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

TL;DR

This paper introduces an ordinal PLS (OPLS) algorithm that effectively handles ordinal variables, especially with few categories, improving upon traditional PLS in social science data analysis.

Contribution

The paper presents a reformulated PLS algorithm tailored for ordinal data with few categories, addressing a gap in traditional PLS methods.

Findings

OPLS performs better than traditional PLS with small category counts

The method is validated through customer satisfaction data analysis

Simulations confirm the advantages of OPLS in practical scenarios

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.