The Dynamic Effects of Perceptions of Dread Risk and Unknown Risk on SNS Sharing Behavior During Emerging Infectious Disease Events: Do Crisis Stages Matter?

Abstract

In response to the increasing prevalence of emerging infectious disease (EID) threats, individuals are turning to social media platforms to share relevant information in ever greater numbers. In this study, we examine whether risk perceptions related to user-generated content have dynamic impacts on social networking site (SNS) sharing behavior in different crisis stages. To answer this question, we applied psychometric analysis to evaluate how dread risk and unknown risk can characterize EID threats. Drawing broadly on the literature of risk perceptions, self-perception theory, and crisis stages, we relied on microblogs collected from Sina Weibo, utilizing the vector autoregression model to analyze dynamic relationships. We found that perceptions of dread risk have a dominant and immediate impact on SNS sharing behavior in the buildup, breakout, and termination stages of EID events. Perceptions of unknown risk have a dominant and persistent impact on sharing behavior in the abatement stage. The joint effect of these two types of risk perception reveal an antagonism impact on SNS sharing behavior, and perceptions of dread- and unknown risk have interaction effects from the buildup to termination stages of EID events. To check robustness, we analyzed keywords related to perceptions of dread- and unknown risk. The results of this study support the empirical application of Slovic’s risk perception framework for understanding the characteristics of EID threats and provide a picture of how perceptions of dread- and unknown risk exert differential time-varying effects on SNS sharing behavior during EID events. We also discuss theoretical and practical implications for the crisis management of EID threats. This study is among the first that uses user-generated content in social media to investigate dynamic risk perceptions and their relationship to SNS sharing behavior, which may help provide a basis for timely and efficient risk communication.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 25 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.