I haven't been able to find the answer to this anywhere online. I may not know the correct terminology to do a proper search.

In the United States - since non-citizens are able to register to vote (even though it's not legal), and are able to then go cast votes, is there any mechanism to check whether a vote cast was indeed cast by a citizen? If there is such a mechanism, what is it and how is it used?

  • 4
    Just to point out how the system is supposed to work, the mechanism that is supposed to keep noncitizens from voting is that only citizens are supposed to register, and only registered voters are supposed to vote. In other words, any checking of eligibility should happen before the vote is cast, not after. – phoog Jan 28 '17 at 4:50

Yes, you could do it, if you were willing to devote some resources to it.

Elections are a state function, so what records are kept varies. But I can speak for my state (Oklahoma).

The names, addresses, party affiliation, and when they have voted for every registered voter in the state are a matter of public record (a lot of people don't realize this). You can ask the state election board for the database, and they will supply it to you for a nominal fee.

The major parties of course do that, and thus have all this information in their own databases. If you are a party operative (like a precinct chair), they will likely even send you a copy for free. That's how I got mine.

If they have some reason to suspect someone (or a lot of someones) who voted isn't legally allowed to for some reason, it would be quite easy to cross-correlate that database with other databases.*

It should also be pointed out that the penalties for lying on a voter registration form are rather severe. In my state(pdf) its a felony worth 5 years in prison and a $50,000 fine. Plus there is essentially no incentive for doing so, as casting a single vote doesn't really help the person doing it all that much. For those reasons, it would be really stupid to purposely register to vote fraudulently.

Of course some people are in fact that stupid, but all evidence is that they are quite rare. Most cases of voter fraud in fact involve faking multiple absentee ballots, and are carried out by political operatives, not random residents.

* - It is true that there's probably no nice database of the nationality of every human being on the planet, but you'd be surprised what can be found out about people by data experts. Many women have discovered marketers knew they were pregnant before they knew.

  • 1
    @Polygnome I think the reason for that is that the US don't really have a list of all citizens and where they are living. – Paŭlo Ebermann Jan 28 '17 at 11:33
  • 1
    @Polygnome In a lot of US states (again, these things are done by state) there IS some kind of ID check. Mine requires you to show either your voter ID card, or some kind of government issued photo ID. The main reason this hasn't been done are historical. Every citizen over 17 has a RIGHT to vote, but not everyone HAS such an ID, and states have a bad history of using any kind of voting restriction for partisan and/or racist purposes. – T.E.D. Jan 28 '17 at 11:58
  • 2
    ... The fact that people who today have no such ID are far more likely to be minority, and to be voters of one particular party, and that there's no real evidence of this particular kind of fraud is actually occurring, AND the fact thst its only the OTHER party that is gung-ho to impose restrictions despite all this, tells you that most likely what is going on where you hear talk about voter registration fraud is a continuation of this same sad racist history. – T.E.D. Jan 28 '17 at 12:10
  • 3
    @AdamDavis, ...I can actually speak to that directly, having previously lived in Texas while married into a lower-income-bracket family. Getting enough time off to spend hours on the bus just getting to and from the DMV was not an easily-rationalized effort for all members of the household. (I owned the car, so my presence changed the calculus a bit, but the ability to observe that change, in and of itself, was eye-opening). – Charles Duffy Jan 28 '17 at 16:19
  • 4
    @AdamDavis, ...there's a damned lot of blindness in the middle class about how the other half (or, rather, the bottom 20%) lives. I didn't observe much by way of laziness -- what I did see was folks who didn't have the money, or the social capital (prior to my presence) to make investments in their future. If you can't put up a deposit, you can't get into an apartment complex that charges market rates for rent, so you end up paying more, despite having less ability to pay than folks in market-rate low-end housing. Etc. – Charles Duffy Jan 28 '17 at 16:23

In the US, one of the simplest ways is to contrast jury summons declinations with votes. Because serving on a jury is a pain and because also it's illegal, often times illegal immigrants when called to serve will apply for jury excusals on those specific grounds. You can then contrast these names with voting registration and voting databases. Jury lists are developed from voter registration rolls. The Wall Street Journal has a good round up on some of these studies.

