Trail of Electronic Voting Uncovered: A Summary of NC Findings

[ignatzmouse]

I've been asked for a summary of my findings in North Carolina. It is important that the key finding is not lost in the length of my report as it may have uncovered national implications for paperless electronic voting. We suspected it. Here it is.

Quite simply, in a comparative analysis of voting systems in the North Carolina 2004 General Election, counties that used electronic voting produced and unexplainable, unavoidable, and distinct early vote disparity with the rest of the state. The effect was SO GREAT that electronic voting gave Bush an enormous 12% point advantage over Kerry while Optical Scan counties with roughly the same number of votes showed only a 2% margin of difference. How does one account for this outside of technological creativity? Here are the numbers:

Absentee/Early Vote

Optical Scan:
Bush: 249,385 51%
Kerry: 235,175 49% -2%

Electronic Voting:
Bush: 280,354 56%
Kerry: 221,235 44% -12%

What is this? Mandate on order? Even though the DRE's had the largest voting block and HIGHEST percentage of early voting, the DRE's had the WORST margin of disparity between the candidates, a 12% LEAD for Bush at 56% to 44%. To further add fuel to the disparity, the rate of early voting within electronic voting counties was much higher than with any other voting system. Electronic voting counties cast 34% of their votes via early and absentee voting whereas the rate was just 29% among optical scan counties. A high early voting turnout is supposed to favor the Democrats' enormous GOTV campaign and yet in paperless electronic voting, this meant a 12% edge for Bush.

Another revealing study found that the 36 electronic voting counties posted leads for Kerry only 5 times, a dismal rate of only 13.8%, but the rate of Kerry early leads for the rest of the state in all other forms of voting was 33.3%, NEARLY TWO AND HALF TIMES HIGHER.

KERRY LEADS STUDY
E-voting: 36 Counties; 5 early leads 13.8%
Optical Scans: 49 Counties; 16 early leads 32.7%
All other: 14 Counties; 5 early leads 35.7%
(note: Lee County did not post early vote counts.)

Although e-voting dominated the early voting and captured 42% of the statewide votes, electronic results ran wildly counter to all forms of auditable voting. Further, within the electronic voting canopy, 3 of the 5 leads were posted by a single vendor, MicroVote General, in just 7 counties. For the rest of the e-voting vendors, the results were an almost non-existent 2 out of 29, 6.9%. There is simply no justification for this. The largest e-voting vendor in North Carolina is ES&S. ES&S serves 13 counties of which Kerry posted an early lead in but one county...by 234 votes! I can hear you thinking... "It must be demographics. It just has to be demographics." Well, let's look.

DEMOGRAPHIC COMPARISON

GROUP A: North Carolina Statewide
DEM 2,582,462 47%
REP 1,903,119 34%
LIB 12,754 0.20%
UNA 1,021,648 19%
Total 5,519,983
White 4,224,098 79%
Black 1,112,959 21%

Group B: ES&S E-Voting (serves 13 counties: 8DEM, 1REP, 4SPLIT)
DEM 295,384 47%
REP 219,759 34.50%
LIB 1239 0.20%
UNA 117,374 18.50%
Total 633,756
White 487,416 78%
Black 134,289 22%

Group C: Microvote General E-Voting (serves 7 counties: 3DEM, 1REP, 3 SPLIT)
DEM 335,529 44%
REP 269,203 36%
LIB 1786 0.20%
UNA 149,551 20%
Total 756,069
White 550,657 75%
Black 179,733 25%

The figures are nearly identical. ES&S has the same Democratic ratio as the rest of the state. They have a hair-split more Republican to Unannounced ratio. They even have 1% higher Black population. ES&S has 3% more Democrats than MicroVote but also a 3% less Black population. ES&S e-voting counties are dominated by Democratic stronghold counties (8 of their 13 counties) and in the one county where Kerry tabulated a 234 vote lead, it swung wildly to a 45/55 Bush success on election day. In comparison, MicroVote General has seven counties of which only three are Democratic strongholds and yet Kerry posted early leads in three counties by 6%, 8%, and 40% percent margins. This is a dire signal that something isn't just wrong, it's got wrong written into it. And if the electronic vote in North Carolina is in question, it must be equally as suspect in the same hidden hands throughout the country.


