Table 1 Results of the generalized linear mixed-effects models described in Equation (2) and (3), with replies’ FmB scores (FmBr, left) and affective polarization scores (Polr, right) as dependent variables

From: Different honesty conceptions align across US politicians' tweets and public replies

 

Dependent variable:

 

FmB_r

Pol_r

 

(1)

(2)

FmB_s

0.075

−0.009

 

(0.067, 0.082)

(−0.011, −0.006)

 

p < 0.001

p < 0.001

I_score

−0.017

0.013

 

(−0.021, −0.013)

(0.012, 0.014)

 

p < 0.001

p < 0.001

PartyRepublican

−0.025

0.020

 

(−0.040, −0.011)

(0.016, 0.024)

 

p = 0.001

p < 0.001

Pol_s

−0.015

0.012

 

(−0.021, −0.010)

(0.011, 0.014)

 

p < 0.001

p < 0.001

Date

−0.011

0.002

 

(−0.017, −0.006)

(0.001, 0.004)

 

p < 0.001

p = 0.006

Pos

−0.006

0.000

 

(−0.010, −0.003)

(−0.001, 0.001)

 

p = 0.001

p = 0.567

FmB_s:I_score

−0.003

−0.000

 

(−0.007, 0.001)

(−0.002, 0.001)

 

p = 0.202

p = 0.471

FmB_s:PartyRepublican

−0.019

0.005

 

(−0.029, −0.009)

(0.003, 0.008)

 

p = 0.001

p = 0.001

I_score:PartyRepublican

0.035

−0.023

 

(0.029, 0.040)

(−0.024, −0.021)

 

p < 0.001

p < 0.001

FmB_s:I_score:PartyRepublican

0.006

0.001

 

(0.001, 0.012)

(−0.000, 0.003)

 

p = 0.023

p = 0.118

Constant

0.042

0.002

 

(0.026, 0.058)

(−0.004, 0.007)

 

p < 0.001

p = 0.553

Observations

97,510

97,510

  1. Estimates are reported together with their 95% confidence intervals. All predictors were scaled by subtracting their means from their values and dividing the result by their standard deviations. The regression was performed with the function lmer from the R library lme467.