Skip to main content
. 2017 Nov 6;12(11):e0187267. doi: 10.1371/journal.pone.0187267

Table 2. Fit indices used to identify number of latent classes.

Latent classes AIC BIC Entropy BLRT Lo-Mendell-Rubin adjusted LRT n (%) of smallest class
1 13652.754 13689.988
2 13560.821 13621.325 0.844 <0.001 0.005 37 (4.77)
3 13476.733 13560.508 0.812 <0.001 0.018 33 (4.25)
4 13455.152 13562.197 0.756 <0.001 0.181 29 (3.74)
5 13424.835 13555.151 0.766 <0.001 0.073 16 (2.06)
6* 13401.426 13555.013 0.719 <0.001 0.740 4 (0.52)

Note. AIC = Akaike information criterion; BIC = Bayesian information criterion; BLRT = parametric

bootstrapped likelihood ratio test; LRT = Lo-Mendell-Rubin adjusted likelihood ratio test; *due to local maxima no stable class structure was found.