Detailed analysis of effects of vote machine fraud in Palm Beach Co. in 2004

[philb]

Detailed analysis of effects of apparent vote machine fraud and voter suppression in minority precincts in Palm Beach County in 2004: Based on voter reported irregularities to the EIRS election hotline(www.voteprotect.org)

Since only a relatively small portion of voters had the knowledge and motivation to report irregularities, only a small portion of total irregularities were reported, but Florida had the largest number of reported irregularities of any state and the thousands of reported irregularities serve as a good paper trail of election fraud and irregularities in the 2004 election.
http://www.flcv.com/fraudpat.html
**
The analysis found that precincts with reported machines that were switching votes from Kerry to Bush had a significantly larger swing of the Dem vote percentage between 2000 and 2004 than the average for precincts without reported vote switching, with a swing of 6.8% on average.
Exit poll data showed that a higher percentage of Democratic voters voted for Kerry in 2004 than for Gore in 2000, over 90%, and that Independent voters preferred Kerry by a significant margin.
**

Summary of precinct data and statistics for affected areas:
http://www.flcv.com/pbvsum.html

Summary of all precinct data and statistics for Palm Beach county precincts:
http://www.flcv.com/pbvdata.html

The official vote results were found to be extremely questionable statistically based on the voter registration data and exit poll data by university researchers such as Hout, Freeman, etc. http://www.flvc.com/flavi04.html

And the authors own similar study http://www.flcv.com/fla04EAS.html

Exit poll data showed that a higher percentage of Democratic voters voted for Kerry in 2004 than for Gore in 2000, over 90%, and that Independent voters preferred Kerry by a significant margin.

But while the Democrats had an 8 to 1 advantage in new registrations in Palm Beach County between 2000 and 2004 and had a major get out the vote campaign, the Republican vote total increased more than the Democratic vote total. Similar occurred in neighboring Broward and Dade counties and in other touchscreen counties.
http://www.flcv.com/fla04EAS.html
http://www.flcv.com/browardo.html
http://www.flcv.com/dadeo.html
http://www.flcv.com/kerrywon.html

Other statistical studies showing many thousands of lost votes or misvotes in Florida in 2004:
Hout et al, Univ. of California-Berkeley:
http://ucdata.berkeley.edu:7101/new_web/VOTE2004/election04_WP.pdf
http://www.prnewswire.com/cgi-bin/stories.pl?ACCT=109&STORY=/www/story/11-18-2004/0002464301&EDATE

[philb]

1. Manipulation of machines in minority precincts had an even larger effect

The analysis found that precincts with reported long lines or machine problems had a significant reduction in official vote turnout from the 2000 to the 2004 elections, with a swing of 7.8% on average for such precincts.

It seems ironic that precincts with extremely long lines correlated with extremely low official turnout compared to other precincts. This would seem to imply that the low official turnout was due to the electoral system rather than voter apathy. This consistently was found to be in minority precincts.

The same thing occurred in minority precincts in Ohio

Documentation in same study as above

Note: Palm Beach County was not the county with the most touch screen switching or other problems in Florida in 2004.
That honor goes to Broward Co, where a new Supervisor was appointed by the Governor just before the election. It appears that as much as 100,000 votes was swung in Broward in 2004 due to switching and manipulation
www.flcv.com/fla04EAS.html

But more detailed documentation was available to the author for Palm Beach for 2000 and 2004.
Also irregularities in Palm Beach Co. alone were enough to swing the election in 2000 from Gore to Bush.

[philb]

2. Note That Getting TouchScreens Replaced does not solve all election problems

Or even the largest problems; its just the easiest way to hack an election where the officials don't take
reports of switching seriously and won't allow audit of switching machines.

It actually required a lot of factors to make it possible to hack elections easily in this manner:

Convincing authorities to use an unreliable and unverifiable system

Convincing the public to go along with use of an unreliable and unverifiable system

Depending on the public and other political parties to not have the will to insist on impounding machines found to
be switching and conducting an audit


Seems someone correctly assessed that this could be accomplished; though that may change in 2006

[Febble]

3. You may want to check your links

The Hout study was widely criticized (rightly, IMO) but, to be fair, was withdrawn from circulation.

