Maximum Difference Scaling Method in the MaxDiff R Package

Abstract

In microeconomics, measurement of consumer preferences is one of the most important elements of marketing research. Accurate measurement of preferences allows to gain an understanding of likes and dislikes of consumers. Using some statistical methods (like e.g. conjoint analysis and discrete choice models) it is possible to quantify preferences and answer the questions: What product will a consumer choose? What attribute of the product is most important? Consumer choice models attempt to answer these questions. This article describes the R package MaxDiff for the Maximum Difference Scaling method to assess consumer preferences from consumer choice experiments. Because practical applications of this method depend on the availability of computer software, this paper describes an implementation of the Maximum Difference Scaling method in the R package MaxDiff. Functions of the MaxDiff R package can be used for the measurement of consumer preferences. MaxDiff supports the design of the experiment (e.g. to build a list of features), encode the alternatives, estimate the models, etc. Some functions of the MaxDiff R package are presented with examples of applications in the empirical analysis of consumer preferences.