South Asians faced as a result of Virtusa’s policies, and he relies on Dr. Hira’s Report to contextualize Virtusa’s actions. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 10 of 44 PageID: <pageID> 11 a. Dr. David Lang Plaintiff retained Dr. Lang7 to analyze employment data provided by Virtusa and opine on whether racial disparities existed in Virtusa’s workforce with respect to hiring, promotion, and termination, and if so, whether those disparities were statistically significant. (ECF No. 144-4 at 3.) Dr. Lang concluded that there was a “significant hiring disparity between White and Asian hiring rates.” (Id. at 15.)8 Specifically, Dr. Lang found that “[r]egardless of the number of applicants for a position, the probability that the hired applicant is Asian is very high (84.6% to 93.2%)” and that “[w]hen the number of applicants for a position is relatively high (11 or more applicants), the likelihood of the hired employee being Asian increases significantly” to 93.2%. (Id.) This disparity also existed within Virtusa’s job Tiers: “Regardless of job [T]ier, the probability that the hired applicant is Asian is very high (59.6% to 93.8%).” (Id.) Dr. Lang provided no data on or analysis of the applicant pool. (See generally id. at 5-6.) As for promotions, Dr. Lang found that “White employees are significantly less likely to receive employee ratings of ‘Exceeds Expectations’ (the second highest rating) compared to Asian employees.” (Id. at 15.) White employees were more likely to receive the highest rating, but the difference was not statistically significant. (Id. at 7.) Based on an analysis of less than 20 white employees, Dr. Lang also found that white employees who started at Tier 4 were significantly less likely to be promoted to Tiers 3, 2, and 1 than were Asian employees. (Id. at 9, 15.) However, 7 Dr. Lang holds a Master of Arts degree and Doctorate in economics from Washington University in St. Louis. (ECF No. 144-4 at 3.) Dr. Lang also holds a Bachelor of Arts degree in economics from Stanford University. (Id.) Dr. Lang is a Full Professor and Chair of the Economics Department at California State University, Sacramento. (Id.) 8 Dr. Lang excluded other races from his analyses because “the number of observations was too small.” (ECF No. 144-4 at 3.) As discussed supra, Dr. Lang focused on Asian employees rather than South-Asian employees—the reference group identified in the FAC, (ECF No. 7), and the proposed classes, (ECF No. 141 at 3.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 11 of 44 PageID: <pageID> 12 there was no overall significant difference in the percentage of white versus Asian employees who were never promoted. (Id. at 8.) Dr. Lang attributed non-significance to a small sample size: there were only approximately 17 white individuals not promoted. (See id.) Finally, with respect to terminations, Dr. Lang found that “White employees at Virtusa were 2.19 times more likely to be involuntarily terminated than Asian employees,” employees with visas were 1.68 times less likely to be involuntarily terminated than employees without a visa, and that “[a]fter accounting for gender, job [T]ier and age, being Asian (as opposed to being White) leads to an 8.2 percentage point reduction in the probability of being terminated involuntarily.” (Id. at 12-13, 15.) Dr. Lang drew from a sample of approximately 18 white individuals, and 101 non-visa holding individuals, who were involuntarily terminated. (See id. at 12.) Dr. Lang conducted his analyses based off of data provided by Virtusa. Sometimes the data included the Virtusa employee’s race and nationality, sometimes it included just race, sometimes just nationality, and sometimes neither, which is why sample sizes for different types of analyses varied. (Id. at 4.) Because the study—and the underlying 42 U.S.C. § 1981 claim—are based on race rather than nationality, Dr. Lang used nationality as a proxy for race in all of his analyses when data about race was unavailable but nationality was available. (Id.) Dr. Lang only implemented this proxy system for employees from India or Sri Lanka because Dr. Lang found that—where race and nationality data was available—99.6% of Indian and 100% Sri Lankan individuals also identified as Asian, whereas American employees were too racially diverse to implement the proxy. (Id.) Beyond this common thread, Dr. Lang employed slightly a different methodology for his hiring, promotion, and termination analyses. To opine on Virtusa’s hiring practices, Dr. Lang employed a “name-matching” method. (ECF No. 144-3 at 10-11.) Based on the data provided to him, 4,420 employees were hired by Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 12 of 44 PageID: <pageID> 13 Virtusa between October 31, 2018 and March 12, 2024. (ECF No. 144-4 at 5.) Dr. Lang wrote that for this dataset, “race and nationality [were] seldom reported,” so he employed name matching. (Id.) Under this method, Dr. Lang compared the first and last names of applicants with the names of employees—whose race was known to Virtusa—that were coded as non-Asian or Asian. (Id.) For example, Dr. Lang testified in his deposition that if an applicant were named “Rajesh Koothrappali,” and there were Asian employees with the first name “Rajesh” and the last name “Koothrappali,” then that applicant would be coded as Asian so long as there was not a non- Asian employee with the first name “Rajesh” or the last name “Koothrappali.” (ECF No. 144-3 at 12-13.) For the applicant to be coded as Asian, there would not need to be an employee named “Rajesh Koothrappali”—there would just need to be at least one employee with the first name “Rajesh” and another with the surname “Koothrappali.” (Id. at 13.) Conversely, Dr. Lang testified that if an applicant were named “Rajesh Smith” but there were non-Asian employees named “David Smith” or “Mary Smith,” then “Rajesh Smith” would not be coded as Asian. (Id.) Defendants raised significant concerns with Dr. Lang’s Report and deposition testimony. First, Dr. Lang testified that because of the incomplete race and nationality data related to hiring, he “was unable to perform” an “error analysis.” (ECF No. 150-3 at 6.)9 Second, Dr. Lang did not “distinguish between South Asian and other forms of Asian ethnicity,” but Dr. Lang stated that this concern was “theoretical[]” because “the vast majority of the data . . . happened to be of two particular [countries: India and Sri Lanka].” (ECF No. 144-3 at 13-14.) Third, Dr. Lang stated that the specific method he employed “does not exist out in the world” but noted that “[o]ther individuals have used name matching techniques.” (Id. at 15.) While Dr. Lang could not point to 9 In Plaintiff’s briefing, Plaintiff cites one study for the proposition that name-matching studies have an error rate of between 4% and 30%. (See ECF No. 150 at 13.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 13 of 44 PageID: <pageID> 14 an authoritative treatise on name matching or specific peer-reviewed studies that have employed name matching, he stated that it was “almost certain” that he had come across peer-reviewed studies that employed name matching given that the articles he reads are almost always peer reviewed. (ECF No. 150 at 7-8.)10 Finally, Dr. Lang did not control for certain variables. (See ECF No. 144-3 at 17, 33.) For example, he did not consider the racial composition of the applicant pools or the racial composition of the job market in his Report. (See ECF No. 144-1 at 14; ECF No. 144-4 at 5-6; ECF No. 150-2 at 12.) As for promotions, employee ratings, and job flow, Dr. Lang drew on data for full-time employees where race information was available or ascertainable based on the proxy system. (ECF No. 144-4 at 7.) For each of these analyses, Dr. Lang used t-tests, “inferential statistic[s] used to determine if there is a statistically significant difference between the means of two variables,” to arrive at conclusions. (Id. at 5.) However, neither in his Report nor in his deposition does Dr. Lang mention controlling for any potentially relevant factors beyond race in conducting his analyses. (See ECF No. 144-3 at 34-36.) Finally, as for termination, Dr. Lang similarly used t-tests to find several statistically significant differences involving employees terminated after July 11, 2014. (ECF No. 144-4 at 11- 13.) Dr. Lang does not appear to control for any other factors in making most of these assessments, though he did conduct one logistic regression that controlled for gender, job Tier, and age in evaluating race-based discrimination. (Id. at 14.) 10 In Plaintiff’s briefing, Plaintiff cites several peer reviewed name-matching studies. (See ECF No. 150 at 12-13.