In the latest edition of the Meet the Expert series, we’ll be speaking to Ms. Patricia Yanes, who has recently joined the Applied Marketing Science, Inc. (AMS) network of testifying survey experts. Ms. Yanes is a Principal at AMS, and serves as an expert witness for civil litigation, with a specialty in conjoint analysis for class actions.
Tell us about your expertise in market research methodology and how you first became involved in conjoint analysis.
I started at AMS back in 2010 supporting all aspects of market research across our Insights for Innovation and Litigation Support practices. I quickly became interested in quantitative methods that closely mirror true consumer behavior, and discrete-choice methods like conjoint analysis were a big part of that. Throughout my last decade or so at AMS, I have worked with companies across a wide variety of industries to apply conjoint analysis to innovation and product development. In 2013, I started working with AMS-affiliated testifying survey expert Steven Gaskin to design a study using conjoint analysis for class action litigation, and I have been working in support of such projects as a basis for determining damages in class actions ever since.
How is conjoint analysis used in class action surveys?
Typically, these are cases where customers allegedly did not receive the value they thought they were getting when they purchased the product or service. Conjoint analysis allows us to create a damages model to calculate the “reduction in market value” or “price premium” (in dollar or percentage terms) due to the discrepancy between what they thought they were buying and what they actually received. For example, if consumers purchase a nutrition snack bar from the grocery store that has a claim stating it contains “no added sugar,” they believe that if they eat the serving size indicated, they will not be consuming added sugar when, in reality, it does contain added sugar. A conjoint analysis survey can determine the price premium they paid for the claim or feature of “no added sugar,” which was not actually delivered by the snack bar. However, it is important to note that during the survey, we never directly ask people how much they would be willing to pay for this claim or feature. Instead, we ask them to make choices between alternative products, at different price points, similar to how they would in the marketplace. Then we use that information to calculate the value of the promised claim or feature, as revealed by analyzing their choices.
What is involved in designing a conjoint survey for class action litigation?
One of the most important parts of designing a conjoint model as a basis for determining damages in class actions is choosing the attributes and levels of the product or service at issue to be included in the survey. “Attributes” are the features that make up or describe a product or service, while “levels” are the specifications of those attributes. For example, the attributes that make up or describe a snack bar could include the brand, flavor, nutritional claims, and price of the product. Examples of levels in a snack bar conjoint could be the various flavors, such as chocolate chip, peanut butter, or blueberry. There is often a tradeoff between how many features are included and how easy it may be for customers to make simulated purchase decisions in the context of the survey. The final design depends on the objectives of the analysis, but for applications within class action litigation, it is generally not necessary to include every attribute. We need to measure the key feature at issue and the price of the product or service. We then include a number of other “distractor” attributes to ensure the survey is not leading. Once we have designed our model, we use Sawtooth Software, a leading conjoint analysis software, to program the survey and run the analysis to calculate the reduction in market value due to the possible absence of a promised feature, or the price premium due to the presence of a feature.
What makes conjoint analysis the right fit for a case?
Conjoint analysis is a well-established discrete-choice model used in quantitative market research that has been around for over 45 years and is the subject of an enormous amount of academic research. It is employed by numerous well-known businesses to determine the optimum price for products and services, and their features. For class action litigation, conjoint analysis can be the right fit when attorneys are looking to calculate the damages consumers allegedly suffered from a product’s or service’s misleading or omitted claim, or from a product defect not disclosed at the time of purchase. For example, a vehicle could be manufactured with a design flaw such as a faulty transmission or exhaust system. For a case like this, the reduction in market value can be calculated, that is, how much less consumers would have paid for the vehicle if they had known about the defect at the time of purchase. Conjoint analysis is well adapted to the benefit-of-the-bargain theory of a class action case, a principle in which any party who breaches a contract must pay the aggrieved party an amount that puts that person in the same financial position that would have resulted if the contract had been fulfilled.
Can you give an example of a case where conjoint analysis was successfully used? What was the case and how was conjoint analysis used?
Conjoint analysis has been used in numerous cases across various industries. Some examples of major cases that have survived a Daubert challenge (if there was one), received class certification, and received a favorable settlement, to the best of my knowledge, include:
- Sanchez-Knutson vs. Ford Motor Company: The SUVs at issue were manufactured with a design flaw that allowed exhaust gases to enter the vehicle cabin under certain conditions.
- Lenovo Adware Litigation: Lenovo was manufacturing laptops that had third-party software installed, without consumers’ knowledge or approval, that experts considered to be malware.
- Khoday & Townsend v. Symantec Corp and Digital River: Customers were misled to believe they had to pay extra for the capability to re-download security software from Symantec and Digital River that was, in fact, available to them for free.
- Kenai Batista et al. v. Nissan North America, Inc.: Some Nissan vehicles had a defect in their transmissions that could lead to a “shuttering” and “juttering.”
What excites you about conjoint analysis and discrete-choice methodologies?
Conjoint analysis allows us to gain powerful, quantifiable insight into consumer decision making. It draws on concepts from mathematics, economics, psychology, etc. to model a real-world buying experience in which people choose products and services the way they do in the marketplace. As a marketing professional with a passion for mathematics, I find it rewarding to apply these methodologies to help my clients solve real-world problems.
For those interested in learning more about conjoint analysis, please view our Conjoint Analysis Resource Guide. For inquiries regarding using conjoint surveys for class action matters, please contact Ms. Yanes via email or call (781) 250-6322.