Conjoint Analysis


Posted February 11, 2010

Conjoint analysis is a statistical technique that helps in forming subsets of all the possible combinations of the features present in the target product. These features used determine the purchasing decision of the product. Conjoint analysis works on the belief that the relative values of the attributes when studied together are calculated in a better manner than in segregation.

For a free consultation on conjoint analysis, click here.

The process provides information about the consumer’s perception about certain characteristics of brands or brand profiles, and they evaluate such characteristics by assigning certain levels to each characteristic. A questionnaire form called the stimuli is presented to the researcher, and this consists of a set of questions that reflects different characteristics of a brand as options that the consumers select as they answer the questionnaires in conjoint analysis.

The stimuli play an important role in conjoint analysis. It is the stimuli that give information to the researchers about the consumer’s preference. With the help of the stimuli, the researchers can perform this method. The researcher should, however, check that the responses are true because the interpretation of the conjoint analysis is dependent on that.

The process of conjoint analysis has found its applications in various disciplines. Such disciplines include branding consumer goods and branding industrial goods, etc. The procedure provides the researcher a flexible opportunity to address certain issues instead of conducting the testing of hypothesis. The researcher should also note that the theory is quite simple and flexible for the researcher to understand, even if he is a non statistical person. The model that is used by the researcher during the procedure is the utility function model. This model is based on the evaluation of conjoint analysis, and is basically a mathematical model. This mathematical model is used by the researcher to express the fundamental relationships between the attributes and the utilities of the attributes that the consumer attaches to it. The dependent variable usually consists of the consumer’s preference or intention to buy a particular brand of product.

There are several procedures for assessing the reliability and validity of conjoint analysis.  A reliability test, called test retest reliability in conjoint analysis, can be used to obtain duplicated judgments that are sometimes involved in data collection. If an aggregate level of conjoint analysis has been done, then the estimation sample can be split into several samples and conjoint analysis is again conducted on each sub-sample. This can assure the researcher that the conjoint analysis being conducted is reliable and valid.

It is important for the researcher to know that conjoint analysis and multidimensional scaling (MDS) are complementary. Both rely on the respondent’s subjective evaluations. The difference between them is that of the stimuli. In conjoint analysis, the stimuli are the combinations of attribute levels, whereas in MDS, the stimuli are the products or brands of the products.

The steps involved while conducting conjoint analysis are the following:

The first and one of the most obvious steps is the formulation of the problem.

The next step is to prepare the stimuli.

The third step is to decide upon the form of data to be input.

The fourth step involves the selection of the procedure.

The next step is to interpret the results obtained.

And the last step is to assess the reliability and validity.


Pin It on Pinterest

Shares
Share This