Little to nothing is being done to so clean up these voter registration rolls or to prosecute voting fraud based upon this approach. The Obama administration sued states to prevent this from occurring on grounds that it was too close to the election and it impacted minorities.

The New York Times also reported on this issue revealing another mechanism: usage of the Homeland Security database which contains legal noncitizens and illegal immigrants who have been detained. The Obama administration made this database off limits to Florida for this purpose.

From the Wall Street Journal article, another tool is post election surveys:

A postelection survey conducted by Americas Majority Foundation found that 2.1% of noncitizens voted in the Nov. 8 [2016] election. In the battleground states of Michigan and Ohio, 2.5% and 2.1%, respectively, of noncitizens reported voting. In 2013, pollster McLaughlin & Associates conducted an extensive survey of Hispanics on immigration issues. Its voter-profile tabulation shows that 13% of noncitizens said they were registered to vote. That matches closely the Old Dominion/George Mason study, in which 15.6% of noncitizens said they were registered.

The Wall Street Journal article concludes with some advice:

The Trump administration should direct the Department of Homeland Security to cooperate with states that want to verify the citizenship of registered voters. Since this will only flag illegal immigrants who have been detained at some point and legal noncitizens, states should pass laws, similar to the one in Kansas, that require proof of citizenship when registering to vote. The Justice Department, instead of ignoring the issue, should again start prosecuting these cases.

  • 1
    The question is about what checks are currently done, not what could be done. – indigochild Jan 27 '17 at 20:00
  • @indigochild The OP asked about mechanisms, this is one. And then how is it used. It was in use, and the mechanism is being prevented from being used, or suspended until US v Florida is decided. – user9790 Jan 27 '17 at 20:04
  • 9
    Any citations that reference "all these illegal aliens applying for jury excusals"? – user1530 Jan 27 '17 at 22:49
  • 13
    It should also be clarified what the Obama administration was stopping by linking to a more reputable news source: nytimes.com/2012/06/13/us/… Your answer implies that the administration was specifically suing to allow illegal aliens to vote. – user1530 Jan 27 '17 at 22:52
  • 5
    @KDog But you imply the lawsuit was to prevent the matching of jury summons declinations with voting history, when the article blip linked makes it clear that the lawsuit was filed to prevent the state from removing registered voters using a list that contained many actual citizens. – Zach Lipton Jan 28 '17 at 7:21

From the research that I did for Who can, in the US, challenge the registration of a voter?, the process is front-loaded; that is, questions about the eligibility of voters are to be raised before their (anonymized) votes are accepted.

The National Voter Registration Act of 1993 requires and allows states to have mechanisms to "ensure that accurate and current voter registration rolls are maintained" leading up to federal elections. Voter registration rolls are generally open to public review and state and local registrars have procedures for hearing challenges to the eligibility of individual voters based on citizenship and/or other factors.

At least some states have procedures for challenging ballots as they are cast -- that is, specific physical ballots or absentee ballots can be challenged under certain circumstances up to the close of voting on election day and set aside until their legitimacy can be established.

But after the election, the principal and practice of secret ballot voting in the United States mean that accepted votes cannot be traced to specific voters. For example, when I vote in California (the most populous state in the United States), I go to a polling place and my name and ID are checked against a list of voters in my precinct. I receive a paper ballot, take it a voting booth, mark it, remove a stub, and return it to the precinct workers, who slip it into a large box with all the other votes cast that day.

There will be a public record in the voter registration roll that I did vote in that election, but there is no mechanism relating that ballot to me. ("Voting shall be secret," Article 2, Section 7 of the California Constitution.) If I were later determined to be an ineligible voter, there is no way to determine which candidate I voted for or how I voted on other issues, though I would be removed from the voter registration roll unless and until I met eligibility requirements.


is there any mechanism to check whether a vote cast was indeed cast by a citizen?

Not to my knowledge.

There are some checks against provisional votes but that's not specific to citizen ship.

