Table A2.
Centrality Bias Slope Regressions: Undirected Networks.
| Variables | Model 1 Degree | Model 2 Bon. Power | Model 3 Closeness | Model 4 Betweeness |
|---|---|---|---|---|
| Intercept | −1.925*** | −2.572 | −2.834** | −1.802 |
| (0.46) | (1.7) | (0.94) | (1.33) | |
| Correlation with Centrality | 0.92*** | 1.199*** | 0.626*** | 0.517*** |
| (0.15) | (0.11) | (0.05) | (0.07) | |
| In-degree Std. Dev. | −0.166*** | −0.174 | −0.075 | −0.041 |
| (0.03) | (0.13) | (0.07) | (0.1) | |
| Log of Size | −0.166 | 0.016 | 0.035 | −0.206 |
| (0.09) | (0.35) | (0.19) | (0.27) | |
| Symmetric Imputation | −40.255*** | −3.407** | −2.313** | −0.397 |
| (0.6) | (1.06) | (0.89) | (0.63) | |
| Simple Model-Based Imputation | −39.492*** | −3.772*** | −2.771* | −1.214 |
| (1.22) | (1.1) | (1.21) | (0.87) | |
| Complex Model-Based Imputation | −39.492*** | −3.31** | −2.712* | −0.703 |
| (1.59) | (1.09) | (1.1) | (0.58) | |
| Correlation with Centrality* Symmetric Imputation | −0.871*** | 0.292 | 0.261*** | 0.68*** |
| (0.21) | (0.16) | (0.08) | (0.1) | |
| Correlation with Centrality* Simple Model-Based Imputation | −0.718*** | 0.225 | 0.267*** | 0.234* |
| (0.21) | (0.16) | (0.08) | (0.1) | |
| Correlation with Centrality* Complex Model-Based Imputation | −0.718*** | 0.232 | 0.291*** | 0.34*** |
| (0.21) | (0.16) | (0.08) | (0.1) | |
| In-degree Std. Dev.* Symmetric Imputation | 0.146** | −0.03 | −0.064 | 0.085 |
| (0.04) | (0.08) | (0.07) | (0.05) | |
| In-degree Std. Dev.* Simple Model-Based Imputation | 0.198* | −0.071 | −0.054 | 0.116 |
| (0.09) | (0.08) | (0.09) | (0.07) | |
| In-degree Std. Dev.* Complex Model-Based Imputation | 0.198 | −0.111 | −0.045 | 0.113* |
| (0.12) | (0.08) | (0.08) | (0.04) | |
| Log of Size* Symmetric Imputation | 0.211 | 0.224 | 0.337 | −0.257* |
| (0.12) | (0.22) | (0.18) | (0.13) | |
| Log of Size *Simple Model-Based Imputation | 0.029 | 0.294 | 0.345 | −0.15 |
| (0.25) | (0.23) | (0.25) | (0.18) | |
| Log of Size* Complex Model-Based Imputation | 0.029 | 0.241 | 0.333 | −0.242* |
| (0.33) | (0.22) | (0.23) | (0.12) | |
| N | 180 | 180 | 180 | 180 |
| Networks | 5 | 5 | 5 | 5 |
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.