Table 3. Fit statistics across LCA models with an ML solution.
| LCA Model | Log Likelihood | Likelihood Ratio Test |
AIC | a-BIC | Entropy | DF | X2 |
|---|---|---|---|---|---|---|---|
| 1-class | -2132.06 | 1308.31 | 4284.11 | 4291.36 | 1.00 | 354.00 | 105778.53 |
| 2-class | -1809.32 | 662.83 | 3660.63 | 3675.85 | 0.83 | 343.00 | 1405.15 |
| 3-class | -1756.60 | 557.41 | 3577.21 | 3600.40 | 0.77 | 332.00 | 1266.21 |
| 2-class DE | -1749.36 | 542.91 | 3558.71 | 3580.45 | 0.74 | 334.00 | 1287.97 |
| 3-class DE | -1735.44 | 515.07 | 3540.87 | 3566.23 | 0.74 | 329.00 | 1180.03 |
Abbreviations: AIC = Akaike’s information criteria; BIC = Bayesian information criteria; a-BIC = sample-size adjusted Bayesian information criteria; DE = direct effects; DF = degrees of freedom; LCA = latent class analysis; ML = maximum likelihood.