TABLE 2.

Model fit indices for measurement models.

	1 Class	2 Classes	3 Classes	4 Classes
T1
Log likelihood	−682.19	−504.19	−445.262	−318.88
BIC	1404.981	1074.35	981.88	754.48
VLMR-LRT	N/A	0.004	0.199	0.416
BLRT	N/A	0.000	0.000	0.000
T2
Log likelihood	−688.43	−497.52	−403.01	−369.76
BIC	1416.62	1059.64	895.47	853.84
VLMR-LRT	N/A	0.067	0.301	0.157
BLRT	N/A	0.000	0.000	1.000
T3
Log likelihood	−467.56	−299.74	−219.31	−183.49
BIC	973.35	661.60	524.64	476.90
VLMR-LRT	N/A	0.135	0.784	0.740
BLRT	N/A	0.000	0.000	0.000
T4
Log likelihood	−418.78	−265.04	−232.77	−188.16
BIC	874.64	590.32	548.97	482.92
VLMR-LRT	N/A	0.201	0.484	0.716
BLRT	N/A	0.000	0.000	0.000

Note: T1 = time point 1; T2 = time point 2; T3 = time point 3; T4 = time point 4. For Waves 2–4, in the 3- and 4-class solutions, the best loglikelihood was not replicated, suggesting that standard errors may not be trustworthy. Additionally, for waves 3 and 4, the 3- and 4-class models each had one class that represented less than 5% of the overall sample.

Abbreviations: BIC, Bayesian Information Criteria; BLRT, Bootstrapped Likelihood Ratio Test; VLMR-LRT, Vuong-Lo-Mendell-Rubin Likelihood Ratio Test.