Too good to check: Statistical analysis shows “Diebold effect” in Hillary NH win?
posted at 2:05 pm on January 16, 2008 by Allahpundit
Someone at Science Blogs did check, though, and pronounces himself amazed at what he found. Trutherism for conservatives (or armed liberals)? Possibly, so let me tear a page from the Sullivan playbook and say that I’m not endorsing it, just “airing it.” In fact, SB has made its data public so consider this an invitation to HA readers to crunch his numbers. Any challenges, debunkings, or — gasp — confirmations will be duly added as updates.
Hillary’s margin of victory in NH: 39% to 36.5%. The alleged Diebold effect? 5.2%.
Blowback
Note from Hot Air management: This section is for comments from Hot Air's community of registered readers. Please don't assume that Hot Air management agrees with or otherwise endorses any particular comment just because we let it stand. A reminder: Anyone who fails to comply with our terms of use may lose their posting privilege.
Trackbacks/Pings
Trackback URL
Comments
Comment pages: 1 2 Next »
Oh that’s just a fairy tale… :-)
D2Boston on January 16, 2008 at 2:07 PM
Knock me over with a straw!
kcluva on January 16, 2008 at 2:10 PM
Further proof that polls, especially if done incorrectly, are worthless.
fourstringfuror on January 16, 2008 at 2:10 PM
Why do people run away and hide at the possibility of the most powerful people in the world cheating to win an election? Why do we act as though it’s absurd to believe it? Why do we completely ignore the fact that it’s been documented many times throughout history that campaigns and individuals who seek power will do ANYTHING to gain it?
Does anyone really believe the Clintons would not do ANYTHING to gain absolute power? Seriously?
Here’s a little something to think about.
Gregor on January 16, 2008 at 2:15 PM
Let’s hurry up and fill this thread with snark before the number crunchers get here!
/ number crunchers :)
RushBaby on January 16, 2008 at 2:16 PM
What if she also has a hand in the NAU?
Allahpundit on January 16, 2008 at 2:17 PM
This should be fun.
libhater on January 16, 2008 at 2:17 PM
C’mon…
Don’t you guys know nothing?
The Clintons didn’t do this. Karl Rove did.
See, the Republicans think they can beat Hillary, and they know they don’t stand a chance against St. Obama, so they oiled up their fixit machine a little for Hillary to get a little head start.
Man…
Keep up with the KosKiddies.
Typhoon on January 16, 2008 at 2:19 PM
Sexism! Help! Help! I’m a victim! Get me a Kleenex!
Waaaaaaaaaaaaaah!
Mean ol’ Men gangin’ up like that!
Waaaaaaaaaaaaaaah!
(Release the Hounds! Clenis! Carville! Sic ‘em!)
heldmyw on January 16, 2008 at 2:19 PM
David Freddoso mentioned this last week. I have wondered what happened to the story.
flyfisher on January 16, 2008 at 2:20 PM
It’s getting close to the point that I really don’t care anymore, but if I said what I thought of that comment you would ban me. I become more amazed each day that you’re allowed to blog on Michelle’s sight.
Gregor on January 16, 2008 at 2:20 PM
site
Gregor on January 16, 2008 at 2:21 PM
New Hampshire Primary recount underway…
bnelson44 on January 16, 2008 at 2:22 PM
Allahpundit on January 16, 2008 at 2:17 PM
Are you suggesting that voter fraud is inconcievable in an American election?
billy on January 16, 2008 at 2:22 PM
On Clenis! On Carville! On Fall Guy and Fall Guy! On Spinster! On Handler! On Race Card and Gender!
BKennedy on January 16, 2008 at 2:22 PM
this has been all over the ron paul forums as well
2 precincts reported they had 0 votes for paul. The town of Sutton reported zero, but had 31 votes; the town of Greenville reported zero, but had 25 votes. The two towns had misreported results affecting exactly the same candidate in exactly the same way.
offroadaz on January 16, 2008 at 2:22 PM
E-mail her. Be sure to tell her how disappointed you are that I’m allowed to mock nutbar conspiracy theories.
