Berkeley Researchers Analyze Florida Voting Patterns

empraptor writes "Researchers at UC Berkeley have crunched numbers and determined that 130,000-260,000 excess votes went to Bush in Florida. They have held a conference and posted their findings online. You can find articles on their research from CNet, Wired News, and many other sources. While the research used statistical analysis based on past elections and demographics, how else do you verify that a paperless voting system is working properly?"

Comments

Is there a choice of what to vote with?

[garcia]

"Without a paper trail, statistical comparisons of jurisdictions that used e-voting are the only tool available to diagnose problems with the new technology," the researchers stated in the report.

WHY WERE THERE NO PAPER TRAILS? Why are we allowing voting to go on in a system that has NOT been proven safe? We aren't allowed to view the code, we aren't allowed to audit our vote except via what is shown to us on the screen, and we have to invest an enormous amount of trust in two large entities that have proven they are NOT worthy of our trust.

Were people permitted to use paper and pencil/pen or more trusted/tried solutions instead of these machines? I certainly would have opted against using one of the e-voting machines knowing what I know and being the paranoid individual I am.

Until the voting machines and their code are open to the public for audit and there is a paper trail I will refuse to use them. This MUST be an option for everyone. I don't see why it can't be the case.

Some places are requiring a paper audit trail by 2006 but that doesn't help the fact that there could have been some hanky panky going on right here in THIS election.

Re:Is there a choice of what to vote with?

[plalonde2]

Let's try this again. There is only *ONE* effective way to run an election:

Vote on Paper

Count each ballot box separately

Count the ballot box at the voting site

Allow every party to have an observer there. Parties that can't pony up a counter/observer per ballot box have deeper problems The advantage is that wide-spread fraud would require widespread efforts, unlike the US's system of centralizing voting processes (central counting machines, central code source bases for voting machines, shipping ballots to counting locations).

Distributed systems are much more robust to fraud.

I'm glad this is happening.

[Jucius Maximus]

Making people realise that paper audit trails are necessary is a lot more important than having your choice of Kerry or Bush for the next 4 years.

Paper trail not enough

[downward dog]

The problem is that if a voting machine is programmed to cheat, it is easy enough to fake a paper receipt. I could cast a vote for A, have the screen verify that I am voting for A, receive a printed receipt that tells me I voted for A, and STILL have that vote count for B within the black box.

The paper trail is a red herring, if you ask me. What is really needed is publicly-available source code that anyone can view.

Re:Paper trail not enough

[j0nb0y]

Certainly recount if there's a problem, but the machines should be randomly audited as well...

Re:Paper trail not enough

[BW_Nuprin]

There should be digital signatures on all voting machines that confirm that the code is the same code as on all other voting machines. Either that or do a binary comparison of the ROM after the fact on machines that are suspected of being tampered with.

Re:Paper trail not enough

[VultureMN]

The problem with this approach is that it'd reward aggressive fraud.

For example, let's say a hypothetical company, call them 'Diecast', has the machines give enough extra votes to candidate, oh, let's call him 'Buck', so that 'Buck' wins the state by 300,000 votes. That wouldn't be enough per machine for anyone to notice, and the other candidate, let's call him 'Harry', isn't going to challenge that big of a statewide spread.

The only way we should trust electronic counting is when the electronics cannot 'know' who goes with what. All they could count is that A got 123 votes, B got 113. Then the pollworkers match up A with Harry, and B with Buck. (and perhaps in other precints, A is Buck and B is Harry.)

Re:Paper trail not enough

[Anonymous Coward]

Isn't it simpler to keep voting on paper? I got a degree in Computer Science and I'm fond of computers, but I recognize when it's the case to use a computer and when it's not. For this application domain paper beats computers.

Re:Paper trail not enough

[Anonymous Coward]

The problem is that we cannot trust a private corporation to do this (vote tabulating). This should NEVER have been outsourced. If we must do it on a computer, then a few universities should get grants to produce open source code to solve the problem.

