Table 3.

Centrality bias regression: beta coefficients from closeness and in-degree simulations.

Variables	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
	In-degree	Out-degree	Total degree	Bon. power	Closeness	Betweeness
Intercept	.09***	.074***	.091***	.09***	.058**	.085***
	(.01)	(.02)	(.01)	(.01)	(.02)	(.02)
Correlation with centrality	.036***	.026*	.035***	.04***	.058***	.012
	(.01)	(.01)	(.01)	(.01)	(.01)	(.02)
Correlation type (0 = closeness 1 = in-degree)	.004***	.005***	.005***	.009***	.005***	.004***
	(.001)	(.001)	(.001)	(.002)	(.001)	(.001)
Correlation with Centrality*Correlation type	.009***	.021***	.016***	.028***	.013***	.008***
	(.002)	(.002)	(.002)	(.004)	(.003)	(.002)
Directed	−.004	.02**	.005	.005	.013	.022***
	(.004)	(.01)	(.005)	(.003)	(.01)	(.01)
Correlation with Centrality*Directed	−.013***	−.012**	−.01**	−.019***	−.019***	−.005
	(.003)	(.004)	(.003)	(.004)	(.004)	(.01)
Log of Size	−.006*	−.006	−.006*	−.005**	.001	−.002
	(.002)	(.004)	(.002)	(.001)	(.004)	(.003)
Correlation with Centrality*Log of Size	−.003*	−.004	−.004*	−.002	−.006**	.001
	(.002)	(.002)	(.002)	(.002)	(.002)	(.003)
In-degree Std. Dev.	−.005***	−.003	−.005***	−.005***	−.003	−.004**
	(.001)	(.001)	(.001)	(.001)	(.002)	(.001)
Correlation with Centrality*In-degree Std. Dev.	−2e−04	.001	−1e−04	−2e−04	−3e−04	−1e−04
	(.001)	(.001)	(.001)	(.001)	(.001)	(.001)
N	96	96	96	96	96	96
Networks	12	12	12	12	12	12

Note: The regression uses the betas slopes from each line as the dependent variable. The betas represent the expected drop in correlation (between the empirical and the observed) for a 10% increase in the amount of missing data. Larger numbers mean larger bias with more missing data. The correlation with centrality takes four values: −.75, −.25, .25, and .75.