Allahpundit on January 16, 2008 at 2:23 PM
This gets old. If you don’t like the results cry the election was stolen. There are examples of stolen elections (Gregoire in Washington state), but I am more inclined to believe exit polls are worthless. Wasan’t it Ohio that had exit polls showing John Kerry beat Bush.
bopbottle on January 16, 2008 at 2:24 PM
DIEBOLD! FRED REALLY WON IN NH! THE NWO AND DIEBOLD RIGGED THE ELECTION! DIEBOLD!!!
Hollowpoint on January 16, 2008 at 2:25 PM
You’re not the only one. Although most narcissistic faux intellectuals are on the left I suppose we should treasure the few we have on the right.
Capitalist Infidel on January 16, 2008 at 2:26 PM
Didn’t Kucinich call for a recount already there? That should clear it up, yes?
Spirit of 1776 on January 16, 2008 at 2:27 PM
Yeah, we know… except they corrected their mistake the next day, which wasn’t enough to stop the Rontards from going insane and calling in threats and accusations of a conspiracy.
Hollowpoint on January 16, 2008 at 2:27 PM
Yes he did.
TroubledMonkey on January 16, 2008 at 2:28 PM
No need. She’s bombarded by email from readers who think you’re an a$$. For you to suggest that election fraud is a “nutbar conspiracy theory” shows that you are completely ignorant of reality. Every year there are CONVICTIONS in the real world of groups and individuals for elections fraud, including individuals responsible for SUPPLYING voting machines. But you want to mock me for it. Why would I be surpised. You make a habit of mocking Christians. I’m sure Michelle just loves that.
*Insert worst possible obnoxious degrading filthy insult here, directed at you*
Gregor on January 16, 2008 at 2:28 PM
One of my favorite traditions on voting day is clearing my mind of any observations & decisions I’ve made in the months, weeks– even hours– before casting my ballot, and sitting down to calculate the statistical probability of my voting for any particular candidate.
Oh, and the other day, there was an 80% chance of rain, but it didn’t rain… DIEBOLD!
a4g on January 16, 2008 at 2:28 PM
Aw. You’re an AP-hater of longstanding, I know. Good for you for hanging in there.
I guess it’s time for the usual reminder of your options:
1. E-mail Michelle and ask her to replace me. If she sees enough demand, she’ll change her supply.
2. Skip my posts and only read Bryan’s.
3. Don’t read the site.
Allahpundit on January 16, 2008 at 2:29 PM
Is she? That must be more of that secret knowledge you’re privy to that’s been withheld from me. Well, I’ll update my resume just in case. Michelle’s not about to sacrifice her business for me, so the axe must be about to drop any minute now.
Allahpundit on January 16, 2008 at 2:32 PM
Forgive my ignorance, but what is a “Diebold Effect”?
Weebork on January 16, 2008 at 2:35 PM
I want to know how you are going to work in “I Tivo The View”
Spirit of 1776 on January 16, 2008 at 2:38 PM
I, for one, am glad fine blogger Allahpundit takes a stand against all nutball conspiracy theorists.
But really Allah, why did you torture us with so many View posts? ;-).
BKennedy on January 16, 2008 at 2:38 PM
I am going to complain about your anti-Yankee bias.
kcluva on January 16, 2008 at 2:38 PM
http://www.youtube.com/watch?v=7-Z8sGp8Zwo
jummy on January 16, 2008 at 2:42 PM
It appears to be a rather luddite suspicion of devilish electronics, fueled by the nonsensical belief that you can weave some magical statistical spell upon tealeaves and deduce conspiracies.
Utter bollocks, of course.
The only way to be sure is to perform a formal integrity/security audit upon the executable code in these machines. I do not know if this has been done, or can be done.
LimeyGeek on January 16, 2008 at 2:42 PM
The good news here is that there are no into-the-ether electronic-only voting machines in New Hampshire — all votes are physical votes that are then read electronically or manually. So the recount will at least has the possibility of showing a disparity. And if there is no disparity, we’ll be able to drive a stake through the Diebold conspiracy theories.
Mark Jaquith on January 16, 2008 at 2:44 PM
LEAVE AP ALLLLLOOOOOOOOONNNNNNNNNNNNNEEEE!!!!!!!
Zetterson on January 16, 2008 at 2:46 PM
Uh, Gregor has a point. Election fraud isn’t out of the realm of possibility.