Arguing about what could be done is a distraction. What was done is what matters. Exit polling showed that Kerry won, and the discrepancy that showed exit polls not agreeing with the actual vote count was in areas where these black boxes were in use.

It is probably already too late. Whatever solution the bright minds might argue about on slashdot is not going to matter, because you can just look at the way the voting was done to realise that their was no interest in paper receipts because there was no interest in allowing voters to chose to kick Bush out of office. The manipulation was rampant and remains pretty much underreported, with the exception of cute little complaints from both sides so that most of us just toss up our hands. How can you or I make them do the right thing? Every time you look away they will pull a trick, and then have a private smirk at what fools the populace is.

Whatever change you want, it will not happen through voting anymore. We will have more black boxes in two years, not fewer. We will have more distractions, and we will have a more compliant and cowardly media to titillate us with more infotainment and little insight.

I recommend that everyone buy gold, or a plow, and plan for exit strategies with groups of friends. Everyone who lived through a disaster must have thought it couldn't happen to them.

Just watch the CEO's offshoring money. I expect that Gold will be rising in value shortly.

You know, I'd love to refute their claims...

[Quinn_Inuit]

but I _can't_, because there's no way to do so. Because of paperless voting, we have no way short of standard polling techniques to tell if these machines were even close to accurate.

Re:Why Berkeley?

[notcreative]

A liberal reputation does not "prove" anything about the argument. Read up on "ad hominem" logical fallacies. Their evidence is publicly available and the research paper makes a statistical analysis. If you want to attack their conclusion, please make a comment relevant to the analysis or its assumptions.

Re:Bad source.

[dema]

If it came from a less biased source I might believe it.

And if electronic voting came from a less biased source I might believe it.

Re:Possible explanation -- the values voters

[csimicah]

Did the Berkeley analysis take into account Hispanics voting Republican because they have compatible values? It may explain the discrepancy that Berkeley claims to have uncovered.

Couldn't even be bothered to read the half-page summary, eh?

The REAL red flags in this debate

[nysus]

What unnerves me more than anything is the simple fact that election officials are so adamantly against paper receipts. There is simply no rational explanation for not wanting them. I've heard it said that cost is a factor. But, really now, how much more can a simple tape register add to the cost of a probably already over-priced voting machine with a CRT, a networked computer, and proprietary software inside? Isn't safeguarding democracy and people's faith in it worth spending a few bucks?

Re:The REAL red flags in this debate

[AnotherBlackHat]

Given that the votes are challeneged and recounted every time, there is no point to going electronic at all.

When I withdrawn a large sum of money from the bank, the teller counts it, then gets a second teller to count it, then they both initial the withdrawl slip.

Since they always get a second teller to count, does that mean that the first count is unnecessary?

I favor electronic counts and paper audits.
If there's a large discrepancy between the two methods, then you investigate everything.

If paying for an audit count is a problem, then make the party asking for it pay for it.

Re:Earth to Berkeley...

[timster]

I've said it before, so I'll probably be called redundant, but criticism of a government during wartime is only treasonous in a totalitarian state. In a democracy, criticism of the government during war is patriotic.

In a totalitarian state, the interests of the government are by nature not aligned with the interests of the people. So criticism of the government damages its effort, because it suggests that interests other than those of the government ought to be considered.

In a democratic state, the government is only operating correctly when it represents the interests of the people. However, it is not easy to act in the proper aggregate interest of the entire population. The only way to do so is for the portions of the populace who see the government actions as contrary to their interests to speak out. Thus the government actions can be modified so that the interests of the people are better fulfilled.

This is pretty basic civics, but for some reason it is repeatedly suggested (during Vietnam, and again recently) that the proper interests of the government are somehow disconnected from those of the people when the government decides to go to war. In my viewpoint that is incorrect, especially when it leads to a government that is eager to use war as a device to promote its own interests, thus leading toward totalitarianism.

