Support for Democracy: Attitudinal versus Revealed-Preference Measures

We assess the performance of the most frequently employed survey measures of support for democracy. We contrast such conventional, attitudinal measures with revealed-preference measures. The latter are based on voting for candidates in experimentally manipulated scenarios that mimic real-world electoral trade-offs between democratic principles and competing political considerations, such as partisan loyalty or policy preferences. We find that all attitudinal measures overstate the depth and breadth of commitment to democracy and several underperform on three baseline criteria that we develop to assess their performance: monotonicity, coverage, and discriminating power. Two popular measures exhibit either no relationship or a reverse relationship to experimentally revealed commitment to democracy, resulting in potentially misleading findings. These conclusions are robust to a range of alternative explanations, statistical techniques, including conventional and machine learning approaches for detecting treatment effect heterogeneity, and hold across more than twenty countries with diverse levels and histories of democracy. We propose a number of recommendations for an improved practice.