Table 2.
Model fit indices for different model specifications, n = 22,374.
| −2 Log likelihood | AIC | BIC | adj BIC | entropy | N (%) of smallest trajectory | |
|---|---|---|---|---|---|---|
| 1 Class | 175,945 | 206,934 | 206,966 | 206,953 | NA | NA |
| 2 Classes | −87945 | 175,907 | 175,971 | 175,945 | 0.84 | 5841 (26) |
| 3 Classes | −82602 | 165,228 | 165,324 | 165,286 | 0.85 | 1989 (9) |
| 4 Classes | −81247 | 162,527 | 162,655 | 162,604 | 0.81 | 1581 (7) |
| 5 Classes | −80652 | 161,345 | 161,505 | 161,442 | 0.79 | 1038 (5) |
| 6 Classes | −80129 | 160,306 | 160,498 | 160,422 | 0.79 | 638 (<3) |
| 7 Classes | −79718 | 159,492 | 159,716 | 159,627 | 0.79 | 274 (1) |
Abbreviations: AIC = Akaike's Information Criterion, BIC= Bayesian Information Criterion, adj BIC = sample size adjusted Bayesian Information Criterion.