PRINCIPAL COMPONENT ANALYSIS OF TAXONOMIC DATA WITH MULTI‐STATE DISCRETE CHARACTERS

Abstract
Summary: It is shown how principal component analysis can be extended to allow for multi‐state discrete characters as well as continuous characters. When all the characters are discrete, the proposed extension reduces to correspondence analysis. A taxonomic example is given to illustrate the method in practice. The technique allows the estimation of a taxonomic distance between objects which have been scored for multi‐state characters. However, taxonomic structure may be better inferred directly from the ordination, rather than by constructing a taxonomic distance measure.