Charles Stewart III, the Kenan Sahin Distinguished Professor of Political Science and head of MIT's Department of Political Science, recently helped to complete the first comprehensive nationwide study that focused exclusively on how American voters experience the administration of elections. The "2008 Survey of the Performance of American Elections," conducted for the Pew Center on the States with support from the AARP and the JEHT Foundation, found that while the vast majority of Americans said that their 2008 Election Day experience went smoothly, many U.S. voters did not cast ballots because of voter registration problems. Stewart took time to discuss some of the survey findings, which were presented in testimony before Congress on Thursday, March 26.
Q. How did this survey come about? What was MIT's role in putting it together?
A. MIT took the lead in proposing and designing the survey. It is the culmination of work that began in 2001 with the Caltech/MIT Voting Technology Project. One of the themes we have harped on since then is that the empirical basis for reforming elections in the United States is really thin. We really don't know how voters experience the election process. When the Pew Make Voting Work program came along, it seemed like the perfect opportunity for us to put into practice what we had been advocating for many years. The team we assembled, all researchers related to the Voting Technology Project, had already been doing the most important work in public opinion about how people vote, but it was mostly in bits and pieces. So, this survey allowed us to pool our collective experience and to study a population that was much larger than we had been able to survey before.
Q. Ninety-eight percent of respondents in the survey said it was "very easy" or "fairly easy" for them to find their polling place; another 98 percent said their polling place was run "very well" or "OK, with only minor problems. Does this mean we "solved" election problems that plagued the process in 2000 and 2004?
A. I would never go so far as to say we've "solved" any problems, but we've made significant strides in addressing the worst problems that were highlighted in 2000, especially the poor quality of voting machines. That problem is well under control, even though there continues to be controversy about electronic voting machines. Having reached the point where the voting machine problem is manageable, we now recognize the much bigger problems with voter registration, voter identification and long lines.
Q. What will you do to build on the results of this study?
A. There are many issues that come out of this study. The first is that we want to be able to continue doing the survey in the future, to provide an ongoing account about how well elections are being run, and to see where we're doing better or falling behind. Second, the survey highlights some problems -- especially registration and voter identification -- that are currently hot topics. Using this survey as a baseline, we'll be able to track whether state and federal efforts to address these issues are making a difference. Third, there continue to be issues of racial disparities that concern a lot of people. For instance, African-Americans had to wait twice as long in line this year to vote, compared to whites. Why is that? It can't be entirely that African-American communities were so excited about voting for Obama that they turned out in historically large numbers and overwhelmed polling sites. We know that because we are discovering evidence that this problem happened in 2006, as well. A series of challenges are currently being made to the Voting Rights Act before the Supreme Court, and the survey can be used to help provide new evidence about the degree to which African-American voters experience greater hurdles when they attempt to vote.
A version of this article appeared in MIT Tech Talk on April 1, 2009 (download PDF).
