New US Count Votes Exit Poll Study Shows Vote Embezzlement is ..

[sunshinekathy]

Hi Everyone.

I've been slaving away to understand all the math and create a new mathematical simulation tool that has allowed us to determine the signature of exit poll bias versus the signature of vote count embezzlement. The paper that US Count Votes will be releasing in a few days clearly shows that the patterns in the Edison/Mitofsky exit polls can only be reproduced by a combination of small Kerry exit poll response bias and significant shifting of votes cast for Kerry to Bush.

Can any of you help us promote this new paper? It will be released, at latest, on Friday at the AAPOR (American Association of Public Opinion Research) in Miami, FL by Ron Baiman.

We did all this so last minute that I haven't even THOUGHT about how to get press attention, or get press to show up a possible small press conference in Miami that Ron Baiman is planning.

The paper is short and readable, full of graphical charts. All the technical stuff has been pulled into the Appendices which are much longer than the actual paper, for those who want to check or replicate our work.

There will also be a downloadable simulation spreadsheet that anyone can download, check our formulas, and play with to see what various rates of exit poll response bias or various vote shifts produce in the patterns of exit polls, and see how close they can come to replicating Edison/Mitofsky's data. (Must do vote shifts from Bush to Kerry, plus small Kerry response bias, it looks like). All the programs will be made publicly available for our programmed simulation as well.

This work uses the same (we think) formulas that Elizabeth Liddle used but clearly show that her conclusions were incorrect.

Please be ready to help us spread this around on Friday.

Here are the current (as yet unedited) conclusion and recommendations at the end of the paper. The conclusions don't have much to do with the rest of the paper, but I'm convinced that people will really listen to any advice given at the end of this most astounding paper yet that US Count Votes has ever produced- which is coming on Friday.

Best,

Kathy Dopp
http://electionarchive.org

Please sign up for our email announcements list and help us spread this new report around to press.
http://uscountvotes.org/fairelection/takeaction.html
(second link on page)

*Conclusions and Recommendations*

Vote counts in America need to be routinely independently audited. It is not enough to require voter verified paper records of ballots. These paper records must be easily "independently" auditable by persons other than the voting machine vendor, preferably without having to hire computer technicians, paper roll advancers, bar code readers, and laptops, as is true with many voting systems on the market today.

3% of randomly selected precincts can be recounted, using the paper record, immediately when polls close, in the precinct, before removing ballots from the precinct. If discrepancies are found, a county-wide recount can be automatically triggered. Additional funding may need to be allocated to state and county election offices to routinely perform independent audits of vote counts.

In order to audit their vote counts and monitor the accuracy of vote counting systems, all state and county election offices should set up election data reporting systems to quickly and easily make publicly available, their precinct-level vote totals, broken out by vote type (i.e. election day, absentee, overseas, provisional, early voting, etc.) If vote counts are not reported down to this detailed level, then padded votes in one vote type can easily "cancel out" under-votes in another type. In other words votes can be subtracted from one candidate in one vote type, while being added for another candidate in another vote type, yet these two problems, when added together, look perfectly normal.

[Carolab]

1. Pardon me, but where do you stand on paper ballots v. trails?

It appears you are recommending "paper records"/trails of ballots. What does that mean? Do you advocate that people vote on ballots (i.e., optiscan) and THAT is the record or are you saying people should vote and there should be a paper record of that vote? There is a huge difference between the two.

[In Truth We Trust]

2. Good question. As for me I prefer Paper ballots and hand counts

[Bill Bored]

3. Self-delete

Have a nice day! :)

[TruthIsAll]

4. Kathy, I look forward to seeing the simulation.

Have you read the latest threads which deal with rBr using voter turnout and Kerry percentage deviation sensitivity analysis from the National Exit Poll (13047 respondents)

What are your thoughts on Voting Identity (VI), which is a linear constraint on the votes that any consumer who has a budget will understand? It's analogous to the basic accounting identity:

TA = TE + TL

The ramifications of VI are profound, partly due to its obvious simplicity:

TV = TG + TB + TN + New

where
TV = 122.26mm votes in 2004
TG = Total Gore 2000 voters STILL ALIVE
TB = Total Bush 2000 voters STILL ALIVE
TN = Total Nader 2000 voters STILL ALIVE

Solving for New = MINIMUM number of NEW voters, who did not vote in 2000: New = 21.15mm.