Real Issue

[notcreative]

I think someone else above made this point: The real issue isn't whether or not Bush won. The real issue is whether we can trust our mechanism of democracy (the ballot box). If we can't, then we don't really live in a democracy, do we (or even a republic for the pedantic)?

I would like to point some facts out:

1. Manufacturers of electronic voting gear had the Opportunity to rig the vote
2. They had the Motive to rig the vote
3. They had the Means to rig the vote

So why do we have a system that allows a company with the Motive, Means, and Opportunity to rig the vote to use software that is unaccountable, and not to provide a backup papertrail? I would be glad to have Bush as prez if I knew for a fact that the vote was counted right, but I don't know that, and NEITHER DOES ANYONE ELSE on this thread. That is scary.

Also, do people think that voter fraud on this level is implausible? Please understand that it has happened before in this country, and if it happened before why can't it happen today?

Ban Chocolate!

[shotfeel]

Researchers have found a correlation between chocolate sales and crime rate. Chocolate must be banned!

Oh, BTW, the years chocolate sales and crime were up also correlate with population.

The above is what I remember of an example of the problem with certain types of statistics. People often see a correlation and jump to the conclusion that there must be some type of causitive effect. That's often not the case and there are often underlying variables (population) that would more readily explain the correlation -if one were to take the time to look for them.

OK, then why have the voting machines at all?

[dave-fu]

What about letting me fill out a paper form with a pencil? No worries about crashing, system malfunctions or crackers, no setup costs, easy recounts, an interface that everyone can understand.
Why the drive for electronic voting? It's an interface that people have never seen before (and won't see again for 2-4 years), is user-unfriendly and overly sensitive and ends up being slower to use for people AND more inaccurate. I got to vote with pencil and paper; voter turnout was far, far higher than I've ever seen it before and I had next to no wait to get in and get out.

Re:OK, then why have the voting machines at all?

[cryptochrome]

Also paper ballots are very scalable - if too many people show up at a polling station you just slap up a new curtain and some markers and you're good to go. Worst case scenario you send for more poll workers to help process the check-ins, and print out/send for some more ballots. Not an option with electronic machines. You don't have enough, you're screwed. We've heard plenty of horror stories in this election of lines being hours long, so this is actually a highly critical problem.

Machines do have one critical advantage against paper ballots though (assuming they are properly designed) - they are tamper-resistant and naturally immune to invalid votecasting. Right now the easiest way to screw with a paper ballot is with a mark or punch made surreptitiously turning a valid vote into an overvote, or a no-vote into a vote. Not to mention the actual overvotes people made by mistake, and misreads. Quick and easy, all you have to do is make sure no one's looking and be ready to call voters incompetent. You can also stuff or lose ballot boxes and invalidate whole districts if necessary.

Computers won't allow you to overvote or make other mistakes, their receipts cannot be plausibly altered or misinterpreted by unscrupulous ballot handlers, and the numbers can be double-checked against the electronic tally in the machines. In fact with the right receipt you could do a second optical scan count and a hand recount, and if all three do not agree start raising giant red flags and sending in lawyers to put asses in jail.

Re:Two things

[Chagrin]

RTFA

Only counties using electronic voting machines showed the increase. Are you claiming that electronic voting machines increase the effect of Bush's post-hurricane visits?

Re:There needs to be some paper trail

[Reducer2001]

Voters swipe a state/government issued ID card
Whoa whoa whoa, stop right there.

Whiskey Tango Foxtrot!

[cryptochrome]

While I find the lack of paper trail alarming, I find the failure to get exact results for elections mind-boggling. Calculating election results isn't brain surgery, IT'S COUNTING FOR FUCK'S SAKE!!! The talent you learned in early elementary school! The thing computers are best at! I can see how some ballots might not register correctly, and god knows there are some shitty ballot designs and systems out there, but there should never be errors made in the actual calculating. Ever.

