Skip to main content

Table 3 Model fit statistics for estimating classes of CAM users through latent class analysis

From: Latent class analysis suggests four distinct classes of complementary medicine users among women with breast cancer

Number of classes

Likelihood ratio G2

AIC

BIC

Sample size Adjusted BIC

% of seeds associated with best fit

1

2903.8

2943.8

3036.5

2973.0

100 %

2

2313.5

2395.5

2585.7

2455.5

100 %

3

2202.8

2326.8

2614.4

2417.5

40 %

4a

2117.9

2283.9

2668.9

2405.3

100 %

5

2070.4

2278.4

2760.8

2430.6

36 %

6

2028.9

2278.9

2858.7

2461.8

13 %

7

1992.5

2284.5

2961.7

2498.1

3 %

  1. Abbreviations: AIC Akaike Information Criterion, BIC Bayesian Information Criterion, CAM complementary and alternative medicine
  2. aSelected model contained 4 latent classes