ABSTRACT

A simple example can be used to illustrate the estimation of the parameters of the Indscal model via the multiple factor analysis. The idea of comparing two groups of variables from their partial cloud suggests representing a group by its matrix of between-individual distances. The eigenvalues of the separate analyses coincide with the variances of the initial variables. The overall evaluation is a supplementary group containing a single variable; its coordinates are the squared coefficient of correlation with the factors. These comments support the individuals’ representation. The two active groups are almost identical from the point of view of the first axis and their coordinate is high: the first axis, that is to say, the opposition between the Florida juices and.