US elections positively reek of either concerted fraud or extreme stupidity, and it's totally unacceptable. Let me point out the glaringly obvious:

1) Not only must there be a paper trail, it should also be hand counted to verify the results from the more rapid machine counting.
2) Makers of "faulty" electronic systems should be indited for treason or fined into poverty.

As for balloting, it's a toss-up between optically counted paper ballots and receipt-printing computer balloting. Paper ballots are cheap and scalable, but can be tampered with (turning a valid ballot into an overvote is as simple as a surreptitious mark/punch from someone handling the ballots in most designs). On the other hand, computer systems are more flexible, prevent under/overvoting, and their paper reciepts are resistant to post-casting fraud and verifiable against the machine tally, but they are expensive and high-maintenance and not scalable at all (though they could have been made a lot cheaper and lower-maintenance - or just used ATMs instead).

Re:Whiskey Tango Foxtrot!

[Incongruity]

Did you read the paper that all of this is coming from? The argument doesn't actually imply that they miscounted this time The authors of the paper would seem to like you to believe that, but it's faulty reasoning, IMHO because of one fundamental reason... The statistical estimates are based primarily on the results of the 2000 election (and the 1996 election using the same voting systems as 2000). Given that the 2000 election in Florida was a primary reason for the major calls for voting technology reform because the previous methods were deemed to be poor methods of reporting the will of the voters, then why in the world should that very same vote count be used as a basis for statistical extrapolation that is then used to argue that votes were either added or misattributed to one candidate in the current election!?

All this does is call in to question ALL the election results. It might well mean that the previous system failed to accurately register votes for President Bush just as much as it might mean that there were additional votes given to him in error this time.

For reference, I didn't vote for either of the major-party-monkeys so don't accuse me of just siding with Bush because I voted for him -- I'm neither siding with him nor did I vote for him. All I am saying is that the logic of the argument doesn't prove vote tampering or incorrect counting in this election. It simply means that based on data collected with the previous balloting systems, one would expect President Bush to have received between 130,000-260,000 fewer votes. So, logically, that either means that the new system is flawed/biased, or that the old one was or that they're both biased but in opposite directions.

The fact that the gains correlate well with the counties in which Bush previously did the most poorly in doesn't mean anything other than supporting the idea that one or both of the contrasted vote collection methodologies is biased.

I mistrust electronic voting, but I also mistrust punch-cards and party ballot box stuffing. Don't think for a moment that the latter never happens.

Evidence of fraud in one or both elections (i.e. eyewitness accounts or other similar evidence) is the only thing that can clear this up at the moment. Otherwise, better electronic voting systems that are more open/reliable/tamper-proof and auditable and a few more elections using those "reliable" measurements as the basis for comparison to this year's and the 2000 election are the only way to differentiate between the various possibilities.

This article is questionable science (and needs to point references to some of its claims, but that's just me knit-picking)-- again, all in my opinion.

Re:Bad source.

[Life2Short]

Less biased? How so? They publish the source of all of their numbers. They are sources that are readily available to anyone with a web browser and a library card. If you think they cheated, go check the numbers. They publish the raw data on the site and tell you the tests they used. If you think they're cheating, do the calculations yourself. As far as the "large gap for a statistical study," again, how so? According to CNN there were 7,507,727 votes cast in FL. The difference between 130,000 and 260,000 gives us a gap of 130,000. Relative to the data set that we have here, that would be a gap of less than 2% (1.73 to be more precise) of the numbers we're talking about. In what sense is that a "large gap for a statistical study?" If you've got a specific criticism - great! let's hear it.

Re:Possible explanation -- You've got no clue?

[mynameis (mother ...]

Dude! Look you obviously just DON'T understand statistical analysis AT ALL, ok?
FTFA:

This multiple-regression analysis takes account of the following variables by county:
- number of voters
- median income
- Hispanic population
- change in voter turnout between 2000 and 2004
- support for President Bush in 2000 election
- support for Dole in 1996 election

After they removed the effects of all of those factors they ended up with 99.0% confidence that e-voting corrolated to extra bush-votes.

