Show of Hands

For one thing, these are not *verified* illegal non-citizens. They are a subset of many thousands of respondents who chose to answer an online survey, who merely self reported as non-citizens. When another population of self-reported non-citizens was queried more extensively over two time points two years apart, many of them reported being citizens at the *earlier* time point but the same individuals said they were *non-citizens* at the *later* time point. Since that doesn't compute at all, it casts doubt on the veracity of reporting of the other group. Maybe they selected non-citizen accidentally in haste? Who knows?

At any rate, 14% of 489 (68 people) is way too small a number of people on which to base this kind of opinion (by extrapolating to millions). And if even a few of them claimed *incorrectly* that they were registered to vote, it would be just crazy.

.

If I ask 489 illegal immigrants, and 68 of them are registered to vote, that's pretty bad. Even if just those 68 illegals happen to be registered (extraordinarily unlikely), that goes completely against the notion that illegal voting is practically nonexistent.

In addition, this particular study actually began as the project of a Pakistani immigrant who was trying to find evidence of voter SUPPRESSION, so I highly doubt that any bias would be in favor of a greater number of illegal voters. I don't know the exact methodology because the actual paper is hidden behind a $20 paywall, but the abstract claims to have used a nationally representative sample.

It also appears that you've immediately assumed that the research *is* bad despite having no reason to believe that, and if you're going to cite the numerous critiques, every research that can lead to conservative conclusions is immediately bombarded with those, because the academic community is every bit as biased as you are.

As I said, I can't confirm how they actually obtained their sample, but they claim to have a nationally representative sample, and I don't believe they would claim that if it was a simple internet survey of the sort found on surveymonkey. Even if there is response error, it would be a miniscule fraction of that 14%, and I am unaware of any follow up study. I don't know how that would even be possible if it was an anonymous internet survey like you claim. Also, this is only the number of illegals who admit to committing voter fraud. Many more could have been afraid of being caught and therefore lied, more than making up the difference for response error.

As I said, I'm not able to read all the associated articles yet. I'm basing some of my comments on the first linked article, which I believe is at this link, and mentions the nature of the survey - importantly, that you *don't know* that these are non-citizens for sure: www.washingtonpost.com/news/monkey-cage/wp/2014/10/27/methodological-challenges-affect-study-of-non-citizens-voting/?utm_term=.0fbeb9c038d1

I've taken many YouGov surveys and nothing in them verifies any of my responses. Additionally, people who take them are self selected - they have access to computers, know how to use them, have facility in English, have leisure time to take surveys, and have interest in taking surveys - all of which sets them as a not necessarily typical subgroup of people anyway (just as being a SOH user does - we are *not* a typical cross section of the US population).

.

You can't claim that I'm *biased* just because I'm saying the study results aren't can't be extrapolated to a much larger population!

I say that because of what I have read of the methodology, not at all because of whether I happen to like what the extrapolated result says! I would say the exact same thing no matter what the result was.

How can this have been an Internet survey of the sort you suggest if there was a follow up study with the participants that you claim also happened? Don't you see the contradiction?

Also, I say you're biased because you said you were going to see "*how* bad" the study was before you even read the first link.

Holy cow I love susanr!

I second that voc!