In a recent APSR article, Gerber, Green, and Larimer (2008) present evidence from a field experiment in which nonpartisan direct mail applying social pressure has a strong and positive effect on voter turnout. In this paper, I test whether the roughly four percentage point effect of social pressure within households from the Gerber, Green, and Larimer (GGL) design remains constant across variations in the electoral environment. I present results from a field experiment conducted during a municipal election in the state of Michigan. A sample of 585,378 households were randomly assigned to receive one of three types of direct mail: 1) a civic duty mailer, 2-3) a mailer with vote history for registered voters within the household for the most recent low(high) salience election. Importantly, I account for what was on the ballot in each municipality. I find results consistent with that of the GGL design. Even in the context of a competitive mayoral or city council race, or both, I observe treatment effects on the order of four to five percentage points.
Mendeley saves you time finding and organizing research
There are no full text links
Choose a citation style from the tabs below