Asymptotic and Finite Sample Characteristics of Four External Criterion Measures
- 1 January 1985
- journal article
- research article
- Published by Taylor & Francis in Multivariate Behavioral Research
- Vol. 20 (1) , 97-109
- https://doi.org/10.1207/s15327906mbr2001_6
Abstract
Finite and asymptotic formulas are derived for the Rand, adjusted Rand, Fowlkes and Mallows, and Jaccard external criterion measures. The formulas were developed for ten clustering situations which have been used previously in the clustering literature. The results indicated that the Jaccard and Fowlkes and Mallows measures were fairly similar. The Rand and adjusted Rand indices formed another similar pair. Weaknesses involving the Rand and the Fowlkes and Mallows indices are noted. Based on these and previous simulation results, it is recommended that the adjusted Rand and Jaccard measures be used in validation studies of clustering algorithms.This publication has 8 references indexed in Scilit:
- The Measurement of Classification Agreement: An Adjustment to the Rand Statistic for Chance AgreementEducational and Psychological Measurement, 1984
- A Method for Comparing Two Hierarchical ClusteringsJournal of the American Statistical Association, 1983
- Characteristics of Four External Criterion MeasuresPublished by Springer Nature ,1983
- The Effect of Cluster Size, Dimensionality, and the Number of Clusters on Recovery of True Cluster StructurePublished by Institute of Electrical and Electronics Engineers (IEEE) ,1983
- A Review Of Monte Carlo Tests Of Cluster AnalysisMultivariate Behavioral Research, 1981
- Mixture model tests of cluster analysis: Accuracy of four agglomerative hierarchical methods.Psychological Bulletin, 1976
- Objective Criteria for the Evaluation of Clustering MethodsJournal of the American Statistical Association, 1971
- A Coefficient of Agreement for Nominal ScalesEducational and Psychological Measurement, 1960