The Site Gap

• Author: Elliott B. Fullmer
• DOI: 10.1177/1532673X14546987
• Published date: 01 March 2015
• Date: 01 March 2015
• Subject Matter: Articles

American Politics Research

2015, Vol. 43(2) 283 –303

© The Author(s) 2014

Reprints and permissions:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/1532673X14546987

apr.sagepub.com

Article

The Site Gap: Racial

Inequalities in Early

Voting Access

Elliott B. Fullmer1

Abstract

In both 2008 and 2012, about one third of the American electorate cast

their votes early. While early voting programs are established by states,

counties have considerable discretion with regard to implementing them.

While some counties offer only a single early voting site, others offer

dozens. Previous research suggests that site density may affect the degree to

which programs increase turnout. I use county-level data from the 2008 and

2012 elections to measure whether a county’s racial and ethnic composition

predicts high (or low) levels of site density. Applying county-level data from

the Election Assistance Commission and American Community Survey, I

find that the percentage of a county identifying as Black has a significantly

negative association with early voting site density. This relationship persists

when numerous demographic covariates are included in ordinary least

squares models. These site disparities suggest that early voting may not be

achieving its full potential in heavily African American communities.

Keywords

early voting, sites, county, turnout, race

As of 2014, 33 states and the District of Columbia allow any registered citi-

zen to vote before Election Day at a physical location(s) somewhere in their

1Randolph-Macon College, Ashland, VA, USA

Corresponding Author:

Elliott B. Fullmer, Randolph-Macon College, P.O. Box 5005; 202 Henry Street, Ashland VA,

23005-5505, USA.

Email: elliottfullmer@rmc.edu

546987APRXXX10.1177/1532673X14546987American Politics ResearchFullmer

research-article2014

284 American Politics Research 43(2)

county (National Conference of State Legislatures [NCSL], 2014). These

early voting programs have expanded considerably in the past two decades.

According to the 1992 Current Population Survey (CPS; U.S. Bureau of the

Census, 1992), only 7% of voters participated before Election Day. In those

days, the vast majority of states only allowed early voting if a registered voter

demonstrated that participating on Election Day was burdensome. Most who

qualified for this exception were elderly, disabled, or deployed in the military.

In the early 1990s, however, early voting entered a “new phase” (Gronke &

McDonald, 2008). New laws began to appear throughout the United States,

predominately in the western part of the country. From 1992 to 2004, the

percentage of early votes among all ballots cast increased by an average of

4.25 percentage points per presidential election, reaching 20% by 2004

(Gronke & McDonald, 2008). In 2012, the figure soared to more than 30% of

the electorate, including a majority of voters in at least 9 states (McDonald,

2012).

Numerous researchers have investigated whether early voting increases

voter turnout by making the process more convenient for citizens. Most

recent studies have reported that the effect of early voting on participation

within states is nonexistent (Fitzgerald, 2005; Primo, Jacobsmeier, & Milyo,

2007;; Wolfinger, Highton, & Mullin, 2005), negligible (Gronke, Galanes-

Rosenbaum, & Miller, 2007), or even negative (Burden, Canon, Mayer, &

Moynihan, 2014; Larocca & Klemanski, 2011). Much of this research, how-

ever, fails to account for important differences in implementation within

states. Early voting administration is largely conducted at the local level and

variation across counties is often considerable. Counties, for example, typi-

cally have the freedom to determine the total number of early voting sites

they offer to voters. Los Angeles, the nation’s most populous county, offered

only one early voting site in 2008 and 2012. Meanwhile, Clark County (Las

Table 1. Top Early Voting States in the United States, 2012 (McDonald, 2012).

State Early voting rate %

Colorado 77.3

Nevada 72.4

Texas 66.4

North Carolina 63.3

New Mexico 62.3

Tennessee 55.6

Florida 52.9

Montana 51.8

Source. United States Elections Project, 2012.