The claim made on Fox and Friends is based on an extrapolation of a controversial study that relied on a very small number of responses. Researchers involved in the underlying survey of voters have cautioned against using their data to reach conclusions about noncitizen voters.
We emailed a spokeswoman for Fox News and did not get a reply; however, the Washington Times article showed that the information came from Just Facts, a think tank that describes itself as conservative/libertarian and was founded by James D. Agresti, a mechanical engineer in New Jersey.Agresti’s conclusions are based on data from a paper by Old Dominion University researchers who used data from the Cooperative Congressional Election Study, or CCES. He multiplied the findings in that data with U.S. Census Bureau estimates of the noncitizen population to come up with a conclusion about the number of noncitizen voters nationwide.
It’s important to note that the CCES researchers have disputed the conclusions Old Dominion researchers reached about noncitizen voters.
Here’s how the studies unfolded: In 2008, the CCES surveyed 32,800 adults nationwide online about their political views. Respondents answered at least 100 questions before they made it to the citizenship question, one of the last questions asked.
The survey showed that 339 identified themselves as noncitizens — about 1 percent of the total respondents. Then of the 339 self-identified noncitizens, 39 of those claim to have voted, said Brian Schaffner, a political science professor at the University of Massachusetts Amherst, one of the main researchers.
That’s 39 respondents out of 32,800 people who are now being used to extrapolate millions of illegal voters. Schaffner has warned that with a subset that small, the responses might be unreliable.
“Survey respondents occasionally select the wrong response by accident—perhaps because they are rushing through and not reading the questions carefully, because they do not fully understand the terminology being used, or because they simply click on the wrong box on the page,” Schaffner wrote in a Politico magazine article after the November election.
Subsequent CCES surveys provide more evidence that some respondents answered the question wrong. There were 20 respondents who identified themselves as citizens in 2010 but then in 2012 changed their answers to indicate that they were noncitizens, which Schaffer said is “highly unrealistic.”
In 2014, researchers at Old Dominion University used the CCES data in 2008 and 2010, as well as voter records in 2008, to conclude that more than 14 percent of noncitizens indicated that they were registered to vote. Their best guess at the portion of noncitizens who voted was about 6.4 percent, or 1.2 million votes cast.
The researchers at CCES (including Schaffner; Stephen Ansolabehere, a Harvard political scientist; and Samantha Luks, a statistician at YouGov) have criticized the methodology used by Old Dominion.
They said it didn’t fully consider the possibility that people responded to the survey inaccurately.
“You are ignoring the measurement error in a very small group which is going to inflate those numbers,” Schaffner said, “then you assume this is a random sample of all noncitizens in the country, which it probably isn’t.”
More than 100 political scientists from universities and colleges wrote an open letter in January disputing the Old Dominion paper as evidence for Trump’s claim that millions of noncitizens voted.
MORE from HuffPo.