The MIT Election Data and Science Lab manages the Elections Performance Index (EPI), an objective, nonpartisan assessment tool evaluating U.S. election administration. Following the recent update to the index with data from 2016, we are dedicating a series of posts to exploring the EPI’s underpinnings.
Our post today was penned by Cameron Wimpy, who is the Research Director at the MIT Election Data and Science Lab.
We do not rank the measures included on the Elections Performance Index (EPI) by importance or influence, but our indicator of data completeness provides the most complete look at data reporting and transparency among the states. How so, you ask? It all comes back to the Election Administration and Voting Survey (EAVS). This survey constitutes the most comprehensive program of data-gathering in the election administration field to date — and it makes up more than half of the indicators for the EPI.
In addition to using the EAVS to calculate many of the other EPI indicators, we also use it to asses the degree to which local jurisdictions and states make available some of the most basic data about election administration in their locale. This is how we measure data completeness in the EPI.
The data completeness indicator is made up 18 survey items from the EAVS. These include:
- New registrations received.
- New valid registrations received.
- Total registered voters.
- Provisional ballots submitted.
- Provisional ballots rejected.
- Total ballots cast in the election.
- Ballots cast in person on Election Day.
- Ballots cast in early voting centers.
- Ballots cast absentee.
- Civilian absentee ballots transmitted to voters.
- Civilian absentee ballots returned for counting.
- Civilian absentee ballots accepted for counting.
- UOCAVA ballots transmitted to voters.
- UOCAVA ballots returned for counting.
- UOCAVA ballots counted.
- Invalid or rejected registration applications.
- Absentee ballots rejected.
- UOCAVA ballots rejected.
Taken together, these indicators asses the reporting completeness of the most fundamental measures of election administration.
The composite measure indicates the degree to which the local jurisdictions in a given state reported the values for these indicators in each election. In more practical terms, taking these indicators together gives a fairly comprehensive sampling of how well states are doing in their reporting of EAVS data to the EAC. Careful readers will also notice that several of these indicators are used as stand-alone measures in our calculations of the EPI.
The figure below presents the distribution of all states across the several elections covered by the EPI. The red lines indicate the average for each year:
It is clear that data completeness has collectively improved over time. In 2008, nearly half of the states were less than 90% complete (86% was the average level of completeness). In 2016, though, only four states had not reached that mark, and the average completeness had risen to 97%. The next figure shows the average trend for data completeness across the states.
Each year there has been at least one state lagging far behind the rest. In 2008, New York did not report these data at the county level and thus the calculation was zero for that election. This has since changed; by 2016, New York had reached 93% coverage. In the intervening years, Alabama has experienced the lowest levels of completeness, primarily because the state does not report voter turnout by mode.
Moving to the other end of the spectrum, a handful of states have been 100% complete in each iteration of the EPI: Alaska, Delaware, North Carolina, and North Dakota.
In general, the trends we see in the EPI suggest that states are becoming more diligent in collecting and reporting these important data to EAVS. It is also likely that the infrastructure needed both to collect the data in the states and then accurately pass it along to the EAVS is improving. As states continue to upgrade their data reporting and storage systems, we should expect this indicator to inch closer and closer to 100% for all states.
Cameron Wimpy is the Research Director at the MIT Election Data & Science Lab.
MEDSL is dedicated to applying scientific principles to how elections are studied and administered, with the aim of improving the democratic experience for all U.S. voters. Follow updates on Twitter and sign up for our quarterly newsletter.