Full study in numbing detail ;-) is posted here:

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=69829&mesg_id=69829

[UL_Approved]

1. Back at it

How many statistics have you gone through yet?

[ignatzmouse]

4. Visions of Statistical Plums

I see them in my sleep. I don't know how to calculate it, but I've been swimming in them. To this point, it's mostly been focused on NC as I was asked by BeFree to look it over, and the deeper I looked, the more was revealed. If I can scratch out the time, I would like to look at another state or two.

[roseBudd]

24. How about Warren County Ohio since that is the county that lied

about a Homeland Security threat and locked the press out during the vote tallying. They went over 70% for Bush and were the last county to report in Ohio.

DU member lizzieforkerry testified at the Ohio Voter suppression hearing about the particulars and she can help if you have questions. I have her cell # and email addy BTW.

[BlueDog2u]

2. Case Closed

Send it to all the media. Let's not hesitate. As Bev says, Be the Media. Incredible work, ignatzmouse. Let them try to explain it.

PS -- Ignatz, are you in touch with Stephen Freeman? You guys are coming at this problem from different but entirely complentary methodologies to reach comparable results. I bet he'd be very interested in your work, and or someone could do a statistical analysis of your findings.

[KoKo]

6. Sending it to Steven Freeman is a good idea. Putting heads together

might give even more weight to this before we send it to the MSM. MSM isn't going to report fairly and will only trash whatever is done by "Internet Folks."

Thanks Ignatz!

[ignatzmouse]

9. Freeman

I am very familiar with his exit poll analysis. Thanks for the suggestion. I've passed it along.

[BlueDog2u]

25. Glad for you two to hook up

We need to build strong teams of people who can help to build the case, critique the weak elements as they be, and move forward with the best data we can gather.

[Action Jackson]

3. THanks for the summary!!

[BeFree]

5. Can't wait

To see how the black box people try and dismiss this.

Hopefully other folks will follow your lead and examine their state's numbers using your structure.

I startd to put together a similar report, but it would have been next year before it was ready. You did it in a week! We all owe you for your hard work and you deserve a great round of applause. Thanks!

[SujiwanKenobee]

10. Mock election testing?

People are investigating after the fact with statistics, but where are these various voting machines being kept and who has access to them after the elections? Are they impounded?

Would it be possible to perform a mock election test with the various voting machines and compare the results with the actual mock votes? Is it possible that certain serial numbers of "defective" machines went to certain districts purposely and that a pattern could be found?

[BlueDog2u]

18. This is a very good question

On the other hand, if I were inserting Trojan horse code to rig an election, I'd make sure the code was set up to eat itself shortly after the election. I'm not a programmer, but I imagine this must be pretty easy to do isn't it? If so this would frustrate any attempt at a forensic audit of the kind you describe. The machine would come up clean unless someone could discover traces of the erased code through careful examination of the hard drive.

[SaveAmerica]

7. Cabarrus County Stats

Thanks so much for all the work you're doing! Do you happen to have registration statistics for Cabarrus County? I've tried to open the form from their web page but it says I have to have a certain disk to open it.

I'm thinking BBV needs to get you in a witness protection program!

[Straight Shooter]

8. Do NC registration increases run parallel with these bizarre results?

Not an artfully worded question, but there's a thread on another anomaly in Ohio which might be another question to add to the pile of questions being asked about NC:

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x72439

This is the text of the post by TennisGuy:

"If you compare the overall increase in voter turnout by county, you get some interesting figures. (These figures aren't certified yet...)

Hamilton County (Cincinatti) revealed an increase in overall votes of just 11% from 2000 to 2004. That's strange.

Cuyahoga County (Cleveland) showed an increase in overall votes of only 18% from 2000 to 2004. Odd.

Warren County, a Bush stronghold and the subject of the now infamous "lockout" showed an increase in overall votes of 36% from 2000 to 2004.

