Table 1.
Model | Unconditional Model | Model with Covariates | ||||
---|---|---|---|---|---|---|
LL | BIC | Entropy | LL | BIC | Entropy | |
2 Class | −4590.08 | 9352.65 | .80 | −4454.23 | 9105.81 | .80 |
3 Class | −4469.58 | 9201.09 | .78 | −4325.71 | 8963.38 | .79 |
4 Classa | −4400.73 | 9152.83 | .77 | −4230.63 | 8887.82 | .78 |
5 Class | −4361.80 | 9164.40 | .80 | −4176.03 | 8893.21 | .80 |
The four-class solution was selected because the BIC was lower than in both the 3- and 5-class solutions. This difference was consistent when including four covariates in the model.
Abbreviations: BIC= Bayesian Information Criterion; LL = log-likelihood