Do problems with government performance change public support for democracy? This study reports results from a 2022 survey experiment of nearly 2,000 residents in Spain that used short writing prompts to make respondents think about two types of poor government performance—corruption or unemployment—and, for corruption, whether it was framed as caused by elites or by the system. The paper examines whether these prompts shift attitudes toward democracy measured both as broad democratic support and as support for specific democratic practices.
Survey experiment; writing-based primes Measures: conceptual democracy vs. specific practices Sample: ~2,000 residents in Spain
Data and methods
The authors conducted a 2022 online survey experiment with close to 2,000 Spanish residents. Respondents were randomly assigned to write about either corruption or unemployment, and a subset received a corruption prompt that specified corruption as driven by elites versus the broader political system. Outcomes come from survey items capturing support for democracy in a general, conceptual sense and separate items about concrete democratic practices (for example, civil liberties and institutional checks).
Main findings
The writing primes reduced respondents' support for democracy when asked about democracy in a broad, conceptual way. By contrast, these same primes did not reduce support for specific democratic practices such as civil liberties or institutional checks on executives. The authors present these results as evidence that government performance can affect democratic attitudes in nuanced ways, distinct from other drivers like partisanship and elite rhetoric.
- Writing prompts about corruption or unemployment lowered support for conceptual democracy.
- No detectable decline in support for concrete democratic practices following the prompts.
