Table 1.
Fit statistics from the latent class analyses.
| Number of classes | Log-likelihood (L2) | BIC-L2 | AIC-L2 | Npar/DF | G2 | L2/df | p-value a | %ERb |
|---|---|---|---|---|---|---|---|---|
| 1-class | 2548.2 | 5148.7 | 5112.4 | 8/238 | 880.3 | 10.71 | 0.0000 | 0.000 |
| 2-class | 2216.3 | 4543.7 | 4466.6 | 17/238 | 472.1 | 9.31 | 0.0000 | 0.087 |
| 3-class | 2060.1 | 4290.3 | 4172.3 | 26/229 | 159.8 | 8.99 | 0.9829 | 0.081 |
| 4-class | 2038.6 | 4306.0 | 4147.1 | 35/220 | 116.7 | 9.26 | 1.0 | 0.800 |
| 5-class | 2022.1 | 4332.0 | 4132.2 | 44/211 | 83.8 | 9.58 | 1.0 | 0.794 |
| 6-class | 2014.8 | 4376.1 | 4135.5 | 53/202 | 69.1 | 9.97 | 1.0 | 0.790 |
| 7-class | 2009.2 | 4423.9 | 4142.4 | 62/193 | 57.9 | 10.41 | 1.0 | 0.790 |
| 8-class | 2005.1 | 4474.4 | 4152.1 | 71/184 | 49.6 | 10.89 | 1.0 | 0.787 |
| 9-class | 2001.3 | 4525.8 | 4162.7 | 80/175 | 42.2 | 11.44 | 1.0 | 0.785 |
| 10-class | 1997.4 | 4576.9 | 4172.8 | 89/166 | 34.4 | 12.03 | 1.0 | 0.784 |
Note: a Significance values can be computed using the Lo-Mendell-Rubin likelihood-ratio test (Lo et al., 2001) allowing for direct tests between models with ‘k’ and ‘k-1’ classes. Low p-values indicate the model with one less class should be rejected in favor of the estimated model. b %ER = percent error reduction in L2 when model is pitted against the null model of complete independence (one-class model).