List Maintenance Legibility
What Indiana Voters Know About How Voter Registration is Maintained in the Hoosier State
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 Mara Suttmann-Lea, an associate professor in the Department of Government and International Relations at Connecticut College, in collaboration with Ball State University’s Voting System Technical Oversight Program (VSTOP), 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—the practice of maintaining and updating voter registration rolls by election offices—has become a contentious process in American election administration. It is technically complicated, and therefore ripe for public misunderstanding. This pattern is only amplified by partisan actors and advocates who watch the process closely for evidence of mistakes or wrongdoing. A routine, systematized requirement of election offices all over the country has become an area of public mistrust and confusion. While public trust in election processes matters in a general sense for the health and quality of a democracy, there are also practical implications of public perceptions of list maintenance for the actual work of election officials. When the public doesn’t understand what list maintenance is for—and the role they themselves can play in ensuring voter rolls are accurate—this can place additional burdens on an already taxed election office workforce
Despite the significance of this procedure to the election process and the importance of public understanding, there is surprisingly little research on what the public actually knows about the process, and whether that knowledge might be bolstered with communication from election officials about the work they are already doing. To address this, I worked with Ball State University’s Voting System Technical Oversight Program (VSTOP) and six local election officials from the state of Indiana to develop a survey to assess Indiana residents’ knowledge of their state’s list maintenance practices.
What I heard from election officials
Before fielding the survey, I spoke with election officials from a range of urban and rural Indiana counties. I asked them questions about the challenges they face reaching voters generally, and what they think the public misunderstands about list maintenance and registration specifically.
A persistent theme was their concerns about a depleted information ecosystem; some described the impact of losing their local radio station or newspaper, others shared stories about holding election information events to receive few, if any attendees, including from members of the media who are invited to cover events.
Specific to this survey, officials also consistently highlighted voter confusion around the list maintenance process. One observed the incorrect belief among some constituents that “once you’re registered, you’re done”; their voters don’t understand that their registration doesn’t follow them when they move. Others said they fear official mail is treated as junk, or as a threat. “People are so wary about anything that they receive in the mail,” one official said. “I don’t even know honestly how to stop that.” They noted that the same voters can receive the same card every cycle and remain confused by it. Officials also reported constituents will sometimes read a voter registration confirmation notice—a postcard sent to voters who officials suspect may have moved—as a cancellation notice. They reported constituents calling their office to ask “why are they cancelling me?”, not understanding that the confirmation notice is meant to verify their registration and preserve their access to voting.
How the survey worked
These insights from officials were used to develop and refine the survey itself. Between August 25 and September 2, 2026, 1153 Indiana adults were surveyed online through a panel from the survey firm Forthright. Partway through, each respondent was shown one of four short messages about the voter registration confirmation notice, a consistent area of voter confusion highlighted by the election officials I spoke to. Each respondent was assigned at random one of three explaining the notice process using different frames, or one unrelated message about how candidates get on the ballot in Indiana, which serves as the comparison baseline. Because the assignment to each message was random, differences between the comparison and the treatment that show up afterward can be attributed to the message itself.
How legible is list maintenance in Indiana?
Before seeing any message, respondents answered a set of questions that gauged their knowledge of a range of election procedures in Indiana. A small majority—53.2%—correctly identified the purpose of list maintenance: keeping registration records accurate and up to date. However, knowledge fell off sharply once the questions turned to the procedures and laws that can actually affect a voter. Only 22% knew what happens when a confirmation postcard is returned as undeliverable, 34% knew that Indiana has a registration deadline before Election Day, and only 12.6% knew that Indiana does not require documentary proof of citizenship to register, while 71.6% believed that it does. That last figure almost certainly reflects ongoing disputes over citizenship documentation requirements in Indiana. Generally speaking, Indiana voters are not required to show proof of citizenship when they register, but two recent policy changes complicate that reality. Certain registered voters can now be flagged for citizenship checks, and select others may be requested to show proof of citizenship if they don’t have a driver’s license or social security number when they register.
Three ways of explaining the same process
Once respondents had answered an initial set of questions, they read their randomly assigned treatment message. Each of the treatment messages described the same voter registration confirmation notice process, but framed it differently. A procedural message simply walked through each step and the reasoning behind them. An access message emphasized that the process protects voters, and that a voter is contacted and given time to respond before anything changes. A stewardship message emphasized that accurate rolls make elections run better for everyone, and that maintaining them is something voters and election offices can do together.
