In his latest step to strike back against media outlets he says have said wronged him, President-elect Donald Trump filed a lawsuit against an Iowa pollster and newspaper that he said intentionally skewed a poll against him to try to help Vice President Kamala Harris in the November election.
A legal expert gave the litigation little chance of succeeding, and press freedom advocates protested it as a another retaliatory measure designed to chill media outlets from making fair assessments of the incoming president — particularly after other Trump lawsuits against media institutions including CBS News, ABC News and the board that oversees the Pulitzer Prizes.
ABC News agreed last week to pay $15 million toward Trump’s presidential library to settle a lawsuit over anchor George Stephanopoulos’ inaccurate on-air assertion that the president-elect had been found civilly liable for raping writer E. Jean Carroll.
Some of Trump’s supporters cheered the Iowa lawsuit and supported its contention that pollster J. Ann Selzer and the Des Moines Register intended to influence the election outcome, though they presented no evidence of malfeasance or collusion intended to favor the Democratic presidential candidate.
Trump’s lawyers filed the action Monday in a state court in Polk County, home to the state capital of Des Moines and the Register, Iowa’s leading newspaper. The president-elect had signaled the action in a news conference earlier in the day, saying: “In my opinion, it was fraud and it was election interference.”
“Defendants and their cohorts in the Democrat Party hoped that the Harris Poll would create a false narrative of inevitability for Harris in the final week of the 2024 Presidential Election,” Trump’s lawsuit contends.
The suit names as defendants Selzer and her polling company; the Des Moines Register; and Gannett, one of America’s largest newspaper chains and owner of USA Today. While Trump has accused other media outlets of defamation, the Iowa action alleges violations of the state’s Consumer Fraud Act, which prohibits deception when advertising or selling merchandise.
Selzer did not immediately respond to the lawsuit. But in post-election interviews she and election analysts familiar with her work rejected Republican conspiracy theories.
Gannett released a statement acknowledging that the final Iowa poll did not reflect the ultimate outcome. It showed Harris leading by 3 percentage points, while Trump won Iowa by more than 13 points. The statement described the voluminous background data the pollster released about how the survey was conducted.
“We stand by our reporting on the matter and believe this lawsuit is without merit,” the Gannett statement said.
Selzer has been one of Iowa’s most trusted pollsters for decades. Her surveys have been watched nationally by journalists and politicos of both parties due to their accuracy, particularly leading up to the primary-season caucuses, which are viewed as crucial in presidential politics.
In 2020, Selzer’s last preelection poll for the Register showed then-President Trump leading former Vice President Joe Biden by 7 percentage points. That proved to be very close to the mark: Trump won the state by a little more than 8 percentage points.
Leading into last month’s vote, the Selzer & Co. poll for the Register came up with a result that the veteran pollster and other observers acknowledged surprised them. It showed Harris held a 47% to 44% advantage over Trump among likely voters, in a state that the Republican had carried handily in the two previous elections.
Publication of the poll in the week before the election heartened Democrats, as a potential sign that Harris had momentum, not just in Iowa, but potentially in other crucial Midwestern battleground states. Republicans expressed doubt about the poll’s accuracy.
Pollsters routinely caution the public that their surveys are only snapshots in time and are not necessarily good at predicting election results. They also urge voters not to use poll results in a single state to extrapolate to other states, which inevitably have different electoral dynamics and demographic makeups.
Still, the Selzer poll’s more than 16-percentage-point miss was wide enough that the pollster conceded she had been troubled by it and had racked her brain for explanations. In recent interviews, Selzer expressed dismay and continuing puzzlement.
In one interview, she described how she and her team had closely scrutinized the poll results, finding no warning signs of a misstep. In fact, a couple of the internal indicators of the sample’s composition seemed to favor Trump, since it included more rural voters and fewer young people than were expected to cast ballots.
