Skip to main content

Table 1 Fit statistics and classification coefficients

From: Patterns of adverse childhood experiences and associations with prenatal substance use and poor infant outcomes in a multi-country cohort of mothers: a latent class analysis

K

par

LL

BIC

SABIC

CAIC

AWE

BFk,k+1

LMR-LRT p-value

BLRT p-value

1

19

-10,401.019

20,936.687

20,876.336

20,879.51

20,889.01

 < 3

 

-

2

39

-9296.006

18,868.394

18,744.515

18,751.04

18,770.54

 < 3

 < .001

 < .001

3

59

-9049.660

18,517.437

18,330.031

18,339.90

18,369.40

 < 0

 < .05

 < .001

4

79

-8919.640

18,399.132

18,148.198

18,161.42

18,200.92

 > 4

 < .001

 < .001

5

99

-8863.024

18,427.635

18,113.173

18,129.74

18,179.24

 > 7

0.60

 < .001

6

119

-8812.199

18,467.719

18,089.729

18,109.64

18,169.14

-

0.74

 < .001

  1. Note: K Number of classes, par parameters, LL Log-likelihood, BIC Bayesian Information Criterion, SABIC Sample-size adjusted BIC, CAIC Consistent Akaike Information Criterion, AWE Average Weight of Evidence Criterion, BFk,k+1 Bayes Factor, LMR-LRT Lo-Mendell-Rubin Likelihood Ratio Test, BLRT Bootstrap Likelihood Ratio Test
  2. Bold indicates best fit for each respective test statistic