Skip to main content

Table 2 Key parameters for determining the optimal number of latent classes

From: Associations of sedentary time, sleep duration and physical exercise with multimorbidity among older adults in Shanghai, China: a cross-sectional study based on national physical fitness surveillance data

Number of latent class

Number of parameters

G2

df

AIC

BIC

X2

Entropy

Maximum log-likelihood

2

19

1642.5

492

93271.7

93414.4

2137.9

3.5

-46616.9

3

29

876.7

482

92525.9

92743.7

1286.5

3.4

-46234.0

4

39

588.1

472

92257.3

92550.1

657.5

3.4

-46089.7

5

49

534.5

462

92223.7

92591.6

614.6

3.4

-46062.8

6

59

470.6

452

92179.8

92622.8

524.5

3.4

-46030.9

7

69

445.9

442

92175.1

92693.1

501.8

3.4

-46018.5

8

79

425.8

432

92175.0

92768.2

486.1

3.4

-46008.5

9

89

439.4

422

92208.7

92876.9

473.8

3.4

-46015.3

10

99

401.0

412

92190.2

92933.4

424.7

3.4

-45996.1