> The advent of large sample surveys, such as the Cooperative Congressional Election Study (CCES), has opened the possibility of measuring very low frequency events, characteristics, and behaviors in the population. This paper documents how low-level measurement error for survey questions generally agreed to be highly reliable can lead to large prediction errors in large sample surveys, such as the CCES. The example for this analysis is Richman et al. (2014), which presents a biased estimate of the rate at which non-citizens voted in recent elections. The results, we show, are completely accounted for by very low frequency measurement error; further, the likely percent of non-citizen voters in recent US elections is 0.
(Richman et al. (2014) is the paper submitted here; also worth noting is that one of the authors is from YouGov, the company that did the large internet survey.)
This was one of my first thoughts--these are extremely, extremely low base-rate events with not much measurement power. People mess with the surveys, or simply fail to understand the questions, or make mistakes in responding. The combination of them makes extrapolating very difficult. This was an internet sample too, and biased in that regard (which the authors acknowledge). It was interesting to me to read, though.
I think the bigger question for me, aside from this, are the benefits and costs associated with any response.
Even if you redid this and found that the likely percent of noncitizen voters was nonzero, you'd have to balance the benefits of any enforcement strategy against its costs.
Given that respondents were also saying that photo id didn't stop them, you'd have to also determine the number of citizens who were unable to vote because of voter identification. To me, the cost of denying a citizen a vote is much greater, and also more certain, than the cost of allowing a noncitizen to vote.
Overall, I'm concerned that the typical "innnocent before proven guilty" logic is thrown out the window when it comes to immigration, citizenship, and voting rights. The government should have to prove to some reasonable level of certainty that I do not have the right to vote, not the converse.
I've been a citizen for years, but if any voting record emerges that shows me voting prior to my naturalization date, the naturalization would be annulled and I would be deported for it.
This is one of the things you get hammered into your head over and over: do not vote if you're just a resident.
It apparently happens: http://chicagoreporter.com/illegal-voting-case-puts-familys-... Remember, US states are required to be quite aggressive about getting people to register to vote under motor voter laws. Also, the people who get caught seem to be the ones who say they voted on their citizenship applications; I'm not sure how actively policed this is.
Let's just remember Obama won the vote by: 69,498,516 to 59,948,323.
It wasn't even close in the EC. As a soon to be Greencard holder, I'm suspicious. Most people who are living here with Visas are super cautious of breaking laws.
Right? The conservative narrative is that undocumented persons can act with impunity and soak up benefits etc, but all the immigrants friends I have are scared shitless of getting in trouble or caught by the police outside a sanctuary city because it puts their whole lives in jeopardy.
North Carolina total population is under 10 million. It doesn't have more than 8 million non-citizen residents, even when one expands beyond the undocumented population.
Also, green cards and permanent residence are the same thing, not to different categories.
Looks like they're saying 432,700 (as of 2008). See table 3 -- reporting that North Carolina had an Obama victory margin of 14,177, 432,700 adult non-citizens, and would've needed 5.1% non-citizen votes to win.
Boy, there are an awful lot of assumptions made there. In a senate race decided by 312 votes, there are dozens of factors that could have swung that outcome. Traffic, the weather, a bad batch of lettuce keeping some people home with food poisoning, etc. To chalk that win up to non-citizen voting requires ignoring an awful lot of other pieces of the puzzle.
- Incredibly clickbait title that is in no way backed up by substantive fact
- It is old (2014) and not tagged as such
- The intellectual contribution is nil. Even the body that put out the dataset used in the study said NOT to use its dataset in this way as it is misleading
What good is a study that "speaks" to important issues when the samples it has are flawed, the methodology is faulty, and the conclusions are wrong? Is this what living in post-truth is like, where as long as something is 'related' or 'speaking' to issues it is counted as genuine material for discourse?
I read/skimmed about half of it, I didn't see much discussion about selection bias. When I first saw "internet survey" I was instantly skeptical.
Selection bias is massive, the types of people who take these polls tend to be highly politically active. I'm not sure you can extend the conclusions from this sample to the general population.
From what I see, they didn't consider untruthy answers on what people voted. Not sure how relevant this is, but considering this year's presidential election's actual votes compared to surveys might significantly change this study's results.
See the second section, "Data", in the linked-to page:
"The data used for this paper is from the 2008 and 2010 Cooperative Congressional Election Studies, based on the files released by Stephen Ansolabehere, 2010 and Ansolabehere, 2011. The 2008 and 2010 Cooperative Congressional Election Studies (CCES) were conducted by YouGov/Polimetrix of Palo Alto, CA as an internet-based survey using a sample selected to mirror the demographic characteristics of the U.S. population. In both years survey data was collected in two waves: pre-election in October, and then post-election in November. The questionnaire asked more than 100 questions regarding electoral participation, issue preferences, and candidate choices."
