Controlled for demographics, the voting anomaly continues to be present

[solarbird]

People at ScienceBlog and a variety of other statistics-literate sources have been doing demographic analysis on the unpredicted/unpolled Clinton/Obama vote swings in machine-counted vs. hand-counted balloting in Vermont. Controlling for a large number of demographic variables - including such outliers as geography - the variation against polling and against hand-counted ballot remains pretty constant. The calculated statistical probability of this being a random effect is p<.001, which is to say, around 1000:1 against. Further analysis is ongoing.

I would like to see similar analysis applied to the Romney surge, which appears to be comparable at the top level.

ETA: I was in a hurry before and forgot to credit cafiorello for the link. Thanks!

Comments

all wrenches are hammers, too

[angharads-house.livejournal.com]

I had a quick-look at this, and yes, the analysis stands up, at least against the geostatistics tools available here (which can of course be applied to these data: all wrenches will serve as hammers, if need be). I got p(Epm) of .0007, fairly corresponding to your p<.001. Hope more comes out on this....

[llachglin.livejournal.com]

If there were systematic counting errors, any statistically valid audited sample of votes will uncover it. Kucinich's challenge is in only a couple of places but that should be enough to quash (or less likely) confirm these rumors.

My guess is that there's some unidentified factor involved here that is perfectly above-board. I don't have the statistics training to figure it out myself, nor do I have the raw data, nor can I say that I can imagine all of the possible factors that might account for the supposed discrepancy. What I can say is that statewide the results were within the margin of error of the polls, and Clinton's lead was consistent throughout the counting process. It strains credibility to think that any fraudulent Diebold counting conspiracy would be able to so smoothly manipulate the results. Thanks to the paper trail, any Diebold counting shenanigans will be easily discernible if they exist, but I would be really surprised if they do.

I don't think it's responsible to jump to the conclusion that something is odd here and claim high odds against the actual result being valid when it's impossible to know all of the confounding variables, and before there's been any comparison with the paper trail. A statistical analysis is only as good as the underlying data, and without verification speculation about voting fraud is not an honest question, it's groundless conspiracy theory. In the absence of evidence--not half-assed statistical analysis--the grander claims of voting fraud are dangerously close to libel. Until there is evidence, this looks like nothing more than a swiftboating of Clinton by her rivals. In its way, it's as sleazy as all the barely-veiled racial and gender politics that have been going on since New Hampshire.

[solarbird]

The results were not within margin of error. They also disagreed with exit polling, as I heard it reported at the time. The results require that essentially all of the large number of undecided voters voted for Senator Clinton, and that Clinton voters were less likely to respond to pollsters than Obama voters. That's possible. It does not help that the same theories were floated by the GOP in Florida in 2000, and those theories, on later study, turned out to be completely false, but it's possible.

any fraudulent Diebold counting conspiracy
It is also possible that something else is going on, either with or without fraud. And election fraud long predates electronic counting, in any system.

A statistical analysis is only as good as the underlying data
The underlying data is public; the demographic information is public. A variety of people have been going over that data; more people are going over it as I type. We'll see what happens.

speculation about voting fraud is not an honest question, it's groundless conspiracy theory. In the absence of evidence--not half-assed statistical analysis--the grander claims of voting fraud are dangerously close to libel.
Why was this not true, then, in 2000 and 2004? Why was it reasonable to see smaller statistical indications of problems in those elections and want to investigate, but not reasonable to do that here? Why is Florida 2000 and Ohio 2004 sensible public oversight and outrage and this "a swiftboating of Clinton by her rivals"?

Is my desire to see a similar analysis of the strange Romney numbers also "a swiftboating of Romney by his rivals?"

For the record, I plan to have very little if anything to do with either party's nominations process. The candidate to whom I donated and who I planned to support in our caucuses pulled 1% in Iowa and dropped out of the race. The other candidate I have considered supporting continues to poll 1% and is barely acknowledged as running at all. I will consider this again once our caucus date approaches, but right now, my plan is to stay home and see who the parties nominate. Arguably, as I'm not a Democrat or a Republican, it's really none of my business anyway. But if something looks funny, I'm still likely to look at it and go, "Hey - that looks funny. What the hell?"