Do you get it yet? Could it be something else that they didn't include as a variabe? Sure, but only if it was somehow specifically different in e-voting areas.

Of course 99% isn't 100%, but lets get real for a minute ok?

Re:Earth to Berkeley...

[timster]

Exactly what meaning is that supposed to have?

We were all taught in 3rd grade that "democracy" meant the citizens voted on everything and that is what they did in Athens or something and that this country was a "democratic republic", but this sort of distinction has nothing to do with anything I said. If you ask President Bush if the USA is a democracy, he will tell you that it is; if you ask Senator Kerry if the USA is a democracy, he will also tell you that it is.

Words evolve over time, and in this case, I am referring to the meaning of "democracy" as a government whose institutions of power are controlled by the citizens. If you look in a modern dictionary, you will find that my usage is not considered incorrect.

Re:Two things

[The-Bus]

Plain and simple, repeat after me: Exit polls showed Bush winning .

The only exit polls which showed Kerry winning were early results from the National Exit Pool, and were only reported on sites like DrudgeReport. The problem was that taking a sliver of the NEP is completely inaccurate. There's a reason major news organizations are very conservative with the data. It's just not 100% accurate. Less so early in the day.

Of course someone will say that late at night Bush's exit poll numbers suddenly jumped up. This too, was caused by a server failure in the NEP which hadn't updated the exit poll information.

Bush did not win by mass conspiracy. There is no cover up.

Either way, the system needs to be changed.

Correlation vs Causality

[bshroyer]

I've read the article. I'll assume for the four students correctly conducted the analysis they've described. The results are compelling: Essentially, net of other effects, electronic voting had the greatest positive effect on change in percent voting for Bush from 2000 to 2004 in democratic counties.

But, the unanswered question is, is there a causal relationship between the presence of e-voting and the "unexpected" change in Bush voting percentage?

A few additional facts:

Of the 67 counties in Florida, 15 used electronic touchscreen voting. (map here)
Of these 15 counties, exactly three (Miami-Dade, Broward, and Palm Beach) were democratic counties. (map here)

The outlying data points, on which the students base their conclusion, consists of three counties. Which happened to have been the focus of the 2000 election irregularities. And which happened to have been heavily campaigned, by both candidates. One could argue that there are a couple of causal relationships here:

a) because the elections in these counties did not go smoothly in 2000, there was pressure to reform the process, and e-voting was installed.
b) because the 2000 election hinged on these counties, the campaigning was extremely heavy there in 2004.

One stimulus (2000 election debacle/recount) may have caused both the e-voting implementation, and the Bush shift.

The authors of the paper go on to say that a similar analysis of Ohio e-voting returns showed no relationship between voting method and change in Bush percentage. Why would the relationship be causal in Florida, but not in Ohio -- or anywhere else that we're aware of?

Re:Bravo

[doom]

Just a few words in defense of "conspiracy theorists"...

daveschroeder wrote:

I didn't mention it in my original message, but my favorite conspiracy theory is that John Kerry's campaign probably decided [...]

"Conspiracy theorists" are always getting beaten up on issues like this, but I'm not sure it's strictly fair... In addition to proposing a hypothetical scenario about Republican corruption, I'm *also* supposed to be a mind-reader, and be able to explain away why John Kerry did what he did? The whole "motives" issue, seems like a lose-lose proposition. On the one hand, if you don't speculate on why so-and-so did such-and-such people will regard the theory as incomplete, too far-fetched. If you do speculate on it, you seem like you're over-reaching, claiming knowledge of things you can't possibly know about.

The journalists all said they'd kill for a juicy election fraud story, but there was none to be found...not even one that might exist but have no "proof".

A lot of us have a lot less faith in the good old "muck-raking journalist", having had to listen to a rather uncritical, monotonous drum-beat during the Iraq war run-up.

