Does Government Rumor Rebuttal Halt Online Collective Action?

• In Event: Digital Media, Repression and State Power

Abstract

The prevalence of Information Communication Technologies (ICT) has elevated the importance of social media platforms as a central arena for the spread of misinformation, often serving as a catalyst for online collective actions. Governments worldwide frequently attempt to intervene in these online protests through strategies such as fact-checking, rumor refutation, denial of “fake news,” and, in some cases, more drastic measures like censorship or even Internet shutdowns (Hobbs & Roberts, 2018; Jansen & Martin, 2015; Roberts, 2018). However, when misinformation serves as the trigger for online collective protests, can government-led rumor rebuttal effectively quell protest behavior? Existing research offers competing implications about whether truth matters to online activists. One stream of literature indicates that rumor refutation can effectively correct people’s misperceptions and subsequently impact political behaviors (Aird et al., 2018; Clayton et al., 2020). The effectiveness of responsive strategies hinges on the assumption that collective protests are driven by facts and that protestors act rationally. This assumption implies that the public has clear, explicit demands related to a specific event, policy, or government, and that collective protests should cease once these demands are met.

An opposing viewpoint suggests that rebuttals can “backfire” by failing to correct people’s beliefs especially among certain ideological groups(Nyhan & Reifler, 2010). This contrasting perspective posits that it is the individual’s preexisting beliefs that shape their perceptions, rather than the content of the rebuttal itself (Webster & Abramowitz, 2017). Netizens with specific political attitudes may initiate protests around issues that are contrary to their viewpoints, irrespective of the factual accuracy of the information. In this view, preexisting attitudes playing a more substantial role in how they interpret information and respond, which may hinder individuals from accepting corrections to their misconceptions. The fundamental distinction between these two competing explanations lies in whether truth serves as the decisive factor in stopping online collective protests.

In this study, we propose a theoretical framework of “influence of presumed influence of rumor rebuttal” (IPIRR). The theory posits that government rebuttal – the provision of facts – can curtail the spread of online collective actions by making them to believe others have accepted the governmental rebuttal and stopped protests, without necessarily changing people’s factual judgement, attitudes, or emotions. We test this theory by analyzing an incident of online collective protest in China as a natural experiment, using a regression discontinuity design to empirically investigate the effect of government-led rumor rebuttal. We collected Weibo posts and comments generated during an online collective action in China. We employ machine learning methods to extract variables such as political attitude expression behavior, judgment expression behavior, emotional expression behavior, dissemination content, and content deviation. This integration allows for the examination of online collective actions in a real-world setting, providing a more dynamic and ecologically valid understanding of the phenomena under investigation. This design also allows for the identification of causal relationships by exploiting naturally occurring variations in large-scale data, addressing concerns related to endogeneity and selection biases that may arise in observational studies.

The findings indicate that government rumor rebuttal did not significantly alter netizens’ expressions of political attitudes, factual perceptions, or emotions. However, government rumor refutation significantly curtailed the information spreading behavior of collectivists. Further analysis of heterogeneity indicates that this effect primarily stemmed from the netizen group with dissent views. The results demonstrate that government rumor refutation did not halt collective action by restoring individuals’ truth perceptions or altering their attitudes, but by diminishing the perceived signal of others’ willingness to participate in protest among the opposition group.

This study challenges the assumption that collective protests are solely driven by factual information and rational decision-making. By introducing a novel theoretical framework of the “influence of presumed influence of rumor rebuttal,” this research adds nuance to the understanding of protester motivations and behavior. This framework sheds light on the social dynamics involved in the acceptance or rejection of government-led refutation efforts.

Authors

• Siyuan Gao, Tsinghua University

• Kaiping Zhang, Tsinghua University