Expressive Responding and Trump's Big Lie

Do surveys measure sincere belief in Donald Trump's "big lie" that fraud decided the 2020 election? We apply a comprehensive approach to detecting expressive responding: three honesty encouragements, a list experiment, two opportunities to express related sentiments, and two opportunities to bet on related predictions about the future. We find that nearly all respondents who endorse the big lie appear to genuinely believe it. These "believers" are evenly split between those who confidently accept the big lie and those who find it plausible but are not deeply convinced. Similarly, those who predicted that evidence of fraud would enable Trump to retain power in January 2021 or be reinstated in August 2021 were overwhelmingly sincere. Our findings indicate that Trump's big lie is unique in terms of the size and veracity of belief differences between Democrats and Republicans. We discuss implications for democratic stability.