These techniques are applied when the rows and the columns of the data table represent the same units and when the measure is a disiance or a similarity. The goal of the analysis is to represent graphically these distances or similarities. Multidimensional Scaling (MDS) is used to represent the units as points on a map such thbt their Euclidean distances on the map approximate the original similarities- (classic MDS, which is equivalent to PCA, is used for distances, nnnmetric MDS for similarities)'. Additive tree analysis and cluster analysis free used to reprcsent the units as "leaves" of a tree with the distance on the tree" approximating the original distance or similarity.