Classification analysis with criterion variable using SPAD

Abstract

This article presents the characteristics, procedure and utility of the technique of classification analysis with criterion variable in a large set of data, using mainly categorical variables. Classification analysis forms part of the techniques commonly known as data mining that analyse relationships or associations between variables. The paper describes step by step how to apply this statistical technique with the support of SPAD software, a statistical package for multivariate analysis, and provides an example of its application. The technique is drawn from the French school of statistics. Despite being little known, it is a very useful classification analysis for working with large amounts of data, a situation that is increasingly common in educational research, and more typical of the secondary analyses that are carried out in our field.