Replication as a Rule for Determining the Number of Clusters in Hierarchial Cluster Analysis

Abstract

A single higher-order cluster analysis can be used to group cluster mean profiles derived from several preliminary analyses. Replication is confirmed when each higher-order cluster contains one cluster mean profile from each of the several preliminary analyses. This study evaluated the utility of replication as a stopping rule in hierarchical cluster analysis. Replication defined by higher-order clustering identifies the correct number of under lying populations that have distinct density regions in the multivariate measurement space. When increased within-population variance obliterates population distinctions, the replication criterion provides an underestimation of the actual number of latent populations. In the case of no true cluster structure or in the case of only two latent populations, chance replication can occur. Thus, replication suggested by higher-order cluster analysis is not a conservative test for the absence of a cluster structure, but it does provide valid evidence concerning the number of latent pop ulations when several are present.

Keywords

Index terms: cluster analysis cluster means hierarchical cluster ing replication in cluster analysis stopping rule in cluster analysis validity of cluster analysis.

Get full access to this article

View all access options for this article.

References

Aldenderfer, M.S. , & Blashfield, R.K. (1984). Cluster analysis. Beverly Hills CA: Sage Publications.

Blashfield, R.K. (1976). Mixture model tests of cluster analysis: Accuracy of four agglomerative hierarchical methods. Psychological Bulletin, 83, 277-288.

Breckenridge, J.N. (1989). Replicating cluster analysis: Method, consistency, and validity. Multivariate Behavioral Research, 24, 147-162.

Davis, D.L. , & Bouldin, D.W. (1979). A cluster separation measure. IEEE Transactions on Pattern Analysis and Machine Intelligence PAMI, 1, 224-227.

Dixon, W.J. (1988). BMDP statistical software manual. Berkeley CA: University of California Press.

Hubert, L.J. , & Schultz, J. (1976). Quadratic assignment as a general data-analysis strategy. British Journal of Mathematical and Statistical Psychology, 29, 190-241.

Imsl (1982). IMSL library reference manual (9th ed.). Houston TX : IMSL Inc.

Jain, A.K. , & Dubes, R.C. (1988). Algorithms for clustering data. Englewood Cliffs NJ: Prentice Hall.

Lorr, M. (Ed.). (1966). Explorations in typing psychotics. New York: Pergamon.

10.

Lorr, M. , Klett, C.J. , & McNair, D.M. (1963). Syndromes of psychosis. New York: Pergamon Press.

11.

McIntyre, R.M. , & Blashfield, R.K. (1980). A nearest-centroid technique for evaluating the minimum-variance clustering procedure. Multivariate Behavioral Research, 2, 225-238.

12.

Milligan, G.W. (1980). An examination of the effects of six types of error perturbation on fifteen clustering algorithms. Psychometrika , 45, 325-342.

13.

Milligan, G.W. (1981). A Monte-Carlo study of 30 internal criterion measures for cluster analysis. Psychometrika, 46, 197-195.

14.

Milligan, G.W. , & Cooper, M.C. (1985). An examination of procedures for determining the number of clusters in a data set. Psychometrika, 50, 159-179.

15.

Mojena, R. (1975). Hierarchical grouping methods and stopping rules: An evaluation. Computer Journal, 20, 359-363.

16.

Mojena, R. , & Wishart, D. (1980). Stopping rules for Ward's clustering method. Proceedings of COMP-STAT 1980. Wurzburg, Germany: Physika-Verlag.

17.

Norušis, M.J. (1986). SPSS/PC+ advanced statistics. Chicago: SPSS Inc.

18.

Overall, J.E. (1974). The Brief Psychiatric Rating Scale in psychopharmacology research. In P. Pichot and R. Oliver-Martin (Eds.), Psychological measurements in psychopharmacology: Modern problems in pharmacopsychiatry (pp. 67-78). Basel : Karger.

19.

Overall, J.E. , Hollister, L.E. , Johnson, M. , & Pennington, V. (1966). Nosology of depression and differential response to drugs. Journal of the American Medical Association, 195, 946-948.

20.

Overall, J.E. , & Rhoades, H.M. (1982). Refinement of phenomenological classification in clinical psychopharmacology research. Psychopharmacology , 77, 24-30.

21.

SAS Institute Inc. (1987). SAS/STAT guide for personal computers: Version 6 edition. Cary NC : Author.

22.

Ward, J.H. (1963). Hierarchical grouping to optimize an objective function. Journal of the American Statistical Association, 58, 236-244.

23.

Winer, B.J. (1971). Statistical principles in experimental design (2nd ed.). New York: McGraw-Hill.

24.

Wishart, D. (1982). CLUSTAN User Manual (3rd ed., Supplement) [Computer program manual]. Edinburgh: Program Library Unit, Edinburgh University.