Table 6.
Results of fitting structural and node attribute model to observed news sources social network.
| ERGM Term |
Null Model |
Node Attributes Only Model (Threshold = 4) |
Structural and Node Attributes Model (Threshold = 4) |
Node Attributes Only Model (Threshold = 3) |
Node Attributes Only Model (Threshold = 5) |
|---|---|---|---|---|---|
| Parameter (Standard Error) | |||||
| edges | -0.61 (.02)*** | -4.23 (0.34)*** | -10.14 (0.12)*** | -3.32 (0.17)*** | -2.62 (0.55)*** |
| gwesp | 4.98 (0.14)*** | ||||
| absdiff.Toxicity | -11.08 (0.59)*** | -11.12 (0.03)*** | -7.25 (0.36)*** | -14.10 (0.91)*** | |
| nodecov.Toxicity | 5.03 (0.35)*** | 5.06 (0.05)*** | 3.21 (0.22)*** | 4.37 (0.52)*** | |
| absdiff.Upvotes | -1.58 (0.18)*** | -1.83 (0.16)*** | -1.09 (0.09)*** | -2.82 (0.34)*** | |
| nodecov.Upvotes | 0.67 (0.14)*** | 0.58 (0.12)*** | 0.53 (0.06)*** | -0.13 (0.26) | |
| absdiff.Comments | -0.00 (0.00)*** | -0.00 (0.00)*** | -0.00 (0.00)*** | -0.00 (0.00)*** | |
| nodecov.Comments | 0.00 (0.00)*** | 0.00 (0.00)*** | 0.00 (0.00)*** | 0.00 (0.00)*** | |
| absdiff.Word_Count | -0.07 (0.01)*** | -0.07 (0.01)*** | -0.07 (0.00)*** | -0.04 (0.01)*** | |
| nodecov.Word_Count | 0.04 (0.01)*** | 0.04 (0.01)*** | 0.04 (0.00)*** | 0.03 (0.01)*** | |
| AIC | 9582.49 | 7789.5 | 7767.06 | 17,905.25 (AIC for null model = 20,939) | 4525.86 (AIC for null model = 20,939) |
| Log Likelihood | -4790.25 | -3885.75 | -3873.53 | 8943.63 (null model = −10468.5) | -2253.93 (null model = −2785.73) |
* indicates p < 0.05, ** indicates p < 0.01, *** indicates p < 0.001