That’s not to say that in this case the NH primary was rigged, just saying it’s not impossible and there is a reason recounts are allowed. How can you judge if you haven’t examined the evidence yet?
BlackCapitalist on January 16, 2008 at 2:46 PM
If it’s shown that the Diebold machines were influenced in some manner, it adds credence to the Democratic Party claim that in the 2004 presidential election Ohio was similarly influenced in Republican favor.
NNtrancer on January 16, 2008 at 2:46 PM
Limey,
Thanks for the explanation. Makes more sense now.
Weebork on January 16, 2008 at 2:47 PM
From what I gathered at reading the site, the analysis was a comparison of actual votes in precincts using Diebold, and those that did not use Diebold. I don’t think it had anything at all to do with exiting polling.
They found that Hilary did much better in Diebold precints than non-Dielbold precincts, even after trying to correct for urban vs rural, closeness of the race in that precinct, etc.
This is worthwhile to talk about, and it doesn’t have to be a conspiracy theory. Maybe certain voters are put off by electronic machines. Who knows. But looking carefully at actual data can certainly be interesting and isn’t a sign of nutbarhood.
The fact is that repeated studies have shown it is painfully easy for someone with the correct data card to adjust vote totals in these electronic machines. And you cannot do recounts in most of these machines, because all you have is an electronic file with the vote totals. There are no paper ballots to recount. We should use episodes like this to push for picture IDs for voters, and for paper ballots that can be recounted later.
Clark1 on January 16, 2008 at 2:52 PM
It’s all just a silly nutbag conspiracy. Absolutely no facts to it. It just never happens. You’re right.
By the way …
Guess who supplied all the voting machines in Louisiana.
Scroll down to the highlighted sections.
Who is Ransom Shoup? He was convicted in 1979 for voter fraud.
But nothing to worry about. Voter fraud doesn’t happen. The Clinton are honest law abiding people. There’s no reason to believe any of our fine government leaders would ever do ANYTHING illegal in order to gain power. Naw!
Gregor on January 16, 2008 at 2:52 PM
Isn’t it obvious? He’s in on it, he’s part of the conspiracy, man!
Basically they found that Hillary’s numbers were higher in districts that used Deibold machines. They’re trying to weed out all the other variables and are currently coming up with a 1 in a 1000 chance that it’s a random result.
trubble on January 16, 2008 at 2:53 PM
I’m a software engineer. It’s easy as pie. Don’t believe me (because you shouldn’t blindly believe anything you read on the internet, you should check it out, ponder it, weigh the evidence, etc.) But we all know someone who works with software or knows someone with experience in the field, just ask them, you’ll see. The KEY to software having integrity are the checks and balances in the programming. As easy as it is to hack software BY THE DEVELOPER, it is that easy to VERIFY whether or not the software has been hacked. One person cannot do this alone I SUCH A WAY THIS IT IS UNDETECTABLE. it can be done, of course, but if it is done, it can be detected. So there would have to be a conspiracy to do this, where everyone involved keeps quiet. Not impossible by any stretch, but goshdarn hard to get away with.
JustTruth101 on January 16, 2008 at 2:53 PM
I don’t know if someone could rig the Diebold machines, but anyone from the Memphis area knows elections can still be stolen. Last year poll workers pled guilty in an attempt to help steal a state senate election for Democrat Ophelia Ford (Harold Ford, Jr.’s aunt). Ford beat the Republican by a whopping 13 votes with the help of dead voters.
flyfisher on January 16, 2008 at 2:54 PM
[d*&% typos...sorry...should say:]
IN SUCH A WAY THAT IS UNDETECTABLE
[i'll start to use the preview...]
JustTruth101 on January 16, 2008 at 2:55 PM
Oh d&mn!
Michelle Malkin must be a nutjob conspiracy theorist. Huh Allah?
Gregor on January 16, 2008 at 2:56 PM
did anyone ever see the “proof” that diebold machines can be hacked (thus “proving” that diebold “lied” when they said they had made their machines secure)?
a couple of stanford students (thus making this a “stanford university study”) made a video showing how “easily” an old model diebold machine can be hacked. it started thusly…
“first, unplug the machine and flip it over. then, remove the eight screws holding the chassis ….”
