Background
Four former Infosys employees and job applicants—Brenda Koehler, Gregory Handloser, Kelly Parker, and Layla Bolten—sued Infosys for discrimination against non-South Asian individuals in hiring and employment decisions. The plaintiffs alleged both disparate treatment and disparate impact discrimination under Title VII and 42 U.S.C. § 1981. Their primary evidence came from an expert report by labor economist David Neumark, who analyzed Infosys’s workforce demographics using a “name-matching” methodology to categorize employees and applicants as South Asian or non-South Asian based on last names.
Neumark’s analysis found that 89.39% of Infosys’s U.S. workforce was South Asian, compared to only 11.45% in the relevant domestic industry. The plaintiffs sought to use this statistical analysis to establish a pattern and practice of discrimination. The four individual plaintiffs also alleged specific employment actions: Handloser’s termination during a reduction in force, Koehler’s rejection from a job opening, Parker’s position ending after a staffing arrangement concluded, and Bolten’s constructive discharge following alleged harassment.
Infosys moved to exclude Neumark’s expert testimony under Federal Rule of Evidence 702. The district court granted the motion in September 2022—nearly six years after it was filed—finding that Neumark lacked qualifications and that his name-matching methodology was unreliable. Without Neumark’s analysis, the district court denied class certification, denied supplemental briefing with demographic data the plaintiffs later obtained, and granted summary judgment on all claims.
The Court’s Holding
The Seventh Circuit affirmed the district court’s decision to exclude Neumark’s expert testimony. The court found that Neumark lacked the specialized knowledge necessary to identify whether names were South Asian or non-South Asian. Although Neumark might have had experience with “western” names based on his upbringing, he admitted at his deposition to having no expertise or training in identifying South Asian names. The court emphasized that even if Neumark could identify western names, the task of determining whether a name could also be South Asian—noting examples like “D’Souza,” “Dave,” and “Joseph” that exist in both South Asian and other populations—required expertise he did not possess.
The court also affirmed the exclusion on reliability grounds. Neumark presented no evidence that his name-matching methodology had been tested, peer-reviewed, validated, or subjected to error-rate analysis. The plaintiffs failed to provide any scientific foundation supporting the methodology’s reliability. Because the plaintiffs had notice of Infosys’s challenge to the methodology and failed to offer evidence supporting its reliability at the district court level, the appellate court deferred to the district court’s exercise of discretion in excluding the opinions.
With Neumark’s expert opinions excluded, the court affirmed the denial of class certification because the plaintiffs’ motion relied heavily on his statistical analysis. The court also affirmed denial of the plaintiffs’ request to supplement their summary judgment briefing with demographic data they obtained after initial briefing, finding that the plaintiffs had been offered opportunities in 2017 to supplement their filings but rejected those opportunities. Finally, the court affirmed summary judgment on all discrimination claims: the pattern-and-practice claims failed because they required class certification; the individual disparate treatment claims failed because the plaintiffs could not rebut Infosys’s nondiscriminatory explanations for its employment actions; and the disparate impact claims failed without statistical support.
Key Takeaways
- Expert testimony in discrimination cases faces strict gatekeeping scrutiny under Rule 702 and Daubert—experts must possess specialized knowledge specific to the task they’re undertaking, and their methodologies must have a reliable scientific foundation with evidence of testing, error rates, and peer review.
- A court may exclude expert testimony sua sponte where the parties have been given notice and an opportunity to respond, and the district court’s gatekeeping role is reviewed only for abuse of discretion.
- Name-matching methodologies that attempt to categorize individuals by ethnicity or national origin based on surname analysis face significant reliability challenges when the expert lacks expertise in the particular ethnic/national origin names being classified.
- Plaintiffs cannot supplement summary judgment briefing with evidence that was available earlier but not used, particularly when they previously rejected judicial offers to supplement their filings and waited years after receiving the data before attempting to use it.
- Pattern-and-practice discrimination claims under International Brotherhood of Teamsters require class certification; without a certified class, individual plaintiffs cannot pursue pattern-and-practice theories.
Why It Matters
This decision reinforces strict gatekeeping principles for expert testimony in employment discrimination cases, particularly for statistical and demographic analyses. Organizations defending against discrimination litigation cannot rely on methodology alone—the foundation for that methodology must be thoroughly established at trial. The ruling signals that name-matching or surname-based analyses are problematic as a means of categorizing individuals by race or national origin, especially when experts lack specific knowledge of the demographics being studied. For plaintiffs’ counsel, the decision underscores the importance of fully developing the record at the district court level, including preserving arguments about expert qualifications and reliability, rather than attempting to supplement briefing years later.
The case also illustrates broader procedural risks for discrimination plaintiffs who rely exclusively on a single expert or analytical approach. The extended litigation timeline—filed in 2013, decided in 2025—prompted the court to note the district court’s obligation to ensure timely resolution of cases, but the delay did not cure the underlying evidentiary deficiencies. The ruling makes clear that a plaintiff’s failure to present reliability evidence supporting an expert’s methodology at the district court level cannot be remedied on appeal, no matter how promising the statistical conclusions might appear.