Table 3.

Odds ratios obtained by logistic regression.

Variable name
Clinton sup.	0.909***	0.884***	0.883***	0.883***	0.883***	0.884***	0.883***	0.883***	0.884***
Cross-group	1.172***	1.156***	1.133***	1.165***	1.117***	1.162***	1.163***	1.122***	1.144***
Clinton sup., cross-group	1.104***	1.162***	1.163***	1.163***	1.165***	1.164***	1.165***	1.162***	1.164***
Avg. score (author)	1.176***	1.174***	1.175***	1.175***	1.175***	1.174***	1.174***	1.174***	1.173***
Avg. score (target)	1.169***	1.151***	1.151***	1.151***	1.151***	1.151***	1.151***	1.151***	1.151***
Diff. avg. score	1.232***	1.208***	1.209***	1.209***	1.210***	1.209***	1.209***	1.209***	1.209***
Frac. positive (author)	1.244***	1.208***	1.211***	1.211***	1.207***	1.209***	1.207***	1.212***	1.208***
Frac. positive (target)	1.192***	1.179***	1.179***	1.179***	1.178***	1.179***	1.178***	1.180***	1.178***
Diff. frac. positive	0.501***	0.488***	0.488***	0.488***	0.488***	0.488***	0.488***	0.488***	0.488***
Same state	1.245***	1.243***	1.241***	1.241***	1.242***	1.239***	1.240***	1.241***	1.242***
Same state, cross-group	0.811***	0.811***	0.810***	0.809***	0.812***	0.815***	0.817***	0.809***	0.814***
Swing state		1.018*
Swing state, cross-group		0.965*
Clinton share			1.005
Clinton share, cross-group			1.026
Trump share				0.997
Trump share, cross-group				0.971*
Non-vote share					0.977*
Non-vote share, cross-group					1.057***
Unemployment						1.020*
Unemployment, cross-group						0.978
Gini coefficient							1.006
Gini coefficient, cross-group							0.977
Median income								1.006
Median income, cross-group								1.046*
High school									1.002
High school, cross-group									1.008

Each column corresponds to a model with a specific set of variables. We indicate with three asterisks the statistically significant correlations ($p<0.001$). Odds ratios and significance are similar with or without the inclusion of the “same state” variables.

Odds ratios. * $p<0.05$, *** $p<0.001$.