Cluster Analysis

Abstract

Cluster analysis is a group of multivariate techniques whose major objective is to combine observations/object/cases into groups or clusters, such that each group or cluster formed is homogeneous or similar with respect to some certain characteristics and these groups should be different from other groups with respect to same characteristics. In cluster analysis, the researcher can classifies objects, such as respondents, products or other entities and cases or events, based on a set of selected variables or characteristics. Cluster analysis works based on certain set of variables, called “Cluster variate”, which form the basis for comparing the objects in the cluster analysis. In cluster analysis, the selection of cluster variate is very important, because in cluster analysis the focus is for comparing the objects in each cluster based on variate, rather than the estimation of the variate itself. This difference makes cluster analysis different from other multivariate techniques. Therefore, the researcher’s definition of the cluster variate plays a crucial role in cluster analysis.