Table 9.

Correlation of missingness robustness scores and network structure: homophily

	Indegree		Outdegree		High Behavioral		Low Behavioral
	Slope β	Var υ	Slope β	Var υ	Slope β	Var υ	Slope β	Var υ
Size	0.6*	−0.42	0.19	−0.08	0.32	−0.32	0.53+	−0.6*
Density, undirected	−0.45	0.28	0.16	−0.24	−0.02	0.04	−0.35	0.42
Indegree Std. Dev.	0.61*	−0.51+	−0.23	0.29	0.11	−0.16	0.5	−0.41
Outdegree Std. Dev.	0.44	−0.37	0.31	−0.28	0.38	−0.36	0.38	−0.27
Degree Std. Dev.	0.65*	−0.51+	−0.16	0.24	0.27	−0.31	0.64*	−0.57+
Mean of true score	0.06	0	0.05	0	0.4	−0.37	0.19	−0.23
Directed Network	−0.31	0.38	−0.21	0.26	0.16	−0.22	0.23	−0.31
Limited Degree	0.03	0.17	−0.22	0.33	0.12	−0.15	0.43	−0.42
% Largest Bicomponent	−0.29	0.36	0.06	−0.01	0.27	−0.31	0.27	−0.29

Note: Positive correlations mean that networks with higher values of that measure (e.g. size) are more robust to measurement error. The direction of the bias is ignored when calculating the correlations.

⁺P<= 0.1;

^*+P<= 0.05,

^**P<= 0.01