COMPARATIVE-EVALUATION OF 2 SUPERIOR STOPPING RULES FOR HIERARCHICAL CLUSTER-ANALYSIS

A split-sample replication stopping rule for hierarchical cluster anal ysis is compared with the internal criterion previously found superior by Milligan and Cooper (1985) in their comparison of 30 different pro cedures. The number and extent of overlap of the latent population dis tributions was systematically varied in the present evaluation of stop ping-rule validity. Equal and unequal population base rates were also considered. Both stopping rules correctly identified the actual number of populations when there was essentially no overlap and clusters occ upied visually distinct regions of the measurement space. The replicat ion criterion, which is evaluated by clustering of cluster means from preliminary analyses that are accomplished on random partitions of an original data set, was superior as the degree of overlap in population distributions increased. Neither method performed adequately when ove rlap obliterated visually discernible density nodes.