Thursday, December 16, 2004

At last, a vote fraud update

It's been too long since I put one of these together...

Kerry/Edwards makes it official: The campaign has formally asked Ohio Secretary of State Ken Blackwell to investigate vote tampering in that state. The request specifically goes to the December 10 testimony regarding machine tampering in Hocking County. This refers to the Triad computer specialist who absurdly claimed that "battery failure" caused massive data loss. (We now learn that the witness to this incident was one Sherole Eaton.)

This AP story discusses the troubling Triad incident in some detail.

Blackwell will no doubt respond with his usual tardiness. If fascists can make trains run on time, Republicans can make elections run slowly. That is the main difference between a fascist and a Republican.

Cliff Arnebeck: I still recall how Randi Rhodes told him (a la "Rocky") -- "Win. Just win." He didn't, alas. His lawsuit alleging voter disenfranchisement was tossed out on a technicality: The complaint contests two separate elections, while state law specifies that the focus should remain on but one at a time.

Fortunately, the complaint can be refiled.

Witness against Diebold: A little-noted aspect of Arnebeck's effort is the claim by one Catherine Buchanan that "Diebold reprogrammed voting machines while present at her local Board of Elections." According to the complaint, "the programming involves deleting information from the memory cards in the central tabulating machine."

This goes to a point many of us have been making since the first week of this scandal: The central tabulators are the most vulnerable link in the electoral chain.

More on Ohio: Extraordinary reporting on the recount drama comes to us via Bob Fritakis, Steve Rosenfeld and Harvey Wasserman of the Free Press:

Jackson said Kerry wanted “forensic computer experts” to examine voting machines, especially those using optical scan technology, because in other states, notably New Mexico, Bush had won all the precincts with that voting system in place. Kerry also wanted to examine 92,000 ballots that recorded no vote for president, and 155,000 provisional ballots that were rejected.
155,000 rejected provisionals? Did I miss something?

(Incidentally, there has been some discussion of the possibility that in New Mexico and Ohio, undervotes were programmed to go into the Bush column.)

More from the Free Press account:

But early responses from the counties to Freedom of Information Act requests for their voting records indicate such an effort may already have been sabotaged. Shelby County officials have admitted to discarding key election data. One county referred requesters to the software company that programmed the county's voting machines, saying the company's permission would be required for access to a recount, as the code is proprietary.
And:

A second brief was also filed Monday, seeking a temporary restraining order to block Republican presidential electors from meeting until the recount was done and the challenge was litigated. It focused on “overwhelming statistical evidence” that pointed to “statewide fraud allegedly conducted at the direction of Secretary of State J. Kenneth Blackwell.”

The TRO filing was primarily based on national and statewide exit poll data, which was the extensive, non-partisan polling done by a consortium of the nation’s major news organizations. Expert affidavits accompanying the brief said an analysis of exit poll data found that the final vote tallies in all but the most contested battleground states mirrored the exit poll’s predictions. The experts said it was unlikely the exit polls could be so accurate in some states while significantly wrong in others. They said election fraud was the only plausible explanation for the discrepancy.

The TRO filing identified exactly when they believe the fraud occurred – at about 12.30 a.m. on Wednesday, Nov. 3. At that time of night, Ohio’s final voting returns were being tabulated at regional and county offices. It was about this time that the Ohio exit poll data – posted on websites such as CNN – put Bush ahead of Kerry, even though the exit polls expected Kerry to win with 52.1 percent of the vote.
And:

Dr. Ron Baiman, a statistician from the University of Illinois, Chicago, confirmed that the odds on vote counts diverting from exit polls as they did the night of November 2 were on the order of magnitude of millions to one. Baiman told freepress.org that the odds of the exit polls being wrong in the key battleground states of Florida, Pennsylvania and Ohio alone were "155,000,000 to one."
And:

In one Cleveland ward, he said, 51% of the provisional votes cast were thrown in the trash, virtually all of them from African-Americans.

Eve Roberson, a former election official from Santa Rosa, California, testified that while working as observer at precinct 354 in Wilberforce, home of Central State University, she witnessed conscious fraud aimed at a student body that went 95% for Kerry. Election officials used an inconsistent, discriminatory set of demands for Wilberforce students to register as opposed to those used in white precincts in Greene County.
I hope I have not quoted too much; read the whole piece for yourself.

Clint Curtis: By now, I presume that most of you need no introduction to this whistleblower. A few straggling newbies may not yet know that Curtis, in a sworn affidavit and while testifying under oath, claims that in 2000, Republican Florida Congressman Tom Feeney asked him -- or rather, asked his former employer at Yang Enterprises (YEI) -- to cobble together a prototype for a vote theft program. If the allegation is untrue, Curtis has made himself open to one hell of a libel suit, a risk undertaken for no fathomable reason.

Obviously, this tale for 2000 impacts the more recent election, not least because any Republican who requests such software indicates his party's contempt for democracy.

After a lengthy delay, Yang came up with a reply to the accusations. Brad Friedman, who has been covering the tale in depth, has published a reply to the reply on his site. For the full story, you'll want to turn to Brad's page. A few high points:

1. Yang accuses Curtis of being a disgruntled employee, an accusation that Friedman easily counters. Curtis was not fired.

2. Yang says that another employer accused Curtis of participation in "a corrupt and dishonest scheme." This reference goes to an unrelated dispute over alleged copyright, in which Curtis and another man sued each other. The matter was settled out of court. Of course, if accusation suffices to prove dishonesty, then Yang itself stands damned.

