Understanding the Scientific Topics in the Chinese Government’s Communication about COVID-19: An LDA Approach

Abstract

The communication of scientific topics can play a key role in the fight against misinformation and has become an important component of governments’ communication regarding COVID-19. This study reviewed the Chinese government’s COVID-19 information sources and identified the patterns of science communication models within them. A corpus of science-related content was collected and coded from 1521 news briefings announced by the Chinese government. An LDA (latent Dirichlet allocation) topic model, correlation analysis, and ANOVA were used to analyze the framing of the scientific topics and their social environmental characteristics. The major findings showed the following: (1) The frames in the Chinese government’s communication of scientific topics about COVID-19 had three purposes—to disseminate knowledge about prevention and control, epidemiological investigations, and the public’s personal health; to make the public understand scientific R&D in Chinese medicine, enterprises, vaccines, treatment options, and medical resources; and to involve citizens, communities, and enterprises in scientific decision making. (2) The frames were correlated with the public and media concerns. (3) The frames varied with the different levels of officials, different types of government agencies, different income regional governments, and different severity levels of the epidemic. (4) The topics concerning sustainability science were more correlated with public and media concern. In addition, we propose several suggestions for building sustainable communication approaches during the pandemic.

The COVID-19 Infodemic on Twitter: A Space and Time Topic Analysis of the Brazilian Immunization Program and Public Trust

“…Xie et al [22] reviewed the Chinese government's information about COVID-19 to identify scientific communication topics. They applied LDA, correlation analysis, and analysis of variance (ANOVA) on a collection of news briefings composing a sciencerelated corpus.…”