Skip to main content
. Author manuscript; available in PMC: 2014 Apr 1.
Published in final edited form as: J Psychiatr Res. 2013 Jan 4;47(4):520–526. doi: 10.1016/j.jpsychires.2012.12.005

Table 1.

Fit statistics of latent growth mixture models of disaster-related PTSD symptoms

Log
likelihood		 BIC SSA-BIC AIC Entropy LRT BLRT
1-class −2067.02 4176.67 4151.32 4150.04 - - -
2-class −2007.53 4073.68 4038.82 4037.07 .965 .0009 <.00001
3-class −1975.70 4025.98 3981.62 3979.39 .933 .20 <.0001
4-class −1955.90 4002.38 3948.51 3945.80 .931 .22 <.00001
5-class −1935.94 3978.44 3915.07 3911.88 .928 .18 <.00001
6-class −1918.53 3959.59 3886.72 3883.05 .932 .19 <.0001

Note. BIC=Bayesian Information Criterion; SSA-BIC=Sample-size adjusted Bayesian Information Criterion; AIC=Akaike Information Criterion; LRT=Lo-Mendell-Rubin Adjusted Likelihood Ratio Test; BLRT=Parametric Bootstrapped Likelihood Ratio Test. Smaller BIC, SSA-BIC, and AIC values, and higher entropy values indicate better fitting models; significant values for the LRT and the BLRT quantify the likelihood that the data can be better described by a model with one less trajectory.