My favorite retort for this one is that all of the corporate media (i.e., all mainstream newspapers, ABC, NBC, CBS, CNN, FNC, etc.) are all in Bush's "pocket", and that even though there is widespread proof of election fraud, the corporate press has ordered all of its staff to "lock the story down" and not speak of it further.

Uh huh... let's roll the clock back, and consider the WMD issue during the Iraq war build-up. Isn't it ridiculous to suggest that the *entire* media could be asleep at the switch for some reason? Certainly if there was some reason to be critical of the administration's claims on this issue, *someone* in the media would be all over it, wouldn't they? I mean, the New York Times is hardly a Republican strong-hold, is it? Are you trying to tell me that Judith Miller has been bought by the Other Side? Oh, please.

The thing about conspiracy theories is that their tautologies: everything can be neatly explained away, no matter how absurd it is, and you can still believe what you want to believe.

And of course, you'd expect that an *actual* conspiracy would be a really clumsy affair, with lots of leaks (Diebold memos, anyone?), lots of funny statistical discrepancies, etc.

Of course it helps that many people will *immediately* reject any suggestion of corruption, tossing it in the "conspriacy theory" bin.

The election was not stolen. Bush won. (I didn't vote for Bush.) Get over it.

Your faith is touching, but why is it supposed to touch me?

Anyway, in the long run, whether or not this election was "stolen" is small beans compared improving the integrity of the voting system to make sure that they can't ever be stolen... there I think we're in agreement.

Canada has always done it on paper...

[necro2607]

Like the topic says: Canadians vote by writing an "X" in a box on a piece of paper next to a party's name and sticking the piece of paper in a cardboard box.

All I have to ask America is: what's the fucking problem?

Why is electronic voting neccesary? That's a rhetorical question - it's NOT neccesary. I'm more wondering why people tolerate whatever the morons in power dictate. Wake up, you're getting fucked with.

http://www.blackboxvoting.org/ - Visit the site.. it's dedicated to revealing any truth behind possible (woops I mean 99% likely) election fraud.

"Black Box Voting has launched a fraud audit into Florida."

"Black Box Voting is also launching a fraud audit in Ohio."

"Black Box Voting is implementing fraud diagnostics on the state of New Mexico. Information we recently received is indicative of widespread vote manipulation."

"Black Box Voting is requesting legal assistance for a specific county in Georgia. Indications of corrupt voting processes, with possible criminal actions by local officials."

"Multiple irregularities. Need people to take affidavits from election workers, statewide."

Just view the page, and read it. Yup, democracy is still strong in the U.S. ...

But hey, don't take my word for it that fraud occured in the US... http://www.votewatch.us/ee/view_observations Just listen to what these thousands of others have to say about their voting experiences... There are some more fun stories here as well: http://www.michaelmoore.com/electionwatch/

Re:Vote Fraud Smoking Gun

[Izaak]

I think you're missing the point that the counties that were using electronic voting were predominantly Democrat 'controlled' counties and as such would have democrat-leaning elections supervisor (they're elected in FL). Therefore the decision to use electronic voting and control of the machines would have been in the hands of Democrats. If the machines were tampered with, the logical beneficiary would be Kerry.

Actually, the decisions on what voting machines are used, how many are assigned to each precict, where the poll locations are, and the people assigned as election supervisors all falls under the authority of the Florida Secretary of State, a Republican appointed by governer Jeb Bush (the president's Brother). The vote counting was in the hands of Republicans, that is an established fact.

As for any tampering more logically favoring Kerry... I will point out again that the 'smoking gun evidence' found so far shows just the opposite. Every single precinct that was examined in Volusia county revealed that votes had been added for Bush. This was discovered by comparing the signed and dated Nov 2nd poll tapes that the election supervisors were caught red handed trying to throw away. This was caught on video tape with police present.

Good grief, what does it take before people acknowlege that we should at least investigate further!