· Q I wish to find a suitable procedure for classifying patients into schizophrenic types by identifying the natural patterns already present in the patient data I have collected. I have recorded data for a considerable number of parameters, including ‘severity of depression’ and ‘severity of anxiety’. Can you recommend an appropriate approach to analysing my data?
A. The most suitable approach here is cluster analysis.
In order to familiarize yourself with the terminology and underlying concepts, have a look at the resource
Details of some of the more common types of cluster analysis which can be performed using SPSS can be found by consulting Chapter 17 of the IBM SPSS Statistics 17 Statistical Procedures Companion. The same chapter provides comprehensive illustrations for the methods described, including pointers to the corresponding SPSS datasets and instructions on how to apply the appropriate menu commands in SPSS.
Should you wish to progress to a more comprehensive study of cluster analysis techniques, you may find it helpful to consult the book chapter Cluster analysis: basic concepts and algorithms.
Cluster Analysis: Arranging Data into Categories to Find a Natural Classification System by Margaret MacDougall is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
