Skip to main content
. 2013 Mar 27;8(3):e59955. doi: 10.1371/journal.pone.0059955

Table 1. Fit Indices for Latent Class Analysis.

1 2 3 4
BIC 13900.997 13419.045 13312.442 13256.482
Entropy 0.950 0.838 0.837
Class >5% No No No No
LMR-LRT Sig Sig NS

Note: BIC = Bayesian information criterion (smaller values indicate better model fit). Entropy at least.80 (higher values indicates well identified classes). Class >5% (any class smaller than 5% not sufficient). LMR-LRT = Lo-Mendell-Rubin Adjusted Likelihood Ratio Test, test of fit between the model of interest (e.g., three-class model) and the model with one less class (e.g., two-class model). Sig = significant. NS = non-significant.