Choice  Conjoint
Home Up About us Organisation Research NEWS Quotation Request INTERNET SURVEY Feedback & Orders


Choice Applications

The main characteristic distinguishing Choice-Based Conjoint analysis from the other types of Conjoint Analysis is that the respondent expresses preferences, by choosing the best concept  out of successive choice set of concepts. That is, instead of making judgments on the likelihood of choosing an alternative (as in Full Profile Conjoint), or the degree of preference between a pair (as in Adaptive Conjoint).

A major advantage of Choice-Based over standard conjoint (Full Profile) is that its task represents more directly the market behavior. After all, consumers do not generally rate alternatives in terms of preferences; they simply choose. On the other hand, Choice-Based Conjoint is less effective than the Full Profile Conjoint for uncovering and defining segments, due to the fact that it requires the analyst to aggregate across respondents, in order to derive stable coefficients. 

Although appealing, Choice is by nature noisy, affected by numerous whims, that cannot be easily controlled and are notoriously difficult to analyze and understand. A discussion about Choice-Based Conjoint will have to focus on the use of experimentally controlled choice versus judgment, as competing tools try  to understand the impact of marketing actions on consumers. For instance, if one wanted to assess the impact of product improvement on sales; then, a judgment-based approach to attack the problem involves describing the improvement and then asking respondents to indicate their likelihood of purchasing it. A Choice-Based approach to the problem replaces the judgment measure with a choice task, involving the product and its close competitors.

The question which one may easily think is, why then, one uses truncated information in choice over the richer, more interval quality of consumer judgments? The answer is that choice mimics better the process in the marketplace: its greater validity more than compensates for its greater noisiness. The choice processes are not like judgments. In making a judgment, attention is focused on each alternative, rather than examining differences among them for each attribute. Choices proceed by screening out unacceptable levels to achieve an evaluation of each alternative. Further, a close examination of market choices indicates that they are truncated and largely heuristic, rather than judgmentally based. It  is generally accepted that we can simulate better what happens in the marketplace, through choices rather than through judgments.

Another integral property of choice is that its sensitivity to marketing efforts depends on the level of those efforts and on the initial predisposition of customer to the brand. In insensitive regions (with probabilities near zero or one), a choice task will be less sensitive than judgment. However, when individual base probabilities or marketing efforts are moderate, choice sensitivities can be far greater, and often more revealing, than judgment sensitivities.

Choice is to be preferred, because it mimics shopping behavior in three ways.

  1. Choice, like shopping, can be characterized as opportunistically scanning the set to find the best, while judgment reflects a more methodical assessment of each alternative
  2. Choice in experimental tasks is differentially sensitive to the consumer's predisposition and the level of the marketing effort, as we would expect it to be in the field
  3. Finally, where it is sensitive, it can be substantially more sensitive than attitude measures to the kinds of context and task conditions (such as clutter and shelf position) that impact on the market choice.

Choice-Based Conjoint analysis has attracted much interest recently and the reasons are:

The task of choosing a preferred concept 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
Choice-Based Conjoint analysis lets the researcher include a "None" option for respondents, who might prefer to choose none of these.
Choice-Based Conjoint data, being analyzed at an aggregate level, rather than for respondents individually, is feasible to quantify for interactions. Thus,  the principal strength of Choice-Based Conjoint analysis is its ability to deal with interactions. However, if interactions do exist, one must be able to measure them, to avoid being misled by them and by measuring interactions it also brings the benefit of discovering cases, in which the value of one benefit depends on the level of another.

(Some parts of the above content are based on the article of Joel Huber, The Importance of Multinominal Logit Analysis of individual Consumer Choices; CBC  by Sawtooth Inc,Sequim,WA)

Quotation Request Form



   Copyright: MARKETECK Co LTD : Design  by John KLOK