The website here:

http://ucdata.berkeley.edu:7101/new_web/VOTE2004/

now has a notice reading:

VOTE 2004

* Working Paper: The Effect of Electronic Voting Machines on Change in Support for Bush in the 2004 Florida Elections
by Michael Hout, Laura Mangels, Jennifer Carlson, and Rachel Best

At the request of the authors, this paper has been removed from this web site.
Comments and Questions may be directed to: [email protected]


Last revision 09/2/2005

For a substantial critique, try here:

http://election04.ssrc.org/research/critique-of-hmcb.pdf

Which opens with the caution:

“It is commonly believed that anyone who tabulates numbers is a statistician. This
is like believing that anyone who owns a scalpel is a surgeon. A statistician is one
who has learned how to get valid evidence from statistics and how (usually) to avoid
being misled by irrelevant facts. It’s too bad that we apply the same name to this
kind of person that we use for those who only tabulate. It’s as if we had the same
name for barbers and brain surgeons because they both work on the head.”

Robert Hooke, How to Tell the Liars from the Statisticians
New York, NY: Marcel-Dekker, Inc., 1983, page 1

None of which is to argue that no fraud happened in 2004 (or indeed in 2006) - I strongly suspect fraud in both elections. However, it is a caution against gung ho use of statistics, and also against giving to much credibility to any one analyst, regardless of how starry their credentials.

There may be good cases to make in 2006 - voter suppression in Virginia for a start, and undervotes in Florida. Let's not dilute good cases with crappy inferences from dodgy data.

[philb]

4. Houts study was consistent with my more detailed study for those counties

And I looked a bit at the critique which was not professional imo. Though I did not assess either study in detail.
I'm a professional statistician. One has to be careful to look to the science rather than the political correctness in such. I think my study will stand up to critiques, though it isn't conclusive "proof".

My links are still ok, as you note some of the other studies aren't. Though there are still copies of them somewhere.

[Febble]

5. Well what were your problems

with the critique? it was harsh, but correct.

At least one of his points struck me as soon as I read the Hout paper, as did the fact that they had not included an interaction term between size of county and technology. Essentially, their whole finding was leveraged by those two big counties - if you took those out, the finding actually went the other way. I thought it was a terrible study.

Which is not to say there was no fraud in Florida in 2004. It would astonish me if there wasn't, and the weird undervote counts in 2006 make this even more likely, as far as I can tell. But I would share your view that you have to look to the science, not the political correctness. Some perceived fraud as anomalous votes for Bush in Optical scan counties in 2004 (me included); Hout came out the other way. I think we were both wrong. Not wrong in the sense that there was no fraud, but wrong in the sense that it could be inferred from those analyses.

I respect Hout for withdrawing that analysis. I ought to try harder to get mine deleted too.

[philb]

8. Early on I tried to discourage that assessment as I never saw any strong case

But my analysis found the most problems to be in the big touch screen counties,
and the EIRS irregularity reports support and explain the reasons for the swings that I calculated.
There is no question about whether there was a lot of illegal dirty tricks, manipulation of registrations, absentees, provisonal, purges, and touch screen switching that swung large amounts of votes. The only question is whether it was enough to swing the election. Both my analysis, which I think is conservative, and the exit poll data indicate that in a fair count Kerry problably won Florida narrowly. www.flcv.com/fla04EAS.html

But there is a strong case that Kerry won Ohio also www.flcv.com/ohiosum.html
and New Mexico www.flcv.com/fraudpat.html

The Miami Herald recount in the county where Live Oak is appeared to give an indication that there might have been
a problem in some of the opti-scan counties as well. For some reason the Herald discontinued that study before it
was finished, but prorating the partial data seemed to imply a significant undercount for Kerry. I probably still have the data from that recount somewhere, and never heard why it was discontinued before it was completed.

