Table 2.
Fit indices for latent profile analysis models with 1–5 profiles
| Number of profiles | Number of free parameters | Log likelihood | AIC | BIC | ABIC | VLMR LRT (p) | Entropy | Smallest class size n (%) |
|---|---|---|---|---|---|---|---|---|
| 1 | 10 | −5300.45 | 10620.89 | 10666.49 | 10634.73 | - | - | 706 |
| 2 | 18 | −5120.69 | 10277.39 | 10359.46 | 10302.30 | 0.151 | 0.87 | 168 (27.8%) |
| 3 | 24 | −5053.64 | 10155.27 | 10264.70 | 10188.50 | 0.261 | 0.90 | 37 (5.2%) |
| 4 | 28 | −5060.71 | 10177.42 | 10305.09 | 10216.18 | 0.390 | 0.79 | 37 (5.2%) |
| 5 | 34 | −5016.57 | 10101.14 | 10256.17 | 10148.21 | 0.460 | 0.81 | 37 (5.2%) |
VLMR LRT and entropy are not calculated for the 1-profile model. Model highlighted in bold represents the best-fitting model
AIC Akaike Information Criterion, BIC Bayesian Information Criterion, ABIC Sample-size Adjusted BIC, VLMR LRT Vuong Lo-Mendell Rubin Likelihood Ratio Test