Skip to main content
. Author manuscript; available in PMC: 2020 May 11.
Published in final edited form as: Infancy. 2020 Feb 25;25(3):264–285. doi: 10.1111/infa.12328

TABLE 5.

Cluster fit statistics

Absolute validation indices
Number of clusters
Fit statistic 3 4 5
Dunn 0.95 0.93 0.84
Avg.silwidth 0.21 0.17 0.13
CH index 114.79 93.22 76.48
Entropy 1.10 1.35 1.59
Comparative validation indices
Cluster comparison
Fit statistic 3 vs. 4 3 vs. 5 4 vs. 5
VI 0.37 0.99 0.68
Corrected rand 0.81 0.49 0.64

Note: For the absolute validation indices, higher scores on Dunn, Avg.silwidth, and CH index indicate better fit, whereas lower scores for entropy indicate a better fitting model. For the comparative validation indices, lower scores on the VI and higher scores on the Corrected Rand indicate better fit.