How allegations of electoral fraud have been charged by unhealthy science


Ansolabehere's conclusion was a milestone, but it was based on something not every pollster does: money. For his research, he signed a contract with Catalist, a provider that buys voter registration data from states, purges it, and sells it to the Democratic Party and progressive groups. Using a proprietary algorithm and data from the CCES, the company validated every self-reported voting claim by comparing individual survey responses with the respondents' voting protocol, their party registration and the method they used to vote. That kind of overhead is not only expensive (the Election Project, a polling information resource run by a professor of political science at the University of Florida, says the cost is about $ 130,000), but also puzzling: Third-party companies can set the terms they want to set, including confidentiality agreements that keep the information private.

In response to criticism of his work, Richman admitted that his numbers may not be accurate. The estimate of 2.8 million non-citizens "is almost certainly too high by itself," he wrote. "There is a 97.5% chance the true value will be lower."

Despite this admission, Richman continued to promote the claims.

In March 2018, he was in a courtroom and testified that non-citizens vote en masse.

Kansas Secretary of State Kris Kobach defended a law requiring voters to prove their citizenship before registering to vote. Such voter ID laws are seen by many as a means of suppressing legitimate votes because many eligible voters – in this case up to 35,000 Kansans – lack the required documents. To reinforce the argument and prove that there is a real danger of non-citizen voting, Kobach's team hired Richman as an expert.

Kris Kobach.


Richman paid a total of $ 40,663.35 for his entry and used various sources to predict the number of non-citizens registered in the state for the vote. An estimate based on data from a county in Kansas that was later found to be inaccurate put the number at 433. Another, extrapolated from CCES data, found 33,104. At the time, Kansas had an estimated 115,000 adult residents who were non-American citizens – including green card holders and visa holders. According to Richman's calculations, that would mean nearly 30% of them were illegally registered to vote. Overall, his estimates were between 11,000 and 62,000. "We are 95% convinced that the real value is somewhere in this range," he said.

The judge eventually ruled that the voter identification laws were unconstitutional. "All four estimates (by Richman), individually or as a whole, are in error," she wrote in her opinion.

Invisible effects

A consequence of this unreliable data – from citizens lying about their voting results to those who mistakenly identify themselves as non-citizens – is that it further distracts the attention and resources of voters outside of traditional constituencies.

"It's a vicious circle for the (little-inclined) crowd," wrote Matt Braynard in his internal memo for the Trump campaign. "They don't get voter love from the campaigns because they don't vote, but they don't vote because they don't get voter contact." It is a persistent state of disenfranchisement. "

Campaigns focus on voters who are likely to vote and likely to give money, says Allie Swatek, director of policy and research for the New York City Campaign Finance Board. She saw this trend firsthand when she returned to New York in time for the 2018 election. Although there have been races for the US Senate, governor, and state congress, “I haven't received anything in the mail,” she says. “And I was like, 'Is it like this if you don't have a voting history? Nobody reach you "

According to the Knight Foundation's poll of non-voters, 39% said they were never asked to vote – either by family members, friends, teachers, political campaigns or community organizations, or in places of work or worship. However, this may change.

Stacy Abrams is running for GA governorStacey Abrams' campaign for the governor of Georgia targeted voters with "low inclinations".


Braynard's mobilization strategy played a role in the 2018 campaign for the governor of Georgia by Democrat Stacey Abrams. She specifically targeted voters with low inclination, especially color voters, and although she ultimately lost that race, more black and Asian voters turned out that year than for the 2016 presidential race. “Any political scientist will tell you that doesn't happened, "wrote Abrams & # 39; former campaign manager in a statement for the New York Times. "Ever."

But even as campaigns and experts try to break these cycles – by cleaning up their data or targeting non-voters – there is a much more dangerous problem at the center of electoral research: it is still vulnerable to those who work with bad faith.

Track claims

I asked Richman earlier that summer if we should trust the sweeping numbers he put in his study or on his testimonial in Kansas. No, he didn't necessarily answer. "One challenge is that people want to know the level of registration and voting of non-citizens with a level of certainty that the data available does not provide," he wrote to me in an email.

In fact, Richman said, he "ultimately agreed" with the judge on the Kansas case, even though she said his evidence was flawed. "On the one hand, I think that voting takes place without citizens and that public order reactions need to be aware of that," he told me. "On the other hand, this does not mean that every public policy response is an appropriate compromise between the different types of risk."

Behind the academic language, it is essentially saying what every other expert on the subject has already said: fraud is possible. So how can we balance security of choice with accessibility? Unlike his peers, however, Richman reached this conclusion by first publishing a paper with alarming results, writing a newspaper article about it, and then stating that the citizen-free voting was widespread, although he later agreed to the decision that he was wrong .


Steven Gregory