Building an Election System that Works

A new tool gives administrators and policymakers crucial benchmarking data they need to make the voting process fair, accurate and convenient.
by | March 27, 2013

President Obama and leaders in both parties, in calling for improving American elections, point to long lines at the polls last year as a significant problem that needs to be solved. And with good reason: Longer wait times can discourage people from voting and fuel the perception that their right to vote is in jeopardy. A post-election poll by the Pew Research Center found that only 55 percent of voters who waited 30 minutes or more to cast a ballot thought that the election was managed "very well," compared with 79 percent for voters who waited less than a half-hour and 83 percent for voters who had no wait.

Long lines, however, are just the tip of iceberg; much more needs to be done. To achieve an election system that is convenient, accurate and fair, state and local leaders need data to review and track their voting processes--from registration to ballot-counting.

This kind of analysis is not easy. Our nation's locally run elections lack a common set of performance measures and a baseline from which reliable comparisons--between election cycles and across jurisdictions-can be made. Accurate data on what leads to better or worse results in any particular area are often scarce.

Now, state and local election officials--policymakers as well as administrators--can access an online tool to help them meet some of these challenges: the Elections Performance Index. This database provides a means for evaluating the management of elections within and across states and from year to year. An advisory group of election officials and leading academics, convened by the Pew Charitable Trusts and partners at the Massachusetts Institute of Technology, guided its development.

The index examines all 50 states and the District of Columbia on 17 indicators of election administration, including polling-location wait times, availability of online voting-information tools, number of rejected voter registrations, percentage of voters with registration or absentee-ballot problems, number of military and overseas ballots rejected, voter turnout, and accuracy of voting technology. While the 17 indicators do not capture every possible measurement of election performance, they do reflect the best currently available data.

The data cover the 2008 and 2010 elections (data from the 2012 elections will arrive later this year) and show that seven states--Colorado, Delaware, Michigan, Minnesota, North Dakota, Washington and Wisconsin--performed well both years. The worst-performing states during those years were Alabama, California, Mississippi, New York, Oklahoma, South Carolina and West Virginia.

The hard questions raised by this assessment are unlikely to have one-size-fits-all answers. There are many possible reasons for long wait times at the polls, voter-registration errors and rejected ballots. Election officials have the important responsibility of identifying the source of the problem and potential solutions, with the understanding that both may vary--not just from state to state but from county to county.

Many states are already taking important steps to find the source of election problems. Florida voters encountered the longest average wait time in the nation last fall. A post-election analysis by Florida's secretary of state identified several causes, including that some early-voting locations were ill-equipped to handle high voter turnout. To address this problem, the report recommends that lawmakers update existing policies to allow election supervisors to use a wider range of facilities as early-voting sites.

In Washington, D.C., where voters in 2012 had the second-longest average wait times, officials attributed much of the problem to the increase in voters casting provisional ballots--almost four times the number in 2008. Outdated precinct boundaries also led to delays and confusion during early voting as poll workers had to manage 694 different ballots, more than could be programmed on a single voting machine.

Data-driven reviews like these can help bridge the partisan and ideological divides that often stymie attempts to improve Americans' voting experience. As two Colorado county clerks--a Republican and a Democrat--observed in a recent column, "To keep legislative discussions from devolving into political rhetoric, credible facts should be brought to bear. That's a gap the Elections Performance Index could fill."

For state and local officials, measurements like those offered in the index can help states see how they compare with their peers while making voting easier, more reliable and more credible. This work is vital to achieving the highest standards of accuracy, cost-effectiveness and security in our election system and to ensuring that when Americans exercise their precious right to vote, the experience is short, easy, accurate and a moment of great pride.

Join the Discussion

After you comment, click Post. You can enter an anonymous Display Name or connect to a social profile.

More from Management Insights