Survey Methodology Employed by Marketing Expert Witness Deemed Admissible

Survey Methodology Employed by Marketing Expert Witness Deemed Admissible

Plaintiff, Anthony Bush on behalf of a class of California consumers, brought a class action against the Defendant, Rust-Oleum Corporation for mislabeling of its “Krud Kutter” cleaning products as “non-toxic” and “Earth friendly,” contending that it violated California consumer-protection laws since the products were, in fact, harmful to humans, animals, and the environment.

The operative complaint consisted of five claims: (1) unlawful, unfair, and fraudulent business practices under the Unfair Competition Law (UCL), Cal. Bus. & Prof. Code §§ 17200-08; (2) deceptive advertising under the False Advertising Law (FAL), Cal. Bus. & Prof. Code § 17500; (3) deceptive practices under the Consumer Legal Remedies Act (CLRA), Cal. Civ. Code §§ 1750-84; (4) breach of express warranties; and (5) unjust enrichment.

The Plaintiff filed a motion to exclude the opinions and survey of the Defendant’s market-research expert Dr. Ran Kivetz while the Defendants moved to exclude the declaration of the Plaintiff’s survey expert, Dr. J. Michael Dennis.

Marketing Expert Witness

Ran Kivetz is a renowned marketing scholar and survey expert holding a Ph.D. in Business from Stanford University Graduate School of Business. He is a tenured, chaired Professor of Marketing at Columbia University Business School, and he has received numerous research awards and nominations from leading marketing and consumer research publications and organizations, including, but not limited to, the Journal of Marketing Research, Journal of Consumer Research and the Association of Consumer Research.

Survey Research Expert Witness

J. Michael Dennis is a nationally recognized expert on survey research methods, with a focus on online surveys and household panels. He has directed hundreds of statistical studies using probability-based and non-probability panels, as well as using telephone and in-person modes of data collection. Dennis is executive director of AmeriSpeak, NORC’s probability-panel owned and operated by NORC. 

Discussion by the Court

With regard to the Plaintiff’s motion to exclude the opinions of Ran Kivetz, the Court observed that Kivetz conducted a experimental design survey which involved a test group and a control group. Kivetz showed the test group the actual label of one of the Krud Kutter products while the control group was shown the same label but without the challenged label claims (“non-toxic” and “Earth friendly”). Kivetz asked two key questions to the respondents. The first question was whether or not they would buy the products shown but for the challenged claims while the second question involved the reasons for the respondent’s purchasing decision which the respondents had to list in open-ended format.

Based on the answers, Kivetz concluded that that the challenged claims were not a but-for cause of purchasing decisions considering the difference between the test and control groups in whether they would purchase the product. Kivetz added that the open-ended responses determined that there were a variety of reasons for consumers’ purchasing decisions.

Plaintiff argued that some of Kivetz’s opinions were irrelevant because under the reasonable-consumer test, the challenged claims could be “material” to purchasing decisions even if the claims were not a but-for cause of the decisions and a variety of factors went into the decisions.

Plaintiff contended that instead of using a proper control stimulus that omitted any references to the “Non-Toxic” and “Earth Friendly” attributes, Kivetz’s control stimulus included several representations that communicated to the control group that the product shown was not only safe or “non-toxic,” but also “earth friendly.” Specifically, he failed to remove the “biodegradability” claim and the Environmental Protection Agency’s “Safer Choice” seal on the front packaging; as well as the word “safely” from the back-packaging claim that the formula “safely and easily” removes various substances. Since the test and control stimuli both showed non-toxic and earth friendly products, Plaintiff concluded that Kivetz’s purported control stimulus was no control at all.

Plaintiff criticized Kivetz’s survey for distorting collected data due to inadequate control over pre-existing consumer attitudes, beliefs, and preferences. The inclusion of branding elements like the Krud Kutter name and packaging were based on the incorrect assumption that respondents could update their preferences despite the removal of Challenged Claims. The survey lacked manipulation checks to determine whether the experimental treatment — removal of the ‘non-toxic’ and ‘Earth friendly’ claims in the control stimulus — was effective.

