Working to find actionable solutions to the nation's key challenges.

Long Voting Times and the 2012 General Election


Tuesday, July 30, 2013

Share on FacebookTweet about this on TwitterShare on LinkedInPin on PinterestShare on TumblrEmail this to someonePrint this page

“I have to go to work. I’ve been here since 8:30 in the morning. I’ve been waiting almost 4 hours to vote.” – 2012 general election voter from Southern Florida.* 

Voters and media observers alike noted unusually long wait times to vote in several jurisdictions in the 2012 general election. Steps were taken on both federal and state levels to address this problem, with the creation of the Presidential Commission on Election Administration and the introduction of numerous initiatives by state legislatures. The Presidential Commission is set to meet in Denver, Colorado on August 8 to further discuss improving election administration.

Wait times varied significantly across the country. The average nationwide wait time was 13 minutes, while statewide averages ranged from Vermont’s two minute wait to Florida’s 39 minute wait**. While these averages display the variation among states, they mask the significant variation within states. Voters within different counties experienced significantly different voting environments, with some voters waiting upwards of six hours to cast their ballot. Which election administration practices contribute to these differences in wait times?

Using recently released election administration data from the 2012 Election Administration and Voting Survey***, and combining it with estimated wait times from the 2012 Survey of the Performance of American Elections****, I determined the effect of various election administration factors on average county wait time. I examined the potential effects of the number of voters per poll worker, the voter to voting booth ratios, the size of polling places and precincts, the type of voting machines used, the percentage of voters voting on Election Day or voting provisionally, and the reported difficulty of finding an adequate number of poll workers.

The ratio of voters to poll workers has a significant effect on wait times. On average, increasing the number of voters per poll worker by 100 increases wait time by approximately ten minutes, even when controlling for the number of voters per voting booth and the number of voters per polling place. Most counties surveyed had between 20 and 200 voters per poll worker, while several had upwards of 300. However, the number of voters per voting booth and number of voters per polling place did not have significant effects on wait times.

One hypothesis suggested by these results is that a significant part of the wait to vote may be occurring at check in, rather than during the actual casting of the ballots. One solution might be to increase the number of poll workers in precincts where there are too few. Another solution might be to invest in electronic poll books, which have been recommended by election improvement commissions in Virginia’s Prince William and Fairfax counties as a way to speed up the check in process.

A second significant effect was found in the percentage of Direct Recording Electronic (DRE) voting machines used. On average, counties that used entirely DREs reported wait times that were 6 minutes longer than counties that used no DREs. There are several hypotheses that could explain these results. Several states, including Virginia, have banned the purchase of new DREs, forcing cash strapped counties to rely on aging voting equipment prone to breakage and malfunction. Several DRE manufacturers have gone bankrupt, making the purchase of replacement DRE parts difficult. Furthermore, the Fairfax Bipartisan Election Process Improvement Commission***** reported that even when DREs worked properly, lines tended to form because voting on DREs tended to take longer. Finally, people sometimes are mistrustful of paper ballots because they fear that these ballots will not be counted, leading them to prefer voting by DRE in precincts where both options are available.

The newly released election administration data can be a valuable resource for election officials and policymakers alike, as it may provide some insight into the causes of election delays and possible solutions to these problems.


** Stewart, Charles, 2013, “Waiting to Vote in 2012”. Journal of Law and Politics, Forthcoming.


**** Stewart, Charles, 2013, “2012 Survey of the Performance of American Elections”, UNF:5:nMKNqnHfGzpAilhPJPvE8g== V2 [Version]


2013-07-30 00:00:00