Skip to main content

Table 1 Mean cluster weights from 2, 3 and 4 cluster solutions using LCA, PAM and Ward methods

From: Addressing preference heterogeneity in public health policy by combining Cluster Analysis and Multi-Criteria Decision Analysis: Proof of Method

      

MEAN CRITERION WEGHTS

 

Clustering Method

Cluster Solution

Cluster Number

N (of 523)

%

Quality

LOSS OF LIFETIME

NEEDLESS BIOPSY

URINARY PROBLEMS

BOWEL PROBLEMS

SEXUAL PROBLEMS

Interpretive Label

Latent Class Analysis (MCLUST)

           
 

4

1

327

62.5

0.24

0.22

0.15

0.20

0.20

0.23

Equals

  

2

53

10.1

0.64

0.88

0.02

0.04

0.03

0.03

Very High Lifers

  

3

121

23.1

0.31

0.53

0.06

0.14

0.15

0.12

Moderate Lifers

  

4

22

4.2

0.39

0.13

0.53

0.11

0.11

0.12

Moderate Biopsers

     

0.31

      
 

3

1

407

77.8

0.29

0.27

0.13

0.19

0.19

0.21

Equals

  

2

92

17.6

0.60

0.78

0.03

0.07

0.06

0.06

Very High Lifers

  

3

24

4.6

0.36

0.16

0.52

0.10

0.11

0.11

Moderate Biopsers

     

0.35

      
 

2

1

493

94.3

0.25

0.36

0.11

0.17

0.17

0.18

Moderate Lifers

  

2

30

5.7

0.36

0.22

0.49

0.10

0.10

0.10

Moderate Biopsers

     

0.26

      

Partitioning Around Medoids (pamk)

           
 

4

1

270

51.6

0.33

0.19

0.18

0.22

0.22

0.19

Equals

  

2

59

11.3

0.63

0.87

0.03

0.04

0.03

0.03

Very High Lifers

  

3

163

31.2

0.26

0.49

0.11

0.14

0.14

0.13

Moderate Lifers

  

4

31

5.9

0.36

0.06

0.06

0.11

0.13

0.64

Very High Sexers

     

0.35

      
 

3

1

301

57.6

0.27

0.18

0.17

0.21

0.21

0.24

Equals

  

2

59

11.3

0.63

0.87

0.03

0.04

0.03

0.03

Very High Lifers

  

3

163

31.2

0.30

0.49

0.11

0.14

0.14

0.13

Moderate Lifers

     

0.32

      
 

2

1

346

66.2

0.40

0.21

0.16

0.20

0.20

0.23

Equals

  

2

177

33.8

0.41

0.64

0.08

0.09

0.10

0.09

Very High Lifers

     

0.41

      

Ward’s Hierarchical (HCLUST)

           
 

4

1

170

32.5

0.34

0.14

0.21

0.23

0.23

0.18

Equals

  

2

38

7.3

0.27

0.08

0.07

0.12

0.14

0.59

Very High Sexers

  

3

60

11.5

0.68

0.86

0.03

0.04

0.04

0.03

Very High Lifers

  

4

255

48.8

0.17

0.42

0.12

0.15

0.16

0.15

Moderate Lifers

     

0.29

      
 

3

1

208

39.8

0.22

0.13

0.19

0.21

0.21

0.26

Equals

  

2

60

11.5

0.68

0.86

0.03

0.04

0.04

0.03

Very High Lifers

  

3

255

48.8

0.23

0.42

0.12

0.15

0.16

0.15

Moderate Lifers

     

0.28

      
 

2

1

463

88.5

0.40

0.29

0.15

0.18

0.18

0.20

Moderate Lifers

  

2

60

11.5

0.76

0.86

0.03

0.04

0.04

0.03

Very High Lifers

     

0.44

      
  1. Also shown are cluster sizes and statistical quality (as measured by average silhouette width). The bold numbers indicate the statistical quality of the cluster solution. N.B. ANOVA showed all clusters to be significant at p < 0.05, except LCA 4/4 (Moderate Biopsers).
Back to article page

Health Economics Review

ISSN: 2191-1991

Contact us