Professor Michael Herron (Dartmouth) and I look forward to sharing our findings on early voting in Florida in the 2008 election at the 2012 State Politics and Policy Conference to be held in Houston, TX on February 16-18, when we present our paper, “The Participatory Impact of Truncating Early Voting in Florida.” It’s pretty timely, given all the attention that Florida US Senators Bill Nelson and Marco Rubio have given to early voting and HB1355.

Here’s our Abstract (tentative):

Over the past two decades, an increasing number of American states have made it more convenient for potential voters to cast early ballots.  Starting with Texas’ adoption of in-person early voting in 1988, 32 states now provide an extended time period prior to Election Day for voters to go to the polls.  Despite the diffusion of and praise by voting rights advocates for early voting, in 2011 the Florida legislature enacted House Bill 1355, which truncated the state’s early voting period from a total of 14 days to eight days and completely eliminated  early voting on the Sunday immediately preceding Election Day.  Critics of the legislation contend the surreptitious goal of the Republican-controlled legislature was to depress African American early voting turnout in 2012.

In this paper, we draw on an original dataset to gauge the potential participatory ramifications of HB 1355 by examining patterns of early voting in the 2008 general election.  By merging the state’s 2008 voter file, comprised of more than 11.3 million registered voters, with the state’s November 2008 early voter file, we are able to assess and study the race and ethnicity, party registration, age, gender, precinct/county registration, and vote history of each registered voter, including those who cast an early ballot, in 2008.

Unlike many studies of early voting in the American states which rely on aggregate-level data, we are able to pinpoint not only which voters were more likely to cast early ballots—specifically their socio-demographic characteristics—but we can also describe on which day during the two-week period in 2008 that they voted.  We employ a variety of multivariate models to test the conventional wisdom that African American voters are more likely than whites to vote early, and vote on Sunday, and that older and partisan voters vote early more often (Stein 1998). In addition, using a voter’s vote history to model early voting, we challenge the growing scholarly consensus—which is based largely on survey data—that early voting merely retains engaged voters (Stein 1998; Neely and Richardson 2001; Berinsky 2005; Kousser and Mullen 2007; Burden, et al. 2011; Gronke, Galanes-Rosenbaum, and Miller 2008) rather than stimulating peripheral voters.