ABSTRACT
Principal component analysis (PCA) is the most widely used factorial method. Analysing the representations of active variables led to interpretations in the form of combinations of initial variables. In practice, factorial analysis is almost always performed with supplementary elements, and in particular, supplementary variables. Analysing the diversity of the individuals’ profiles or the correlations between variables means examining clouds with the same inertia, a property induced by the fact that the points of one are the same as the dimensions of the space within which the other evolves. Intuitively, a quantitative variable and a qualitative variable are linked if the individuals of the same class have similar values for the quantitative variable. The representation of sensory variables shows a strong relationship between acidity and sensory description. All of the p-values for the PCA factors are highly significant: the high school variable is undoubtedly linked to all of the dimensions from the PCA.