Further down, they point out that some non-citizens had the right to vote in about half the states around 1900; that non-citizens legally voted in every presidential election until 1928; and that their sample was dependent on people's self-reported citizenship status, but the characteristics of self-reported non-citizens who voted matched those of self-reported non-citizens who didn't vote.
Additionally there is no discussion about selection bias here.
For example, we know that since it's an internet survey, everyone on the survey has the internet or an ability to use the internet (eg: public library access?). That rules out a non-trivial portion of the population.
Additionally, not everyone answers or engages in these surveys. So there's an additional bias.
A lot of immigrants take a low profile, and dont vote, and don't engage civically, such as surveys. So missing those.
In the end, the paper is assuming the people who self selected into taking this survey are representative of the entire population, and that might be true for the more common things, but for less common things, not so true.
A good example is to fill up a jar with white marbles. Then put another 50 red ones on top. Now take the top 100 marbles as a "sample". You'd conclude from this that the jar is 50% red, when its actually like 5% red! This is the classic sampling bias/error. Since the paper doesn't touch this at all, I assume the authors know their conclusion is bogus and they don't want to go here because it'd destroy their headline: "NON CITIZENS GAVE OBAMA THE PRESIDENCY".
So much shit in this paper, yet it masquerades as science.
> roughly one quarter of non-citizens were likely registered to vote
> If more than 5.1 percent of non-citizens residing in North Carolina turned out to vote in 2008, then the vote margin they gave Obama would have been sufficient to provide Obama with the entirety of his victory margin in the state. Since our best estimate is that 6.4 percent of non-citizens actually voted...
Stefan Molyneux had an interesting video on this subject. Not so much about the statistics of non-citizens voting, but on the incentives encouraging some politicians to allow it:
I'm not sure how much the federal government can do: A state has a lot of leeway in how it chooses its electors and Congressmen. It's probably more effective for people to demand strict voter ID laws in each state, especially if your state allows citizen referendums.
The article mentions it could be ineffective if your state allows non-citizens to get ID cards, so there should probably be a stamp or icon or something to indicate that you can vote, to reduce the risk of the pool worker being confused.
Normally I'd worry about the privacy risks, but it's so important that I wonder if it's worth taking photos of the state ids associated with each ballot, and also requiring people to write their id number on the ballot. That could be very dangerous if that database gets leaked, though. Are there other alternatives that don't leak information as easily?
I'm not sure how reliable this particular study is, but it's a common enough worry that it's worth making the voting systems more robust just to avoid even the appearance of fraud or tampering.
>I'm not sure how reliable this particular study is, but it's a common enough worry that it's worth making the voting systems more robust just to avoid even the appearance of fraud or tampering.
And thus we witness the power of rhetoric and 'truthiness'. Voter fraud, especially non-citizen voter fraud, is simply not an issue in the US election system, but the fears that it is allow restrictive and disenfranchising legislature to be passed http://www.scholarsstrategynetwork.org/brief/misleading-myth...
I personally am less worried about individual voter fraud, and more worried about fraud at the county level. Perhaps in a small county with fewer safeguards, someone could throw out 10% of the paper ballots. Or the voting machines could have errors that discard votes. It'd be nice to see something that can be end-to-end validated.
In Oregon, we do vote-by-mail. You have to register ahead of time, and you can receive an email when your ballot is sent and when it is received. That seems like enough to protect against people dropping ballots. Invalid ballots are different, but it might be rare if what you're saying is correct. I don't trust any other states, but Oregon at least seems to be run fairly well.
Ultimately, it's up to the citizens of each state to decide how stringent they want their state to be.
> The advent of large sample surveys, such as the Cooperative Congressional Election Study (CCES), has opened the possibility of measuring very low frequency events, characteristics, and behaviors in the population. This paper documents how low-level measurement error for survey questions generally agreed to be highly reliable can lead to large prediction errors in large sample surveys, such as the CCES. The example for this analysis is Richman et al. (2014), which presents a biased estimate of the rate at which non-citizens voted in recent elections. The results, we show, are completely accounted for by very low frequency measurement error; further, the likely percent of non-citizen voters in recent US elections is 0.
(Richman et al. (2014) is the paper submitted here; also worth noting is that one of the authors is from YouGov, the company that did the large internet survey.)