VISUALIZING INTERACTION AND SEQUENTIAL DATA IN ANIMAL BEHAVIOR - THEORY AND APPLICATION OF CLUSTER-ANALYSIS METHODS

The results of behavioral studies can be difficult to assimilate, often because such enormous quantities of data are involved. Some methods of cluster-analysis are presented which help to produce visual representations of such data. A single-link dendrogram (which is a way of representing a single-link cluster-analysis or SLCA), a maximum spanning-tree (MST), the properties of non-metric multidimensional scaling (MDS) and B(2) cluster-analysis are described. Because methods of cluster-analysis often involve fewer assumptions, they are used rather than factor analysis, principal-component analysis or canonical variate analysis. The results of factor analyses are compared with those of cluster-analyses. The methods described are illustrated with reference to the times 11 free-living chimpanzees were seen together, analysed by SLCA, MST and MDS. B(2) analyses are used for the chimpanzee data and their application to data about sequences of phrases in a blackbird''s song are shown. Constructing a particular kind of similarity commits one to a particular way of interpreting the data. Moreover, a method of constructing similarities can conceal differences of precision. The problems posed for cluster-analyses by asymmetric tables such as occur when elements precede other elements more often than vice versa, or when an individual in a social group grooms another for longer than vice versa are examined. How methods of cluster-analysis help the interpretation of association data for 11 chimpanzees is disucssed and the consistency of the chimpanzees'' patterns of association through 2 consecutive 4 mo. periods is examined. The association data of the whole community to which the 11 belong is examined and 3 consecutive analyses show how the males in the community began to split into 2 sub-groups. Association data for 3 captive groups of rhesus monkeys are compared and one group was compared in 12 consecutive monthly analyses to discover periods of change in that group''s social organization. Statistical methods for comparing dendrograms are discussed.