We calculate TG, TB, TN by applying the U.S. 0.87% annual death rate to determine approximately how many 2000 voters have died and subtract this number from the vote totals.

THE VI IS A LOGICAL CONSTRAINT ON THE SOURCE OF 2004 VOTES.
The ramifications are profound. It is analogous to the basic accounting identity: TA = TE +TL

1) It puts the lie to the IMPOSSIBLE 43%/37% Bush/Gore weighting as stated in the Final National Exit Poll of 13660 respondents. The MAXIMUM Bush percentage is 39.82%, after deducting Bush 2000 voters who have dies. His 2000 vote was 50.456 million. Subtracting the approximate 1.77mm who died, he is left with a MAXIMUM 2000 voter turnout of 48.69mm. Dividing 48.69/122.26 = 39.82%.

2) It puts the LIE to the rBr hypothesis.

Even assuming 100% Bush 2000 voter turnout (impossible) and 84% Gore voter turnout (impossible), Kerry is STILL the winner by 630K votes in the "How Voted in 2000" Demographic.

Assuming 90% turnout and a 1% reduction in the Kerry vote, he's STILL the winner by 170k votes. The margin of error is 1.77% for the 3168 who responded in that demographic category.

The MoE is 0.86% for the 13047 respondents, but the MoE is near ZERO for the maximum number of 2000 voters.

Assuming 100% turnout of ALL 2000 voters, Kerry won by 5.9 million votes. Using a 1% MoE, that means the chances are greater than 97.5% that Kerry won over 5.84 mm votes (5.90 - .059).


Links to the analysis and a graph are provided in this thread:

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x368143

[TruthIsAll]

37. Kathy, here is my updated thread of links (as of 5/14)...

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x369024

[Peace Patriot]

5. Questions - and some valuable Miami & other news info...

Please explain the following in layman's terms. It's very important that people understand this--and I don't fully.

"The paper that US Count Votes will be releasing in a few days clearly shows that the patterns in the Edison/Mitofsky exit polls can only be reproduced by a combination of small Kerry exit poll response bias and significant shifting of votes cast for Kerry to Bush."

What does "small Kerry exit poll response" mean? Small WHAT (voter response to exit pollsters? low level of response from Kerry voters? ???). And then what does, "small Kerry exit poll response BIAS" mean? (--bias toward or against what? bias with what result? ???)

The meaning of "significant shifting of votes cast for Kerry to Bush" seems clear, but what does "a combination of" mean? How are you combining these things, and what does it mean that "the patterns in E/M" can "only be reproduced" by this combination?

Staticians often speak in this weird backward way, that requires a lot of translation. I cannot seem to get the plain meaning of this.

------

MEDIA

Here's the MEDIA BLASTER. Easy emailing to all media. Just go there, select the format, and then copy each of the email groups onto the BCC line of your email program, put your own email on the "to" line, compose your subject line and message, and blast away!

Be sure to use a catchy subject line with date, time, place (i.e., "Scientists cry foul on '04 election TODAY, 1 pm, Dolphin Hall..."--or whatever)

Media Blaster at: http://www.independentmediasource.com/voteintegrity2_12.htm

------

Please take advantage of all these news room fax numbers below. They are not easy to obtain--lots of research. And faxes have much higher attention value than emails. (Phone calls are the best--get editor to commit, talk to reporter).