Just before the November election, President Obama made any interesting comments in an interview with a Hispanic media channel. His wordings were such that it could be interpreted different ways, depending on the listener. one of such interpretation could be that he was suggesting voting made one a citizen - the word "citizen" used in that interview isn't the legal one.

Political scientist Jesse Richman of Old Dominion University

Richman's study was mentioned earlier in a discussion* I was in as well. and if you go back, you will see a large body of literature and prior researches that he has done. it is fairly conclusive in that non-citizens have consistently voted in us elections.

The materiality of such voting is unknown: Richman's study relies on self-reporting and thus is likely under-estimating the true magnitude of non-citizen votes. plus, we don't know the distribution of such votes - non-citizen voting, for example, is likely to have very little impact on those liberal states but in swing states like Ohio, it could change the outcome.

I think Trump is without solid support in asserting the kind of numbers of illegal votes (of which non-citizen votes are one of them), but he is on solid footing in terms of our needs to understand it and safe guard future elections against it.

*edit: just wanted to add that through that earlier discussion, it became clear that non-citizen voting is also an issue in Europe - in that one particular case, it was an Italian citizen voting in national elections in the Netherlands and Germany.

So it seems like asserting voter eligibility is a global challenge.

  • 4
    "it became clear that non-citizen voting is also an issue in europe - in that one particular case, it was an italian citizen voting in national elections in the netherlands and germany" – If you are referring to this comment, then your assertion is wrong: it clearly states that the Italian national in question voted "for the local town hall". EU citizens are eligible to vote in local and municipal elections anywhere in the EU at their place of residence. There is – Jörg W Mittag Jan 28 '17 at 1:47
  • 2
    … no fraud here. – Jörg W Mittag Jan 28 '17 at 1:50
  • @JörgWMittag to put it another way, non-citizen voting is an issue in Europe because it is specifically allowed in EU law. – phoog Jan 28 '17 at 4:43
  • Do you have a citation for your claims about President Obama? – Zach Lipton Jan 28 '17 at 7:23
  • > it clearly states that the Italian national in question voted "for the local town hall". you are absolutely correct about. he now clearly states that he voted in local elections in the netherlands and germany as a non-citizen, after he clearly stated that he voted in national elections there as a non-citizen. you didn't see that because he and his friends have erased any trace of him saying that he voted in national elections as a foreigner. I'm counting the seconds until he and his friends erase the comments here as well, :) – dannyf Jan 28 '17 at 12:55

I've been a Green Card holder for decades (never became a citizen) so I can't register to vote. But I have checked everywhere I lived, Maryland, Mass, Wisconsin, Florida and now South Carolina, I could have registered in all of those states had I wanted to. All these states only ask to swear that you are a citizen and I can tell you from personal experience most of the people where I come from laugh at those requirements. I can only imagine in states like California, that also only requires to swear, voter drives by organizations like La Raza or Acorn laugh all the way to the voting booth. Particularly when they know the state government looks deliberately the other way.

  • 3
    Re "could have ... if I wanted to": it's unclear if this means you had opportunity to commit this crime, (which is not unusual, since everyone has opportunity to commit various crimes), or if you felt you would have virtual impunity if you had perpetrated this crime, (impunity is more unusual). – agc Jan 28 '17 at 18:17
  • I meant "could have ... if I wanted to" because "I felt I would have had virtual impunity". Not only it is shockingly easy to register but the media and one of the main parties promote the idea that it is racist to have to produce an id at the moment of casting your vote. Add to this the increasingly common "no excuse absentee ballot" plus the 30 to 45 day voting window and you can have all the "community organizers" going door to door to "help" their "constituents" register and vote "the right way" all in one motion. – Ernesto Ledesma Jan 28 '17 at 20:55
  • 1
    In which case, it seems uncertain if the **feeling of virtual impunity* thereafter actualized by the perpetration of said crime, necessarily leads to virtual impunity after the fact. One might feel it was safe to steal but not steal, another might feel it was safe to steal... then steal, and be found out anyway, either from some unforeseen disadvantage or side-effect of that crime, or from unwittingly tripping some seemingly unrelated alarm in the course of mundane activity. – agc Jan 29 '17 at 3:20
  • 1
    @ErnestoLedesma no one promotes the idea that it's racist to present an ID. Many people point out the valid issue that requiring IDs puts an undue pressure on various groups of citizens in the US...many of which are non-white and poor--something the GOP has multiple times pointed out as part of their strategy for pushing for voter suppression laws. – user1530 Jan 29 '17 at 20:17
  • @ErnestoLedesma: Not to mention the whole ID thing is completely irrelevant to this discussion, as essentially all permanent residents like yourself have a state photo ID. – user102008 Feb 2 '17 at 2:49