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 14 of 44 PageID: <pageID> 15 b. Dr. Ronil Hira Plaintiff retained Dr. Hira11 to analyze Virtusa’s business model and discuss how that model impacts Virtusa’s employment practices. (ECF No. 146 at 3.) In preparation for his Report, Dr. Hira reviewed 60 documents produced by Virtusa, deposition transcripts, and publicly available data, including government visa data and Virtusa’s Securities and Exchange Commission (SEC) filings. (ECF No. 146 at 2, 26-28.) Dr. Hira began his Report by providing an overview of Virtusa. Among other things, he noted “[f]or its onsite workforce, Virtusa relies heavily on importing foreign employees from India and Sri Lanka who require a visa.” (Id. at 5.) Dr. Hira recounted that employers are required to pay those employees with H-1B visas as much as they pay similarly experienced non-visa workers in the United States. (Id. at 5-6.) Only 85,000 new H-1B visas may be issued each year, and demand has far exceeded that cap. (Id. at 6.) Thus, the United States has implemented a lottery system for those spots. (Id.) Dr. Hira opined, based on publicly available data and documents obtained during discovery, that Virtusa submits more visa applications than it needs, with the knowledge that only a portion of them will be successful. (Id.) Dr. Hira reported that Virtusa is one of a series of employers that “improperly attest[] on the H-1B visa petition that there is an 11 Dr. Hira holds a doctorate in Public Policy from George Mason University. (ECF No. 146 at 30.) He also holds a Master of Science from George Mason and a Bachelor of Science from Carnegie-Mellon University, both in Electrical Engineering. (Id.) Dr. Hira is a professor of political science at Howard University and a research associate at the Economic Policy Institute. (Id. at 2.) Dr. Hira has published two books, three peer reviewed articles, six book chapters, and testified seven times before the United States Congress and twice before the Maryland State House of Delegates on the topics of business models and employment practices of information technology service companies. (Id.) Dr. Hira has also published dozens of policy briefs on these topics. (Id.) Dr. Hira served as an expert in Palmer v. Cognizant Technology Solutions Corp., Civ. No. 17- 06848 (C.D. Cal.). (Id. at 2-3.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 15 of 44 PageID: <pageID> 16 actual open position in the U.S. for the sponsored employee to fill when, in reality, no such position exists.” (Id. at 9.) Dr. Hira then explained why companies like Virtusa employ these visa policies. Dr. Hira concluded that “from [his] experience, IT outsourcing firms like Virtusa have adopted workforce strategies designed to predominantly fill U.S. positions with visa workers from countries in South Asia, rather than hiring Americans locally, because of deep-seated racial and ethnic preferences, rather than any business necessity.” (Id. at 11.) Dr. Hira drew this conclusion based on several facts. First, he opined that management and senior members of IT outsourcing firms are generally comprised of South Asians, like Virtusa, and they prefer to hire employees who are racially and ethnically similar to them. (Id. at 12.) Second, Dr. Hira stated that management prefers to hire visa workers because of the bargaining power: H1-B employees will likely not leave Virtusa because the employees would have to return to 10% wages in India or Sri Lanka, as compared to United States wages, so management can exert control over their working condition in an “indenturing”-like manner. (Id. at 12-13.) Indeed, as Dr. Hira opined, Virtusa requires visa employees to sign an “HR bond” before processing their visas, and if the employee leaves the company within less than two years, the employee is required to re-pay the bond amount. (Id. at 13.) Thus, Dr. Hira opined that even though Virtusa is required to pay H1-B visa employees at equal rates to U.S. employees, Virtusa pays them less. (Id. at 14.) Dr. Hira noted that these cost- saving measures came from McKinsey & Company recommendations. (Id.) II. LEGAL STANDARD A. Expert Testimony Federal Rule of Evidence 702 governs the admissibility of expert testimony. The Rule reads as follows: Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 16 of 44 PageID: <pageID> 17 A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if the proponent demonstrates to the court that it is more likely than not that: (a) the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert’s opinion reflects a reliable application of the principles and methods to the facts of the case. Fed. R. Evid. 702. Under Rule 702, “[d]istrict courts are tasked with a ‘rigorous gatekeeping function,’” to ensure that expert testimony is both relevant and reliable. Cohen v. Cohen, 125 F.4th 454, 460 (3d Cir. 2025) (quoting Elcock v. Kmart Corp., 233 F.3d 734, 744 (3d Cir. 2000)). Notwithstanding, “Rule 702 . . . has a liberal policy of admissibility.” Pineda v. Ford Motor Co., 520 F.3d 237, 243 (3d Cir. 2008) (quoting Kannankeril v. Terminix Int’l, Inc., 128 F.3d 802, 806 (3d Cir. 1997)). Expert testimony will be admissible if: (1) “the expert is qualified”; (2) “the proposed testimony is reliable and concerns matters requiring scientific, technical, or specialized knowledge”; and (3) “the expert’s testimony is ‘sufficiently tied to the facts of the case,’ . . . so that it ‘fit[s]’ the dispute and ‘will assist the trier of fact[.]’” Cohen, 125 F.4th at 460 (quoting Daubert, v. Merrell Dow Pharms., Inc., 509 U.S. 579, 591 (1993)). The party proposing the expert testimony must show that each prong is satisfied by a preponderance of the evidence. Oddi v. Ford Motor Co., 234 F.3d 136, 144 (3d Cir. 2000). Importantly, the “gatekeeping function is necessarily ‘flexible,’ . . . granting district courts ‘latitude in deciding how’ these requirements are met.” Cohen, 125 F.4th at 460 (first quoting Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 17 of 44 PageID: <pageID> 18 Daubert, 509 U.S. at 594, and then quoting Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152 (1999)). “Such ‘discretionary authority’ permits a district court to ‘decide whether or when special briefing or other proceedings are needed to investigate reliability.’” Id. (quoting Kumho Tire Co., 526 U.S. at 152). 1. Prong One – Qualification “Qualification refers to the requirement that the witness possess specialized expertise.” Schneider ex rel. Est. of Schneider v. Fried, 320 F.3d 396, 404 (3d Cir. 2003). The basis for this specialized expertise “can be practical experience as well as academic training and credentials.” Elcock v. Kmart Corp., 233 F.3d 734, 741 (3d Cir. 2000) (quoting Waldorf v. Shuta, 142 F.3d 601, 628 (3d Cir. 1998)). While courts have interpreted the special expertise requirement “liberally,” “at a minimum, a proffered expert witness . . . must possess skill or knowledge greater than the average layman.” Elcock, 233 F.3d at 741 (quoting Waldorf, 142 F.3d at 625). Importantly, however, an expert does not have to “be the best qualified” or “have the specialization that the court considers most appropriate.” Pineda, 520 F.3d at 244. 2. Prong Two – Reliability The “reliability requirement ensures that an expert’s testimony is ‘based on the methods and procedures of science, not on subjective belief and unsupported speculation.’” Cohen, 125 F.4th at 461-62 (quoting Karlo v. Pittsburgh Glass Works, LLC, 849 F.3d 61, 80-81 (3d Cir. 2017)). “[A]dmissibility does not hinge on ‘whether a particular scientific opinion has the best foundation, or even whether the opinion is supported by the best methodology or unassailable research.’” Id. at 462 (quoting In re TMI Litig., 193 F.3d 613, 665 (3d Cir. 1999)). Rather, courts look to whether the expert’s testimony is supported by “good grounds.” Id. (quoting UGI Sunbury LLC v. A Permanent Easement for 1.7575 Acres, 949 F.3d 825, 834 (3d Cir. 2020)). In conducting this analysis, courts are to consider “all aspects of an expert’s testimony: the methodology, the Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 18 of 44 PageID: <pageID> 19 facts underlying the expert’s opinion, [and] the link between the facts and the conclusion.” ZF Meritor, LLC v. Eaton Corp., 696 F.3d 254, 291 (3d Cir. 2012) (quoting Heller v. Shaw Indus., Inc., 167 F.3d 146, 155 (3d Cir. 1999)). “The court’s focus must be on the principles and methods, rather than on the ultimate conclusions reached.” Milltown-Ford Ave. Redevelopment Agency v. SB Bldg. Assocs., L.P., Civ. No. 19-21494, 2024 WL 3690776, at *16 (D.N.J. Aug. 7, 2024) (citing In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 742 (3d Cir. 1994)). Because, however, “‘conclusions and methodology are not entirely distinct from one another,’ . . . the court may conduct ‘at least a limited review of the expert’s conclusions “in order to determine whether they could reliably flow from the facts known to the expert and the methodology used.”’” Id. (first quoting General Elec. Co. v. Joiner, 522 U.S. 136, 146 (1997), and then quoting In re Human Tissue Prods. Liab. Litig., 582 F. Supp. 2d 644, 656 (D.N.J. 2008)). As the Third Circuit recently reiterated, “there is no ‘definitive checklist or test,’” for courts to employ when analyzing what constitutes “good grounds.” Cohen, 125 F.4th at 462 (quoting Daubert, 509 U.S. at 593). Nevertheless, along with “any other[ factors] that are relevant,” courts consistently consider the following factors in determining the reliability of a proposed expert: (1) whether a method consists of a testable hypothesis; (2) whether the method has been subject to peer review; (3) the known or potential rate of error; (4) the existence and maintenance of standards controlling the technique’s operation; (5) whether the method is generally accepted; (6) the relationship of the technique to methods which have been established to be reliable; (7) the qualifications of the expert witness testifying based on the methodology; and (8) the non-judicial uses to which the method has been put. In re Paoli, 35 F.3d at 742 & n.8 (citing Daubert, 509 U.S. at 591-94; United States v. Downing, 753 F.2d 1224, 1238-39 (3d Cir. 1985); see also Cohen, 125 F.4th at 462 (noting that “some analysis of these factors is necessary” (quoting UGI Sunbury, 949 F.3d at 834)). Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 19 of 44 PageID: <pageID> 20 “Although these factors are neither exhaustive nor applicable in every case, they provide a convenient starting point for analyzing” reliability. Kannankeril, 128 F.3d at 806-07. But in non- scientific cases, the relevant reliability concerns “may focus upon personal knowledge or experience,” rather than “scientific foundations.” Kumho Tire Co., 526 U.S. at 150; see also United States v. Price, 458 F.3d 202, 212 (3d Cir. 2006) (“It is settled law that an expert may testify about common behavior patterns in a profession or subculture.”); Wolfe v. Allstate Prop. & Cas. Ins. Co., Civ. No. 10-800, 2012 WL 12929813, at *5 (M.D. Pa. July 23, 2012) (“While rate of error, publication,[] peer review[,] and testing are imperative in ascertaining the reliability of a new scientific methodology . . . their application to testimony regarding customs and practices in [a business] industry” may be “both illogical and irrelevant.”); S.Y. v. Roman Cath. Diocese of Paterson, Civ. No. 20-2605, 2024 WL 1231333, at *6 (D.N.J. Mar. 21, 2024) (“[I]n nonscientific cases, the relevant reliability concerns will focus upon personal knowledge and experience of the witness and the methodology used will be applying that experience to the facts of the case.”) (internal quotation marks and citation omitted). Importantly, “good grounds” do not require that an expert’s opinion be perfect. Matter of Soued, 689 F. Supp. 3d 1, 11 (D.N.J. 2023). “Indeed, ‘an expert opinion is not inadmissible because it may contain flaws, nor is it excludable because it provides testimony regarding only one facet or aspect of an action but does not prove the whole case; such vulnerabilities affect the weight of the testimony, not its admissibility.’” Id. (quoting N.J. Dep’t of Env’t Prot. v. Amerada Hess Corp., Civ. No. 15-6468, 2019 WL 4052431, at *6 (D.N.J. Aug. 28, 2019)). If good grounds are shown, an expert’s testimony “should be tested by the adversary process—competing expert testimony and active cross-examination—rather than excluded from jurors’ scrutiny for fear that they will not grasp its complexities or satisfactorily weigh its inadequacies.” United States v. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 20 of 44 PageID: <pageID> 21 Mitchell, 365 F.3d 215, 244 (3d Cir. 2004) (quoting Ruiz-Troche v. Pepsi Cola of P.R. Bottling Co., 161 F.3d 77, 85 (1st Cir. 1998)). 3. Prong 3 – Fit “The third prong of admissibility, fit, concerns whether the expert’s testimony will be helpful to the trier of fact.” In re Johnson & Johnson Talcum Powder Prods. Mktg., Sales Pracs. & Prods. Litig., 509 F. Supp. 3d 116, 132 (D.N.J. 2020) (citing Fed. R. Evid. 702(a)). The Third Circuit has explained that “admissibility depends in part on ‘the proffered connection between the scientific research or test result to be presented and particular disputed factual issues in the case.’” In re Paoli, 35 F.3d at 743 (quoting Downing, 753 F.2d at 1237). “Fit is an issue of relevance and simply means that scientific validity of the method or principles applies to the issues at hand.” Geiss v. Target Corp., Civ. No. 09-2208, 2013 WL 4675377, at *5 (D.N.J. Aug. 30, 2013) (citing United States v. Ford, 481 F.3d 215, 220 n.6 (3d Cir. 2007)). “Whether an expert’s testimony meets the fit requirement ‘is not always obvious, and scientific validity for one purpose is not necessarily scientific validity for other, unrelated purposes.’” In re Johnson & Johnson Talcum Powder Prods. Mktg., 509 F. Supp. 3d at 132 (quoting In re Paoli, 35 F.3d at 743). B. Class Certification “The class action is an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only.” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 348 (2011) (internal quotation marks and citation omitted). The “party proposing class-action certification bears the burden of affirmatively demonstrating by a preponderance of the evidence [its] compliance with the requirements of Rule 23.” Byrd v. Aaron’s Inc., 784 F.3d 154, 163 (3d Cir. 2015), as amended (Apr. 28, 2015) (citing Comcast Corp v. Behrend, 569 U.S. 27, 33 (2013)). “Rule 23 does not set forth a mere pleading standard. A party seeking class certification must affirmatively demonstrate his compliance with the Rule—that is, he must be prepared to Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 21 of 44 PageID: <pageID> 22 prove that there are in fact sufficiently numerous parties, common questions of law or fact, etc.” Dukes, 564 U.S. at 350. A court must engage in a “rigorous analysis” to determine whether all of Rule 23’s prerequisites have been satisfied. Id. (quoting Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 161 (1982)). This rigorous analysis has “three key aspects”: First, the court must find that the requirements of Rule 23 are met and any factual determinations supporting Rule 23 findings must be made by a preponderance of the evidence. Second, the court must resolve all factual or legal disputes relevant to class certification, even if they overlap with the merits. Third, the court must consider all relevant evidence and arguments, including expert testimony, whether offered by a party seeking class certification or by a party opposing it. In re Lamictal Direct Purchaser Antitrust Litig., 957 F.3d 184, 191 (3d Cir. 2020) (internal citations and quotations omitted) (cleaned up). “Merits questions may be considered to the extent—but only to the extent—that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied.” Williams v. Jani-King of Philadelphia Inc., 837 F.3d 314, 322 (3d Cir. 2016) (quoting Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455, 466 (2013)). III. DISCUSSION A. Motions to Exclude Plaintiff’s Experts 1. Dr. Lang Defendant raises two arguments in support of excluding Dr. Lang’s Expert Report and corresponding testimony.12 First, Defendant contends Dr. Lang’s methodology was unreliable. (ECF No. 144-1 at 7-20.) Second, Defendant argues Dr. Lang improperly offered a legal 12 Defendant also seeks to exclude Dr. Lang’s Expert Reply Report, but the Magistrate Judge has already granted Defendant’s motion to strike that Report on other grounds. (See ECF No. 166). Therefore, the Daubert motion is moot with respect to this Report. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 22 of 44 PageID: <pageID> 23 conclusion. (Id. at 20-22.) Plaintiff disagrees on both counts. (See ECF No. 150.) There is no dispute over whether Dr. Lang is qualified to opine or whether his opinions are relevant. (See generally ECF No. 144-1.) a. Reliability Defendant argues Dr. Lang’s methodologies are unreliable for two reasons. First, Defendant challenges Dr. Lang’s use of the “name-matching” method in support of hiring statistics. (ECF No. 144-1 at 7-11.) Second, Defendant argues that Dr. Lang failed to control for relevant factors in his hiring, promotion, and termination analyses. (Id. at 11-20.) The Court is unpersuaded that Dr. Lang’s Report should be excluded on these grounds.13 i Name Matching First, Defendant argues the name-matching method is unreliable because “[i]t has not been subjected to peer review, publication, or scrutiny by Dr. Lang’s peers. . . . Nor did Dr. Lang perform an analysis of this methodology to determine its error rate.” (Id. at 7.) Plaintiff argues “name- matching is a well-established methodology that has been subject to peer review, publication, and scientific scrutiny.” (ECF No. 150 at 12.) As for peer review, it appears that the specific type of 13 Defendant appears to raise a third argument: Dr. Lang’s proxy coding mechanism for race of Virtusa employees—separate from the name-matching method used for the hiring analysis— inflated the number of persons deemed to be “Non-Asians suffering from the employment practices complained of by Plaintiff, that is, non-South Asians not being hired, receiving poor review ratings, not receiving promotions, and being involuntarily terminated.” (ECF No. 144-1 at 9 (citing ECF No. 140-1).) The document Defendant cites to in support of this argument is a declaration from Defendant’s rebuttal expert, Dr. Paul White. In that declaration, Dr. White raises the inflation concern. (See ECF No. 140-1 at 3.) However, the concern only arises in response to Dr. Lang’s stricken Expert Reply Report, (id. at 2.), so the Court need not address this third argument. To the extent there are issues in the original Report that remain unaddressed because the Expert Reply Report is stricken, the “identification of such flaws in generally reliable scientific evidence is precisely the role of cross-examination” and should not serve as grounds for inadmissibility. Quiet Tech. DC-8, Inc. v. Hurel-Dubois UK Ltd., 326 F.3d 1333, 1345 (11th Cir. 2003) (collecting cases, including In re TMI Litig., 193 F.3d 613, 692 (3d Cir. 1999)). Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 23 of 44 PageID: <pageID> 24 name-matching Dr. Lang employed is not peer-reviewed, but that name-matching as a broader methodology is peer-reviewed. (See ECF No. 144-3 at 15.) While Dr. Lang could not point to specific peer-reviewed, name-matching articles in his deposition or a treatise in support of name matching, he testified that he was “almost certain” articles existed, (ECF No. 150-3 at 7-8), and Plaintiff’s brief highlights several instances, (ECF No. 150 at 12-13.) For example, Plaintiff cites a peer-reviewed article that itself reviews 13 articles that have employed name matching. (Id. at 12 (citing Pablo Mateos, A Review of Name-based Ethnicity Classification Methods and their Potential in Population Studies, 13 Population, Space and Place 243 (2007)). In that article, Dr. Mateos sought to “extract[] methodological commonalities” in name-matching studies. Mateos at 243. First, the 13 studies all included “a name reference list [that was] independently built . . . .” Id. at 249 (emphasis in original). “Generally, [the studies] all used some type of ‘ethnic origin information’ in the reference population, such as self-reported ethnicity, [or] country of birth or nationality, to classify individuals into ethnic groups” in the reference list. Id. at 250. Second, the studies included “a separate target population” and individuals from that target population were “classified into ethnic groups.” Id. at 249 (emphasis in original.) There were two common approaches for classifying such individuals: manually creating a “rich number of ‘fuzzy rules’” that the classifier applies “in order to decide the group into which an individual should be assigned”; or “apply[ing] an automated algorithm to search for the name of each individual in the target population against the reference list, and then assign . . . [an] ethnic group for that name to the individual.” Id. at 252. Third, “all of the studies measure the accuracy of the name-based classification by comparing it to a ‘gold standard’ for the ethnicity of the individuals in the target population, which had to be previously known through an independent source.” Id. In other words, the authors had access to information confirming the Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 24 of 44 PageID: <pageID> 25 ethnicities of the individuals in the target populations, so they could compare their results to that information and assess their approach’s accuracy. The study found that the 13 articles’ approaches had a positive predictive value—“the proportion of persons classified” into an ethnic group that “were correctly classified as such”—of between 70% and 96%. Id. at 254. There were “no substantial differences between the accuracy of the manual [versus] automatic classification methods.” Id. Dr. Mateos notes that one limitation of the studies was that “[n]ames usually only reflect patrilineal heritage” and thus name matching is “incapable of identifying mixed ethnicity or women’s ethnicity in mixed marriages.” Id. at 255. Nonetheless, the study concluded that while “this methodology cannot entirely replace self assigned ethnicity information, it provides a sufficient level of classification confidence to be used in the measurement of inequalities . . . .” Id. at 256. On the whole, Dr. Lang’s Expert Report survives Rule 702’s “liberal policy of admissibility.” Pineda, 520 F.3d at 243 (citation omitted). Many of the Paoli factors are sufficiently present. See 35 F.3d at 742 n.8. For example, Dr. Lang is qualified as an expert and employed hypothesis testing to arrive at his conclusions. Further, the method of name-matching has been subject to peer review, and Dr. Lang’s approach is substantially similar to those in Dr. Mateos’s article. Dr. Lang created a reference list based on the known ethnicities of employees and classified applicants, based on a “rich number” of rules, in reference to that list. (ECF No. 144-3 at 12-13; ECF No. 144-4 at 5.) Dr. Lang could not perform an error analysis for this particular study, (ECF No. 150-3 at 6), but Dr. Mateos identified an error rate between 4% and 30% in 13 name-matching articles, Mateos at 254. Cases in which courts have considered name matching in the context of Daubert motions are also instructive. In Palmer v. Cognizant Technology Solutions Corporation, the court Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 25 of 44 PageID: <pageID> 26 considered an expert’s analysis that employed name-matching to categorize employees as South Asian. See Civ. No. 17-6848, 2022 WL 18214014, at *11 (C.D. Cal. Oct. 27, 2022). Where available, the expert used birth country and passport information to code employees as South Asian. Id. To determine if other employees were South Asian, the expert analyzed their last names—i.e., employed name matching. Id. The court found the methodology permissible even if “somewhat imprecise.” Id. at *12 n.17. By contrast, the court deemed name-matching inadmissible in Koehler v. Infosys Ltd. Inc., 628 F. Supp. 3d 835, 862-63, 891 (E.D. Wis. 2022). There, the plaintiff’s expert employed name-matching as follows: [The expert] considered an Infosys employee to be Indian or South Asian if (a) the employee was born in India or South Asia, or (b) the employee’s last name matched the last name of another Infosys employee born in India or South Asia and was not one of the sixty- three “Western” names removed from [a list created by the expert]. [The expert] considered an applicant to be Indian or South Asian if the applicant’s last name matched the last name of an Infosys employee born in India or South Asia and was not one of the sixty- three “Western-sounding” names removed from the list. If an employee or applicant did not meet one of the preceding criteria, that employee or applicant was considered non-Indian or non-South Asian. Id. at 856-57 (emphases in original). The court concluded the methodology was not reliable because the record reflected that it was “created for this litigation, not tested, not peer-reviewed, [and had] no known history of error rates.” Id. at 891. But many of those potential issues are not apparent here. Dr. Lang testified that he employed a “conservative approach . . . to mitigate potential errors.” (ECF No. 150-3 at 6.) And unlike in Infosys, this Court has been presented with evidence that name matching is a peer-reviewed method with known error rates. While it is true that Dr. Lang created his specific name-matching method for purposes of this litigation, the method was designed in response to the unique data with which he was presented. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 26 of 44 PageID: <pageID> 27 ii Controlling for Other Variables Defendant’s argument that Dr. Lang’s Report is unreliable because it does not control for other variables is similarly unpersuasive. With respect to hiring, Defendant argues that Dr. Lang “simply counted the number of allegedly ‘Asian’ and allegedly ‘Non-Asian’ employees . . . hired and found the gap between the two significant.” (ECF No. 144-1 at 13-14.) But Dr. Lang did not account for, among other things, “the relative demographic proportions of qualified, interested, and available candidates who presented themselves for hiring consideration” nor “did he consider the factors that were considered by Virtusa’s hiring managers” such as skills the customer required in Virtusa employees or whether the applicant was open to travel or relocating. (Id. at 14.) As for performance and promotions, Defendant argues that Dr. Lang again employed a binary analysis— he tested “whether an individual was ‘Asian’ or ‘Non-Asian’ . . . and whether they were promoted.” (Id. at 17.) But, Defendant argues, Dr. Lang “made no effort to determine what factors were considered by Virtusa’s supervisors in performance reviews or promotions,” or “which employees were interested in or sought a promotion.” (Id. at 18.) Finally, Defendant contends Dr. Lang’s termination analyses failed to account for factors such as Virtusa supervisors’ considerations in making termination decisions, changes in the job market, and tenure and disciplinary records of terminated employees. (Id. at 19-20.) Plaintiff argues Dr. Lang was not required to control for other explanatory factors in his hiring, promotion, or termination analyses. (ECF No. 150 at 16- 30.) The Court agrees that Dr. Lang’s methods are sufficiently reliable under Daubert. As the Supreme Court stated, “[w]hile the omission of variables from a regression analysis may render the analysis less probative than it otherwise might be, it can hardly be said, absent some other infirmity, that an analysis which accounts for the major factors must be considered unacceptable Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 27 of 44 PageID: <pageID> 28 as evidence of discrimination . . . . Normally, failure to include variables will affect the analysis’ probativeness, not its admissibility.” Bazemore v. Friday, 478 U.S. 385, 400 (1986) (internal citation omitted). Following Bazemore, courts in and out of this Circuit have continued to admit simple statistical analyses under Daubert that failed to control for other variables. See Adams v. Ameritech Servs., Inc., 231 F.3d 414, 425, 428 (7th Cir. 2000) (finding report that examined “correlations that existed between the ages of employees and the companies’ decisions to terminate” admissible even when expert did not “run a multiple-regression analysis” because disputes over inclusion of variables “went to the weight of the evidence . . . not to its admissibility”); Obrey v. Johnson, 400 F.3d 691, 695-97 (9th Cir. 2005) (finding study “based entirely on statistical disparities” between “the race of managers selected. . . compared to the race of those who applied for managerial positions” admissible because “objections to a study’s completeness generally go to the weight, not the admissibility of the statistical evidence . . . and should be addressed by rebuttal, not exclusion.” ) (internal quotation marks and citation omitted); Gutierrez v. Johnson & Johnson, Civ. No. 01-5302, 2006 WL 3246605, *5 (D.N.J. Nov. 6, 2006) (“Generally, decisions regarding what control variables to include in an expert report go to weight, not admissibility.”)14 While Defendant may have raised serious concerns with Dr. Lang’s statistical analyses of hiring, promotion, and termination data, those potential weaknesses go to the weight of Dr. Lang’s 14 The cases cited by Defendant are inapposite because they concern the type of statistical evidence needed to prove causation rather than the variables needed to be admissible under Daubert. See Bruno v. W.B. Saunders Co., 882 F.2d 760, 767 (3d Cir. 1989); Bell v. Lockheed Martin Corp., Civ. No. 08-6292, 2010 WL 3724271, at *7 (D.N.J. Sept. 15, 2010). The question here is “not whether the report[] proffered by the plaintiff[] prove[s] the entire case; it is whether [it was] prepared in a reliable and statistically sound way.” Adams, 231 F.3d at 425 (7th Cir. 2000); see also Obrey, 400 F.3d at 694. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 28 of 44 PageID: <pageID> 29 opinions rather than admissibility.15 Dr. Lang’s report and corresponding testimony “should be tested by the adversary process . . . rather than [be] excluded from jurors’ scrutiny for fear that they will not grasp its complexities or satisfactorily weigh its inadequacies.” Mitchell, 365 F.3d at 244 (citation omitted.) b. Opinion on the Ultimate Question Defendant argues that Dr. Lang’s conclusion should be excluded because it “invad[es] the exclusive province [of] the jury in applying the law to the facts” and “is an improper legal conclusion” that Dr. Lang is unqualified to make. (ECF No. 144-1 at 20.) Dr. Lang’s challenged conclusion is as follows: Based on these findings, it is my expert opinion that there is significant evidence that non-Asians employees are being discriminated against at Virtusa in terms of hiring, employee ratings and promotions, and through terminations. Further, there is significant evidence that visa employees are being favored at Virtusa with lower involuntary termination rates. 15 It is true that “[t]here may . . . be some regressions so incomplete as to be inadmissible as irrelevant.” Bazemore, 478 U.S. at 400 n.10. These analyses would face challenges under Daubert’s “fit” rather than “reliability” prong. Indeed, Defendant appears to make such an argument with respect to the hiring analysis. Defendant argues it is “egregious” that Dr. Lang failed to control for “the racial composition of the applicant pools.” (See ECF No. 144-1 at 14.) If Dr. Lang found the applicant pool was predominately Asian, Defendant argues, then it would not be surprising that the hires were predominately Asian, so any analysis that does not factor in this information would be irrelevant for finding disparate treatment. Id. But a statistical showing of discrimination need not be “based on [an] analysis of the characteristics of actual applicants” because the “application process might itself not adequately reflect the actual potential applicant pool, since otherwise qualified people might be discouraged from applying because of a self- recognized inability to meet the very standards challenged as being discriminatory.” Dothard v. Rawlinson, 433 U.S. 321, 330 (1977) (citing Int’l Brotherhood of Teamsters v. United States, 431 U.S. 324, 365-367 (1977); see also Buchanan v. Tata Consultancy Servs., Ltd., Civ. No. 15-01696, 2017 WL 6611653, at *11 (N.D. Cal. Dec. 27, 2017) (finding expert’s failure to consider applicant pool was 40% South Asian went to weight, not admissibility). Therefore, this is an issue more appropriately directed to the weight of Dr. Lang’s opinions as opposed to admissibility. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 29 of 44 PageID: <pageID> 30 (Id. (quoting ECF No. 144-4 at 16.) Plaintiff argues that this conclusion is permissible because Dr. Lang “does not offer the opinion that Virtusa engaged in a pattern or practice of race discrimination, only that his statistical analyses offer evidence of discrimination.” (ECF No. 150 at 31 (emphasis in original) (citing ECF No. 144-4 at 16).) And in any event, Plaintiff submits, Dr. Lang, “will not offer the legal conclusion that Virtusa engaged in intentional race discrimination” but instead will testify that “his statistical analyses show disparities that are not explained by chance and not explained by factors other than race and are therefore suggestive of discrimination.” (Id. at 32.) “An opinion is not objectionable just because it embraces an ultimate issue.” Fed. R. Evid. 704(a). But the “use of legal terms of art or legal conclusions by an expert” are inadmissible because they “could lead the jury to be prejudicial against the opposing party.” O’Bryant v. Johnson & Johnson, Civ. No. 20-2361, 2022 WL 7670296, at *6 (D.N.J. Oct. 13, 2022) (internal quotation marks and citation omitted). The “term ‘discrimination’ has a specialized legal meaning that is more precise than the lay understanding of the term.” United States v. Xue, 597 F. Supp. 3d 759, 774 (E.D. Pa. 2022) (quoting Burkhart v. Washington Metro. Area Transit Auth., 112 F.3d 1207, 1212 (D.C. Cir. 1997)). Accordingly, courts have held that statistical experts may not draw conclusions about whether a party acted in a discriminatory manner. See, e.g., Chen-Oster v. Goldman, Sachs & Co., Civ. No. 10-6950, 2022 WL 814074, at *10 (S.D.N.Y. Mar. 17, 2022) (“[T]o the extent that [the expert] uses the term ‘sex discrimination,’ which is a legal term of art, he is making a legal conclusion that is improper as it usurps the role of the jury. . . . [T]he expert may opine . . . so long as he does not use the term ‘sex discrimination.’”) (citation omitted); Simpson v. Dart, Civ. No. 18-0553, 2021 WL 1906469, at *2 (N.D. Ill. May 12, 2021) (finding expert could not opine on “any . . . legal conclusion about ‘discrimination’ as understood as a legal Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 30 of 44 PageID: <pageID> 31 term of art”); Chicago Tchrs. Union, Loc. v. Bd. of Ed. of City of Chicago, Civ. No. 12-10311, 2020 WL 914881, at *13 (N.D. Ill. Feb. 25, 2020) (“[Expert] cannot opine on whether the Board discriminated against employees nor whether there is statistically significant evidence of racial discrimination.”). Dr. Lang’s conclusion must be excluded. Dr. Lang’s Report opines on the legal term of art—“discrimination.” It is no matter that Dr. Lang caveats his written conclusion by noting there is “evidence” of discrimination rather than that there plainly is discrimination. See Chicago Tchrs. Union, at *13. The same goes for his anticipated testimony that his analyses “are suggestive of discrimination.” Therefore, Dr. Lang’s conclusion must be excluded to the extent it uses the word “discrimination,” and should Dr. Lang testify, he may not testify that his findings are “suggestive of discrimination.” Rather, Dr. Lang’s conclusions must be limited to his statistical findings. Defendant’s Motion is therefore granted in part and denied in part. 2. Dr. Hira Defendant argues Dr. Hira’s Expert Report and corresponding testimony should be excluded on three Rule 702 grounds. First, Defendant argues Dr. Hira is not qualified to opine on the topics in his Report. (ECF No. 145-1 at 10-11; ECF No. 160 at 9-12.) Second, Defendant contends Dr. Hira’s report is not reliable because Dr. Hira did not employ a proper methodology and of the 36,000 documents produced in discovery, Dr. Hira only reviewed 60 that were picked by Plaintiff. (ECF No. 145-1 at 11; ECF No. 160 at 5-9.) Third, Defendant argues Dr. Hira’s report is irrelevant because it would not assist the Court in deciding whether to certify a class. (ECF No. 145-1 at 8-9.) In addition to these Rule 702 considerations, Defendant argues Dr. Hira made improper legal conclusions that should be excluded. (ECF No. 145-1 at 9; ECF No. 160 at 12-18). The Court finds these arguments unpersuasive. In making this determination, the Court Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 31 of 44 PageID: <pageID> 32 finds particularly instructive the decision in Cognizant, 2022 WL 18214014, at *15-17 (denying motion to exclude Dr. Hira’s testimony concerning business model of Virtusa’s competitor). a. Qualification Defendant’s initial argument goes to the first prong of the 702 inquiry—qualification. To be sufficiently qualified, the individual must have “specialized expertise” on the topic of their testimony, and the basis for this expertise can be “academic training and credentials.” Elcock v. Kmart Corp., 233 F.3d 734, 741 (3d Cir. 2000) (internal quotation marks and citation omitted). This is a “liberal” standard. Id. (“[A]t a minimum, a proffered expert witness . . . must possess skill or knowledge greater than the average layman.”) (internal quotation marks and citation omitted). Here, Dr. Hira’s report states that he is opining on “business models of information technology services companies.” (ECF No. 146 at 3.) With a Ph.D. in political science, Dr. Hira has “published two books, three peer reviewed articles, six book chapters, and testified seven times before the U.S. Congress and twice before the Maryland State House of Delegates on” the topic of business models and employment practices of information technology services companies. (ECF No. 146 at 2; see also id. at 32-47.) This topic is therefore “well within his area of expertise” and “he is qualified to opine.” Cognizant, 2022 WL 18214014, at *17. b. Reliability Defendant’s next argument goes to the second prong of the 702 inquiry—reliability. Defendant raises two concerns. First, Defendant contends Dr. Hira did not employ a “legitimate, structured analytical method” and instead relayed a factual narrative as opposed to a legitimate expert opinion. (ECF No. 145-1 at 5, 9-11, 14.) Second, Defendant argues that Dr. Hira’s opinions are not reliable because he only reviewed a small portion of allegedly cherry-picked documents. (Id. at 11; ECF No. 160 at 6.) As to the first point, Defendant’s argument is unpersuasive because Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 32 of 44 PageID: <pageID> 33 in cases involving “specialized,” rather than scientific knowledge, the relevant reliability concerns “may focus upon personal knowledge or experience,” rather than “scientific foundations.” Kumho Tire Co., 526 U.S. at 150. Indeed, standard reliability considerations such as rate of error, publication, and peer review are “illogical and irrelevant” when it comes to opinions “regarding customs and practices” in a particular business industry. Wolfe, 2012 WL 12929813, at *5. So long as Dr. Hira applied his “experience to the facts of the case,” then his conclusions are sufficiently reliable to be admissible. S.Y., 2024 WL 1231333, at *6 (internal quotation marks and citation omitted). Here, Dr. Hira drew on his experience studying offshore business models in providing his analysis. This was sufficient. See Cognizant, 2022 WL 18214014, at *16 (stating that Dr. Hira “appears to rely primarily on his own expertise and understanding in opining on the industry generally”). And while courts have held that “[a]cting simply as a narrator of the facts” does not pass muster under Daubert, see SEC v. Tourre, 950 F. Supp. 2d 666, 675 (S.D.N.Y. 2013), Dr. Hira does more than that. In paragraphs 16 to 20 of his Report—the section challenged by Defendant (ECF No. 145-1 at 14)—Dr. Hira draws on his experience studying offshore business models to explain why, in his opinion, other companies predominately fill their positions with South Asian visa workers and the ways in which those companies are similar to Virtusa. (ECF No. 146 at 11- 14.)16 16 Defendant argues this same section of Dr. Hira’s Report is also improper because it “focuses singularly on the question of Virtusa’s motivation.” (ECF No. 145-1 at 16.) It is true that “experts may not provide testimony concerning . . . . the intent, motives or states of mind of [defendant] corporations . . . .” Krys v. Aaron, 112 F. Supp. 3d 181, 203 (D.N.J. 2015) (internal quotation marks and citation omitted). This is because such testimony “offers inappropriate legal conclusions and invades the province of the jury.” TAKTL, LLC v. IWR, N. Am., LLC, Civ. No. 18-1546, 2024 WL 4343141, at *11 (W.D. Pa. Sept. 30, 2024). But an “expert may testify about common behavior patterns in a profession or subculture.” United States v. Price, 458 F.3d 202, 212 (3d Cir. 2006). And “it is permissible to testify that facts and circumstances surrounding that Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 33 of 44 PageID: <pageID> 34 As to the second point, while it is true that Dr. Hira reviewed a hand-picked, small portion of the documents received during discovery, his Report “is based on decades of research about the industry and the visa regime under which [Virtusa] operates, in addition to the non-public, [Virtusa]-specific documents produced in this action. Dr. Hira also relied on public information about [Virtusa], including [Virtusa]’s own 10-K and USCIS data about H-1B visa approvals.” Cognizant, 2022 WL 18214014, at *16. (See also ECF No. 146 at 27-28.) To the extent it is noteworthy that Dr. Hira has not reviewed the vast majority of the documents produced in discovery such that it is possible that those documents therefore contradict his findings, this fact goes to the weight of Dr. Hira’s testimony rather than its admissibility. See S.Y., 2024 WL 1231333, at *7; Cognizant, 2022 WL 18214014, at *16. c. Fit Defendant’s subsequent argument goes to the third prong of the 702 inquiry—fit or relevance. “The third prong of admissibility, fit, concerns whether the expert’s testimony will be helpful to the trier of fact.” In re Johnson & Johnson Talcum Powder Prods. Mktg., Sales Pracs. & Prods. Litig., 509 F. Supp. 3d 116, 132 (D.N.J. 2020) (citing Fed. R. Evid. 702(a)). Defendant argues that Dr. Hira’s Report will not be helpful for class certification because it does “nothing to behavior may be ‘indicative’ of a certain intent.” Eddystone Rail Co., LLC v. Bridger Logistics, LLC, Civ. No. 17-00495, 2022 WL 22885376, at *2 (citing United States v. Farrish, 297 F. App’x 162, 166 (3d Cir. 2008). Nowhere in the challenged section of Dr. Hira’s Report does Dr. Hira usurp the role of the factfinder by opining on Defendant’s intent. Instead, Dr. Hira opines that other companies with similar practices to Defendant have engaged in outsourcing “because of deep-seated racial and ethnic preferences, rather than any business necessity.” (ECF No. 146 at 11.) Because the fact finder is left with the ultimate task of deciding whether Virtusa’s intent was discriminatory, the section of the Report is proper. Cf. Palmer v. Cognizant Tech. Sols. Corp., Civ. No. 17-6848, 2023 WL 4155403, at *12 (C.D. Cal. June 1, 2023) (declining to exclude analysis when “Dr. Hira provide[d] context from which the jury could properly make informed inferences about [defendant]’s intent or motive.”) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 34 of 44 PageID: <pageID> 35 answer any common question concerning hiring, termination, or promotions; do[es] not establish any policy or practice; and do[es] not help Plaintiff avoid the obvious fact that hiring, termination, and promotion decisions at Virtusa are highly individualized and predicated upon the discretion of hundreds or thousands of Virtusa employees and clients.” (ECF No. 145-1 at 9 (emphasis in original).) But the question on a Daubert motion is not whether the expert will make or break a subsequent decision but rather whether it will be “helpful” in making those decisions. In re Johnson & Johnson, 509 F. Supp. 3d at 132. Here, “Dr. Hira’s conclusions about how a business model reliant on H-1B visa holders would help maximize profits lend support to [Plaintiff’s] case for commonality because they help to explain why a company might pursue” certain visa policies and “thus substantiate [Plaintiff’s] contention that [Virtusa] had such a policy.” Cognizant, 2022 WL 18214014, at *15. Additionally, “Dr. Hira’s opinion that cultural preferences—not only unconscious bias, but also supervisors’ preferences relating to the relationship between supervisors and supervisees—might lead managers to develop a preference for [South Asian] workers even in local hiring is obviously relevant to [Plaintiff’s] claims.” Id. at *16. And while Dr. Hira’s conclusions about Virtusa are more limited than his opinions about the industry as a whole given the dataset Dr. Hira reviewed, the Court will not exclude his opinions on that ground but acknowledges that “these limitations may limit their weight.” Id. d. Opinions on the Ultimate Question Finally, Defendant challenges certain statements on the grounds that they are legal conclusions. (ECF No. 145-1 at 9 (quoting ECF No. 146 at 9, 12-14, 24).)17 These statements 17 Defendant introduces additional statements in its reply brief, (see ECF No. 160 at 12-13), but it “is axiomatic in federal practice that arguments raised for the first time in a reply brief should Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 35 of 44 PageID: <pageID> 36 are: (1) “some IT outsourcing firms have wrongfully used the H-1B visa program to secure visas for future or fictitious work, improperly attesting on the H-1B visa petition that there is an actual open position in the U.S. for the sponsored employee to fill when, in reality, no such position exists,” (ECF 146 at 9); (2) the “indenturing nature” of the H-1B visa program, (id. at 12-13); and (3) that underpayment “constitutes wage theft.” (Id. at 14.)18 The first statement is proper. Courts have upheld similar statements concerning compliance with laws or regulations when the statements concerned business customs. See, e.g., United States v. Leo, 941 F.2d 181, 196-97 (3d Cir. 1991); Bartoli v. Novartis Pharms. Corp., Civ. No. 13-0724, 2014 WL 1515870, at *7 (M.D. Pa. Apr. 17, 2014). And the use of the word “wrongfully” does not make the statement excludable. Cf. Gov’t Emps. Ins. Co. v. Adams Chiropractic Ctr., Civ. No. 19-20633, 2023 WL 4248817, at *7 (D.N.J. June 29, 2023) (finding expert “report d[id] not improperly instruct the jury on what result to reach simply by employing the word ‘legitimate.’”) The second statement is proper because it concerns industry customs rather than legal conclusions. See Price, 458 F.3d at 212 (3d Cir. 2006) (“It is settled law that an expert may testify about common behavior patterns in a profession or subculture.”). The third statement is arguably improper. See Cognizant, 2022 WL 18214014, at *16 n. 25 (“Dr. Hira’s specific conclusion that Cognizant engages in ‘wage theft’ does appear to be beyond the scope of his expertise.”). However, Dr. Hira has agreed to withdraw the phrase, so the point is moot. (See ECF No. 151 at 14 n.3) be disregarded.” Wilson v. Parker, Civ. No. 18-2954, 2018 WL 6696783, at *5 (D.N.J. Dec. 20, 2018) (internal quotation marks and citation omitted). 18 Defendant also challenges Dr. Hira’s assertion that Virtusa’s visa policies are not the result of “business necessity.” (ECF No. 145-1 at 9 (citing ECF No. 146 at 24).) This statement is admissible for the same reason Dr. Hira’s similar statement was proper. See supra n.16 (citing ECF No. 146 at 11.) Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 36 of 44 PageID: <pageID> 37 Therefore, Defendant’s motion to exclude Dr. Hira’s Expert Report is denied. B. Motion for Class Certification Under Rule 23(a)(1)-(4), a class may be certified only if: (1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class. Fed. R. Civ. P. 23(a). These requirements are respectively referred to as the numerosity, commonality, typicality, and adequacy requirements. See id. “[E]very putative class action must satisfy the four requirements of Rule 23(a) and the requirements of either Rule 23(b)(1), (2), or (3).” Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 590 (3d Cir. 2012). Here, Plaintiff seeks certification under Rule 23(b)(3). (ECF No. 142 at 37.) Where a party seeks to certify a class under Rule 23(b)(3), the moving party must satisfy three additional requirements. See Byrd, 784 F.3d at 163. First, the party “must prove by a preponderance of the evidence that the class is ascertainable.” Id. Second, the party seeking certification must show that “questions of law or fact common to class members predominate over any questions affecting only individual members.” Fed. R. Civ. P. 23(b)(3). Third, the party must show “that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” Id. These additional requirements are, respectively, referred to as the ascertainability, predominance, and superiority requirements. See, e.g., Byrd, 784 F.3d at 161 n.4, 162-64; Johnson v. Comodo Grp., Inc., Civ. No. 16-04469, 2024 WL 2933195, at *3 (D.N.J. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 37 of 44 PageID: <pageID> 38 Jun. 10, 2024). A district court “must first determine whether the putative class satisfies the . . . provisions of Rule 23(a), and then proceed[] to analyze the . . . provisions of Rule 23(b)(3).” In re Citizens Bank, N.A., 15 F.4th 607, 612 (3d Cir. 2021). Plaintiff argues that pursuant to Rule 23(b)(3), common questions of law and fact predominate over individualized questions in this matter, and that a class action is superior to other available methods of adjudicating the claims. (ECF No. 142 at 37-46.) Plaintiff proposes the following three classes: Hiring Class: All non-South Asian applicants who applied for a position with Virtusa in the United States between February 28, 2014 and the date of class certification and who were not hired. Promotion Class: All non-South Asian employees who worked for Virtusa in the United States between February 28, 2014 and the date of class certification and who were not promoted. Terminations Class: All non-South Asian employees who worked for Virtusa in the United States between February 28, 2014 and the date of class certification and who were involuntarily terminated. (ECF No. 141 at 3.) The Court addresses whether class certification is appropriate for each of the three classes. Citizens Bank, 15 F.4th at 612. 1. Rule 23(a) a. Numerosity The numerosity requirement is disputed by the parties. (ECF No. 142 at 23-24; ECF No. 153 at 22-27; ECF No. 157 at 5-8.) Numerosity is met where “the class is so numerous that joinder of all members is impracticable.” See Fed. R. Civ. P. 23(a)(1). Although there is no minimum number of members needed to satisfy numerosity, courts hold as a general rule that if the proposed members exceed forty people, the first prong of Rule 23(a) has been met. See Stewart v. Abraham, 275 F.3d 220, 226-27 (3d Cir. 2001). Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 38 of 44 PageID: <pageID> 39 i Hiring Class Plaintiff estimates—based on Dr. Lang’s declaration—that the Hiring Class is in the tens of thousands with a cap of 23,589. (ECF No. 142 at 24 (citing ECF No. 142-1 ¶¶ 2-4).) Dr. Lang reviewed data provided by Virtusa which revealed 34,806 individuals who applied to and were rejected from Virtusa during the class period. (ECF No. 142-1 ¶ 2.) Dr. Lang subtracted instances in which individuals applied multiple times and were rejected multiple times and arrived at 23,589. (Id. ¶ 3). Dr. Lang stated that race and national origin data were seldom available for application data but would be able to “ascertain the race of Virtusa’s applicants by utilizing a names-based analysis similar to what” he employed in his Expert Report. (Id. ¶ 4-5.) Defendant argues that Dr. Lang, on whom Plaintiff solely relies, “presents no actual evidence of [the class size number]. Rather, he simply counts the total number of applicants, regardless of race.” (ECF No. 153 at 25 (emphasis in original).) In other words, the only information Dr. Lang provided is that 23,589 individuals were rejected and an allusion to future name matching is insufficient. Plaintiff replies that “common sense” dictates that the numerosity requirement is met. (ECF No. 157 at 7 (citing Marcus, 687 F.3d at 596-97.) The question therefore boils down to whether the fact that 23,589 individuals were rejected by Virtusa is enough to establish, by a preponderance of the evidence, that at least 40 non-South Asian individuals were rejected. The Court finds that this is not enough. The Third Circuit has given the numerosity requirement “real teeth” in recent years. Mielo v. Steak‘n Shake Operations, Inc., 897 F.3d 467, 484 (3d Cir. 2018) (citation omitted). “[W]here a putative class is some subset of a larger pool, the trial court may not infer numerosity from the number in the larger pool alone.” Hayes v. Wal-Mart Stores, Inc., 725 F.3d 349, 358 (3d Cir. 2018.) It is true that when a plaintiff “show[s] sufficient circumstantial evidence specific to the products, problems, parties, and Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 39 of 44 PageID: <pageID> 40 geographic areas actually covered by the class definition . . . . the court [may] rely on ‘common sense’ to forgo precise calculations and exact numbers.” Marcus, 687 F.3d at 596-97. But the Third Circuit has repeatedly found plaintiffs have not satisfied numerosity when applying this principle. See Marcus, 687 F.3d at 595-97 (numerosity unsatisfied when 740,102 vehicles leased nationwide but no evidence in the record about vehicles leased in New Jersey—the geographic limitation of the class); Hayes, 725 F.3d at 357-58 (numerosity unsatisfied when “the only conclusion that can be drawn from the evidence presented to the trial court is that the number of class members would be equal-to-or-less-than 3,500 and equal-to-or-greater-than zero”); Mielo, 897 F.3d at 486 (numerosity unsatisfied when evidence of “between 14.9 million to 20.9 million persons with mobility disabilities who live in the United States” but “no evidence that would permit [the Third Circuit] to use ‘common sense’ to determine—rather than speculate about—the portion of those disabled individuals who have actually patronized a relevant Steak ‘n Shake restaurant, let alone the portion who have experienced or will experience an ADA violation at one of those restaurants.”). In these type of cases, the “high[] likel[ihood] that at least 40 individuals” from the larger pool would be in the sub-class is insufficient. Mielo, 897 F.3d at 486. And even though it may be “tempting to assume [that the] class meets” the numerosity requirement based on the larger pool, the Court must not “giv[e] in to such temptation” and “excuse speculation.” Id. at 485 (internal quotation marks and citation omitted). Ultimately, it is a “plaintiff's burden to demonstrate numerosity by a preponderance of the evidence.” Hayes, 725 F. 3d at 358. And future calculations are of no import because courts “cannot take a wait-and-see approach to numerosity or any other requirement of Rule 23.” Id. Here, Plaintiff has not established numerosity. The only evidence Plaintiff cites is the size of the larger pool of rejected applicants. Even though it is likely at least 40 members of this pool Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 40 of 44 PageID: <pageID> 41 are not South Asian, Plaintiff provides no evidence for the Court to make this inference. Indeed, Dr. Lang’s Expert Report—while admissible under Daubert and not cited by Plaintiff in support of numerosity—only provides race estimates for the applicants who were hired, (ECF No. 144-4 at 5), so the Court has no way of determining, beyond speculation, the number of non-South Asian individuals who were not hired. And even if the numerosity issue would be resolved if Dr. Lang applied name matching to the applicant data, the Court cannot accept a “wait-and-see approach.” Hayes, 725 F.3d at 358. “[T]he only conclusion that can be drawn from the evidence presented to the trial court is that the number of class members would be equal-to-or-less-than [23,589] and equal-to-or-greater-than zero.” Id. at 357-58. The Court must therefore deny certification with respect to the Hiring Class. See Fuentes v. Super Bread II Corp., Civ. No. 18-6736, 2020 WL 7237942, at *6 (D.N.J. Dec. 9, 2020) (“Because a failure to meet the numerosity requirement is dispositive of [plaintiff’]s motion to certify a class, the [c]ourt will not address the other three requirements under Rule 23(a)—namely, commonality, typicality, and adequacy of representation—or the requirement[s] under Rule 23(b).”19 19 Even if Plaintiff provided enough information such that the Court could apply “common sense” and find numerosity satisfied, Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 596-97 (3d Cir. 2012), class certification would still be denied with respect to the Hiring Class because Mr. Sugg was in fact hired by Virtusa, (see ECF No. 143-4 at 4). It is axiomatic that a class representative must be a member of the proposed class. See Amchem Prods., v. Windsor, 521 U.S. 591, 594-95 (1997) (“Representatives must be part of the class . . . .”); In re Pett Food Prods. Liab. Lit., 629 F.3d 333, 343 (3d Cir. 2010) (“Class representatives must be part of the class . . . .”) (internal quotation marks and citation omitted). Courts have discussed this principle in relation to both the typicality and adequacy prongs. See E. Tex. Motor Freight Sys., Inc. v. Rodriquez, 431 U.S. 395, 403-04 (1977) (adequacy and typicality); Amchem Prods., v. Windsor, 521 U.S. at 594- 95 (adequacy); In re Pet Food Prods. Liab. Litig., 629 F.3d at 343 (adequacy); Martin v. Ford Motor Co., 292 F.R.D. 252, 269-70 (E.D. Pa. 2013) (adequacy and typicality). Here, it is uncontested that Mr. Sugg was hired, (see ECF No. 143-4 at 4; ECF No. 157 at 18), so Plaintiff fails the adequacy and typicality requirements. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 41 of 44 PageID: <pageID> 42 ii Promotion Class Plaintiff estimates that the Promotion Class is approximately 269 members. (ECF No. 142 at 24 (citing ECF No. 142-19 at 16).) Defendant argues that this number was derived from a faulty analysis in Dr. Lang’s Expert Reply Report—an analysis which resulted in dozens of individuals allegedly in the class who appeared to be South Asian and several individuals who were promoted or held a position from which they could not be promoted. (ECF No. 153 at 26-27.) Plaintiff responds that Dr. White’s opinion—a declaration filed with Defendant’s opposition brief—cannot be used because it was not disclosed as part of his expert report and is untimely. (ECF No. 157 at 6-7 (citing Fed. R. Civ. P. 37(c)(1)).) In any event, Plaintiff argues other evidence in the record allegedly identifies the requisite number of class members. (Id. (citing ECF No. 153-3 ¶ 22; ECF No. 157-1 ¶ 12).) Plaintiff has failed to demonstrate numerosity by a preponderance of evidence. Plaintiff’s chief hurdle is that Dr. Lang’s Expert Reply Report has been stricken by the Court. (See ECF No. 166.) And the only source Plaintiff relies on in his brief is that Report. (ECF No. 142 at 24 (citing ECF No. 142-19 at 16).) On reply, Plaintiff cites a declaration by the Vice President of Human Resources for Virtusa, (ECF No. 153-3 ¶ 22), along with Dr. Lang’s declaration filed with the reply brief, (ECF No. 157-1 ¶ 12.) But the employee’s declaration is an analysis of the Expert Reply Report and Dr. Lang’s declaration is a response to that analysis. An analysis of evidence struck from the record cannot sustain Plaintiff’s burden. Plaintiff is left with Dr. Lang’s original Expert Report, which reveals approximately 17 non-South Asian individuals who were not promoted. (See ECF No. 144-4 at 8 (Table 6 depicting that 23.3% of 73 white employees not promoted).) The Court has not been presented with any reason why joinder of these individuals would be impracticable, so the Court finds numerosity is unsatisfied. See Zangara v. Zager Fuchs, Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 42 of 44 PageID: <pageID> 43 P.C., Civ. No. 17-6755, 2019 WL 6310056, *2-3 (D.N.J. Nov. 25, 2019) (stating that as a “general rule[] . . . a class of twenty or fewer members is usually insufficiently numerous,” but “a class of twenty-one to forty members may or ‘may not meet the numerosity requirement depending on the circumstances,’” and finding class of 26 insufficient) (quoting In re Modafinil Antitrust Litig., 837 F.3d 238, 250 (3d Cir. 2016)). The Court must therefore deny certification with respect to the Promotion Class. See Fuentes, 2020 WL 7237942, at *6. iii Termination Class Plaintiff estimates that the Termination Class is approximately 249 members based on Dr. Lang’s Expert Reply Report. (ECF No. 142 at 24 (citing ECF No. 142-19 at 19).) Defendant argues Plaintiff cannot rely on the Reply Report, and the original Report only records significantly less than 40 individuals in the Class. (ECF No. 153 at 23-24.) Plaintiff fails to establish numerosity of the Termination Class for the same reason they fail in the Promotion context. Without the Reply Report, which the Court has stricken, the only evidence is Dr. Lang’s original Expert Report which puts the number of potential class members at approximately 18. (See ECF No. 144-4 at 12 (Table 9 depicting that of 103 white employees terminated, 17.7% were terminated involuntarily).) The Court has been provided with no explanation for why joinder of these individuals would be impracticable, and the Court therefore finds this number insufficient. See Zangara, 2019 WL 6310056, at *2-3. The Court accordingly denies certification with respect to the Termination Class. See Fuentes, 2020 WL 7237942, at *6. Therefore, class certification must be denied for each of the three proposed classes.20 20 Plaintiff requests that Plaintiff’s counsel, Kotchen & Low LLP and Jonathan Rudnick, be appointed class counsel. (ECF No. 141 at 4.) Plaintiff’s request is moot because the Court denies Plaintiff’s Motion for Class Certification. Case 3:18-cv-08036-GC-JTQ Document 167 Filed 01/13/26 Page 43 of 44 PageID: <pageID> Case 3:18-cv-08036-GC-JTQ Document167 _ Filed 01/13/26 Page 44 of 44 PagelD: <pagelD> IV. CONCLUSION For the foregoing reasons, and other good cause shown, Defendant’s Motion to Exclude the Expert Reports of David Lang, Ph.D. (ACF No. 144) is GRANTED in part and DENIED in part; Defendant’s Motion to Exclude the Expert Report of Ronil Hira, Ph.D. (ACF No. 145) is DENIED; and Plaintiff's Motion for Class Certification (ECF No 141) is DENIED. An appropriate Order follows. Dated: January 13 2026 (| onaee Catia GEORGETEE CASTNER PED STATES DISTRICT JUDGE 44
Vous vous demandez comment ce texte s’applique à votre situation ?
Consultation gratuite, sans engagement
Décrivez votre situation. Notre équipe vous recontactera par e-mail pour votre consultation.
Poursuivez vos recherches dans ChatGPT ou Claude
Connectez Omnilex pour rechercher dans le corpus juridique depuis votre assistant IA.