…and on and on proving that you’d need to do everything but break out the soldering gun to get these things to take malllicious code.
and of course, the only people who have been proven to have hacked a diebold machine have been the “black box voting” nutters.
election fraud does happen. but i’ve worked several elections in the sorts of places where i might exppect to see fake registrations claimed and the like, but the worst i’ve seen are precinct captains passing out pluggers too close to the polling place.
jummy on January 16, 2008 at 2:58 PM
Voter fraud and ‘Diebold effect’ are two totally different things. I have no problem believing in nefarious schemes to get dead people voting, voting multiple times etc. These are human corruptions though, not electronic.
Do the Diebold machines not produce a paper trail also? I thought they did. Pretty easy to manually recount the printouts if so.
Otherwise, audit the code. See if it is possible to exploit the machines.
I’m not discounting the theory, but statistical mumbo-jumbo simply does not hold water. More rigour is required before anyone starts throwing accusations around.
LimeyGeek on January 16, 2008 at 2:58 PM
Hillary cried! Diebold lied!
ronsfi on January 16, 2008 at 2:59 PM
I guess THIS VIDEO is just complete nonsense. The entire demonstration must have been created through CGI.
Gregor on January 16, 2008 at 2:59 PM
why are you retarded?
what does the fake registrations thing have to do with voting machines being rigged in a massive conspiracy?
jummy on January 16, 2008 at 3:00 PM
When the fraud topic came up on Real Time the other night, Maher pretty much said to Tony Snow the Pubbies were helping
Hillary win. Snow couldn’t believe it. Instead of bringing up old relaible Gore. He brought up Muskie for some reason. So the playbook has been set in motion.
PowWow on January 16, 2008 at 3:00 PM
What year is it again? Isn’t LA electronic now? And you are talking about Shoup?
lorien1973 on January 16, 2008 at 3:02 PM
Can I ask a dumb question? What percentage of voting machines in the state were using Diebold voting machines? And, were all the Diebold voting machines in NH the same model and ballot design? (I don’t know that much about electronic voting machines to know.)
It just seems awfully silly to me that there is this odd “Diebold” effect that just happens to have a 5% Hillary bias, a -4.2% Obama bias, and (presumably) a -0.8%-ish Edwards bias. And is there a Diebold bias on the Republican side?
Outlander on January 16, 2008 at 3:02 PM
Did anyone think that perhaps , if the two options of voting machines in NH are Diebold and paper, that perhaps on the paper ballots people were stuffing it for Obama?
It could be “the other way around.” I think there is more chance for crime with paper ballots than machines.
Vincenzo on January 16, 2008 at 3:03 PM
Some do, and some don’t, and the question isn’t whether it’s possible to do a recount. The question is why they continually REFUSE to allow it. Be sure to let me know when you hear that New Hampshire is going to allow the paper ballots to be looked at.
Gregor on January 16, 2008 at 3:04 PM
Yeah! That Northern Arizona University thing is suspicious.
ronsfi on January 16, 2008 at 3:04 PM
Fred really won. lol
Drunk Report on January 16, 2008 at 3:06 PM
If you remove the Diebold Hillary bias, it seems the polls were on the money after all.
flyfisher on January 16, 2008 at 3:08 PM
oh yes! diebold explains whatever results you’re unwilling to accept. in fact, diebold explains how bush won in 2004. it even explains “anomolies” in elections which used punchcards, like florida 2000 prior to which this one nut cllaims that he was approached by an associate of bush to rig the electronic voting machines ….which florida didn’t even decide to purchase until 2003!
jummy on January 16, 2008 at 3:08 PM
Unfortunately security audits are done in private, so it’s just as faith-based as trusting that the code is secure to begin with.
The first step would be to make the source code to these executable binaries public. Anything else is “security by obscurity,” and makes it much more likely that a flaw will go undetected. This is a major contributing factor towards why 80% of Windows PCs have viruses/malware and why the Open Source GNU/Linux platform is used so extensively for Internet servers (including this one). Bugs and flaws detest sunlight, and thrive in dark corners.
The second step is to produce a physical record of the vote that the voter can personally verify. I’d go a step further and have it print two copies — one for the voter to take home.
There are also issues of physical security of electronic votes. For instance, some machines do not take adequate steps against physical tampering, which could make it possible for someone to swap out a memory card.
