Temporal Dynamics of the Information Marketplace and the Politicization of Science

• In Event: Science, Knowledge & Technology Section Roundtables
  In Refereed Roundtable (60 minute): Table 4: (Infra)structures

Abstract

Recent years have seen the politicization of science globally and in the US. Much has been researched on the macro historical causes of politicization, as well as the communication strategies that enable readers to understand scientific information. What remains unknown, however, is how the timing in the communication of science, particular in a dynamic and evolving scientific controversy, affects whether a scientific issue becomes politicized. While the knowledge deficit model would recommend that scientific information be communicated early and often, the dialogue model suggests that science communicators are better off communicating after the public has settled on their main sources of information need. To test these hypotheses, I investigate the effect of a temporal mismatch between the supply and demand of scientific information on the politicization of science during the Covid-19 pandemic. To measure supply, I create topic models of various Covid-related topics (e.g. vaccines, lockdown, wastewater surveillance) using the ProQuest News dataset. To measure demand, I trace search trends of these topics on Google Trends. Finally, to measure politicization, I train dynamic word embeddings using the ProQuest News dataset. Preliminary results from the pilot study suggest that “demand-first” Covid topics are more likely to be politicized, while “supply-first” topics are less likely. Future analysis will expand to include a more comprehensive dataset, more in-depth analyses of supply, demand, and politicization patterns, and a regression to quantify effects of temporal mismatch on politicization. As science controversies are increasingly becoming political controversies, findings from this study can help science communicators decide on best the communication strategies for maintaining science’s authority and autonomy.


Keywords: The politicization of science, information marketplace, science communication, the sociology of knowledge, natural language processing

Author

• Hsin-Keng Ling, University of Michigan-Ann Arbor