[Image courtesy of MIT]
This guest post by Charles Stewart, Kenan Sahin Distinguished Professor of Political Science at MIT, ran in yesterday's electionlineWeekly.
Pollsters talk to voters all the time, but it's usually with one thing in mind -- to find out how they voted. Pollsters rarely probe a topic that is near and dear to the hearts of election geeks -- what the voters experienced on Election Day.
What do voters experience when they go to vote?
To help answer this question, I have led a public opinion project, the Survey of the Performance of American Elections (SPAE), for the past six years. I gave a glimpse into the results from the 2012 edition at Pew's recent conference, Voting in America, on December 10. Here are some highlights from that talk.
The SPAE is the nation's only large-scale public opinion project that comprehensively studies the experience of voters in presidential elections. In 2012, we interviewed 10,200 registered voters, 200 from each state and the District of Columbia, in the days right after the November election.
A survey as comprehensive as the SPAE will take weeks to digest; but as an initial peek at the data, I offered some observations at the Voting in America conference along four topics: (1) the overall voter experience, (2) voter identification, (3) waiting in line to vote, and (4) partisan divisions over election reform issues.
The overall experience
Despite the fact that there were well-publicized pockets of voting difficulties in 2012, the overall portrait painted by the 2012 SPAE was one of satisfaction with voting.
The main performance statistics associated with 2012 were very similar to those reported after the 2008 election. In 2012, 86 percent of Election Day voters reported it was "very easy" to find their polling place when they went to vote, only 3 percent said they encountered a registration problem, and 2 percent encountered an equipment program.
All told, 78 percent said their polling place was run "very well," 65 percent said the performance of poll workers was "excellent," and 67 percent said they were "very confident" their vote was counted as cast. Finally, the average voter waited 13 minutes to vote, most of which was spent waiting to check in. Very similar answers were given by those who voted early in-person or by mail.
While answers like these do show room for improvement -- what happened to the 22 percent of voters whose polling places weren't well run, for instance? -- they suggest that for the vast majority of people, the direct experience of the electoral process is positive, and that reform efforts likely need to be targeted to the specific places where things didn't go so well. (More on this below.)
In the months leading up to the 2012 election, photo voter identification was the hottest issue in the election administration world. The 2012 SPAE confirms that, as a general matter, photo ID laws are popular with the general public -- 69 percent of respondents expressed a favorable opinion of requiring all voters to show a photo ID when they voted, including 53 percent of Democrats and 58 percent of African Americans. (The comparable figures for Republicans and white voters are 91 percent and 71 percent, respectively.)
However, compared to 2008, overall support for these laws dropped seven percentage points, with this decline mostly due to support among Democrats falling by 11 percentage points, and support among African Americans dropping by 16 points. There is also preliminary evidence in the survey that party and racial divisions have widened in many states in which the issue was contested actively following the 2010 midterm election.
Voter identification laws are frequently justified on the grounds they will increase confidence in the electoral process. There is no evidence in the 2012 survey that these laws have had this affect, however. For instance, voters living in states with photo ID laws are just as likely to state they think it is common for people in their city or county to commit voter impersonation fraud as are voters in states without photo ID laws.
As the early voting period unfolded and Election Day came and went, attention shifted from photo ID to long lines at the polls. Evidence from the SPAE documents that waiting in a long line to vote was uncommon in most states in 2012.
About one-third of all voters reported they did not wait at all to vote; the average wait time on Election Day clocked in at 13 minutes. (The average was 20 minutes during early voting.)
However, waiting in long lines was quite common in a few states. The identity of the five states whose voters reported the longest lines won't surprise anyone who paid attention to the news on Election Day. They were Florida (44 average minutes waiting to vote), D.C. (34 min.), Maryland (32 min.), Virginia (27 min.), and South Carolina (27 min.).
This contrasts with the states with the shortest waiting times: Vermont, at 2 minutes, and Maine, Alaska, South Dakota, and Wyoming, all at 4 minutes. While these "fast five" are all small, mostly rural states, it should be noted that the average waiting time in California was 6 minutes, including an average wait time of 3 minutes in Los Angeles County.
The states with the longest wait times in 2012 tended also to have the longest wait times in 2008. This suggests that the long lines observed in some places in 2012 were due to chronic situations that draw their source from state laws and practices, rather than representing one-off problems that are fundamentally local in origin.
Finally, as in 2008, the factors associated with waiting in longer lines are easy to identify: early voters wait longer than Election Day voters, city dwellers wait longer than all others, and African Americans and Hispanics wait longer than white voters.
Partisan divisions in election administration
Attitudes about election-related policy issues have always elicited slightly different responses from adherents of the two parties, but the politics of election administration over the past four years have pushed these attitudes even further apart, compared to where they were in 2008.
In the most general assessment of the quality of election administration, Democratic identifiers expressed much greater confidence their votes were counted as cast (78 percent said they were "very confident") than Republicans (53 percent). This 25-point gap between the parties contrasts with the mere 7-point gap (76 percent vs. 69 percent) in 2008.
The gap in confidence extends up through levels of government, when we ask about attitudes concerning vote counting in the respondent's county, state, and nationwide. Asked about vote counting nationwide, 37 percent of Democrats and 13 percent of Republicans stated they were "very confident" votes were counted as cast.
Democrats and Republicans also differ significantly in the degree to which they believe certain problems with fraud are common in their city or county. For instance, 35 percent of Republicans, but only 9 percent of Democrats, say it is "very common" for non-citizens to vote in their city or county; 9 percent of Democrats and 17 percent of Republicans say that ballot tampering is very common.
Similarly, there are significant gaps in support for proposed election reforms. The partisan gap in support for photo ID has already been mentioned, but it also extends to other reforms that garner majority support among all voters, such as automatically allowing registration to follow a voter when one moves (supported by 87 percent of Democrats and 62 percent of Republicans), automatically registering 18-year-olds to vote (77 percent vs. 38 percent), and making Election Day a holiday (72 percent vs. 44 percent).
The only popular reforms that don't show partisan divisions are requiring electronic machines to have a paper backup (84 percent vs. 82 percent) and non-partisan election officials (64 percent vs. 64 percent).
For the typical voter, the direct experience in 2012 was quite positive. The most visible operational problem, long lines, was focused on a few states that had problems in 2008.
Looking ahead, a few election reforms, such as portable registration for the residentially mobile, are embraced by a wide variety of voters from both parties, and could be the focus of bipartisan agreement in most states that wanted to work on such issues.
On most other issues facing election administration, partisan opinions are moving apart. This will make addressing many election-related issues a serious challenge for those who wish to work in a bipartisan fashion to address the problems that continue to afflict America's voters.
NOTE: The blog will be taking a short holiday break; posts will be infrequent until the blog formally returns on Wednesday, January 2, 2013. Happy New Year!