Table 2.
Fit Statistics for Latent Class Solutions Two through Five
| Number of Profiles | Loglikelihood | AIC | BIC | Entropy | LMR | LMR p | BLRT | BLRT p | n of smallest class |
|---|---|---|---|---|---|---|---|---|---|
| Indicator Variances Freed Across Profiles | |||||||||
| 2 | −851.74 | 1753.48 | 1821.44 | 0.93 | 194.38 | <.001 | 197.55 | <.001 | 33 |
| 3 | −811.11 | 1698.22 | 1801.53 | 0.87 | 79.95 | 0.64 | 81.25 | <.001 | 25 |
| 4 | −780.89 | 1663.77 | 1802.41 | 0.90 | 59.48 | 0.02 | 60.45 | 0.11 | 13 |
| 5 | −756.12 | 1640.24 | 1814.23 | 0.92 | 48.73 | 0.29 | 49.53 | 1.00 | 10 |
Class indicators were standardized prior to running the analyses; indicator variances were constrained across classes; AIC = Akaike Information Criterion, BIC = Bayesian Information Criterion, LMR = Lo-Mendell-Rubin Test, LMR p = Lo-Mendell-Rubin Test p value, BLRT = Bootstrap Likelihood Ratio Test, BLRT p = Bootstrap Likelihood Ratio Test p value. The bolded class was selected as the best fitting class solution