Back to letting people keep a physical receipt… the receipt could have a randomly generated and anonymous number associated with the vote, and printed on it. All votes would be stored along with this number. Simple increments are made for tampering. What’s more, the voting machines could, when printing out the physical receipt, electronically transmit the vote (and the random number) to a State tabulation computer, whose numbers could be physically accessible to third parties. So the Republican Party could set up a “verify your vote” site where you’d enter your receipt number and you could verify that your vote was counted correctly (it could also store things like the time of your vote, and the location). Same for Democrats. (The reason you’d likely have partisan sites is for trust reasons, but other third parties could do this too). Picture this: I cast my vote, and get my receipt number. I whip out a cell phone and call some third party verification service. I enter in my number, and right then and there get third-party verification that my vote has been correctly counted.
What you’d end up with is an anonymous unique tally of votes that voters could use to verify their vote. It could also be used to guard against ballot-stuffing (have observers count people at polling locations and compare to the electronic tally of voters for the polling location).
The essence of what is needed is openness. The process (including source code and physical procedures) needs to be open to intense scrutiny. The votes need to have a verifiable physical record.
I’m not saying that nefarious things are happening… it’s just that with black box systems, there is no way to prove that they’re not happening. So the seeds of doubt are perpetually sown, and the fruits of malice are a tempting treat for would-be riggers.
Mark Jaquith on January 16, 2008 at 3:09 PM
Is there any consistency regarding ‘automatic recount’ rules? Within 5%? 2%? Can any candidate elect to pay privately for a recount if an automatic recount is not warranted? I imagine it would be expensive, so you’d better be sure of your theory beforehand!
LimeyGeek on January 16, 2008 at 3:10 PM
You’re asking why “I’M” retarded? Is it really tough to figure out that my point is that election fraud is happening EVERY SINGLE day of EVERY SINGLE election? Are you really suggesting that those who would commit election fraud follow some type of unwritten rule of principles? You know? Maybe they have a list of acceptable election fraud procedures, and a list of those that are NOT acceptable?
It’s really quite simple. History has shown without a doubt, that people will do ANYTHING to gain power. ANYTHING! As I’ve asked before … are you saying that you believe the Clintons would not do anything within their power to gain power?
I’ve provided links that show some of the actual individuals who manufacture and supply these voting machines are convicted felons, having been found guilty of election fraud. But you don’t seem to think that means anything. And I’m retarded. Yep.
Gregor on January 16, 2008 at 3:10 PM
4. Keep reading AP’s posts like a tool. After all, much of Rush’s ratings are due to liberal audience members just waiting for that outrage fix. Why should AP be any different?
calbear on January 16, 2008 at 3:11 PM
Absolutely correct. Let us not forget Florida 2000 either where paper ballot votes were subject to interpretion of elections officials- based on the pregnancy of the chad.
Electrons are what they are and can’t be pregnant, hanging, or otherwise be open to interpretation.
highhopes on January 16, 2008 at 3:12 PM
are the diebold voting boxes tied into a network?
offroadaz on January 16, 2008 at 3:13 PM
In New Hampshire, a candidate can request a recount as long as they’re willing to pay for it. Kucinich has done so, and has given the state $27,000 toward the $70,000 it costs to do a statewide recount.
Slublog on January 16, 2008 at 3:15 PM
Watch the video I posted. People keep asking questions that indicate they simply refuse to follow links. The video also answers the question of what happens if someone actually questions the results, and who is responsible for verifying them.
Gregor on January 16, 2008 at 3:17 PM
Is the Kucinich recount going forward? If so, we’ll hopefully gain some hard data to judge the ‘Diebold’ theory by.
LimeyGeek on January 16, 2008 at 3:17 PM
It describes the inexplicable effect the word “Diebold” has to cause overcaffinated, childish, moms basement dwelling WOW junkies to think they are master sleuths and glaze over in tantric spasms of overestimated self importance.
Tragic really.
Syn. Haliburton, Cheney, Wolfowitz, Single Bullet Thoery.
ronsfi on January 16, 2008 at 3:17 PM
theory, sorry
ronsfi on January 16, 2008 at 3:18 PM
Yup. Just started in Manchester.
