A Partial Least Squares Algorithm Handling Ordinal Variables also in Presence of a Small Number of Categories

Abstract

The partial least squares (PLS) is a popular modeling technique commonly used in social sciences. The traditional PLS algorithm deals with variables measured on interval scales while data are often collected on ordinal scales: a reformulation of the algorithm, named ordinal PLS (OPLS), is introduced, which properly deals with ordinal variables. An application to customer satisfaction data and some simulations are also presented. The technique seems to perform better than the traditional PLS when the number of categories of the items in the questionnaire is small (4 or 5) which is typical in the most common practical situations.