Editor’s note: Faina Shmulyian is the vice president of insights at Big Village. This is an edited version of an article that originally appeared under the title “Alternative-Specific Conjoint and its Applications.”
Choice-based conjoint (CBC) analysis or discrete choice modeling (DCM) is a widely used and trusted tool for product development and price optimization in modern survey-based research. In a discrete choice experiment, each respondent is presented with a series of tasks or screens. Each task is a set of products or concepts that a respondent must choose from. CBC tasks are generated using principles of experimental design to make sure the maximum amount of information is elicited from respondents to obtain the most accurate modeling.
One of the obvious advantages of CBC is that tasks in a discrete choice experiment are mimicking a natural behavior of consumers in a marketplace. A respondent is considering different products in a competitive context, evaluating the products’ features and prices and making reasonable trade-offs. The data collected in a discrete choice exercise can be utilized in a Hierarchical Bayesian estimation to model preferences individually for each respondent and accurately describe the market heterogeneity. Using CBC, researchers can simulate consumer behavior in hypothetical scenarios, take into account interactions between different product attributes, simulate various scenarios on the market, optimize product features and understand price sensitivity.
To generate a choice task for a CBC, researchers define a set of attributes describing products or concepts they want to evaluate. For example, if the researcher is studying laptops, the attributes could include brand, type, screen size, processor, memory, battery life, price, etc. Each attribute is presented on different levels, for example, for a laptop brand, it could be Apple, HP, Dell, Lenovo, etc. With a stand...