Slublog on January 16, 2008 at 3:19 PM
Has anyone considered that the diebold votes may be accurate and that the hand count tallies were manipulated to give Obama an edge? This demonstrates the gender bias that is ingrained in society. There’s suspected hanky panky afoot at the Circle K and everyone assumes that the woman did it.
rw on January 16, 2008 at 3:20 PM
“Back to letting people keep a physical receipt…”
why?
70% would stay in the voter’s possession for precisely as long as an atm reciept does.
are you going to recall these receipts in case of a recount?
if the boe doesn’t get 100% of these second priontouts returned to them, do you think the “injured” party in the election who demanded the printout will understand, or use it as an opportunity to further debase the election?
i understand the meaning of the printout. i think it’s a solid second firewall. but the people involved in this black box voting nonsense will never be satisfied with anything because their actual aims lie with disrupting and discrediting u.s. elections rather than safeguarding them.
i had this put to me rather specificly once. this smelly hirsuit hippie came to the table once, noticed the sequoia machine set up and immediately started talking nutty about diebiold and how bush was going to change his vote around. i told him that it was equipment manufactured by a company in which the venezuelan government is a partial stakeholder. he actually said, “oh, well, then at least we know its going to count my vote right then.”
jummy on January 16, 2008 at 3:22 PM
no they are not. they broadcast, one way, the election results to the board of election at the end of the night when the election workers configure it to do so.
men in black vans with transmitters do not and cannot go around inserting phony votes or mallicious code remotely, as the nutters assert.
jummy on January 16, 2008 at 3:27 PM
Because all of our leaders are honest and would never do anything illegal to gain power.
Gregor on January 16, 2008 at 3:27 PM
Yeah- since SOME voter fraud allegations have merit, ALL voter fraud allegations have merit.
How about we just let the facts come out instead of acting like an annoying whiner who automatically believes the almost completely unfounded allegation of intentional fraud, OK, Mr. Black Helocopter?
It’s possible that the vote count was accurate. It’s possible that the vote count was inaccurate but the result of negligence rather than malice. It’s possible that there was in fact fraud occurring. Right now, we have no way of knowing which is the case. Well, at least us rational people don’t.
Hollowpoint on January 16, 2008 at 3:29 PM
this is arguing with gregor:
YOU – you know, i really find kittens to be cuddly and adorable.
GREGOR – YOUR STUPID! ALIGATORS ARE SCALEY SLIMY CARNIVOROUS MENACES!!!
YOU – what does that have to do with kittens?
GREGOR – U SED THAT AMIMALS ARE CUTE AND THET THERE NEVER SCALEY OR DANGEROUS@!! WHOZE THE LIER NOW???
jummy on January 16, 2008 at 3:35 PM
Please show me where I suggested this. I simply jumped on Allah because he mocks the “possibility” that the election MIGHT have been effected by fraud. Nowhere have I said that it DID happen.
Fixed it for you.
Thank you for agreeing with me. You might want to duck! Allah should be mocking you any moment for admitting there’s a possibility of fraud. You nutjob you!
Gregor on January 16, 2008 at 3:39 PM
JustTruth101 on January 16, 2008 at 2:53 PM
Exactly. To update the tables storing the vote count would require a conspiracy of DBAs, application developers, security administrator and possibly network security guys. I assume the Diebold system has audit tables and calculations that compare sub-totals to final vote tallies and kicks out audit reports when the numbers don’t add up. It wouldn’t be as simple as updating the final tally. You’d have to update all the sub-systems and joining records. Also the conspirators would have to be very careful with the math not to create impossible situations, such as more HRC votes cast in one polling station than voters, or more votes cast in a time frame than is possible. Even a basic audit would catch all but the most well-planned and flawlessly executed conspiracy.
This ain’t Ocean’s 11. One lone geek can’t hack the database over an iPhone and give HRC the election. Traditional voter fraud involves dead voters and ballot box stuffing but directly accessing the database records and updating the values to throw the election is a whole different animal.
Vote Sauron 08 on January 16, 2008 at 3:41 PM
At least you’re not bitter and thats what counts.
ronsfi on January 16, 2008 at 3:43 PM
You would be 100% wrong, but I guess we should just accept your “assumption” as fact. You might want to take the few minutes to watch the video I posted which you obviously ignored before posting your “assumption” as fact.