[llachglin.livejournal.com]

Obama got all of the votes the polls said he would get (not counting the party internals or exit polls, which both use demographic weighting factors that are prone to mistakes--in this case they assumed turnout overall and for independents on par with Iowa, which did not happen.) Clinton got what she polled, plus the undecideds, plus or minus the margin of error. The results were consistent with a late surge by Clinton that was too late to show up in the polls, combined with a lack of turnout by the demographics that favored Obama in Iowa. In other words, they're not the least bit suspicious until you start counting Diebold counted vs. hand-counted precincts, and that could easily be explained by something else.

You point out that election fraud existed before electronic voting. That's true, but the alleged fraud in this case is electronic. No one is suggesting ballot box stuffing, the old way of fixing elections, because if that happened you wouldn't see a correlation with Diebold counted precincts. All precincts use the same optical scanning paper forms so it's no easier to stuff Diebold-counted ballots than other ballots.

Serious statisticians seem to be saying that you can't draw any conclusions about fraud from a discrepancy like this because you can't account for all the other factors that might produce the discrepancy. On the other hand, if there was counting fraud, any audit of suspect counties would quickly reveal the fraud. That's the advantage of having a paper trail.

I have always had problems with the analyses from 2000 and 2004 that use similar statistical techniques to suggest fraud without corroborating evidence (I think I've called them conspiracy theorists). On the other hand, there are differences. In 2000 the butterfly ballot and hanging chads resulted in a lot of missed votes. A hand count showed that the intention of voters was to deliver a majority to Gore, and had the whole state been recounted, Gore would have won. There was also serious evidence of vote suppression (as there always is by Republicans). Finally, the Supreme Court judged the case on political grounds. That's all quite different from a statistical anomaly in a few precincts. 2004 was all about voter suppression in Ohio, electronic voting without a paper trail, and the failure of the GOP Secretary of State in Ohio to do anything about it. Again, none of that is relevant in the case of the New Hampshire primary.

I suspect that this claim originated with the knee-jerk anti-Clinton right, which has a history of smearing the Clintons without evidence. That Andrew Sullivan is one of the people who got this started only confirms that. I basically think Sullivan is full of shit even when I agree with him. The other people pushing it are anti-Clinton liberals and Democrats who have bought into right-wing propaganda, if they're not open misogynists themselves. Again, I'm not a Clinton supporter, but I know the history of her enemies and I have much more respect for her than for them.

I have no problem with your requests that this and Romney's numbers be investigated. The problem I have is with the tone of your posts, which seem to assume that something fraudulent is happening even though there's zero evidence of fraud to date. (And no, statistical discrepancies are not evidence.)

[cafiorello.livejournal.com]

Sheesh. Just raising the question that something seems odd, and doing statistical analyses (which, BTW, if you don't have the statistics training to figure out yourself, you don't have the statistics training to classify as half-assed), is not libel, sleazy, or swiftboating. Did you read the original articles? Generally everyone, solarbird included, is saying "we don't know what this means." Hell, the one voter fraud claim I've seen is the speculation that the Repubs did it on purpose to make Clinton look bad--nobody I have seen has claimed that Clinton did it. Yeah, there might well be another confounding variable, but knowledgeable people keep pointing out possible ones, they redo the analyses, and the results continue to be anomalous. NOT evidence of fraud, but anomalous.

Cathy

[llachglin.livejournal.com]

I'm classifying them as half-assed because the person doing them admits to not being a statistician, because statisticians have pointed out that you can't jump to conclusions in cases like this because you can't identify all of the relevant factors, and because there's literally ZERO corroborating evidence for the outlandish claims of the electoral fraud theorists.

For the record (and as I've already stated), I support an audit of the discrepancy. What I dismiss is the tone of people, typically Obama supporters or Hillary haters or both, who assume that a discrepancy is evidence of fraud and perpetuate a slander against the Clinton campaign just because they don't like her. I am an Edwards supporter who agrees more with Kucinich on matters of policy than with any other Democrat. I like Obama's style and politics, but the messianic fervor of many of his supporters and their willingness to engage in slurs against Clinton over stuff like this is leading me to consider supporting her over him if it comes down to that choice. (Right now, in part because of the tone of his campaigning in the last couple of days, Obama still has the edge over Clinton for me.) And as a Democrat who genuinely likes all candidates running and wants to defeat the Republicans, I have a real problem with slander of any of them. There's lots about Clinton that's worthy of withering criticism, so it's a shame that something like this is getting so much attention.

