(Nearly) Final 2008 Early Voting Statistics
Last updated: May 1, 2010
In the presidential election of 2008, approximately 39.7 million or 30% of all votes were cast prior to Election Day, November 4, 2008. This is a significant increase from 20% in 2004 and part of the upward trend experienced since 1992, when 7% of all votes were cast early. These numbers are likely to increase in subsequent presidential elections as more states adopt early voting and more voters become comfortable with the practice. A summary of these early voting laws can be found at Paul Gronke's excellent Early Voting Center. I discuss the bright future of early voting - among other trends - in this article available at the on-line political science journal, The Forum.
I say "approximately 30%" because there are a number of technical issues that arise when discussing these early voting statistics.
Perhaps most importantly, some mail-in ballots are dropped off on Election Day at a polling place. There were 1.1 million such drop-off ballots in California alone, which encourages this practice. There is some question as to whether or not these drop-off ballots count as early votes since they are technically cast on Election Day, though one could imagine a voter filling out their ballot the night before the election. Most states do not report their mail ballot statistics in such fine detail, so I do not attempt to make a distinction in the statistics reported here. Depending on how one would count these drop-off ballots, and depending on other details I discuss below, the percentage of early votes could be higher or lower than 30%.
Second, there are a number of states that report incomplete statistics or are not yet reporting, noted below. Sometime mid-2009, I hope to update these numbers with statistics provided by the states and localities to the U.S. Election Assistance Commission as part of their survey of election administration in the United States, known as the Election Day Survey. (I am a consultant to the EAC on the 2008 Election Day Survey and a co-author of the 2004 report.)
Third, most states report the total number of early votes cast and early votes are expressed as a percentage of the total ballots cast in these states, while a few states noted below report statistics for the number of early votes cast by office. I report for these states the total number of early votes cast for president, or what is known generically as Vote for Highest Office (a designation that has more meaning in a non-presidential election year). I report early votes as a percentage of the vote for president in these states.
Fourth, for a few states, a small number of localties have incomplete reporting. Here, the number of early votes is estimated by extrapolation.
Where available, I post the percentage of early votes that were cast by mail, rather than cast in-person at a special early voting polling place. For some states, such in-person early votes may be cast in-person at a local election administration office.
The Associated Press Elections Research and Quality Control Group graciously provided many of these statistics. All errors are my own. Please be sure to read the notes at the bottom of the page if you believe there is an error. During the run-up to the 2008 general election, I posted pre-election 2008 early voting statistics that for some states have links to early voting data and summary statistics by party registration and other demographics. I have also posted 2008 voter turnout rates, some of which data is used in the table below.
|Dr. Michael McDonald
Department of Public and International Affairs
George Mason University
4400 University Drive 3F4
Fairfax, VA 22030-4444