Gregor on January 16, 2008 at 3:43 PM
I’m going to make a bold prediction that the recount will show the Diebold-ballots were counted more accurately than the hand-counted ballots. If the Diebold effect is real, I think it says more about the subconcious of a typical Democrat voter than it does about any possible voter fraud.
RightOFLeft on January 16, 2008 at 3:48 PM
Ooohh! DIEBOLD, I shudder at the thought of its evil. ;)
SCGOPgirl on January 16, 2008 at 3:48 PM
What?!
I’m returning this i-Phone.
Spirit of 1776 on January 16, 2008 at 3:49 PM
You don’t “hack the database.” The memory card is pre-programmed on the machine to spit out whatever result the programmer desires, regardless of the votes that are actually input. There is no “hacking.” There is no “computer data base.”
By the way …
Election officials have absolutely no access to these cards and have no ability to check the programming. The only person with access for programming is the representative of the machine provider. As the video I posted shows, many of these “providers” have prior fraud convictions.
The ONLY way to verify the machine’s accuracy is to “run a test” or to check the actual paper ballots. Many machines HAVE NO PAPER BALLOTS, and the only person allowed to run a test of the machine is … the manufacturer.
Gregor on January 16, 2008 at 3:56 PM
Just taking a glance at it, there are a few issues with the data analysis… first, they are using linear regression, which is not well-suited to this type of study for a number of reasons (for one, linear regression assumes a continuous dependent variable, something that can theoretically go from negative-infinity to positive-infinity… Clinton or Obama’s values only range from .00 to .58 or so). This can lead to a lot of problems in doing a quick ANOVA table (which I did and got results fairly similar to the ‘Diebold effect’). There are a lot of diagnostics that need to be done to clean up even a basic analysis, especially since the key variables have little variation (the Diebold variable is binary… either they have it or they don’t in the region, and the aforementioned lack of range in the dependent variable). When the two variables have such limited variation, the natural inclination of the ANOVA results is to see the small variables to be comparable and thereby linked. As an aside, the “Diebold effect” for Kucinich is about -2.5% (whose range was .00 to .17), which seems pretty silly on its face but is consistent with the “effect” of these variables overall. Without looking at anything else (though I might if I have the time later), I would say this is an artifact of the data rather than a true effect.
Wineaholic on January 16, 2008 at 4:02 PM
Can I suggest that the reason why this story deserves snark and mockery is that the Left has proven themselves to be absolutely bug-eyed crazy when it comes to election day paranoid hysteria. They did it to us in 2000 and 2004, and it’s amusing to see them start to eat their own with the same delusional mania in 2008.
Yes, we all know that election fraud occurs. In some cases (1960), it even gave the White House to the rightful loser, but the real reason why the Left goes ape-poop about it is because they’re so full of themselves and an unshakeable faith in the righteousness of their desires rather than because there’s a sensible reason for them to bleat so crazily.
That’s why this is funny.
Kensington on January 16, 2008 at 4:05 PM
Gregor, err, hate to break it to you buddy, but Vote Sauron 08 is right. The fact that you actually typed:
is beyond ridiculous…your very own video refutes that…there has to be a database to record the votes for the count fraud to be based on.
And just out of curiosity, what is your expertise in software development? Have you ever worked on a software development team writing interdependent modules?
JustTruth101 on January 16, 2008 at 4:06 PM
Well, this has been a fun thread to read, I’d like to vote for the following one-liner as my personal favorite:
But seriously, I did a little reading about this primary election last week, and while I don’t have exact figures in my head, the disparity between exit polls and the actual vote was greater than ever before in the history of exit polls.
And only on one side.
The Bush/Kerry exit polls that had the left so bent out of shape was only off by about 3 points…the Clinton/Obama vote was off around 11 points.
Very interesting, indeed.
