Skip to main content

Table 4 Summary of results from clustering analysis methods applied

From: Cluster analysis and its application to healthcare claims data: a study of end-stage renal disease patients who initiated hemodialysis

Clustering Approach

Linkage Type

Number of Clustersa

Cluster Sample Size (Smallest in Bold)

Hierarchical

Average

3

18,376; 3; 1

 

Average

4

18,376; 2; 1; 1

 

Average

5

18,312; 64; 2; 1; 1

Hierarchical

Centroid

3

18,365; 14; 1

 

Centroid

4

18,351; 14; 14; 1

 

Centroid

5

18,351; 13; 14; 1; 1

Hierarchical

Single-Linkage

3

18,378; 1; 1

 

Single-Linkage

4

18,377; 1; 1; 1

 

Single-Linkage

5

18,376; 1; 1; 1; 1

Hierarchical

Complete-Linkage

3

18,367; 7; 6

 

Complete-Linkage

4

18,118; 249; 7; 6

 

Complete-Linkage

5

18,118; 249; 6; 6; 1

Hierarchical

Flexible-Beta

3

13,416; 3,732; 1232

 

Flexible-Beta

4

13,416; 3,732; 1059; 173

 

Flexible-Beta

5

8,919; 4,497; 3,732; 1,059; 173

Hierarchical

McQuitty’s Similarity

3

18,373; 6; 1

 

McQuitty’s Similarity

4

18,367; 6; 6; 1

 

McQuitty’s Similarity

4

18,205; 162; 6; 6; 1

Hierarchical

Ward’s Method

3

15,718; 2,315; 347

 

Ward’s Method

4

15,718; 2,315; 284; 63

 

Ward’s Method

5

15,718; 2,315; 239; 63; 45

Non-hierarchical

N/A

3

336; 17,909; 135

 

N/A

4

113; 16,624; 1,554; 89

 

N/A

5

116; 594; 16,162; 48; 1,460

  1. N/A not applicable. aNumber of clusters in the model