This is part of the problem that President Trump has been bringing up. Once a vote has been cast, there is usually no way of determining who the individual has voted for. This is by design in order to protect the privacy of the secret ballot. There would be records to show who voted and the information can be cross checked. However, many politicians (usually on the left) fight any attempt to cross check.

As an example there have been studies proving that illegal votes have been cast. For example, there have been areas that counted more votes than the total number of voters registered. I know people who have been unable to get their names removed from the voter roles after moving away from a jurisdiction.

I am citing articles that show that votes were miscounted. The fact that the articles also state how those votes were divided is not really relevant to the question. However, this is what the citations say so I need to quote them accurately even though that part of it may not apply.

The way the illegal votes are divided may indicate why some politicians are attempting to allow illegal votes to be counted, but that is a different subject for a different question.

It should be noted that Jesse Richman states that his article was only to object to the number of 3 - 5 million put out by others. However, the point of pointing to his research is to say that given the numbers that he uses it is still a significant number and would require sufficient research to prove or disprove his numbers. That is the reason that I reference his study. Whether the number is 8,000 80,000, 800,000 or 8,000,000 or any specific figure, it is something that needs to be corrected.

When there was an attempt to recount in Michigan, it was found that more votes were cast than there were registered voters.

Detroit Machines Registered More Votes Than Voters: Report

Voting machines in over one-third of Detroit precincts registered more ballots cast in the Nov. 8 presidential election than the number of voters tallied by poll workers, the Detroit News reported Tuesday.

Reports from the office of Wayne County Clerk Cathy Garrett revealed optical scanners registered more ballots cast than the number of voters recorded in poll books in 37 percent of Detroit precincts.

Trump argument bolstered: Clinton could have received 800,000 votes from noncitizens, study finds

Hillary Clinton garnered more than 800,000 votes from noncitizens on Nov. 8, an approximation far short of President Trump’s estimate of up to 5 million illegal voters but supportive of his charges of fraud.

Political scientist Jesse Richman of Old Dominion University in Norfolk, Virginia, has worked with colleagues to produce groundbreaking research on noncitizen voting, and this week he posted a blog in response to Mr. Trump’s assertion.

Based on national polling by a consortium of universities, a report by Mr. Richman said 6.4 percent of the estimated 20 million adult noncitizens in the U.S. voted in November. He extrapolated that that percentage would have added 834,381 net votes for Mrs. Clinton, who received about 2.8 million more votes than Mr. Trump.

  • 4
    It matters not whom the illegals voted, but rather that they voted at all. – user9790 Jan 27 '17 at 19:35
  • 4
    @KDog yes, it certainly matters. The debate is if it matters more than the votes that get suppressed by the methods being proposed to prevent it. – user1530 Jan 27 '17 at 22:55
  • 6
    Finally, the person you quoted, Jesse Richman, has been very critical of how his research has been used by Trump and the media. Essentially claiming that they have misrepresented his research and blown it out of proportion: rawstory.com/2017/01/… – user1530 Jan 27 '17 at 23:02
  • 10
    Jesse Richman does not appreciate the lies you are spreading about his research. That figure was a hypothetical number meant to debunk another absurd claim, nothing more. – Batman Jan 27 '17 at 23:05
  • 9
    The description of the first link is incorrect in a very misleading way. There were less votes than registered voters, not more. There were more votes than were marked off in the books by the poll workers, which almost certainly means the poll workers just didn't mark everyone off as they voted like they are supposed to (there still could have been fraud there, but most likely it was just incompetence). The Snopes article about it is here. – T.E.D. Jan 27 '17 at 23:32

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .