Acceso usuarios
Michael W. Sances, Charles Stewart III
To what degree are evaluations of political processes affected by political outcomes? In this paper, we explore this question by combining 30 national U.S. surveys from 2000 to 2012, improving on previous analyses in three ways. First, our measure asks directly about the counting of votes, rather than broader democratic processes. Second, we control for endogeneity by comparing the same respondents pre- and post-election, and by comparing respondents whose preferred candidate barely won to those whose candidate barely lost. Third, we reveal previously unknown within-country variation in this effect. We find losers are significantly more likely to believe votes were improperly counted, an effect that has grown over time and that is stronger for more remote levels of government.
