Table 2.
Comparison of latent profile analysis fit statistics.
| Model | df | LL | AIC | BIC |
|---|---|---|---|---|
| 1 Class | 8 | −241898.2 | 483812.3 | 483878.7 |
| 2 Classes | 13 | −240272.1 | 480570.1 | 480678.0 |
| 3 Classes | 18 | −237034.1 | 474104.2 | 474253.5 |
| 4 Classes | 23 | −236320.7 | 472687.5 | 472878.3 |
| 5 Classes | 28 | −236141.3 | 472338.5 | 472570.8 |
| 6 Classes | 33 | −236141.3 | 472348.5 | 472622.3 |
df, degree of freedom; LL, log likelihood; AIC, Akaike Information Criterion; BIC, Bayesian Information Criterion.The bold values indicated the best-fitted model because the least AIC or BIC values indicate the optimal number of classes.