Most of the slander is coming from Clinton's political opponents, but I detected a whiff of it in solarbird's original post at http://solarbird.livejournal.com/594810.html.

[risu.livejournal.com]

It's irresponsible to turn away from anomaly.

I am willing to assume that it's not anomalous for some people, based on the order of received information. If one encounters the allegations for the first time in a framework of "just one more of the things people are throwing around desperately hunting for something, anything to smear Hillary with," then fine, I can see why one would rationally think it's not worth caring about.

Let me offer my perspective.

I was aware of the news sources hawking a massive Obama win before it didn't happen. I was pleased at the idea of *someone* getting a massive win in the primaries, but, because of what the blogs have been calling "the Tweety effect," I was kind of rooting for Hillary. And then she won, and I was pleased, and also a bit nervous. "Yay, I got my wish. Wait, is this going to be bad?"

And then I reassured myself by thinking again of Pharyngula's article on how even female liberal feminists can underestimate the competence and general caliber of individual women (http://pharyngula.org/index/weblog/comments/the_cost_of_being_a_woman_in_science/), which is what makes me believe that she'd be a much better President than I think.

I was only aware of the exit poll thing later, and kind of vaguely. I knew that it was out there but not whether it was meaningful. I'm still not.

Then I was pointed to an analysis that correlates Diebold machines with the shift.

My first reaction was that my stomach fell like a stone, my mouth went dry, and the world went shaky. I had to ask Kirby, who was showing anime at the time, to pause it while I went into the bathroom and seized.

And it wasn't because of fear of Senator Clinton. Honestly, I'd rather guess a false flag operation. And it wasn't because of the fear that it would tar her campaign, although I admit that that was part of it. I caught myself speculating aloud about things like who would have done it and what would happen, but *those things aren't important*.

What's important is procedural validity.

[llachglin.livejournal.com]

I don't want to turn away from anomaly. The vote should be audited.

What I have a problem with is excitable language like "horrified," " desperately stupid," "sloppy," "rage," and "disaster" used to describe a statistical anomaly that without corroborating evidence of fraud is nothing more than a mathematical curiosity (these were all in solarbird's original post about this on 1/9, and are mild compared to what people were saying in other places where this story was being propagated). I also think it's irresponsible to pass on information from Andrew Sullivan without pointing out that he has a well-known grudge against the Clintons.

[solarbird]

I stand by horrified. If the electoral system is fucked up, and there's any whiff of it, that should be horrifying. A rigged election would be a disaster, and rage is the right tag for something like that, as far as I'm concerned. Why do you think I was posting asking for another explanation? Because these numbers are statistically meaningful (and btw, the original exit polling put the "undecided" vote as split between Obama and Clinton, not going almost entirely to Senator Clinton as you propose) and looking at what the numbers say has historically been really, really illuminating in my historical experience.

On the other hand, slapping me with the use of "desperately stupid" and "sloppy" when I was using those as words to describe why it's UNLIKELY Senator Clinton did something crooked is Not Fair Play as evidence for accusations of slander at me.

[llachglin.livejournal.com]

FYI, the recount was completely on Monday and there was no fraud: http://www.dailykos.com/storyonly/2008/1/21/83051/3514/129/440226 (which has links to the official NH recount results). I'm posting here because it missed my attention until now (in a story about Kucinich dropping out today) and I haven't seen anything from you on this, so I figured you missed it too.

By the way, in regard to quoting some of your language, I wasn't intending to make a personal attack but just trying to point out why to me your post read as accusing, even if that's not what you intended. There seemed to be an implicit bias toward the possibility of fraud being real, despite a complete lack of evidence, and even though your language was conditional it still ended up characterizing the Clinton campaign for something there was no evidence had happened, and which we now know did not happen, at all. There's enough actually wrong with Clinton's substantial positions and campaign that there's no need to raise the alarm about things that didn't happen. The general tone of the campaign has really devolved over the last month, and I really wish we could stay focused on substance rather than rumor, mudslinging, accusations, and counter-accusations.

