Skip to main content

Table 2 Measures of cluster partitions quality for different numbera of clusters

From: Sequence analysis of sickness absence and disability pension in the year before and the three years following a bicycle crash; a nationwide longitudinal cohort study of 6353 injured individuals

  PBCb HGc HGSDd ASWe ASWwf CHg R2h CHsqi R2sqj HCk
2 clusters 0.74 0.99 0.99 0.75 0.75 2063.58 0.25 4404.48 0.41 0.01
3 clusters 0.66 0.84 0.84 0.68 0.68 4232.97 0.57 5975.31 0.65 0.07
4 clusters 0.67 0.84 0.84 0.70 0.70 3589.46 0.63 6458.03 0.75 0.07
5 clusters 0.71 0.92 0.92 0.72 0.72 3419.08 0.68 6821.89 0.81 0.03
6 clusters 0.73 0.97 0.97 0.76 0.76 3502.94 0.73 6901.61 0.84 0.02
7 clusters 0.73 0.98 0.98 0.77 0.77 3247.65 0.75 7341.25 0.87 0.01
8 clusters 0.73 0.98 0.98 0.77 0.77 3039.62 0.77 7073.04 0.89 0.01
9 clusters 0.73 0.98 0.98 0.77 0.77 2786.74 0.78 6716.10 0.89 0.01
10 clusters 0.73 0.99 0.99 0.78 0.78 2587.42 0.79 6513.68 0.90 0.01
11 clusters 0.73 0.99 0.99 0.78 0.78 2410.38 0.79 6289.63 0.91 0.01
12 clusters 0.73 0.99 0.99 0.76 0.76 2324.17 0.80 6086.74 0.91 0.01
13 clusters 0.73 0.99 0.99 0.77 0.77 2270.38 0.81 5883.08 0.92 0.00
  1. aThe here selected number of clusters market in bold
  2. bPoint Biserial Correlation
  3. cHubert’s Gamma
  4. dHubert’s Somers’ D
  5. eAverage Silhouette Width
  6. fAverage Silhouette Width (weighted)
  7. gCalinski-Harabasz index
  8. hPseudo R2
  9. iCalinski-Harabasz index squared
  10. jPseudo R2 squared
  11. kHubert’s C