Table 2.
Class enumeration fit indices and qualities for longitudinal latent class analyses, stratified by gender: Growing Up Today Study, 2003–2010.
| Classes | Free parameters | BIC | BF | cmP | AWE | Entropy |
|---|---|---|---|---|---|---|
| Women | ||||||
| 1 | 144 | 200031 | <1 | 0.00 | 201 895 | NA |
| 2 | 289 | 177553 | <1 | 0.00 | 181 296 | 0.875 |
| 3 | 434 | 170 637 | <1 | 0.00 | 176 256 | 0.848 |
| 4 | 579 | 168055 | <1 | 0.00 | 175 553 | 0.828 |
| 5 | 724 | 167337 | <1 | 0.00 | 176 712 | 0.817 |
| 6 | 869 | 167 004 | > 10 | 1.00 | 178 257 | 0.798 |
| 7 | 1014 | 167091 | > 10 | 0.00 | 180 222 | 0.779 |
| 8 | 1159 | 167369 | > 10 | 0.00 | 182 376 | 0.777 |
| 9a | 1304 | 167 720 | NA | 0.00 | 184 605 | 0.778 |
| Men | ||||||
| 1 | 144 | 111 869 | <1 | 0.00 | 113 791 | NA |
| 2 | 289 | 99 323 | <1 | 0.00 | 103 180 | 0.860 |
| 3 | 434 | 95437 | <1 | 0.00 | 101 228 | 0.838 |
| 4 | 579 | 94 504 | <1 | 0.00 | 102 229 | 0.818 |
| 5 | 724 | 94 380 | >10 | 1.00 | 104 040 | 0.779 |
| 6 | 869 | 94518 | >10 | 0.00 | 106 113 | 0.762 |
| 7 | 1014 | 94953 | NA | 0.00 | 108 484 | 0.754 |
Models were estimated with samples of 7470 women and 5033 men. Bold type indicates the model we selected for each gender (i.e., the 6 class model for women, and the the 5 class model for men). BIC = Bayesian information criterion; BF = Bayes factor; cmP = correct model probability; AWE = approximate weight of evidence criterion; NA = not applicable. The likelihood ratio χ2 goodness-of-fit tests could not be computed in Mplus because the latent class indicator model was too large. Bootstrap likelihood ratio test could not be estimated because we employed the COMPLEX command in Mplus to adjust for non-independence within sibling clusters.
Model was not well identified.