BUSINESS RESEARCH APPLICATIONS

Business research focuses on assessing the preferences and choices of consumers and potential consumers. The Business Research Application (BRA) of SAS/STAT software is an intuitive point-and-click interface that provides statistical and graphical techniques for Business research data analysis. The Business Research Application provides easy access to five important types of data analysis commonly used in market research. These are:

Conjoint Analysis Correspondence Analysis Discrete Choice Modeling Multidimensional Scaling Multidimensional Preference Analysis

Conjoint Analysis
Conjoint analysis is used to evaluate consumer preference. If products are composed of attributes, conjoint analysis determines which combination of attribute levels are most preferred by consumers. Consumers indicate their preferences by ranking a number of different combinations of attribute levels. Conjoint analysis assumes that consumers make purchases by simmultaneously considering several attributes of a product. The ability to analyze several attributes at once distinguishes conjoint analysis from traditional market research methods where each attribute is studied separately. Conjoint analysis helps you discover how consumers make trade-offs between the various possible attribute combinations available. Usually, conjoint analysis consists of a main-effects analysis of variance with ordinally scaled dependent variables. Consumer preferences are the dependent variables, and product attributes are the independent variables. The following are some of the questions that can be answered with a conjoint analysis.

How important is each product attribute to consumers? Which existing products do consumers prefer? What combination of product attributes do consumers prefer most? How well will my product do in the current market?

Correspondence Analysis

Market research frequently involves categorical data. Correspondence analysis characterizes the associations between the levels of two or more categorical variables by performing a weighted principal component analysis of a contingency table. The result of a correspondence analysis is a two-dimensional graphical representation of the association between rows and columns of the table. The plot contains a point for each row and each column of the table. Rows with similar patterns of counts produce points that are close together, and columns with similar patterns of counts produce points that are close together. Simple correspondence analysis analyzes a contingency table made up of one or more column variables and one or more row variables. In multiple correspondence analysis, only column variables are used and a Burt table is produced. The following are some of the questions that can be answered with a correspondence analysis.

Who are my customers? Who else should be my customers? What new products should I create? Who should I target with my new products?

Discrete Choice Modeling

Discrete choice analysis is used to evaluate consumer choice. In conjoint analysis, consumer preferences for a product are evaluated by having a number of consumers rank different combinations of attribute levels. In discrete choice analysis, each consumer chooses one combination of attribute levels from a particular set of attribute level combinations. A set of attribute level combinations is called a choice set. Each consumer may be asked to choose one combination from each of several different choice sets. The attributes of the product that define the choices are called choice attributes to distinguish them from other attributes that may be of interest but do not contribute to the definition of the choices. For example, you may want to include demographic variables related to the consumers in the analysis. The following are some of the questions that can be answered with a discrete choice analysis.

Which combination of attribute levels is most likely to be chosen by consumers? Which combination of attribute levels is least likely to be chosen by consumers? What is the probability of my product being chosen by a consumer?

Multidimensional Scaling
Multidimensional scaling (MDS) analysis takes consumer judgments of similarity (or difference) of pairs of products and produces a map of the perceived relationship among the products. Each consumer evaluates the similarity (or difference) of each pair of products. MDS determines the relative similarity perceived by consumers among all the products. The results enable you to identify products that consumers see as similar. The following are some of the questions that can be answered with a multidimensional scaling analysis.

Which products do consumers see as similar to my product? Which products do consumers see as different from my product?

Multidimensional Preference Analysis

In a conjoint analysis, consumers indicate their preferences for products that are composed of attributes. Sometimes in market research, the available data consist of consumer preferences for products for which attributes are not defined. Multidimensional preference analysis (MDPREF) is used to analyze such data. MDPREF analysis is a principal component analysis of a data matrix with columns that correspond to consumers and with rows that correspond to products. The analysis results in a plot that reveals patterns of consumer preference for the products. The following are some of the questions that can be answered with a multidimensional preference analysis.

Who are my customers? Who else should be my customers? What new products should I create?