In May 2025, Plaintiff Richard Paul Merrell filed the operative putative class action complaint against Defendant Ralph Lauren Corporation. Plaintiff is visually impaired and legally blind, and he uses screen-reading software to read website content using his computer.
Screen-reading software (or a “screen reader”) allows blind and visually impaired users to access websites using a combination of a keyboard and “software that vocalizes the visual information found on a computer screen.”
Plaintiff alleged that Defendant’s website, including but not limited to https://www.ralphlauren.com/, “is not fully or equally accessible to blind and visually impaired consumers.”
According to Plaintiff, “Defendant’s failure to properly code its website” to be accessible means that he and other class members “have been and are still being denied equal and full access to Defendant’s retail stores and the numerous goods, services, and benefits offered to the public through Defendant’s website in conjunction with Defendant’s brick-and-mortar retail store locations.”
Defendant filed motions to exclude the testimony of Plaintiff’s experts, Dr. Jon A. Krosnick and Dr. William C. Easttom, II. Plaintiff filed a motion to exclude the testimony of Defendant’s expert, Aaron Cannon.
Psychology Expert Witness
Jon Alexander Krosnick is a Stanford professor and a research psychologist at the U.S. Census Bureau with a degree in psychology from Harvard University and an M.A. and Ph.D. in social psychology from the University of Michigan.
He has taught classes and trainings on research methods, received awards for research and psychology, authored or co-authored hundreds of articles and presentations, and otherwise conducted extensive research on survey research methods.
Cybersecurity Expert Witness
Dr. William Charles Easttom, II has Ph.Ds in technology and computer science, a Doctor of Science in cyber security, and multiple related master’s degrees. He has authored 44 computer science books and dozens of research papers and is an inventor of 27 computer science patents. He is as an adjunct lecturer at Georgetown University and Vanderbilt University.
Easttom has published three books specifically on JavaScript (a coding language for web development), holds a Certified Professional in Accessibility Core Competency certification, and has taken two accessibility courses.
Web Accessibility Expert Witness
Aaron Cannon has “been working in the field of web and mobile app accessibility for over seventeen years and [has] been developing software for twenty-nine years.” He is a “Certified Professional in Web Accessibility” and holds multiple accessibility-related certificates.
Cannon is totally blind and has been using screen readers since 1988, including JAWS for Windows, NVDA, TalkBack, VoiceOver, and the Orca screen reader. He also co-founded and served as chief accessibility officer for a company that provided accessibility auditing of websites and mobile apps.
Discussion by the Court
A. Dr. Jon A. Krosnick
Based on his review of the complaint, literature and documentation about methodologies employed in past surveys of blind people, past surveys of visually impaired people, studies of accessibility challenges faced by blind people when using computers, and other information, Krosnick conducted a survey and concluded that “the number of legally blind people living in California who used a screen reader and were prevented from doing business with a physical store due to problems related to accessibility with the Ralph Lauren webpage is 3,243.”
Defendant argued that Krosnick’s testimony should be excluded in full because (1) his opinions are unreliable and not based on sufficient facts or data; (2) his survey results are inadmissible hearsay; and (3) his survey is irrelevant and will not help the trier of fact.
Analysis
First, Defendant argued that Krosnick’s survey did not use a random probability sample and instead “recruited non-representative survey respondents from Facebook groups for visually impaired people.” But Krosnick made clear that he combined a primarily randomized probability sample with some respondents from Facebook groups oriented toward visually impaired people, and that this kind of “blended” probability sample “is a widely accepted practice in contemporary survey science.”
Defendant also contended that Krosnick’s survey had a low response rate, leading questions, self-interest bias, recall bias, and a lack of verification of unreliable answers.
The Court also is not persuaded that the survey involved leading questions or self-interest bias just because it asked participants if they wanted to “help improve the experiences of blind people using the Internet,” and then listed potential problems the respondents may have experienced.
Even if the survey’s wording did implicate some level of bias, this would not be so substantial as to render the survey unreliable.
B. Dr. William C. Easttom, II
Despite his extensive background, Defendant argued that “Easttom is not qualified to opine on website accessibility,” since general education in computer science is not interchangeable with competence in website accessibility. Even if Easttom has a relatively limited amount of experience with website accessibility, he plainly is an expert on website design and computer science, and the Court is not persuaded that his extensive background is insufficient to qualify him as an expert.
Defendant also argued that Easttom’s testimony should be excluded in full because (1) his report relied on the wrong website; (2) his methodology is unreliable; and (3) he impermissibly offers legal conclusions.
Analysis
First, Defendant argued that Easttom failed to test a screen reader version of its website, “despite being prompted to enable accessibility features.”
The fact that Easttom didn’t test an alternative, enhanced version of the website may undercut the weight of his methodology or Plaintiff’s claims against Defendant, but it did not render Easttom’s opinions completely irrelevant or unreliable.
Methodology
Second, Defendant raised various attacks on Easttom’s methodology, including that Easttom (1) did not use two tools he used in another case and (2) makes a “huge inferential leap” that Defendant denied access to its goods and services from the fact that the store locator wasn’t rendered as a clickable website attribute.