And Allen County, another Bush stronghold, which is run by Keith Cunningham, showed an increase in overall votes of... 135% from 2000 to 2004."

[BlueDog2u]

19. Interesting that should mention Keith Cunningham....

On this thread you will read that Cunningham has taken the lead to frustrate attempts at a recount:

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x72710

I smell a big rotten fish here.

[BlueDog2u]

20. On second thought, that number for Allen County has been

questioned. The CNN links provided show only an 11% increase in Allen county 2004. TennisGuy appears to be wrong about this figure. Tennis Guy, what is your source?

[wiggs]

11. Next step?

What do you want the next step to be? How did this happen? Is there any way to audit the vote at the machines? Did it happen at the individual voting machine or did it happen at the county tabulation computer? Is there any way to prove tampering in NC with a recount or getting a log of the modem entries to see if it was hacked?

Is there no way to show tampering, except by assuming that democrats could not have voted for Bush and in greater numbers at the end of the day than in the beginning?

This is like a missing persons report. We have a dead body, no proof of foul play and no murder weapon. We know someone who has a motive for the murder, but can't tie him to the crime. In fact, we have no solid evidence that a crime was committed.

In the case of the election, we don't need to pin the crime on someone in particular...we just need to SHOW that a crime was committed. We need to prove it. We need to see the software that did it or get a whistleblower to come out and tell us how it was done. Anthing else would be a temporary headline.

Here's what I don't get: There are plenty of questions. Plenty of anomalies. Plenty of doubt. However, political leaders, law enforcement, and media seem to be waiting for something. We aren't hearing what the DNC, FEC, FBI, or Robert Byrd think about the Ohio election office being locked down. We aren't hearing what the FBI thinks of your stats, of which they are surely aware and probably figured out earlier. Using the murder analogy, it is as if no one in authority was even talking about the dead body, much less doing anything about it or reporting on it. Bloggers are....and I'm hoping they can get the proof that is needed, but it would be easier if the people whose job it is to do this would help.

[go west young man]

12. Is that you John Kerry?

;-)

[understandinglife]

23. At least the GAO has joined the investigation....

....as noted here:

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=73210&mesg_id=73210

[jhgatiss]

13. Tabulator Comparison

Much has been made of the GEMS central tabulator system thanks to Bev's demonstration to Howard Dean that is revisited in the Votergate movie. My thought is this: If it seems from some different analyses that the optical scanning is at fault perhaps through the central tabulator (Ida Briggs in NH, other Florida analyses, Bev's finding in Valushia county) but your analysis of NC seems to implicate the DRE's, then is it possible that the central thread here may be that these counties are using the same central tabulator software system. I'm aware of verifiedvoter.org and its excellent website regarding the individual type of machines county by county, but is there a similar reference where we can find out the type of tabulator system in use in each county. Perhaps that would tie all of this together?

[gapower]

14. Is it possible

to compare the number of votes by county (or even better by precinct) between 2004 and 2000 to see if the differences differ by voting type (optical scan versus electronic) ? You could do an analysis of covariance (ANCOVA).

[BeFree]

15. Here's a start, gapower

I've begun going thru each county (will take forever) and breakdown the #'s like this.

**********

CHEROKEE 2004
total 10999 - 9544= 1455 new votes
12.7% increase from 2000
Bush 7382, + 1077 new votes = 67% = 74% of new votes
Kerry 3617 + 378 new votes = 33% = 26% of new votes

CHEROKEE 2000
total votes 9544
Gore 3,239 34%
Bush 6,305 66%


**************

Needed: A list of the voting machines used in NC to add to each county's database.

Some help with the breakdown of other counties. Anyone? Ignatzmouse has done his fair share, who wants in?

[ignatzmouse]

16. NC Voting machines

Here's a list of voting equipment by county, though I don't have specific machine differences recorded. When there was a question, I verified the information with various county elections officials. VerifiedVoting is a good start but is not 100% accurate.