3. Curtis affidavit also refers to a matter related only tangentially to the vote fraud accusation: A Taiwan national named Henry Nee, a.k.a. Hai Lin Nee, worked for Yang (a firm whish receives a number of government contracts) was involved with the illegal export of missile components to China. Nee was, in fact, convicted of this charge on October 7, 2004. Yang denies that Nee ever worked for YEI. Brad Friedman, however, publishes documentary evidence that such employment did, in fact, occur.

4. YEI notes that no Florida elections used touch-screen systems until 2001. But Curtis never claimed that his program was anything other than a prototype, one which might well have applied to a central tabulating system designed to count (say) optical scan ballots.

So far, YEI's response is as weak as pint of near-beer.

Danger signs? Right-wingers who profit from allegations of Democratic wrongdoing never seem to worry about covert retribution. Carlton Sherwood (the hack-tacular Moonie "journalist" who attacked John Kerry as a baby-killer) will no doubt die in bed at an advanced age -- as will John O'Neill, Larry Nichols, Linda Tripp, Lucianne Goldberg, and everyone who ever dripped pus all over the editorial pages of the Wall Street Journal.

By contrast, those who would undermine right-wing operations tend to find life difficult and, occasionally, short.

We note (to cite but the most recent example) the sad case of Gary Webb (with whom I once briefly corresponded), who committed "suicide" by firing two shots into his head. Needless to say, only "conspiracy nuts" consider that situation odd.

Those with longer memories may recall such names as Karen Silkwood, Danny Casolaro, Anson Ng, Don Bolles, Dr. David Kelly, James Hatfield, Dorothy Kilgallen, Steven Carr...well, one could list names well into tomorrow.

Yesterday, we discussed the harassment of Katrina Sumner, the Ohio recount activist who has been followed and run off the road. Her report reminds us of the sad fate of Athan Gibbs, inventor of the TruVote system -- a "clean" e-vote machine with a paper trail (not to mention a company owner wedded to the idea of democracy, not Christian Dominionism). Gibbs died last March 12, when an 18-wheeler ran his Chevy Blazer off the road.

When a similar accident took the life of Karen Silkwood, investigators heard rumors that truckers had been used as assassins on previous occasions. No-one has ever proven those rumors, of course.

Wayne Madsen has boasted of his protection by Mr. Smith and Mr. Wesson. Good to hear -- but let's also keep an eye out for Mr. Peterbilt.

3 comments:

Anonymous said...

You say Ron Baiman is a 'statistician' but have you checked on this? I can find nothing published by him in the field of statistics.
http://www.acorn.org/index.php?id=215 lists him as: "Research Assistant Professor at the UIC Center for Urban Economic Development".
However it seems that even that is not now correct. if you go the staff page of the UIC Center of Urban Economic Development, at
http://www.uic.edu/cuppa/uicued/nAbout/team/nteamtext.htm
he is not listed at all.
Perhaps he has now moved to the field of statistics, but before you call him a statistician, I think you should check his credentials.

Furthermore, I should point out that Dr Steven Freeman is also not a statistician. He is a professor of strategy, organization and entrepreneurship, specialising in resilience. As far as I can tell, he has never produced a publication in anything close to the field of statistics. He's researching: "effects of loss and adversity on organizations and individuals". And he is teaching a course entitled "A Systems Approach to Crisis Preparation and Building Organizational Resilience"
In other words, he appears to have simply thrown a pebble into the crowds of disappointed Democrats and is watching what happens.

This is all easily discovered at his home page at:
http://www.appliedresearch.us/sf/

There are, of course, thousands of able and experienced statisticians around. Producing two non-statisticians as evidence makes the case look not only weak but downright stupid.

Anonymous said...

.
.
.
The fact that Dr. Freeman is not a statitian is irrelevant. Dr. Freeman's work has held up to the scrutiny of academics everywhere. He was challenged by a few to re-evaluate his data to include certain biases, but his final conclusion has NOT yet been challenged -- The discrepancy between the exit polls in Ohio, Florida, and Pennsylvania and the actual polling results could not be attributed to statistical error. In other words:

1. There is systematic bias of some kind which is overwhelmingly skewing the exit polls towards Kerry

or

2. There has been COUNTING ERROR

------------------------------------

With respect to the thrown-out case contesting the Ohio presidential results:

This story's bits and pieces (including the AP report) is all tied-together here:

<<<< Newsclip Autopsy >>>> FOCUS: VOTERGATE

HALF-TRUTH: AP Fails to Report Diebold Vote-Tampering Testimony & Moyers Conflict of Interest

http://newsclipautopsy.blogspot.com/2004/12/half-truth-ap-fails-to-report-diebold.html
.
.
.

Anonymous said...

You've got to be kidding. Dr Freeman's paper is about a first year undergraduate level piece of calculation which no academic statistician would bother to read. If you have an academic statistician of any standing who supports this stuff, please tell me who it is.
And yes of course it matters if Dr Freeman is a statistician or not, because that is what he has been described as. If it isn't important, why have people described him as something he isn't?
Basically, as Dr Freeman has no idea what the sample design of the exit polls was (as we haven't been told) he can make no calculations regarding the possibilities of the results. He is assuming pure random sampling which is about as basic an error as you can possibly make. It is so basic that I'm sure that Dr Freeman must know this.