Why Studying State Voting Laws Is Not Enough

Abstract

Early studies of the effects of voter laws on turnout often showed that early voting, absentee, and mail voting had limited impacts on voter turnout, with only same day registration consistently linked to higher turnout. Much of the previous research measured these laws in isolation (although most states have combinations of the laws), omitted measurement of election administration, did not account for possible selection bias in state adoption of the laws, focused on overall voter turnout rather than that for disadvantaged groups, and did not measure the effects of the laws on campaign mobilization strategies. Census data used in previous studies omitted variables (e.g., political interest and partisanship) known to influence voting decisions. Building on research from 2000s and 2010s, Chapter 3 emphasizes how causal inference research design and national voter files can lead to more precise estimations of the effects of convenience voting laws and election administration on voter turnout.