Dorvillian on January 16, 2008 at 4:23 PM
nonsense.
the cards are removed from the machine and sent with the opscan ballots to the boe.
not only do you not know what you’re talking about, you’ve clearly never worked with these machines before.
jummy on January 16, 2008 at 4:26 PM
You’re misrepresenting the conversation. Of course there’s a database, but it’s not tied to the machine as far as actually reading the votes. The database mentioned in the video refers to the information on the election, such as the questions and candidates. As far as actual “votes” are concerned, they are all stored on the memory card and again … the card can be programmed to spit out any result desired. You can check the “database” all you want and it’s not going to verify actual votes. The only way to verify actual votes would be to compare the physical paper ballots to the results spit out by the machine.
As you can see by the video, the votes are actually meaningless if the card is pre-programmed to spit out a pre-determined result. And remember that many of the new touch screen machines do not have paper ballots to check.
Gregor on January 16, 2008 at 4:31 PM
Once again, you’re misrepresenting the conversation.
My comment was referring to “access to program or verify” the card’s accuracy. The on-site poll workers only handle the cards as far as inserting them into an envelope and lockbox, depending on local rules. They do not program the cards. They do not verify the cards. Any and all questions or troubles are referred to the machine’s supplier.
Yes, I know exactly what I’m talking about and I have family members who have worked polling locations using machines for the last two elections. Have YOU?
Gregor on January 16, 2008 at 4:37 PM
As I continue to tool around with this (instead of doing my actual work), I am starting to see two major things — first, linear regression is really bad for this, and second, there is likely omitted variable bias. The models I get have a tendency to be over-fitted too quickly. That is indicative of unobserved heterogeneity in the sample. There are other more complicated statistical models probably better for this, given what I am seeing, that would not over-estimate the significance of the Diebold variable. It goes back to a point I noticed before… there is not much actual variation in the dependent variable, so everything has a tendency to be sensitive to even slight changes. Independent variables with large range of values would be less impacted, but independent variables with small range would be highly prone to overestimation. Keep in mind, the Diebold variable that I have seen only has two values, yes or no. So again, I would say that these results are more a function of the data application itself and not of an actual link (let alone a causal link).
Wineaholic on January 16, 2008 at 4:39 PM
During 2004, I did a lot of research on these claims of computer voting fraud for an article that was ultimately never published. I spoke with a lot of the experts and campaign officials about this, including the head of the FEC.
Yeah, in theory these machines are susceptible to vote-rigging, but there are processes in place to prevent it. The computer experts knew about the machines, but not about the processes.
Because the votes are tabulated from each individual machine before they are sent to any main computer, you would have to have access to a significant number of machines in order to rig the vote in a meaningful way (and not be detected). It would take a pretty huge conspiracy to pull-off a voter fraud in that manner.
The question I kept asking was this: since you need to have access, what’s the difference between rigging a vote this way and stuffing a ballot box? And, given the complexities involved with rigging a machine, wouldn’t it be easier to stuff a ballot box?
No one could effectively answer that question.
Spolitics on January 16, 2008 at 4:44 PM
i was a pollster in High School. they paid us a small amount for each household polled..door to door. this was 1979, so it must have been around 50 cents per household we polled. i was pretty broke, so on occassion if no one was home, or if the door was answered by a kid, or if the person answering the door was not interested in being polled, guess what? i filled the responses in for them anyway. that way i still was paid for that interview. i’m not proud of that, but the methodology really was a joke. i gaurentee the same thing goes on today.
DrW on January 16, 2008 at 4:47 PM
VERDICT: New Hampshire is unable to document its chain of custody properly, lacks written procedures, its secretary of state has said he doesn’t know where its memory cards are, and LHS has been encroaching on state elections with near-total control.
Gregor on January 16, 2008 at 4:50 PM
Sounds like a problem no matter what voting system they use. Remember all those milling ballots in 2000?
Spolitics on January 16, 2008 at 4:53 PM
Wrong. It’s the “cards” that are tampered with. Not the individual machines. The cards can be pre-programmed and the provider is the only person able to verify that programming.
Gregor on January 16, 2008 at 4:56 PM
re: VERDICT
BlackBoxVoting.org yikes, do some research on these folks, lot’s of threads at DU and conservativeunderground
it’s my understanding that these are electronic scanners which can surely be hacked but they do have actaul paper ballots that can be checked.
a level headed summary can be found here:
http://www.correntewire.com/even_more_voting_follies_nh_ed
windansea on January 16, 2008 at 5:00 PM
Comment pages: 1 2 Next »