Skip to main content

Table 5 The Estimated Result of ZINB model

From: Identifying features of source and message that influence the retweeting of health information on social media during the COVID-19 pandemic

Predictor

B

S.E.

P

IRR [95% CI]

Sig.

Source Feature

Followers(log)

0.297

0.019

0

1.345 [1.295 1.398]

***

Governmental level

2.055

0.121

0

7.809 [6.161 9.897]

***

Message Feature

  Structure

    Hyperlink

-1.272

0.077

0

0.280 [0.241 0.326]

***

    @

0.099

0.121

0.413

1.104 [0.871 1.399]

NS

    #

0.225

0.083

0.007

1.252 [1.0645 1.475]

**

    Picture

0.476

0.078

0

1.610 [1.382 1.875]

***

    Video

0.588

0.146

0

1.800 [1.352 2.397]

***

Style

  Q&A(?)

-0.153

0.108

0.155

0.858 [0.694 1.060]

NS

  Emotional(!)

0.419

0.088

0

1.521 [1.279 1.808]

***

Content

  Severity

1.983

0.095

0

7.267 [6.029 8.759]

***

  Reassurance

0.532

0.106

0

1.703 [1.383 2.096]

***

  Efficacy

0.609

0.01

0

1.839 [1.513 2.235]

***

  Uncertainty

-16.193

1.962

0

-16.194 [-20.038 -12.349]

***

  Action

14.274

0.827

0

14.274 [12.654 15.894]

***

  Technology

2.767

1.702

0.104

2.7668 [-0.5683 6.1018]

NS

  1. Note: NS Not significant; * p < 0.05; **p < 0.01; ***p < 0.001