Stand up for the facts!
Misinformation isn't going away just because it's a new year. Support trusted, factual information with a tax deductible contribution to PolitiFact.
I would like to contribute
If Your Time is short
A report from a conservative group tried to use a disparity between registered voters and voting-age adults in Iowa as evidence of potential voter fraud in the caucuses. But Iowa’s secretary of state said the numbers are wrong.
Comparing voter data to population estimates is not a good way to measure registration rates.
One day before the Iowa caucuses, a conservative activist started making claims on Twitter about alleged voter fraud.
"It’s been revealed that EIGHT Iowa counties have more adults registered to vote than voting-aged adults living there," tweeted Charlie Kirk, founder of Turning Point USA, Feb. 2. "Don’t let voter fraud steal the 2020 election."
The allegations from Kirk and another conservative organization aim to stir up concerns about voter fraud with a questionable methodology that leads to unsubstantiated conclusions. On Caucus Day, Iowa election officials issued public statements denouncing their numbers and the narrative.
The current data released by Iowa election officials don’t support Kirk’s tweet.
Kirk’s viral tweet makes the same claim as a Feb. 3 report published by Judicial Watch, a conservative legal group.
The Judicial Watch report says "eight Iowa counties have more voter registrations than their eligible voting-age population." Two of them, Johnson and Scott counties, are the locations of Davenport and Iowa City, respectively.
The report was based on its analysis of data from the U.S. Election Assistance Commission and the Census Bureau’s American Community Survey.
"Dirty voting rolls can mean dirty elections and Iowa needs to undertake a serious effort to address its voting rolls," said Judicial Watch President Tom Fitton in the news release.
Iowa officials refuted the report. A Feb. 3 news release from Iowa Secretary of State Paul Pate said his office’s data "shows this information to be false."
"It’s unfortunate this organization continues to put out inaccurate data regarding voter registration, and it’s especially disconcerting they chose the day of the Iowa Caucus to do this," Pate said in the statement. "My office has told the organization, and others who have made similar claims, that their data regarding Iowa is deeply flawed and their false claims erode voter confidence in elections."
We reached out to Kirk and Judicial Watch for comment, but neither has responded with one. It appears they were using old voter registration numbers..
We took a look at the most recent voter registration and population data from Iowa’s government and the Census Bureau. The numbers don’t support what Kirk said, and experts say comparing them is flawed.
Among the counties identified by Judicial Watch, none have more active registered voters than adults.
For example, Scott County — the largest county included in the Judicial Watch story — has an estimated population of 131,961 people ages 18 and older, according to 2018 data from the Census’s American Community Survey. According to the Iowa secretary of state, there are 127,737 total voter registrations, 115,393 of which are active.
Judicial Watch’s story looked at "total registrations," which includes inactive voters. In Iowa, inactive voters are those that have not responded to notices from the state asking them to verify their registration and residence.
Only one county, Lyon, has more total registered voters than adults: 8,490 to 8,430. But that data is from 2018. The Iowa secretary of state’s office says 8,117 of voters in Lyon County are active.
Either way, experts say comparing voter rolls and Census estimates isn’t an accurate measurement of registration rates.
"That’s an apples-to-oranges measure. They measure different things, at different times, with different types of levels of certainty," said Justin Levitt, a Loyola Law School professor, in an email. "And so it’s really bad science to claim that a disconnect between those two numbers means that there’s fraud."
Even if there were more registered voters than voting-age adults in some Iowa counties, experts told us it would not amount to voter fraud.
"For there to be voter fraud in connection to bloated voter rolls, you’d have to see people going to the polling places and see them voting in places of people who have moved or died," said Richard Hasen, a law professor at the University of California at Irvine. "Not only would that be a difficult thing to do … you’d have to do this on a wide enough scale to sway an election without being able to verify how the people you’ve paid to do this have voted."
Voter fraud remains rare in American elections. A database maintained by the conservative Heritage Foundation has documented 17 confirmed instances of voter fraud in Iowa dating back to 2011.
A tweet from Charlie Kirk claims that "EIGHT Iowa counties have more adults registered to vote than voting-aged adults living there."
Voter registration data and a statement from Iowa’s secretary of state, as well as population estimates from the Census Bureau, refute that claim. And experts say comparing voter data to population estimates isn’t a good measure of registration rates in the first place.
Kirk’s tweet is inaccurate. We rate it False.
Election Law Blog, "Here come the latest purge demands," Jan. 3, 2020
Email interview with Justin Levitt, a law professor at Loyola University, Feb. 3, 2020
Facebook post, Feb. 2, 2020
Heritage Foundation, Election Fraud Cases: Iowa, accessed Feb. 3, 2020
The Hill, "Iowa Secretary of State disputes viral misinformation about voter registration," Feb. 3, 2020
Interview with Richard Hasen, a law and political science professor at the University of California at Irvine, Feb. 3, 2020
Iowa Secretary of State, Voter Registration List Requests, accessed Feb. 3, 2020
Statement from Iowa Secretary of State Paul Pate, Feb. 3, 2020
State of Iowa Voter Registration Totals, Feb. 3, 2020
Tweet from Charlie Kirk, Feb. 2, 2020
U.S. Census Bureau, American Community Survey 5-Year Data: 2018, accessed Feb. 3, 2020
Read About Our Process
In a world of wild talk and fake news, help us stand up for the facts.