Multivariate Statistics. II. Using Different Types of Multivariate Statistics to Identify Types of Psychopathology Cluster Analysis and Multidimensional Scaling

Abstract

This is the second of four papers considering the application of multivariate statistics in psychiatric research. The identification of relatively homogeneous sub-groups of types of behavioural deviance is illustrated. First, cluster analysis is applied, and the effects of different distance metrics and clustering methods are discussed. The ability of multidimensional scaling to recover the relative spatial arrangement of towns from distances between pairs of towns only is presented. Finally, the complementary relationship among factor analysis, cluster analysis, and multidimensional scaling in revealing different aspects of the same set of data is highlighted.