Table 3.

Performance of methods for regression of (continuous) distress at Wave II with missing dieting indicator

% complete data	Method	Diet (β1 = −0.101)				Distress (β2 = 0.554)				Health (β6 = 0.042)
		Bias	SE	StdBias	Coverage	Bias	SE	StdBias	Coverage	Bias	SE	StdBias	Coverage
90%	CCA	0.001	0.067	0.019	94.2%	−0.006	0.036	−0.154	94.9%	−0.001	0.032	−0.020	95.1%
	MVNI-skew0	0.005	0.065	0.074	94.6%	0.001	0.034	0.020	94.7%	−0.001	0.030	−0.043	95.3%
75%	CCA	−0.001	0.075	−0.014	95.0%	−0.014	0.040	−0.360	94.9%	<0.001	0.036	0.009	95.5%
	MVNI-skew0	0.008	0.070	0.119	96.1%	<0.001	0.034	0.013	95.2%	−0.001	0.030	−0.042	95.6%
50%	CCA	−0.006	0.097	−0.065	94.2%	−0.028	0.049	−0.577	91.8%	0.004	0.047	0.086	94.6%
	MVNI-skew0	0.014	0.080	0.175	95.6%	0.001	0.034	0.032	94.6%	<0.001	0.030	−0.002	96.3%
25%	CCA	−0.011	0.147	−0.076	94.6%	−0.043	0.072	−0.603	91.9%	−0.001	0.071	−0.014	94.3%
	MVNI-skew0	0.018	0.104	0.172	96.1%	<0.001	0.034	−0.009	93.6%	−0.002	0.030	−0.071	94.2%
10%	CCA	−0.020	0.250	−0.081	95.8%	−0.053	0.118	−0.452	92.2%	0.009	0.120	0.077	94.7%
	MVNI-skew0	0.011	0.149	0.071	95.0%	−0.001	0.035	−0.042	95.9%	<0.001	0.031	−0.014	95.4%

Measures of performance are mean values in the estimation of the β parameters from Equation 1 across the 1000 simulated datasets of 1000 observations (compared to the true values from the synthetic population of 971,327). CCA = Complete Case Analysis; MVNI = Multivariate normal imputation; StdBias = standardised bias; SE = average (estimated) standard error across the 1000 datasets.