Evaluating Voter Fraud Research Is NOT A Popularity Contest

01/27/2017 01:50 pm ET Updated Jan 27, 2017

Scholarly research earns its credibility because it must pass through a rigorous double-blind peer review process before it is published. Reviewers with established expertise in the subject matter area of an article submitted for publication are assigned to review, comment and recommend the article or book. The reviewers do not know who the author is and the author does not know who the reviewers are. Reviewers evaluate the methodology and findings of the submitted research in order to assess its validity and contribution to our body of knowledge. This process is designed to protect against bias and to ensure a high level of rigor before accepting anything for publication. It is NOT a popularity contest.

When it comes to voting fraud, there is no research that has made it through the review process that demonstrates voter fraud on anything close to the scale Trump suggests has happened. Even the one study that suggests there may be extensive fraud has been called into question by the lead scholar who collected the data and by the author of the article using that data.

Despite the lack of evidence that voter fraud beyond a small number exists in US elections, generally, or the 2016 election, specifically, Donald Trump says it is his opinion that 3 to 5 million votes were fraudulently cast. Where did he get this opinion? From Alex Jones, who presented on his show the allegations of Gregg Phillips of VoteStand. Let’s leave aside the fact that Alex Jones is the purveyor of what has been badly labeled as “fake news” (if it is fake, it is not news… for example, Jones alleges that the shooting at Sandy Hook was fake). If we simply look at Gregg Phillips’ own statements about VoteStand’s research, we can see that it is simply too soon to give it any credence.

Phillips made his initial claim on November 10, when he made his as-yet unsubstantiated claim that 3 million votes were illegally cast in 2016. Despite the fact that he would not release his methodology or data, nor has he since, the claim spread across social media like wildfire.

Raising additional concerns are his inconsistent statements about his research. He claims to have the names of the 3 million fraudulent voters, yet he says he has not released his findings because he is rechecking his algorithms.

Rather than publishing things that might be wrong, we not only just want to do a quality check on our own algorithm, we want to do an internal audit, if you will.” – Gregg Phillips

This raises a serious red flag. Algorithms are used to make extrapolations beyond the evidence in hand. Yet Phillips claims he has the names of the fraudulent voters. If so, he would have no need for an algorithm. This suggests that he does not have real evidence of voter fraud, but just theoretical projections that some list of people might have voted illegally. These are very different beasts and his and Trump’s claims that this is evidence of 3+ million illegal votes cast would never pass a blind peer review test.

More likely, based on the way Trump and Phillips talk about voter fraud, they are conflating it with mistakes in the voter rolls. While it is true that many people who have moved are listed on more than one voting jurisdictions’ rolls, and it is true that many on the rolls are now deceased, neither of these are examples of voter fraud. Voter fraud happens ONLY when someone casts an illegal vote. From all accounts available, the 3 to 5 million number corresponds to people erroneously listed on the voter rolls. This is NOT voter fraud. Unless the research can identify that non-citizens actually casted votes or that someone cast multiple votes, there is no fraud.

Further, Phillips refuses to release his data, findings and methodology to anyone, preferring to wait and release it all at once to the public.

“I committed from the outset to publish all of this data to the public. I’m gonna let the public see everything we’ve done. Our analysis, everything, will be published. We will also give copies to the federal government. We’re gonna publish the entire data set.” – Gregg Phillips

While I am a big fan of democracy and have a lot of faith in the public’s sense of what people around them feel, the public is not qualified to evaluate research methodology and findings. Phillips’ intention to release this to the public all at once will create a national kangaroo court of untrained opinions assessing the validity of his research. Any expert assessment of his research would be drowned out by the loud, untrained public who supported Phillips and Trumps claims before they ever saw his research.

This is exactly what is meant by the phrase “mob rule.” And, it is very likely the intention of Trump and Phillips to create mob rule in order to ram through unjustified voter fraud prevention rules (regulations) that will be designed to suppress legal votes by a disproportionately Democratic voter base.

Trump and the GOP’s constant harping on unproven allegations of voter fraud have always been a red herring to distract Americans away from their very real efforts to suppress Democratic voters through legal hoops, negative advertising and voter purges. We cannot fall prey to these tactics. We must never allow anyone to whip up public discord over something that does not exist. We must protect our nation from those who seek to confuse the public into thinking that voter fraud is a problem by giving them a research results that use bad methodology to conflate real voter fraud with activity that is neither fraudulent nor illegal. We must preserve our ability to let experts evaluate research based on the requirements of good research. Anything less turns science into a popularity vote. And that is how we destroy knowledge and institutionalize ignorance. In other words, that is how we can create a new Dark Age, where science is banned and ignorance reigns supreme.


Alan Rosenblatt, Ph.D. is Director of Digital Research at Lake Research Partners and SVP of Digital Strategy at turner4D. He is also a long-time political science professor, currently serving as an adjunct professor at George Washington, Johns Hopkins and American Universities. He previously served on the faculty at George Mason University, as well as serving as an adjunct professor at Georgetown and Gonzaga Universities. He has a Ph.D. in Political Science from American University, an M.A. in Political Science from Boston College and a B.A. in Political Science and Philosophy from Tufts University.