County Voting Equipment
MACON E-Voting: Danaher Controls
MADISON E-Voting: Danaher Controls
CHEROKEE E-Voting: Danaher Controls
HENDERSON E-Voting: Danaher Controls
MOORE E-Voting: Danaher Controls
LENOIR E-Voting: Danaher Controls
BLADEN E-Voting: Danaher Controls
WILSON E-Voting/Special Category: Shoup/Shouptronic 12 42
GASTON E-Voting: Diebold
CRAVEN E-Voting: ES&S
JACKSON E-Voting: ES&S
POLK E-Voting: ES&S
DAVIE E-Voting: ES&S
PAMLICO E-Voting: ES&S
RUTHERFORD E-Voting: ES&S
TRANSYLVANIA E-Voting:ES&S
BRUNSWICK E-Voting:ES&S
GREENE E-Voting:ES&S
STANLY E-Voting:ES&S
GUILFORD E-Voting:ES&S
PERQUIMANS E-Voting:ES&S
PASQUOTANK E-Voting:ES&S
ALLEGHANY E-Voting: Fidlar Doubleday
CATAWBA E-Voting: Hart InterCivic
MCDOWELL E-Voting: MicroVote General
ALAMANCE E-Voting: MicroVote General
WARREN E-Voting: MicroVote General
CASWELL E-Voting:MicroVote General
MECKLENBURG E-Voting:MicroVote General
PENDER E-Voting:MicroVote General
DAVIDSON E-Voting:MicroVote General
LEE E-Voting:MicroVote General
BUNCOMBE E-Voting:Sequoia
PITT E-Voting:Sequoia
UNION E-Voting:Sequoia
BURKE E-Voting:Unilect Corporation
CARTERET E-Voting:Unilect Corporation
SCOTLAND Lever: AVM
SWAIN Lever:Sequoia
CHOWAN Lever:Sequoia
HOKE Lever:Sequoia
CAMDEN Optical Scan:Airmac
JONES Optical Scan:Airmac
YANCEY Optical Scan: Diebold AccuVote
MITCHELL Optical scan: Diebold/Central Count
MARTIN Optical scan: Diebold/Central Count
RICHMOND Optical scan: Diebold/Central Count
HARNETT Optical scan: Diebold/Central Count
MONTGOMERY Optical scan: Diebold/Central Count
NORTHAMPTON Optical scan: Diebold/Central Count
WASHINGTON Optical scan: Diebold/Central Count
ASHE Optical Scan: Diebold/Precinct
ANSON Optical Scan: Diebold/Precinct
AVERY Optical Scan: Diebold/Precinct
DARE Optical Scan: Diebold/Precinct
EDGECOMBE Optical Scan: Diebold/Precinct
FRANKLIN Optical Scan: Diebold/Precinct
COLUMBUS Optical Scan: Diebold/Precinct
HALIFAX Optical Scan: Diebold/Precinct
ROCKINGHAM Optical Scan: Diebold/Precinct
ROWAN Optical Scan: Diebold/Precinct
IREDELL Optical Scan: Diebold/Precinct
ROBESON Optical Scan: Diebold/Precinct
GATES Optical Scan:ES&S/Central
YADKIN Optical Scan:ES&S/Central
PERSON Optical Scan:ES&S/Central
ORANGE Optical Scan:ES&S/Precinct
CHATHAM Optical Scan:ES&S/Precinct
CLEVELAND Optical Scan:ES&S/Precinct
WAKE Optical Scan:ES&S/Precinct
LINCOLN Optical Scan:ES&S/Precinct
WILKES Optical Scan:ES&S/Precinct
STOKES Optical Scan:ES&S/Precinct
ALEXANDER Optical Scan:ES&S/Precinct
CALDWELL Optical Scan:ES&S/Precinct
CLAY Optical Scan:ES&S/Precinct
HAYWOOD Optical Scan:ES&S/Precinct
RANDOLPH Optical Scan:ES&S/Precinct
CUMBERLAND Optical Scan:ES&S/Precinct
GRANVILLE Optical Scan:ES&S/Precinct
NEW HANOVER Optical Scan:ES&S/Precinct
CURRITUCK Optical Scan:ES&S/Precinct
DURHAM Optical Scan:ES&S/Precinct
NASH Optical Scan:ES&S/Precinct
SAMPSON Optical Scan:ES&S/Precinct
BEAUFORT Optical Scan:ES&S/Precinct
JOHNSTON Optical Scan:ES&S/Precinct
WAYNE Optical Scan:ES&S/Precinct
HERTFORD Optical Scan:Fidlar Doubleday
SURRY Optical Scan:Fidlar Doubleday
BERTIE Optical Scan:Fidlar Doubleday
GRAHAM Paper Ballots: Hand Counted
TYRRELL Paper Ballots: Hand Counted
HYDE Paper Ballots: Hand Counted
CABARRUS Punch Card: ES&S
WATAUGA Punch Card: ES&S
FORSYTH Punch Card: ES&S
ONSLOW Punch Card: ES&S
DUPLIN Punch Card: ES&S
VANCE Punch Card: ES&S

