Table 2.

Latent profile analysis model fit indices on the screened sample (k = 1 to 6; N = 489).

Profiles (k)	Log-likelihood	AIC	BIC	aBIC	Entropy
1	−1387.7	2783.5	2800.3	2787.6	-
2	−1258.0	2534.1	2571.8	2543.2	0.61
3	−1207.3	2442.6	2501.3	2456.8	0.73
4 (selected)	−1187.6	2413.2	2492.9	2432.6	0.65
5	−1172.6	2393.3	2493.9	2417.7	0.70
6	−1161.6	2381.1	2502.7	2410.7	0.78

AIC, Akaike Information Criterion; BIC, Bayesian Information Criterion; aBIC, sample-size-adjusted BIC. Lower values indicate better fit; entropy closer to 1 indicates clearer classification. The BIC was minimized at k = 4 (narrowly relative to k = 5, ΔBIC = 1.0); the aBIC and AIC continued to decrease through k = 6; entropy was moderate at k = 4 and somewhat higher at k = 5 and k = 6. The four-profile solution was selected on BIC, parsimony, and interpretability (Section 3.3).