ABSTRACT
The term cluster analysis is generally used to describe a set of numerical techniques for classifying objects into groups based on their values on a set of variables. The intent is to group objects such that objects within the same group have similar values on the set of variables and objects in different groups have dissimilar values. The objects classified into groups are most typically persons and the variables used to classify objects can either be categorical or continuous. Cluster analysis can be used as a data reduction technique to reduce a large number of observations into a smaller number of groups. It can also be used to generate a classification system for objects or to explore the validity of an existing classification scheme. Unlike other multivariate techniques, such as logistic regression or MANOVA (see Chapters 16 and 25, respectively), group membership is not known but instead imposed on the data as a result of applying the technique. Because objects are classified into groups even if no groups truly exist, additional planned analyses beyond cluster analysis are essential. Researchers should use these analyses to provide support for the replicability and validity of a particular cluster solution.
