Skip to main content
. Author manuscript; available in PMC: 2013 Jan 1.
Published in final edited form as: Alcohol Clin Exp Res. 2011 Oct 17;36(1):153–161. doi: 10.1111/j.1530-0277.2011.01595.x

Table 2.

Model fit statistics showing a confirmatory factor analysis with covariates (MIMIC) model, latent class analysis (LCA) 2-class and 3-class models with and without binge drinking and a factor mixture model in 361 college drinkers.

log
likelihood
df AIC BIC Adjusted
BIC
MIMIC * 10 criteria, no binge drinking
(1 factor model)
−1132.4 22 2308.9 2393.4 2323.6
MIMIC * 10 criteria plus binge drinking
(2 factor model)
−1328.5 26 2709.0 2809.0 2726.4
LCA (10 criteria no binge drinking)
 2 classes (290, 71)
 3 classes (217, 114, 30)
−1227.2
−1186.4a
21
32
2496.3
2436.7
2578.0
2561.2
2511.4
2459.7
LCA (10 criteria plus binge drinking)
 2 classes (268, 93)
 3 classes (221, 112, 28)
−1444.8
−1393.5a
22
35
2935.7
2857.0
3025.1
2993.1
2952.2
2882.1
Factor Mixture Model (10 criteria, no
binge drinking)
 2 classes (376, 12), 1 factor
−1209.5 32 2482.9 2609.7 2508.1
*

Adjusted for age at initiation of alcohol drinking and sex.

a

p<0.001 comparing 3-class solution to 2-class solution