Table 3.
Model fit statistics for LPA models
| Number profiles | Convergence problems | Lowest LLH replicated | BIC | Entropy | Avg class probabilities | Smallest class size | LMRT | BLRT |
|---|---|---|---|---|---|---|---|---|
| 1 | No | Yes | 27344 | -- | -- | -- | -- | -- |
| 2 | No | Yes | 26097 | 0.88 | 0.93–0.98 | N=238 (21%) | p<.05 | p<.05 |
| 3 | No | Yes | 25731 | 0.81 | 0.86–0.94 | N=192 (17%) | p<.05 | p<.05 |
| 4 | No | Yes | 25466 | 0.76 | 0.82–0.93 | N=190 (17%) | p<.05 | p<.05 |
| 5 | No | Yes | 25192 | 0.79 | 0.84–0.92 | N=82 (7%) | p=.15 | p<.05 |
| 6 | No | Yes | 25027 | 0.80 | 0.81–0.93 | N=69 (6%) | p=.15 | p<.05 |
| 7 | Yes | No | 24988 | 0.77 | 0.74–0.92 | N=59 (5%) | p=.36 | + |
Note. Convergence problems were noted when the majority of solutions failed to converge. BIC, sample size–adjusted Bayesian Information Criterion; BLRT, bootstrapped likelihood ratio test; LLH, log likelihood; LMRT, Lo-Mendell-Rubin test; LPA, latent profile analysis.
+
BLRT test evidenced convergence issues.