Active, Inactive, and Everything in Between
Active Voter Registration Rates and Policies for Making Voter Registration Records Inactive
The MIT Election Data and Science Lab helps highlight new research and interesting ideas in election science, including through research grants under our ongoing Practice-Oriented Research program.
Our post today comes from Christopher Mann, April Tan, and Kira Flemke, from the Center for Election Innovation and Research, based on their ongoing research funded by this program. The information and opinions expressed in this column represent their own research, and do not necessarily represent the opinions of the MIT Election Lab or MIT.
List maintenance for voter registration rolls has recently been under intense scrutiny. List maintenance practices vary widely between states, making comparison difficult and risking confusion or misinformation. Responding to this scrutiny requires robust understanding of list maintenance procedures across states.
The Center for Election Innovation & Research (CEIR) is conducting research to provide a foundation for discussing voter list maintenance procedures across states, with a specific focus on record status change from “active” to “inactive”. This ongoing research will culminate in a report on all states’ processes for registration record inactivation and the relationship between inactivation processes and state active registration rates.
CEIR began by capturing a key outcome measure of voter list maintenance, active voter registration rates, directly from official state sources. Historically, the Election Administration and Voting Survey (EAVS) has been the most widely used national data source for state registration. However, EAVS data quality has been limited by: (1) incomplete responses from jurisdictions within states which result in incomplete state totals; (2) variation in question interpretation which results in values that are not comparable between states; and (3) a lack of response dates which make it impossible to know the exact timeframe represented by the registration totals and further undermines comparisons between states.
CEIR has addressed these limitations by creating a dataset that relies on direct state sources for registration totals proximate to the 2024 general election. These data on active voter registration rates for the 2024 election represent a more complete, time-stamped, and verified list of states’ active registration rates.
To capture active voter registration rates in each state around the 2024 general election, the CEIR research team collected the number of active registered voters from official public sources in each state dated as close as possible to the November 2024 general election. If a source could not be identified or the data was unclear (e.g. active records only vs active and inactive records), CEIR reached out to state election offices to confirm their values. Outreach was necessary in nearly half of all states to ensure accurate data on active registration rates.
Each state’s active registration rate is calculated as the percentage of active voters out of the Citizen Voting Age Population (from the United States Census Bureau) as an estimate of the eligible population. These values provide a simple estimate of how many eligible voters have registration records with up-to-date information and serve as an approximate measure of a voter list’s accuracy and completeness.
CEIR’s analysis found that the national average active registration rate was approximately 85% at the 2024 general election. Within this dataset, each state’s active registration rate reflects differences in population mobility, demographic patterns, voter registration policies, and – importantly for CEIR’s research project – list maintenance practices. Active registration rates in 25 states fall within five points of the 85% average, and 39 states and DC fall within 10 points of the average. Together, this suggests that, in most states, list maintenance practices tend to keep pace with natural changes in voters’ lives.
This dataset provides researchers with a way to begin assessing voter list completeness and comparing list maintenance across states. CEIR plans to continue this line of work by repeating this data collection process around the 2026 general election.
In order to contextualize state registration rates, CEIR is collecting and analyzing detailed information on list maintenance policy and practice from all 50 states and DC.
So far, CEIR researchers have conducted a 50 state + DC review of statutes, regulations, and other publicly available guidance to identify and categorize the conditions and processes that trigger the process to inactivate voter records. CEIR researchers are now working on rigorous quality control, interpretation, and categorization of these state policies. Currently, CEIR researchers are in the process of drafting summaries of the inactivation processes in each state to inform discussions with states and the final report.
