Understanding the mechanics of democracy : evaluating and improving election administration in the U.S.

Since the controversial 2000 presidential election, there has been an increasing demand for information about improving the conduct of American elections. With only a decade-and-a-half of sustained attention by political scientists, our understanding of election administration has grown greatly. Most notably, research has focused on politically salient issues like turnout, residual vote rates, voter identification, and voter suppression. Although these issues are important and contemporaneous, persistent less visible problems plague the system and attract scant scholarly attention.

Given these major concerns about the American voting system, we know little about the relationship between election administration and participation. This dissertation examines how variation in election administration is related to the costs associated with voting and confidence in our electoral institutions at both the aggregate and individual levels. In this dissertation, I describe a theory about the relationship between turnout in elections and the public value local election administration. I operationalize public value through measures of investment of resources into local elections. Although imperfect, spending on voting systems is a proxy for the quality of an election system. In support of my theory, an original data set was created using variables related to the costs and quantity of voting equipment. This data on voting equipment costs is the first of its kind in scholarly election administration research. Throughout the three empirical chapters of this dissertation, evidence is presented for how high-quality election administration can have both short-term and long-term effects on voters' propensity to turnout.