Department of Linguistics, University of Washington, Seattle, WA, United States of America.
Department of Public Policy, University of California-Los Angeles, Los Angeles, CA, United States of America.
PLoS One. 2021 May 6;16(5):e0250784. doi: 10.1371/journal.pone.0250784. eCollection 2021.
Scholars have offered multiple theoretical resolutions to explain inconsistent findings about the relationship of state repression and protests, but this repression-dissent puzzle remains unsolved. We simulate the spread of protest on social networks to suggest that the repression-dissent puzzle arises from the nature of statistical sampling. Even though the paper's simulations construct repression so it can only decrease protest size, the strength of repression sometimes correlates with a decrease, increase, or no change in protest size, regardless of the type of network or sample size chosen. Moreover, the results are most contradictory when the repression rate most closely matches that observed in real-world data. These results offer a new framework for understanding state and protester behavior and suggest the importance of collecting network data when studying protests.
学者们提出了多种理论解决方案来解释国家镇压与抗议之间关系的不一致发现，但镇压与异议的难题仍然没有得到解决。我们模拟了抗议在社交网络上的传播，以表明镇压与异议的难题源于统计抽样的性质。尽管该论文的模拟构建了镇压，使得镇压只能减少抗议的规模，但镇压的力度有时与抗议规模的减少、增加或不变相关，而与所选择的网络类型或样本大小无关。此外，当镇压率最接近实际数据中观察到的情况时，结果最具矛盾性。这些结果为理解国家和抗议者的行为提供了一个新的框架，并表明在研究抗议时收集网络数据的重要性。