First, any questions about the tools used go to weight, rather than admissibility. Second, the Court disagreed that Easttom’s conclusions are that large of an inferential leap, or that they are irrelevant to answering the question of whether Defendant’s website violated the ADA. And to the extent another expert said that visually impaired customers could locate Defendant’s brick-and-mortar stores through its website, that is a battle of the experts to be resolved by the trier of fact.
Legal Conclusions
Third, Defendant argued that Easttom improperly offered legal conclusions. In particular, Defendant took issue with Easttom’s statements suggesting that Ralph Lauren’s website is not ADA compliant.
The Court agreed with Defendant: Easttom’s opinion and testimony that Defendant’s website did not meet ADA compliance requirements, is an improper legal conclusion, and the Court granted the motion as to these opinions.
However, one of the purported “legal” conclusions is actually a conclusion about whether Defendant’s website complies with the Web Content Accessibility Guidelines (“WCAG”), which are “a set of international standards designed to make web content more accessible.” This is not a legal conclusion just because Easttom made reference to tools that are used for scanning ADA compliance.
Fourth, Defendant argued that Easttom “conflates Ralph Lauren’s alleged non-compliance with WCAG 2.1 AA guidelines, and alleged non-compliance with the ADA,” and “there is no authority at all for the proposition that websites must comply with the WCAG guidelines.”
Reliability
Finally, Defendant argued that Easttom improperly used the Internet Wayback Machine to determine if historical versions of Defendant’s website were inaccessible because the Wayback Machine does not necessarily include full JavaScript functionality and may “not allow a user to evaluate [a website’s] full functionality.” These critiques—and critiques about the sample of Wayback Machine pages that were chosen—go to weight and do not fundamentally undermine the reliability of the opinion. This is especially true where, as here, Defendant has offered no reason to believe that these individual webpages had missing JavaScript functionality.
C. Aaron Cannon
Plaintiff first argued that portions of Cannon’s declaration should be excluded as untimely. Plaintiff sought exclusion under Federal Rules of Civil Procedure 26 and 37.
Rule 26
Defendant argued that Cannon disclosed that he had new opinions in his deposition, yet Plaintiff “did not question him about those additional opinions.”
The Court found that Defendant has not shown harmlessness. Defendant’s argument appears to be that Plaintiff’s counsel should have pushed harder at the deposition, ignoring the fact that Cannon evasively and vaguely mentioned he had some unspecified other opinions, but would not explain what they were. This is especially true where Defendant had already asserted that it would not disclose its rebuttal opinions to Easttom’s report, which it did not perceive to be related to class certification. As a result, the Court rejected Defendant’s suggestion that Plaintiff had the same “free rein” to get additional discovery on these untimely opinions as in Defendant’s cited cases.
Defendant’s other arguments regarding a lack of prejudice or surprise are unpersuasive. Defendant summarily stated that Plaintiff only cites a “single distinguishable case” supporting his claims of prejudice, and “Plaintiff has also not identified any additional expenses caused by Defendant’s alleged failure to disclose, because there is none.” This flips the burdens: it is Defendant’s burden to demonstrate harmlessness or substantial justification, and it has not done so. Accordingly, the Court will strike the identified portions of Cannon’s declaration, though those portions would have been irrelevant to the remainder of the Court’s order here.
Rule 702
Plaintiff next argued that Cannon’s testimony should be excluded in full because (1) his opinions are not based on sufficient facts or data; (2) he did not apply a reliable, testable method to the case; (3) his class certification opinions are unhelpful; (4) his opinions offer impermissible legal conclusions; and (5) he is not qualified to offer “socioeconomic and population-level opinions or reliability engineering opinions.”
In particular, Plaintiff first argued that Cannon’s opinion should be excluded because he did not “gather any site-specific facts about Ralph Lauren’s website” and instead “offered generalized assumptions based on experience, which Rule 702(b) does not permit.”
The Court found that Cannon’s opinions about what individualized technical problems may impede a user with a screen reader from accessing a website are adequately based upon his extensive personal knowledge and experience. To the extent Plaintiff believes Cannon’s testing was flawed or didn’t consider enough data or site-specific information, those critiques go to weight, not admissibility.
The same is true for Plaintiff’s arguments that Cannon did not deploy a reliable methodology.
Cannon is applying his extensive experience with accessibility-related software and screen readers to make high-level observations about, for example, the differences in hardware, operating systems, and screen reader software, in order to explain why this Court would have to engage in individualized inquiries about each class members’ experience.
The Court also disagreed that Cannon’s opinions “substitute legal judgment for technical analysis.” The fact that Cannon is identifying individualized issues in a user’s experience with accessibility features—a central part of the predominance inquiry—did not mean that he is impermissibly making legal conclusions.
Held
- The Court denied the motion to exclude the testimony of expert Dr. Jon A. Krosnick.
- The Court granted in part and denied in part the motion to exclude the testimony of William C. Easttom.
- The Court granted in part and denied in part the motion to exclude the testimony of Aaron Cannon.
Key Takeaway
When evaluating specialized or technical expert opinion testimony, the relevant reliability concerns may focus upon personal knowledge or experience.
Case Details:
| Case Caption: | Merrell V. Ralph Lauren Corporation |
| Docket Number: | 4:23cv6669 |
| Court Name: | United States District Court, California Northern |
| Order Date: | July 09, 2026 |
Leave a Reply