Table 3 OLS and mixed effects logistic regression on support China.

From: Chinese online nationalism as imaginary engagement: an automated sentiment analysis of Tencent news comments on the 2012 Diaoyu (Senkaku) Islands incident

 

Dependent variable: Support China

 

OLS

Mixed Effect

 

Model 1

Model 2

Model 3

Model 4

Intercept

0.559***

0.512***

0.566***

0.509***

 

(0.126)

(0.128)

(0.047)

(0.034)

Restrictive

−0.268*

−0.283**

−0.199***

−0.205***

 

(0.107)

(0.108)

(0.047)

(0.034)

Number of Words

0.058***

0.047***

0.061***

0.050***

 

(0.004)

(0.004)

(0.003)

(0.003)

Number of sentences

0.063

0.054

0.034

0.028

 

(0.056)

(0.056)

(0.033)

(0.028)

Happiness of comments

 

0.361*

 

−0.380***

  

(0.170)

 

(0.034)

Sadness of comments

 

0.068

 

0.099**

  

(0.210)

 

(0.034)

Love of comments

 

0.503***

 

0.480***

  

(0.08)

 

(0.032)

Anger of comments

 

1.016*

 

1.013***

  

(0.510)

 

(0.034)

Disgust of comments

 

0.284***

 

0.297***

  

(0.072)

 

(0.030)

Surprise of comments

 

−0.704

 

−0.597

  

(0.737)

 

(0.750)

Fear of comments

 

0.567

 

0.557***

  

(0.350)

 

(0.034)

SD (Intercept news)

  

0.476

0.462

Observations

6398

6398

6398

6398

Adjusted R2

0.18

0.2

  

R2 Marg.

  

0.797

0.825

R2 Cond.

  

0.81

0.835

AIC

  

3459.2

3405.9

BIC

  

3493

3487

ICC

  

0.1

0.1

RMSE

  

0.27

0.27

  1. Model 1 and Model 2 are OLS logistic regression models. In comparison, Model 3 and Model 4 are mixed effects logistic regression models using news as a group variable.
  2. *p < 0.05; **p < 0.01; ***p < 0.001.