Sawtooth Software: The Survey Software of Choice

What is Choice-Based Conjoint?

Choice-Based Conjoint (CBC) is used for discrete choice modeling, now the most often used conjoint-related method in the world. The main characteristic distinguishing choice-based from other types of conjoint analysis is that the respondent expresses preferences by choosing from sets of concepts, rather than by rating or ranking them. The choice-based task is similar to what buyers actually do in the marketplace. Choosing a preferred product from a group of products is a simple and natural task that everyone can understand.

If you are having trouble deciding which conjoint method might be best for your specific situation, try our Interactive Advisor.

CBC is often used to study the relationship between price and demand.  It is especially useful when the price-demand relationship differs from brand to brand. One of the strengths of CBC is its ability to deal with interactions, such as when different colors work better with different styles. Most conjoint methods are based on "main effects only" models that ignore the existence of such interactions. In contrast, CBC may be used to evaluate all two-way interactions.

The CBC System provides everything needed to conduct a choice-based conjoint study via Web, CAPI (devices not connected to the Web), or paper-based surveys.

CBC data can be analyzed in a number of ways. First, the relative impact of each attribute level can be assessed just by counting "wins." In randomized CBC designs, each attribute level is equally likely to occur with each level of every other attribute. Therefore, the impact of each level can be assessed by counting the proportion of times concepts including that level are chosen. This "counting" method can be used for main effects as well as for two- or three-way interactions. For a second type of analysis, CBC includes an easy-to-use module to perform multinomial logit estimation. This analysis results in a set of conjoint "utilities," but which differ from standard conjoint in that they describe preferences of a group rather than for an individual. CBC's Logit module can estimate main-effects and two-way interactions.

Two advanced analysis modules are included with the base CBC System (Lighthouse Studio V8 and later): the Latent Class Segmentation Module and the Hierarchical Bayes Module. An Advanced Design Module is also available as an extension of the base CBC System for accommodating alternative-specific designs, shelf-display, and partial-profile CBC.

Read More