Yilin Su

Yilin Su

PhD Candidate at
University College London

Posts by Collection

portfolio

Portfolio item number 1

Short description of portfolio item number 1

Portfolio item number 2

Short description of portfolio item number 2

publications

How Do Fake News Laws Impact Online Opposition in Nondemocratic Regimes? Evidence from Singapore

Su, Y., Working Paper, 2025

Abstract: Restrictive Internet regulations enforced in nondemocracies, such as “fake news” laws, have been widely criticised by media and human rights defenders for their repressive potential and the likelihood of inducing self-censorship online. However, there is limited empirical evidence on their impact on online opposition. This study investigates how state-sponsored disinformation labels influence opposition behaviour online. I analyse 8,751 social media posts created by dissident actors before and after the introduction of Singapore’s fake news law, as well as after specific posts were officially labelled as false. Using k-means clustering and dictionary-assisted sentiment analysis, I find that after the online disinformation law was enacted, online activists tend to decrease discussions on domestic policies while engaging more with geopolitical issues and transnational mobilisation. In political content, their tone grew more neutral following the passage of the law, although it did not change further after being targeted. These findings provide empirically grounded new insights into digital activism under authoritarian rule.

When Do Nondemocracies Legalize Online Information Control?

Su, Y., Working Paper, 2025

Abstract: Why do some authoritarian regimes implement restrictive Internet laws to control online information while others do not? Building on the literature on online repression and authori- tarian politics, I argue that nondemocratic governments use restrictive Internet regulations as a form of preventive repression in the digital space. The primary goal is to produce a chilling effect that deters opposition coordination and encourages self-censorship among the population, leading them to withdraw support for state challengers. Using cross-national panel data and a survival analysis, I demonstrate that nondemocratic regimes treat legal and extralegal tools of digital repression as complementary and incorporate restrictive Internet laws into their control strategies, particularly when they have high regulatory capacity and face increasing threats from offline mobilisation. Moreover, the findings identify a pattern of “opportunistic oppression”, where regimes exploited the COVID-19 pandemic as a window of opportunity to accelerate the legalisation of online control. Even after accounting for this exogenous shock, the pre-existing online regulatory capacity and street protests remain strong incentives for the adoption of restrictive Internet laws. This research highlights the conditions under which autocracies use a legal façade to undertake nonviolent repression in the digital era. My work also has significant implications for research on democratic backsliding that focuses on how populists weaponise legal instruments to shape mass attitudes–a topic of timely relevance given that attacks on media freedom are often the very first step in eroding democracy.

Fake News Labels and Public Opinion in Nondemocracies: Evidence from Singapore

Su, Y., Working Paper, 2025

R&R with Comparative Political Studies

Abstract: In democracies and nondemocracies alike, the share of content flagged as “fake” on social media is on the rise, yet we know little about how fake news labels shape public opinion. This study examines the use of fake news labels in nondemocratic contexts and argues that authoritarian regimes employ them to reduce alignment with and trust in opposition messages. Using a pre-registered survey experiment conducted in Singapore, I find that government-sponsored fake news labels lower agreement with evidence-oriented posts from opposition channels, especially among government supporters. Meanwhile, rebuttals produced by targeted opposition accounts are ineffective in counteracting this negative effect. In addition, fake news labels do not significantly alter perceptions of the content’s trustworthiness or the reliability of its sources. I conclude that the opinion effects of government-sponsored fake news labels vary depending on the attributes of targeted posts, as these labels have limited influence when applied to politically sensitive messages that already attract low support from the masses or when individuals may hesitate to reveal their true opinions. This research contributes to a deeper understanding of the strategies used by authoritarian regimes to crack down on online dissent and their consequences. It offers timely insights into how modern autocracies adapt to the digital age and informs debates on democratic backsliding, where attacks on media freedom often serve as the first step in eroding democracy.

talks

Talk 1 on Relevant Topic in Your Field

Updated:

This is a description of your talk, which is a markdown file that can be all markdown-ified like any other post. Yay markdown!

Conference Proceeding talk 3 on Relevant Topic in Your Field

Updated:

This is a description of your conference proceedings talk, note the different field in type. You can put anything in this field.

teaching