DES MOINES — The claims of electoral fraud were false, proved untrue by public data and the state’s top election official.

That didn’t stop them from going viral, as right-wing activists took to Twitter over the weekend to spread specious allegations of malfeasance on the eve of Iowa’s first-in-the-nation caucuses.

The episode showcased the perils of conducting elections in the age of social media, where volume is more important than veracity.

The Iowa Democratic Party, in partnership with national Democratic officials, has labored to make the caucuses more transparent and to fend off the sort of confusion and conspiracy theories that marred the process in 2016. The Democratic National Committee has its own unit tracking viral disinformation and flagging falsehoods to campaigns, as well as to technology companies that have pledged to clean up their platforms after they were enlisted by Russian actors four years ago to boost Donald Trump in his campaign against Hillary Clinton.

But their efforts falter when users with massive online audiences push falsehoods, which social media platforms often refuse to remove, arguing that they should not serve as the Web’s arbiters of truth. On Monday, Twitter affirmed its mostly hands-off approach, maintaining that the false claims about Iowa’s voter rolls did not qualify as a form of voter suppression.

“The tweet you referenced is not in violation of our election integrity policy as it does not suppress voter turnout or mislead people about when, where, or how to vote,” said spokeswoman Katie Rosborough.

The claims came from a pair conservative activists.

Tom Fitton, president of the conservative activist group Judicial Watch, wrote Sunday morning that “eight Iowa counties have more voter registrations than citizens old enough to register.”

That notion, based on a Judicial Watch report purporting to find similar irregularities in hundreds of counties across the country, is false, according to state officials and a Washington Post review of the most up-to-date data.

Of the eight Iowa counties listed by Judicial Watch, a single one — Lyon County — has more registered voters (8,490) than adult residents (8,430), based on five-year estimates released by the Census Bureau in 2018. The estimates, however, do not account for population growth over the past two years. And the total number of registered voters includes both active and inactive voters.

“Their data is flawed, and it’s unfortunate that they’ve chosen caucus day to put out this deeply flawed data,” said Kevin Hall, a spokesman for the Iowa secretary of state.

Flaws in the data did not stop other conservative activists from pushing the misleading conclusion. Charlie Kirk, the founder of Turning Point USA, a group mobilizing young conservatives, followed up Sunday afternoon to proclaim, “One day before the Iowa Caucus, it’s been revealed that EIGHT Iowa counties have more adults registered to vote than voting-aged adults living there.” He asked users to retweet to show their support for a national voter-identification law.

And retweet they did. By Monday, the two tweets together had more than 100,000 interactions, meaning retweets, likes and replies. Among the users amplifying the falsehood were Kelli Ward, the chairwoman of the Arizona Republican Party, and Mimi Walters, a Republican former congresswoman from California. Analysis by VineSight, a group tracking online falsehoods, said that some of the amplification came from accounts exhibiting signs of automation and that few of the users appeared to be from Iowa.

Presented with figures that contradicted his findings, Fitton stood his ground.

“It’s all very interesting and curious, but the fact is our data shows eight counties over 100 percent,” he said in an interview.

A clearer picture will emerge once new analysis is available from the Election Assistance Commission, he added. “Things may have changed, and certainly things will get better or worse over the next two years, but that’s the data we’re relying on,” he said. He called his organization’s efforts to raise alarm about voter rolls a “public service.”

Turning Point USA declined to provide an on-the-record statement.

Early Monday, Iowa’s secretary of state, Republican Paul Pate, weighed in to debunk the allegation.

“False claim,” he wrote. “Here is a link to the actual county-by-county voter registration totals. They are updated monthly and available online for everyone to see.”

He included a link to his office’s website, as well as the hashtag #FakeNews.

Pate’s post gained virtually no amplification.

“The truth actually gets retweeted almost never, and the things that are the most inflammatory get the most play,” said Ann Ravel, the director of the Digital Deception project at MapLight, which tracks money in politics. She previously served on the Federal Election Commission.

Ravel accused tech companies of failing to grapple with what she says is a form of voter suppression. She said such tweets have the effect of casting doubt on the legitimacy of the political process.

“People do not have trust in institutions anymore,” she said. “This augments that.”

On Facebook, a similar post from Judicial Watch was referred to the company’s third-party fact-checkers for review, after it had been shared roughly 11,000 times on the site. By late Monday, one fact-checking organization determined it was “false information getting spread around by conservative activists,” and Facebook then began labeling it as such for its viewers.

Top tech companies have maintained that they are not “arbiters of truth,” in the words of Facebook chief executive Mark Zuckerberg, adopting a hands-off approach to most speech, even outright lies. But the companies have sought to stake out a more aggressive approach to content considered to be voter suppression. Generally, Facebook, Google and Twitter prohibit users from misrepresenting how, when and where to vote, or from sharing posts, photos and videos designed to discourage people from turning out on Election Day.

Under Twitter’s rules, updated in April, the company also bans tweets that include “misleading claims about voting procedures or techniques which could dissuade voters from participating in an election.” Those who run afoul of its standards are locked out of their accounts until they delete their tweets and risk permanent suspension if they do not change their behavior.