S-PLUS code for ordinal correspondence analysis

Correspondence analysis is a popular graphical tool used to analyse contingency tables. In the past, it has commonly been performed by applying a singular value decomposition to a transformation of the data in the contingency table. A recent advance in its theory is to perform a bivariate moment decomposition instead. This approach is especially useful for the detection of linear and non-linear associations between ordinal variables; a feature not readily available using singular value decomposition. This paper outlines S-PLUS code that will perform correspondence analysis using bivariate moment decomposition. It also includes a simple plotting function that will enable the graphical interpretation of the different levels of association.