[Febble]

11. I agree about NM

I did one analysis of NM here:

http://uscountvotes.org/index.php?option=com_content&task=view&id=70&Itemid=43

Still not convinced by Ohio and Florida (or not by exit poll evidence in Florida) but I'd like to read your analysis (and will). Although I am certainly convinced that more people tried to vote for Kerry in Ohio than had their votes cast and counted.

cheers

Lizzie

[philb]

12. Its clear from the EIRS data that Florida was a bigger problem than Ohio in 2004

but its more clear that the swing was enough to swing the election in Ohio
Richard Phillips has new evidence that makes the case for major fraud even stronger, lots of people should be in jail
in both Ohio and Florida.
The balance of evidence of mine and Richard's studies of Ohio, as supported by the exit poll data is that it appears Kerry would have won Ohio in a fair count. I don't think it would have been very close.

Just like Gore won by a lot in a fair count in 2000, as documented by the Media recount, along with other evidence. The exit polls in 2000 were correct, that election was not at all close, and one didn't need to look at chad to decide the winner. That was a distraction. I knew where Gore's uncounted votes were the day after the election and told lots of officials but they were never counted, even though they were clearly legal votes under Florida law excluded illegally by Florida officials who mounted a major and effective coverup. This was confirmed after the election by the Media recount.

The exit poll data and my analysis both showed Florida in 2004 extremely close toss-up too close to call without the manipulations that occured. But the predominance of the evidence supports a Kerry win in Ohio and New Mexico and likely Florida, and that a lot of officials and others committed malfeasance and illegal acts that were documented but havent' been acted on. I don't think that issue is over in Ohio.

[philb]

14. Per my analysis virtually all of the significant problems in Florida were in the big counties

The only serious effort to look into the strange results in the small Dem majority rural counties that voted strong Repub was abandoned in the middle of it for some unknown reason by the Miami Herald.

[Blue Shark]

13. Hi Liz...Good Points.

...You might wish to try a little harder now to prove Election Fraud now that there is some money in it.

...http://www.livereal.org/blog/250000-reward-for-voter-fraud-evidence/>

...and http://www.velvetrevolution.us/>

...$750k would be a nice payday indeed.

[rumpel]

6. thanks, philb

what do you think is the percentage voters whose votes may be wrongfully rejected due to the HAVA mandated registration database upgrades, during the canvass?

Here is one such example - I am looking at LA County.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x457986#458007

[rumpel]

7. remember this?

Next problem: DID YOUR MAIL-IN BALLOT ARRIVE AT THE ELECTIONS DIVISION?

Some jurisdictions allow voters to confirm whether or not their ballot arrived (but this doesn't confirm whether their signature was accepted). In other jurisdictions, there is no easy way to find out whether the ballot you mailed in ever got to the elections division.

In Broward County, Florida, an extraordinary citizen named Ellen Brodsky spent months trying to track down over 50,000 missing mail-in ballots. In King County, Washington, bags of ballots were once found years after they were supposed to be delivered. Also in King County, incoming ballots were being taken from the U.S. Post Office to a private company called PSI Group, without an accounting of how many arrived at the Post Office, how many arrived at PSI Group, vs. how many arrived at the Los Angeles County Elections division.

WHAT TO DO: Call your local jurisdiction to find out the procedures for you to verify that your ballot was received. If your county cannot provide you with this information, contact Black Box Voting and also take action to change this policy (but that won't help you in the Nov. 2006 election).

http://www.opednews.com/articles/genera_bev_harr_061028_election_bulletin_3a_m.htm

[philb]

9. According to my analysis there was about a 100,000 vote swing in Broward

from Kerry to Bush due to irregularities including the major absentee ballot problems.
There were registration purges, widespread illegal dirty tricks by Repub operatives to prevent minority votes from being counted, manipulation of absentees, touch screen switching from Kerry to Bush and to blank, etc.

www.flcv.com/browardo.html
www.flcv.com/broward.html
www.flcv.com/EIRSfla.html

You may remember that Bush appointed a new SOE in Broward just before the election.
Broward has the most Dems in Florida and had the most irregularities by a lot.

