U.S. Election Administration Remained Strong in 2024

About the EPI 

The Elections Performance Index was launched in 2013 by the Pew Charitable Trusts as a project exploring ways to provide nonpartisan, objective measures of election administration in the United States. When it was first established, the project convened a working group of academics and election officials from 14 states to identify key metrics used to evaluate election administration in the United States. In collaboration with Pew, this working group identified 40 potential indicators as possible ways to measure the effectiveness of United States elections. (More information about these initial indicators was summarized in a report authored in 2012, Election Administration by the Numbers, which provided a general framework for evaluating election administration in the United States.)

The advisory group put these indicator candidates through strong validity and reliability testing, eventually narrowing the possible metrics for the EPI down to 17. To be selected, each of the final indicators had to meet a few types of strict quality standards and criteria. First, every indicator needed to be grounded in high quality data that the advisory group could be reasonably certain would be available and consistent over time as well as across all states. Second, the indicators were also meant to reflect a comprehensive understanding of election policy and administration. 

To organize their approach on that front, the advisory group developed a rubric that allowed them to group candidate indicators by three key areas of election administration: registration, voting, and counting. The rubric also allowed the group to sort possible indicators by whether they measured convenience for voters or security of the process. Each of these factors was important to include, and the original 17 indicators were meant to provide a look across each of them. 

Pew passed management of the index over to the MIT Election Lab in 2017, and we have run it following every federal election since. In 2020, we added two more indicators, bringing the total number of indicators in the EPI up to 19. We still refer back to the original rubric developed by Pew’s advisory committee, but we also organize the 19 indicators in the Election Performance Index into four broad categories overlapping with different areas of election administration: electoral environment, election infrastructure and general administration, mail voting, and in-person voting. 

These categories can be a useful framework for exploring different aspects of the EPI. You can also use the website to filter by the indicator(s) you are most interested in seeing. Examining the individual indicators on a national level can illuminate the ways election management can be impacted by both policies and outcomes in a set period; when averaged together for each state, they provide insight into how election administration has fared in that particular state over time. (More information on the indicators and how they work to measure election administration outcomes to come in future blog posts!)

Why the EPI Matters

The EPI gives us a consistent record of measurable, historical data from both midterm and presidential elections. The greatest advantage of this is allowing us to clearly see how election management and administration has changed in the United States since 2008—the first year captured by the index. 

Since 2008, election officials have dealt with unprecedented challenges like the COVID-19 pandemic, as well as increasing public scrutiny and significant changes in election policy and funding. But by focusing on the outcome of specific factors in election administration, the EPI shows us that elections in the United States are well-run, and continue to get better every year. 

Since 2008 election administration in the United States has improved across the board. 

The indicators we measure in the index do not only paint a picture of each election year; they also work together to illustrate how election management in the U.S. has changed over time.  Access to online tools has largely improved since 2008, making it easier for voters to make an informed decision at the ballot box. And, safeguards like post-election audits are becoming standard across the nation, helping ensure the accuracy of elections even after Election Day has ended. More recently, we saw very little dramatic change between 2020 and 2024, there wasn’t a single indicator that changed dramatically in any state in any direction, leveled out after a period of rapid improvement. 

Being able to see this administrative consistency is important, showing that election management continues to improve, even as the broader environment that election officials navigate has changed.

What’s Next for the EPI 

Every two years, the EPI lets us step back and take stock of election administration retrospectively. It measures finite and trackable outcomes across the full timeline of election activities—from initial voter registration through the final audit after Election Day—with attention to both security and convenience. It’s meant to provide a look at the full picture of election administration in the U.S., beyond specific policy changes, but we know there is always room for improvement. Beyond the two new indicators we added in 2020, are there more potential measures that we now have reliable data on, which we didn’t a decade ago? Should certain metrics be retired or revised in order to reflect now-commonplace improvements made to American election systems? What would those changes look like? What improvements could we make to provide an even more accurate measure of election administration?

None of these are easy or quick questions to answer; here at the MIT Election Lab, we will continue to do some deep, careful thinking about them. In the meantime, though, we look forward to bringing you some more detailed looks at the 2024 index and indicators! We’ll be publishing those right here on our blog; stay tuned, or explore the 2024 index yourself through the map or by visiting our methodology.