ABSTRACT
The publication of Sokal and Sneath’s book (1963) gave rise to the testing and use of the three basic agglomerative procedures in hierarchical clustering: single linkage, complete linkage, and average linkage. The third one was coined UPGMA for unweighted pair group method of aggregation (Sokal and Michener, 1958), construction ascendante hiérarchique selon la distance moyenne in French. In many practical cases it appeared useful to associate such methods with correspondence analysis (CA).
