Minimum detectable effect size for main outcomes (accounting for sample
design and clustering)
The main test compares the control condition with the civic education arm. We are to be sufficiently powered for a minimum detectable effect ranging in 0.1 to 0.2 standard deviations in an outcomes index with 80% power and 95% confidence. For instance, recent studies of civic education's effects on voter turnout have found effects of about 6 - 6.6 percentage points, or about 0.13 standard deviations. We have 84% power to detect a 6pp increase in voting, 76% to detect a 3pp effect on affective polarization, 96% to detect effect on whether democracy is the best form of government, 60% for IAT.