Skip to main content
. Author manuscript; available in PMC: 2013 Oct 1.
Published in final edited form as: J Adolesc. 2012 Jun 13;35(5):1351–1360. doi: 10.1016/j.adolescence.2012.05.003

Table 2.

Model Fit Statistics by Number of Classes (N = 7508)

Fit Statistics No. of classes
2 3 4 5
Degree of freedom 19 29 39 49
Log likelihood −27250.6 −25497.5 −24856.1 −24675.2
Akaike (AIC) 54539.2 51052.9 49790.2 49448.4
Bayesian (BIC) 54670.8 51253.7 50060.3 49787.6
Sample-Size
Adjusted BIC 54610.4 51161.5 49936.3 49631.9
Entropy .789 .862 .844 .810
VLMR LRT < .0001 < .0001 .0003 0.145 a
LMR adjusted LRT < .0001 < .0001 .0004 0.148 a

Note.

AIC = Akaike’s Information Criteria; BIC = Bayesian Information Criteria; VLMR LRT = Vuong-Lo-Mendell-Rubin Likelihood Ratio Test; LMR adjusted LRT: Lo-Mendell-Rubin Adjusted Likelihood Ratio Test.

a

The tests compares the 4-class model and the 5-class model. As the 5-class model didn’t significantly increase the model fit, the 4-class model was selected as the optimal model.