[solarbird]

Yeah, I'd been following it on the Secretary of State page (tho' hadn't seen a final tally yet, I've been busy), and also, this paper provided a rational possibility for why there would be a non-demographically-visible difference. (Picosummary: Boston exurbs were more likely to vote for Clinton than Obama. They're also (for unknown reasons) more likely to have machine-counted ballots.) So that's good.

What's new since I lived there is the rest of the state having demographics comparable to the Boston-exurb areas. (Which, apparently, it does.) Either I was wrong about that at the time, or that's a significant shift.

[risu.livejournal.com]

What's *important* is that Diebold machines *not* be giving different results than hand-counting in the bloody *Democratic primary*. Or, if they do, that that *be* looked into, as a matter of course. What's important is that there be some part of the political process where procedural integrity and trustable outcomes is there. And I'd been kind of stupid to think that it *would* be, in the Democratic primary, but, you know, I had.

What's important is that it be responsible to assume a legitimate result---

*Not* because questioning the result is bad, but because the methodology *supports* that assumption, *supports* being held to a high standard.

What's important is not the legitimacy of adherence to some indeterminable "truth on the ground," but that the *system be structurally legitimate* and *practiced with integrity*.

You can't ever defend perfectly against fraud. You can't ever protect yourself against that deep and murky conspiracy that is indistinguishable from unknown facts.

But you can do things proper.

And it makes me want to *cry* to think that this isn't.

I kept wanting to shout, when Banner was around and talking about how the people in Guantanamo were terrorists,

"No. If you don't hold a trial, you *must* assume that they are *innocent*. If you don't accept 'their captors refuse to try them fairly' as a *proof of their innocence,* you invalidate the basis of the law."

But I didn't, because he is Banner.

It's the same thing. If we don't have *reason* to trust in these machines, then . . .

Right now the primaries, along with all the rest of the system, are corrupt, befouled, *until* the recount. The damage is *here*, not speculative. It *cannot* be the responsible thing to do to assume that things are fine, because that assumption facilitates things *not* being fine. The responsible thing *has* to be vigilance.

*This discrepancy*, the *fact that it is out there and not explained*, has to be seen as an inherently bad thing that must be addressed. Ideally by an investigation that finds that all is okay.

And it terrified me, to swing around the blogs later that night, looking for reactions, looking for insight into what was going on, to only find some guy on DailyKos using argument from authority to say: "Because I am smart and cool and know what I'm doing, these allegations are ridiculous nonsense. I don't even need to prove it."

To find that echoed so many places. Not just from Hillary supporters, where at least I could understand it, but from people who were worried about the cost to New Hampshire and the way that a recount might take food from the plates of starving children there. (And I am certain there *are* starving children in New Hampshire, and if New Hampshire winds up deciding not to help them because it spent all this money on a primary recount, then I'll feel pretty bad, but Jesus God, if the state isn't going to pay for procedural validity of elections, then what the hell is it doing in the business of stating things?)

And when Kucinich stepped forward---

Well, one step back towards being able to stay in this country, as long as there are philanthropic multimillionaire primary candidates with a concern about election integrity and no hope of winning, I guess.

May all the bright and good powers of the universe bless him, and I hope that nothing bad turns up.

[llachglin.livejournal.com]

I think every election should be, at the very least, audited against a paper trail. Ideally, every election should be hand-counted for the official results. That's how it's done in most Canadian elections and in the UK, and those countries are pretty much the gold standard for proper election procedure.

Like you, I admire Kucinich's willingness to stand up for principles. I might end up voting for him, even though I'm currently an Edwards supporter (mainly because he's the guy--to use your example--who's actually thinking about children starving in New Hampshire, who might actually have a chance of winning).

[twoeleven.livejournal.com]

you may be interested in this blog entry pointing to a premliminary paper:

Voting Technology and the 2008 New Hampshire Primary: Herron, Mebane, and Wand don't find any problems (http://www.stat.columbia.edu/~cook/movabletype/archives/2008/01/voting_technolo.html)

[solarbird]

Thanks, I'll poke at that more later.