Source: Dissertation Abstracts International, Volume: 73-05, Section: A, page: .

Adviser: Donald P. Green.

This dissertation studies the effects of social processes of attracting voters to the polls. Specifically, it focuses on social methods of voter mobilization and Election Day efforts that embody interpersonal, communal, and convivial attributes. I find that research on voter mobilization and voter turnout has not examined sufficiently the way in which eligible voters respond to these interrelated social characteristics.

As a guide to a potentially effective social voter mobilization strategy, I explore the history of Election Day in the United States, particularly during the eighteenth and nineteenth centuries when voting resembled our modern-day (and historical) celebratory Fourth of July experience. I outline when and how the convivial approach to voting and Election Day changed and voter turnout fell dramatically. I present the puzzle, which motivates this dissertation, of how voter turnout fell from levels routinely at 70 percent or higher in the late nineteenth century to under 50 percent in the 1920s. What would be the resulting effects on voter turnout if we brought back the social elements of the past?

I adapt relevant theories from psychology and political behavior, particularly theories of social influence, to traditional voting models and develop a model of voting behavior that accounts for the social experience of voting, incorporating both the emotional and instrumental benefits to this process---the Social Vote Model. I posit that maximizing these social benefits can be an effective strategy in driving higher levels of voter turnout, particularly given that lowering the costs to voting, long the subject of scholarly focus, has not produced overwhelming results and that U.S. voter turnout remains relatively low and narrowly based.

In addition, beyond these theoretical claims about the importance of social attributes to the voting process, this dissertation contributes to the study of voting behavior by empirically testing the effectiveness of a social voter mobilization effort through a blend of randomized field experimentation and direct participant observation. The first series of experiments involves an Election Day festival conducted at the polling site. The second series of experiments involves a first-time voter program for high school seniors. These experiments indicate that voter turnout increases substantially when voter mobilization and Election Day efforts incorporate social experiences for the voter. Election Day festivals increase voter turnout by as much as six percentage points with more widely attended festivals generating larger increases in voter turnout, suggesting that the social dynamics are integral to the estimated effects. Attending the First-Time Voter Program increases the voter turnout of high school seniors by over nine percentage points, a significant finding given that this segment of the population has the lowest voter turnout rate. Overall, the magnitude of the estimated treatment effects in these studies exceeds the increases in voter turnout typically observed in voter mobilization efforts that focus solely on lowering the costs to voting. The findings in this dissertation highlight the relevance of the Social Vote Model to potentially explaining historical trends in voter turnout, to developing new voter mobilization and Election Day experiences, and ultimately, to motivating higher and more broad-based levels of voter participation.