Plaintiff argued that Kivetz’s open-ended questions were unreliable because such questions tend to measure only what comes first to a respondent’s mind while close-ended questions were better suited for qualitative research. Moreover, Kivetz allegedly failed to properly represent the class or replicate the marketplace.

Kivetz designed a coding frame and employed two blind coders to read and categorize the responses, reconcile any discrepancies between them, to quantify the open-ended responses and determine, in his view, whether a significant number of respondents identified the “non-toxic” and “earth friendly” features as a reason for their purchase decision. Kivetz failed to present the Plaintiff with the data upon which Kivetz relied, including Kivetz’s coding frame, the blind-coders coding, and the reconciliation of discrepancies. Consequently, Plaintiff was deliberately prevented from evaluating how verbatim responses were categorized.

The Court held that the Plaintiff’s attacks concerned the weight that should be accorded to Kivetz’s survey and opinions and determined that the survey methodology employed by Kivetz was within the bounds of accepted principles.

As for the data that Kivetz did not provide, it was seen that while responses were being categorized to the open-ended questions, the coding company excluded certain responses because the respondent spent too little time on it. Some of the data at issue, such as “all starts and metadata reflecting the excluded interviews and basis for their exclusion” were never actually provided to Kivetz.

The Court, citing Republic of Ecuador v. Mackay, 742 F.3d 860, 869-70 (9th Cir. 2014), held that the data an expert “considered” under Rule 26(a) refers to data the expert “was provided or otherwise exposed to in the course of developing his or her opinions.

As for the “pricing data that Kivetz collected to select the $9.47 price point in his survey,” the Defendant contended that it was public. This indicated that any failure to produce was harmless. The Court thus denied the Plaintiff’s motion to exclude the opinions of Kivetz, but this portion of the motion (regarding the alleged failure to produce data) was denied without prejudice to its being refiled as a separately noticed motion if warranted.

The Defendant filed a motion to exclude the declaration of the Plaintiff’s survey expert, J. Michael Dennis, alleging that Dennis repeated methodological errors found in previously excluded surveys from other cases. Dennis aimed to assess the extent to which reasonable consumers perceived the challenged claims regarding product harmlessness. His survey presented a hypothetical label without the Krud Kutter brand name, asking respondents if they believed the claims conveyed the stated meaning.

The Court, once again, noted that the Defendant’s arguments concerned the weight of Dennis’ testimony instead of its admissibility.


The Court denied both the Plaintiff’s as well as the Defendant’s respective motions to exclude the opinions of Ran Kivetz and J. Michael Dennis.

The Court has not arrived on an outcome for this case since the remaining issues involved in this case still await resolution.

Key Takeaways:

The case involved expert testimony from both parties, with the Plaintiff seeking to exclude the opinions of Ran Kivetz and the Defendant attempting to exclude the declaration of the Plaintiff’s survey expert, J. Michael Dennis. Kivetz’s experimental design survey with test and control groups, aimed to assess the impact of certain label claims on consumer purchasing decisions. However, the Plaintiff criticized Kivetz’s methodology, arguing that the control stimulus used was flawed as it still contained elements suggesting product safety and environmental friendliness. Additionally, the Plaintiff raised concerns about the incompleteness of expert disclosures. Despite these objections, the Court found that the methodology employed by Kivetz was generally acceptable within the bounds of established principles. As for Dennis’ testimony, the Defendant challenged its admissibility based on methodological errors found in previous surveys. However, the Court reiterated the distinction between challenges to weight versus admissibility of expert testimony, ultimately allowing both Kivetz’s and Dennis’ opinions to be admitted.

Case Details:

Case Caption:Bush V. Rust-Oleum Corporation
Docket Number:3:20cv3268
Court:United States District Court, California Northern
Citation:2024 U.S. Dist. LEXIS 23728
Order Date:February 8, 2024


Leave a Reply

Your email address will not be published. Required fields are marked *