Table 1.
# classes | MLL | AIC | BIC | aBIC | Entropy | LMR-LRT1 |
---|---|---|---|---|---|---|
1 | -173262.52 | 347521.10 | 350343.31 | 116525.09 | -- | -- |
2 | -54319.42 | 108846.80 | 109435.84 | 109105.41 | 0.894 | 7661.74 |
3 | -53391.99 | 107099.98 | 107994.80 | 107492.81 | 0.931 | 1854.86 |
4 | -52849.69 | 106123.39 | 107324.03 | 106650.48 | 0.937 | 1084.60 |
5 | -52437.69 | 105407.38 | 106913.84 | 106068.73 | 0.937 | 824.00 |
6 | -52122.50 | 104885.01 | 106697.30 | 105680.63 | 0.916 | 630.38 |
7 | -51822.60 | 104393.21 | 106511.33 | 105323.09 | 0.920 | 599.80 |
Notes. MLL = Maximum log-likelihood, AIC = Akaike information criterion, BIC = Bayesian information criterion, aBIC = adjusted Bayesian information criterion, LMR-LRT = Lo-Mendell-Rubin adjusted Likelihood Ratio Test. 1 LMR-LRT k(-1) Model comparisons were all significant at p < .001. The four-class solution was identified as optimal and highlighted in boldface