The pollster said it was possible that she had used too stringent a “screen” to weed out voters whom she deemed unlikely to vote. But she said critics had tarred her for something far more nefarious.
“They’re saying that this was election interference, which is a crime,” she said last week during a panel discussion. “So, the idea that I intentionally set up to deliver this response, when I’ve never done that before — I’ve had plenty of opportunities to do it — it’s not my ethic.
“But to suggest without a single shred of evidence that I was in cahoots with somebody, I was being paid by somebody — it’s all just kind of … hard to pay too much attention to it, except that they are accusing me of a crime.”
On one thing both sides in the furor agree — the Iowa poll results got major coverage in the media. Selzer speculated in last week’s interview that the finding that Trump was trailing Harris might have actually spurred more of his voters to the polls in Iowa.
“There might have been something that happened between when we finished polling on Thursday night and election day,” she said. “Contrary to what was being charged, it could be that the release of that [poll] got the Republicans more hustled up, and bringing more people to the polls … had the impact of actually inflating the Trump vote.”
But she acknowledged that there was no proof of that speculation. “I don’t have the data for that,” she said.
Selzer also noted previous occasions when her polls mirrored the election victory, and one when it did not.
In 1988, her poll showed a large lead for Democrat Michael Dukakis over Vice President George H.W. Bush, a finding that also went against conventional wisdom. Journalists at the Register even discussed whether they should publish the finding. They did, and Dukakis went on to win the state by more than 10 percentage points, much as Selzer’s poll forecast.
In 2004, by contrast, Selzer’s poll showed Democratic Sen. John F. Kerry (D-Mass.) with a narrow lead over President George W. Bush ahead of the election. But Bush eked out a victory of under 1 percentage point. Selzer recalled that Iowa’s GOP governor, Terry Branstad, had later told her that the poll helped prod more Republicans to vote.
In his Monday news conference in Florida, Trump acknowledged Selzer’s positive reputation. His implication seemed to be that such an adept pollster could not have been so off the mark unless she intended to favor his opponent.
“You know, she’s got me right, always. She’s a very good pollster,” Trump said. “She knows what she was doing.”
Election law expert Rick Hasen wrote on his blog: “I don’t expect this lawsuit to go anywhere.”
In an interview, Hasen noted that defamation cases surrounding public figures require plaintiffs to demonstrate “actual malice.” He said he expected that standard would be applied in Iowa, even if state law does not specifically enumerate it.
“This is 1st Amendment activity, a speech activity, and therefore she’s protected,” he said. “She and the publishers of the poll are protected.”
Hasen predicted Trump’s lawyers would face other hurdles.
“It doesn’t appear that there was any false statement. And there’s no evidence that the pollster deliberately manipulated the results,” he said. “Also, it’s not clear that this [state] statute applies to something like a poll, as opposed to a consumer product, or typical protecting consumers from bad products or lies about products.”
Much of the reaction to the lawsuit fell along predictably partisan lines.
“She wasn’t part of any conspiracy, there was no conspiracy. She was just wrong,” progressive commentator Cenk Uygur wrote on X. “So, can Hillary [Clinton] sue all the pollsters who said she was going to beat Trump? Most importantly, can all the slimy politicians sue everyone now for criticizing them or even doing a poll that shows them losing?”
One Trump loyalist, former national security advisor Michael Flynn, posted a compilation of a series of preelection polls that mostly showed Harris with a narrow lead over Trump.
“Powerful data,” Flynn wrote. “Clearly shows how organized polling is meant to influence and not inform.”
“I hope it doesn’t have a chilling effect on news gathering, but it might,” said Barbara Kingsley-Wilson, a lecturer and media advisor with the Department of Journalism and Public Relations at Cal State Long Beach. “It’s a financially difficult time for journalism organizations in general, and the well-heeled forces that try to bully and intimidate know that.”
She said she would tell student journalists to “be thorough, be fair and don’t be intimidated by threats of meritless lawsuits.”