Multiple Correspondence Analysis

Identify Associations between Levels of Categorical Variables

Multiple Correspondence Analysis (MCA) takes multiple categorical variables and seeks to identify associations between levels of those variables. MCA extends correspondence analysis from two variables to many. It can be thought of as analogous to principal component analysis for quantitative variables. Similar to other multivariate methods, it is a dimension reducing method; it represents the data as points in 2- or 3-dimensional space.

Multiple correspondence analysis is frequently used in the social sciences particularly in France and Japan. It can be used in survey analysis to identify question agreement. It is also used in consumer research to identify potential markets for products.

Figure 7.1 Multiple Correspondence Analysis 

Contents

Example of Multiple Correspondence Analysis

Launch the Multiple Correspondence Analysis Platform

The Multiple Correspondence Analysis Report

Multiple Correspondence Analysis Platform Options

Correspondence Analysis Options

Show Plot

Show Detail

Show Adjusted Inertia

Show Coordinates

Show Summary Statistics

Show Partial Contributions to Inertia

Show Squared Cosines

Cochran’s Q Test

Cross Table

Additional Examples of the Multiple Correspondence Analysis Platform

Example Using a Supplementary Variable

Example Using a Supplementary ID

Example of Cochran’s Q Test

Statistical Details for the Multiple Correspondence Analysis Platform

Details Report

Adjusted Inertia

Summary Statistics

Partial Contributions to Inertia

Cochran’s Q Statistic