I could not do a detailed analysis comparing 2000 and 2004 election details by precinct such as this one for Palm Beach because the Broward SOS was uncoopertive in providing me the 2000 data from their archieves. But Broward had a lot more of the things I assessed in the PB study than Palm Beach did. All 3 big S. Florida counties had a lot of problems however.

[philb]

10. One of my links(VU) no longer works it seems, I think this might be that info

Broward County. www.votersunite.org/electionproblems.asp?sort=state&selectstate=FL&selectproblemtype=ALL

1. 58,000 ballots that were supposed to mailed out on Oct. 7 and 8 are late and appear to be missing.
2. 2505 absentee ballots were mailed on Saturday (October 30). Election is on Tuesday(Nov 2).
Most had been requested long before the deadline for absentee ballot requests.
3. Absentee voting problems were a "disaster." Voters were unable to confirm the status of their request for an absentee ballot. In some cases, by the time they realized a ballot wasn't on its way, it was too late. Hundreds of voters couldn't vote because their early orders for ballots disappeared
4. 94% of 78,861 absentee ballots were recorded in favor of Amendment 4, which passed by a thin margin in the county. The votes in question were counted late on election night after a glitch was discovered in the computers tallying absentees.
5. Nine of 14 early-voting polling sites had trouble linking laptops to the main computers that confirm voter eligibility.
6. After waiting in line for several hours, several voters were told late Sunday evening that they would have to come back another day to be able to cast their ballot. The machines had broken down.
7. At least 21 voting machines in Broward County malfunctioned and were replaced Tuesday. Most of them had been used by some voters before being taken out of service
8. An improperly calibrated machine at the polling place at 2501 Coral Springs Dr. in Coral Springs was used by an undetermined number of voters before it was replaced. If they did not review their ballots, it is possible that some votes were recorded inaccurately. Possibly other malfunctioning machines were also miscalibrated.
9. Several touch-screen voting machines in Broward County malfunctioned this morning when their batteries went dead.
10. ES&S vote tabulating software used for absentee ballots "is not geared to count more than 32,000 votes in a precinct. So what happens when it gets to 32,000 is the software starts counting backward. Amendment 4 passed in Broward County by more than 240,000 votes rather than the 166,000-vote margin reported Wednesday night. ES&S has known about the problem for two years and done nothing about it. The same software is used in Martin and Miami-Dade Counties
11. Florida voters can't use provisional ballots except in their home precincts, the Florida Supreme Court unanimously ruled Monday. 90 polling places were destroyed by the hurricanes.
Secretary of State Glenda Hood applauded the decision because it will help ensure an orderly election.
12. Broward County accepted on 2428 provisional ballots out of a "ton" of them. In many cases, clerks at the polling places gave provisional ballots to people who could have voted regularly, deputy registrar Salas said. She thinks many clerks had trouble with last names of voters with compound names -- they simply couldn't find their names in the precinct registers, even though they were there.

13. The vast majority of provisional ballots -- voters' last chance to have their voices heard -- were rejected. The majority of rejected ballots were cast by people who simply were not registered to vote. Other reasons: voting in the wrong precinct, signatures that didn't match those on file at the elections office and lapsed registrations because voters hadn't responded to address-verification requests and hadn't voted in at least four years.
14. Broward and Miami-Dade counties both have about 1,058,000 registered voters, but Miami-Dade has 20 early-voting sites, compared with Broward's 14. Every Miami-Dade site is equipped with at least 20 voting machines, while some in Broward have fewer than 10. Some voters wait 4-1/2 hours in Broward to vote.
15. Republicans and their attorneys say they are arming themselves with lists of voters whose registrations appear flawed, preparing to challenge voters on election day. But they refuse to give the list to officials who want to correct errors ahead of time.
16. University of California's Berkeley Quantitative Methods Research Team released a statistical study - the sole method available to monitor the accuracy of e-voting - reporting irregularities associated with electronic voting machines may have awarded 130,000-260,000 or more excess votes to President George W. Bush in Florida in the 2004 presidential election.
***************