Skip to main content

CEP/STICERD Applications Seminars

Voter Misperceptions and Preferences for Democratic Institutions

Ceren Baysan (Univeristy of Toronto), joint with Daron Acemoglu, Cevat Aksoy, Carlos Molina, and Gamze Zeki

Monday 26 February 2024 12:00 - 13:30

Many of our seminars and public events this year will continue as in person or as hybrid (online and in person) events. Please check our website listings and Twitter feed @STICERD_LSE for updates.

Unless otherwise specified, in-person seminars are open to the public.

Those unable to join the seminars in-person are welcome to participate via zoom if the event is hybrid.

About this event

Whether voters appreciate and support democratization in countries with persistent democratic backsliding and authoritarian leadership is unclear. After years of worsened democratic institutions, censorship, and state repression, voters may not prefer democracy because of misperceptions about the current state of democratic institutions and their value. Even if voters correct their misperceptions, the fear of being ruled by the opposition (Padró i Miquel, 2007) may inhibit voters from supporting electoral turnover. We evaluate large-scale online and field experiments to explore these factors of low support for democracy at a critical juncture. The experiments were implemented just before the general and presidential elections held on May 14th, 2023, in Türkiye, when the incumbent leader was up for re-election after 20 years in power. In the online experiment, we establish that an information treatment on the state of democracy and the value of democracy reduces voters’ misperceptions and increases their valuation of democratic institutions, as measured by reduced support for the incumbent. These effects are driven by respondents who underestimated the deterioration of democracy in Türkiye from 2000 to 2022 and self-reported authoritarianism as sometimes preferable at baseline. Building on preliminary results from the online experiment and to measure revealed preference for democratic institutions using electoral data, we evaluate a door-to-door campaign randomized at the neighborhood level and with a sample of approximately 830,000 voters. The campaign contained analogous information to the online experiment. Finally, we cross-randomized whether the information campaigns were implemented by partisan canvassers from across the ideological spectrum or non-partisan canvassers to isolate the impact of information from party persuasion. The information campaigns increased the opposition’s vote share by 1.9 percentage points (3.3%) overall and by 3.1 percentage points (6.4%) in neighborhoods in the lower median of the 2018 opposition vote share distribution. We cannot reject the null hypothesis that the campaigns delivered by non-partisan canvassers have the same treatment effect as partisan canvassers. Finally, we find a zero effect of the campaign on voter turnout, suggesting that the information campaign treatment effects on vote share are driven by voters who would have otherwise supported an authoritarian leader.

Applications (Applied Micro) Seminars are held on Mondays in term time at 12:00-13:30 in SAL 3.05 in person.

Seminar organiser: Maitreesh Ghatak

For further information please contact Sadia Ali:

Please use this link to subscribe or unsubscribe to our mailing list (applications).