Skip to main content
. Author manuscript; available in PMC: 2017 Jan 1.
Published in final edited form as: Psychol Violence. 2015 Sep 7;6(1):145–155. doi: 10.1037/a0039713

Table 3.

Model Fit Indices for Exploratory Latent Class Analysis

Model Log-Likelihood # of Parameters BIC LRT BFK, K+1 cmPK
1-Class −16436 28 33080.94 <.01 <.10 0.0
2-Class −15800.8 57 32027.04 .16 <.10 0.0
3-Class −15586.6 86 31815.02 .27 <.10 0.0
4-Class −15435.7 115 31729.73 .17 <.10 0.0
5-Class −15300.1 144 31674.97 .11 >10 1.0
6-Class −15236.3 173 31763.81 .77 >10 0.0
7-Class −15181.4 202 31870.5 .77 >10 0.0
8-Class −15132.8 231 31989.63 .78 >10 0.0
9-Class Not well-identified

Note. BIC is the Bayesian Information Criteria, LRT is the p-value from the Lo–Mendell–Rubin adjusted likelihood ratio comparing the K class model to the K+1 class model, BFK, K+1 is the Bayes Factor comparing the K class model to the K+1 class model. cmPK denotes correct model probability for the K class model. Boldface type denotes the best fitting model for each fit index.