[BeFree]

17. Thanks

Some of the DRE machines vary in style and manner of recording votes. I guess we need people from around the state to give us details.

I have an inkling of what it takes to gather all this info... one must wade thru a bunch of stuff to come up with such a list. The state BoE hasn't even done a good job of it, in fact, you've done the best job so far.

[Critical Thinker]

21. Once again, real good stuff here...

I find the most intriguing aspect to be the 10% disparity between votes as recorded by electronic vs. opscan machines during the early voting. Ignatz, does this disparity also exist within the election-day voting?

[ignatzmouse]

26. Two Sides of the Coin

The answer is complex because if E-voting changed results on both sides of the election, then there will be low totals on both sides -- there's no place from where to drop. Indeed, that is the case. E-voting totaled the greatest election day disparity at 58/42 Bush, but Optical Scans showed a huge drop to 57/43 Bush. The question is if they are nearly the same on election day, why are they so different in the early/absentee voting? The clue may be that early voting comprised 34% of the vote in e-voting counties but only 29% in optical scan counties mostly weighted by Diebold's paltry 23% in their 20 counties. The question becomes whether the early optical scan vote was suppressed or the early E-voting vote was inflated? An inflated early Bush e-vote would answer both sides of the equation, but there is also the real possibility of "spoiled" optical scan ballots.

[Chomp]

22. Exits

Do the exit polls reenforce your numbers? If so, their addition to the tables could add to presentational impact.

[eowyn_of_rohan]

27. Anyone heard of "Research Triangle Institute" in NC? whoa...

Try a Google search: "Research Triangle Institute" elections
Here is a sample from among the FASCINATING links:

The Guardian
www.guardian.co.uk/comment/story/0,3604,1144705,00.html

Democracy and robbery -
Washington wants to outsource Iraqi sovereignty, but its grip on the country is growing weaker
Naomi Klein
Tuesday February 10, 2004

If you believe the White House, the future government of Iraq is being designed in Iraq. If you believe the Iraqi people, however, it is being designed in the White House. Technically, neither is true; Iraq's future government is being engineered in an anonymous research park in suburban North Carolina.

On March 4 last year, with the military campaign just 15 days away, the United States agency for international development asked three American firms to bid for a unique job; after Iraq had been invaded and occupied, one company would be charged with setting up 180 local and provincial town councils in the rubble.

This was newly imperial territory for firms that were more accustomed to the friendly NGO-speak of "public-private partnerships", and two of the three companies decided not to apply. The "local governance" contract, worth $167.9m in the first year and up to $466m in total, went to the Research Triangle Institute (RTI), a private non-profit-making body best known for its drug research. None of its employees had been to Iraq in years.

At first, RTI's Iraq mission attracted little public attention. Next to Bechtel's inability to turn the lights on, and Halliburton's wild overcharging, RTI's "civil society" workshops seemed rather benign. No more. It now turns out that the town councils RTI has been quietly setting up are the centrepiece of Washington's plan to hand over power to appointed regional caucuses - a plan that has been so widely rejected in Iraq, it could end up bringing the occupation to its knees.