Abstract
Conjoint analysis is a statistical procedure often used by marketing researchers to measure the relative importance of various characteristics of a product or service as perceived by consumers. During the past ten years, conjoint analysis has been used to estimate consumers' preferences for many different types of products and services including educational services. In a conjoint analysis study, a researcher must determine whether the product factor estimates, which are used to measure consumer preferences, should be calculated and interpreted for each respondent or the respondents collectively. The purpose of this article is to demonstrate how a researcher can use multiple regression models to determine whether it is appropriate to analyze and interpret the aggregate data by examining the factor-respondents interaction effects. A hypothetical example is used to clarify how this technique can be used.
Recommended Citation
Fraas, John W. and Newman, Isadore
(1996)
"The Use of Multiple Regression Models to Determine if Conjoint Analysis Should Be Conducted on Aggregate Data,"
Mid-Western Educational Researcher: Vol. 9:
Iss.
4, Article 7.
Available at:
https://scholarworks.bgsu.edu/mwer/vol9/iss4/7
