Multiple Correspondence Analysis

Multiple Correspondence Analysis (MCA) is an extension of correspondence analysis which can handle a large set of categorical variables.

The MCA implementation in BrandMap uses as input a "Burt Matrix," which can be described as a typical "banner" of multiple variables, each of which is split into some number of groups. For instance, we may have four variables, a b, c and d. Each of them could be split into 4 groups. If we cross tabulate a banner of these 4 variables by itself, we would get the following table of frequencies, and we would inform the program in the dialog box that we had 4 variables.

Copyright © 1994 - 2017 WRC Research Systems, Inc.

Still need help? Please  email  us with any questions  .