The messages are short, plain, and unremarkable, but had a discernable effect. Among respondents who read the unrelated comparison message, 30.5% could correctly say what an inactive registration status means. That number rose to 63% after the procedural message, 55.2% after the access message, and to 52.1% after the stewardship message.
Figure 1. One paragraph roughly doubled the share of respondents who could say what inactive registration status means
Figure 1. One paragraph roughly doubled the share of respondents who could say what inactive registration status means
The gains were similarly large on what may be the single most consequential fact about inactive voter status: that an inactive voter is still eligible to vote. Only 17.3% of the comparison group knew this. After the procedural or access message, about half did, while around 40% did after the stewardship message.
A complication in need of further investigation
The messages also produced an unexpected result, and complicate how we think about what short explainers can do for multi-step processes like list maintenance.
After receiving the messages, respondents were shown a list of nine situations and asked which ones can move a voter to inactive status. They could select as many as they wanted. Respondents who had been assigned any of the treatment messages were substantially more likely to select each of the three situations the treatment messages had described as a reason for being moved to inactive status—returned undeliverable mail, not responding to a confirmation mailing, and not having voted in two election cycles. For the six situations the messages never mentioned, there was no discernable uptick in the respondents selecting these options.
Figure 2. The three items the messages described all rose. The six they never mentioned did not.
Figure 2. The three items the messages described all rose. The six they never mentioned did not.
This suggests the messages handed respondents a vocabulary for list maintenance, without clarifying the sequence. In other words, readers came away holding the right ingredients, but weren’t able to say which procedures produced the outcome of inactive status.
A further wrinkle is how the responses voters were handed were treated. Those who selected "not voting in the past two election cycles” were treated as an incorrect answer, on the reasoning that non-voting alone cannot move a voter to inactive status. Technically only an unanswered notice does that. But under Indiana law as amended in 2025, not voting in the two most recent general elections is itself a trigger that requires the county to begin the confirmation process (IC 3-7-38.2-2).
Respondents who selected non-voting as a reason for being moved to inactive status were pointing at something real even if they may have not understood it. The distinction is precisely the one the messages failed to convey. Non-voting is one of several things, presently, that can initiate the confirmation process. It is not, on its own, what completes it—only not responding to a confirmation notice moves a voter to inactive status, and federal law is explicit that no one may be removed from the rolls for not voting alone.
What this means
These results reflect how the complexities of list maintenance are understandably lost on the average voter, but also show how a short, plain explanation-- the length of something that would fit on a mailer, or be adapted into a social media post—can move knowledge of a central list maintenance concept in meaningful ways. The determination of inactive status is not a small thing to teach, and one paragraph roughly doubled the share of people who understood it, and understood that voters with an inactive status are still able to vote regularly.
As a part of this research, we were also interested in understanding the effects of these messages on public attitudes towards list maintenance and election offices, and on this score, the outcomes were decidedly muted. While all three messages produced a small but consistent increase in confidence that Indiana’s rolls are accurately maintained, only the stewardship message bolstered respondents’ sense of connection with their local election office. Presumably this is because the message explicitly described list maintenance as a cooperative effort between offices and voters. Trust that county officials conducted list maintenance fairly did not move at all.
This finding sits alongside what officials told me about their perspective on how trust actually gets built between an election office and their constituents. Every official I spoke with spoke of a similar approach: bring people in. Recruit them as poll workers. Invite critics to attend public tests. Hold feedback sessions after the election. One official described the people who attend these kinds of events or who volunteer as poll workers as force multipliers who carry accurate information back into the community. This survey suggests that while a mailer might be sufficient for moving information, translating knowledge to trust requires community engagement. These observations reflect other research documenting the effects of directly observing and experiencing election processes on public confidence and trust in elections. The list maintenance process—technical and not easily lent to public observation—may be more difficult for the public to connect with and understand through seeing.
The problem of reach deserves mention. While over 50% respondents say they would turn to an election office for clarification if they encountered confusing information, around 90% reported not actually seeing anything from their state or local election office over the last two months. A message can be well written, but have no impact if it is not received. This finding speaks to the depleted information ecosystem about which the election officials I spoke with expressed concern.
Finally, these respondents were surveyed just before the start of the general election season. The same respondents will be surveyed again closer to Indiana’s voter registration deadline, a period during which the information environment around them will shift as the general election moves into full swing. This may lend itself to shifts in Indiana residents’ exposure to information from election offices, and knowledge about these practices, a pattern we will be keeping a close eye on.
