Voting studies misrepresented

Even after being sworn in as president, Donald Trump isn’t letting go of his unsubstantiated belief that millions of people voted illegally in November, costing him the popular vote.

Now, he’s calling for a “major investigation” into voter fraud, he wrote Wednesday in a tweet.

We have debunked Trump’s claims of massive voter fraud again and again and again over the past year. Trump has not yet produced any evidence that supports these claims — because none exists.

In the daily press briefing Tuesday, a reporter asked White House press secretary Sean Spicer how Trump came to believe that it’s possible illegal votes were to blame for his popular vote loss.

“I think there’s been studies,” Spicer responded. “There’s one that came out of Pew in 2008 that showed 14% of people who voted were noncitizens.

“There’s other studies that have been presented to him. It’s a belief he maintains.”

Spicer is conflating a couple different studies that have been erroneously used to prop up claims that noncitizens have swayed elections by voting illegally.

There is no study that shows 14% of the votes cast in 2008 were cast by noncitizens. That would have added up to more than 18 million fraudulent votes — an implausible assertion, considering the total noncitizen population was about 22.5 million in 2010.

Pew study misinterpreted

The study that “came out of Pew in 2008” actually came out in 2012, and it’s about outdated voter rolls, not fraudulent votes.

The 2012 Pew study — titled, “Inaccurate, Costly, and Inefficient: Evidence That America’s Voter Registration System Needs an Upgrade” — makes no mention of noncitizens voting or registering to vote.

Rather, it found that about 24 million, or one in every eight, voter registrations in the United States are inaccurate or no longer valid, but it did not find evidence of actual voter fraud. The study was about recordkeeping that is badly managed and in disarray.

Here’s what the former director of Pew’s election program, David Becker, said about the study in a tweet in November: “We found millions of out of date registration records due to people moving or dying, but found no evidence that voter fraud resulted.”

Old Dominion study contested

The study that shows “14% of people who voted were noncitizens” has been widely criticized for its methodology, and Spicer cites the findings incorrectly.

The study Spicer likely meant to reference was conducted by professors at Old Dominion University and published in 2014. The researchers wrote in The Washington Post that about 14% of noncitizens in their sample said they were registered to vote, and they deduced that about 6% of noncitizens actually voted in the 2008 election. (That would mean somewhere in the ballpark of 1.3 million votes, using census estimates of the noncitizen population.)

So the study doesn’t show the statistic Spicer says it does. Beyond that, many credible researchers have panned the study as methodologically unsound for using an opt-in internet poll originally designed to survey citizens and not considering possible survey response error.

In a recent blog post, one of the authors, Old Dominion professor Jesse Richman, said he stands by his study, but “our results suggest that almost all elections in the U.S. are not determined by noncitizen participation, with occasional and very rare potential exceptions.”

There might be a problem of noncitizens voting, but it’s not nearly as large as the statistic Spicer cites.

These studies do not prop up Trump’s belief that millions of illegal votes cost him the popular vote.

Hillary Clinton won almost 3 million more votes than Trump. So erasing Trump’s popularvote deficit requires the assumption that all 3 million of these votes were cast illegally, and every one of them went to Clinton and not Trump.

Further, no official sources or reputable academic sources have produced any evidence of large-scale voter fraud in the 2016 election, and numerous studies have shown that voter fraud is rare.

Our rating

Spicer said, “There’s one (study) that came out of Pew in 2008 that showed 14% of people who voted were noncitizens.”

Spicer is both conflating and misquoting two studies: A 2012 Pew report about outdated voter rolls and a 2014 Old Dominion University study that found 6% of noncitizens surveyed voted in 2008. The Old Dominion University study has been widely criticized for its methodology.

No study has found the statistic Spicer cites. If 14% of all voters in 2008 were noncitizens, that would have to mean that more than 80% of America’s noncitizen population voted.

Spicer’s claim is False.