Table 4.

Simulation results for LCA with a continuous distal outcome: Bias in mean given latent class membership

		Moderate Measurement					Strong Measurement
		Latent Class					Latent Class
Method	ES	1	2	3	4	5	1	2	3	4	5
		n = 500
Model	Zero	0.004	−0.002	0.002	−0.001	0.002	−0.000	−0.001	−0.001	−0.007	−0.006
Assign	Zero	0.003	−0.001	0.001	−0.000	0.002	0.000	−0.002	−0.002	−0.006	−0.005
P-C	Zero	0.002	−0.000	0.001	−0.002	−0.001	−0.000	−0.001	−0.002	−0.006	−0.005
Model	Sm	−0.005	0.005	−0.034	−0.052	−0.010	−0.004	−0.001	−0.022	−0.018	0.010
Assign	Sm	0.018	0.019	−0.046	−0.073	−0.057	0.006	0.004	−0.035	−0.031	−0.009
P-C	Sm	0.022	0.018	−0.052	−0.078	−0.065	0.006	0.004	−0.037	−0.034	−0.011
Model	Med	−0.016	0.005	−0.056	−0.145	−0.018	−0.015	−0.004	−0.059	−0.042	0.012
Assign	Med	0.044	0.038	−0.131	−0.199	−0.147	0.012	0.009	−0.098	−0.082	−0.040
P-C	Med	0.054	0.040	−0.138	−0.212	−0.176	0.014	0.009	−0.105	−0.087	−0.047
Model	Lg	−0.015	0.012	−0.111	−0.218	−0.043	−0.014	0.001	−0.067	−0.045	0.029
Assign	Lg	0.082	0.061	−0.224	−0.314	−0.251	0.028	0.024	−0.165	−0.127	−0.053
P-C	Lg	0.097	0.065	−0.240	−0.334	−0.298	0.032	0.023	−0.177	−0.140	−0.066
		n = 1000
Model	Zero	0.000	0.002	0.001	0.003	−0.006	−0.001	0.005	−0.001	−0.000	0.004
Assign	Zero	−0.000	0.001	0.001	0.001	−0.003	−0.001	0.004	0.000	−0.000	0.003
P-C	Zero	−0.000	0.001	−0.000	0.003	−0.003	−0.001	0.004	−0.000	−0.000	0.004
Model	Sm	−0.006	−0.001	−0.028	−0.040	0.002	−0.008	−0.003	−0.019	−0.013	0.005
Assign	Sm	0.018	0.011	−0.048	−0.061	−0.043	0.003	0.004	−0.032	−0.028	−0.013
P-C	Sm	0.022	0.012	−0.052	−0.066	−0.058	0.003	0.003	−0.035	−0.031	−0.017
Model	Med	−0.019	−0.001	−0.065	−0.076	0.003	−0.014	−0.004	−0.037	−0.021	0.017
Assign	Med	0.044	0.034	−0.135	−0.155	−0.117	0.015	0.013	−0.096	−0.072	−0.034
P-C	Med	0.054	0.035	−0.146	−0.172	−0.154	0.016	0.012	−0.104	−0.080	−0.043
Model	Lg	−0.027	−0.002	−0.067	−0.136	0.017	−0.016	0.002	−0.033	−0.031	0.022
Assign	Lg	0.074	0.053	−0.225	−0.266	−0.190	0.027	0.027	−0.156	−0.128	−0.057
P-C	Lg	0.090	0.053	−0.243	−0.292	−0.254	0.029	0.025	−0.170	−0.138	−0.075

Note: Model = model-based approach; Assign = maximum-probability assignment rule; P-C = multiple pseudo-class draws.