Table 4.

Biases (“bias”), ratios of empirical SEs (“ratio SE”), ratios of mean estimated SEs (“ratio empSE”), and relative efficiencies (%, “rel. eff.”) of six MI methods when N = 100 and p_miss = 0.25. Ratios and relative efficiencies are calculated relative to the corresponding complete-data estimators. Each reported ratio or relative efficiency is the average over three ratios or relative efficiencies: one from each of the MCAR, MAR-A and MAR-B scenarios. Reported biases are the signed average absolute bias over these three scenarios.

	βcat				βconA
	bias	ratio SE	ratio estSE	rel. eff.	bias	ratio SE	ratio estSE	rel. eff.
M=1
Complete data	0.038	1.000	1.000	100.0	0.018	1.000	1.000	100.0
Match var: FCS	0.048	1.199	1.227	69.7	0.025	1.164	1.192	73.5
Normal	−0.017	1.068	1.186	88.3	0.027	1.150	1.185	75.2
Latent norm	0.066	1.193	1.219	69.9	0.008	1.080	1.178	86.4
Match set: FCS	0.041	1.221	1.277	67.3	0.022	1.234	1.231	65.7
Normal	0.029	1.160	1.309	74.7	0.051	1.321	1.295	55.6
Latent norm	0.099	1.304	1.301	57.9	−0.013	1.085	1.203	85.3
M=4
Complete data	−0.015	1.000	1.000	100.0	0.003	1.000	1.000	100.0
Match var: FCS	−0.018	1.200	1.217	69.5	0.005	1.130	1.183	78.3
Normal	−0.065	1.099	1.184	80.3	0.008	1.124	1.176	78.9
Latent norm	−0.009	1.191	1.210	70.7	−0.008	1.073	1.172	86.6
Match set: FCS	−0.037	1.157	1.213	74.3	−0.011	1.091	1.183	83.7
Normal	−0.040	1.136	1.228	76.7	0.012	1.172	1.211	72.5
Latent norm	−0.010	1.234	1.237	65.8	−0.026	1.079	1.185	83.6