MIAMA

The Miami Herald is a pretty kick-ass newspaper. Definitely fax and call their news room. Don't have at my fingertips what reporters have covered elections, but a quick Google search should turn them up.

http://www.miami.com/mld/miamiherald/

Miami Herald
305-350-2111 or 1-800-HERALD5 (1-800-437-2535)
News FAX: [phone]
One Herald Plaza
Miami, FL 33132

The Miami Herald
General Information
305-350-2111 Miami-Dade
954-538-7000 Broward
1-800-HERALD5 (1-800-437-2535)
from outside South Florida
Dade Bureau
[e-mail]
[phone]
Features
[e-mail]
[phone]
World News
[e-mail]
[phone]
National News
[e-mail]
[phone]
State Desk
[e-mail]
[phone]
Investigations
[e-mail]
[phone]
News Photo
[e-mail]
[phone]
Letters to the Editor
[e-mail]
[phone]
Photo department
[e-mail]
[phone]

http://www.miami.com/mld/miamiherald/contact_us/

-------

The Miami Hurricane (student paper, U of Miami, Coral Gables)
Phone numbers:
Business Office: 305-284-4401
Newsroom: 305-284-2016
Fax: [phone]

Important E-Mail Addresses:

* Editors ([e-mail])
* Advertising ([e-mail])
* Webmaster ([e-mail])

http://www.thehurricaneonline.com/
submit letter to ed: http://www.thehurricaneonline.com/main.cfm?include=submit

-------

El Nuevo Herald - Spanish-language daily with emphasis on Miami and Latin America news, business, and entertainment.

Contactos de El Nuevo Herald
Contactos en la redacción:

El Nuevo Herald
One Herald Plaza
Miami FL 33132-1693
Información general
[phone] (gen. offc fax: [phone])
News FAX: [phone]


-------

TELEVISION

WFOR TV - CBS -5 News
tel: [phone]
News FAX: [phone]

WPLG TV ABC-10
tel: [phone]
News FAX: [phone]

WB39 News at 10 - WBZL - NBC-6
office tel: [phone]
news tel: [phone] (ext. 1 breaking news)
News FAX: [phone]
Also FAX: [phone]
Email news (ok): [e-mail]

WLRN tv-17, and 91.3 fm (community access tv and radio--news)
tel: [phone]
FAX; 305-995-2299
Email: [e-mail]

WPBT Channel 2 (public tv - does daily news show
tel: [phone]
FAX: [phone]
Email: [e-mail]

WSVN-TV 7 - Fox
tel: [phone]
News FAX: [phone]
Email: [e-mail]

-------

RADIO

Miami Radio

* WDNA -- 88.9 FM- Miami, FL (9.67 miles, & (CP) (23.56 miles) -jazz
WDNA-FM 88.9 (more music than other public sta)
Phone: [phone]
FAX: [phone]

* WMCU -- 89.7 FM- Miami, FL (23.56 miles)
Phone: [phone]
Fax: [phone]

* WVUM -- 90.5 FM- Coral Gables, FL (6.64 miles)
U of Miami - Coral Gables
Phone: [phone]
Fax: [phone]

* WLRN -- 91.3 FM- Miami, FL (14.26 miles)

Phone: [phone]
Fax: [phone]

* WLVE -- 93.9 FM- Miami Beach, FL (13.35 miles)-Clear Ch., jazz
Phone: [phone]
Fax: [phone]

* WZTA -- 94.9 FM- Miami Beach, FL (13.35 miles) -alt. rock
Phone: [phone]
Fax: [phone]

* WPOW -- 96.5 FM- Miami, FL (13.35 miles) -hip hop
Phone: [phone]
Fax: [phone]

* WFLC -- 97.3 FM- Miami, FL (13.43 miles) -hot ac
Phone: [phone]
Fax: [phone]

* WEDR -- 99.1 FM- Miami, FL (13.35 miles) -hip hop
Phone: [phone]
Fax: [phone]

* WKIS -- 99.9 FM- Boca Raton, FL (15.19 miles)-country
Phone: [phone]
Fax: [phone]

* WHYI - 100.7 FM- Fort Lauderdale, FL (13.35 miles) - Top 40
Phone: [phone]
Fax: [phone]

* WLYF - 101.5 FM- Miami, FL (13.36 miles)-music
Phone: [phone]
Fax: [phone]

* WMXJ - 102.7 FM- Pompano Beach, FL (13.35 miles) -oldies
Phone: [phone]
Fax: [phone]

* WHQT - 105.1 FM- Coral Gables, FL (13.43 miles) -music
Phone: [phone]
Fax: [phone]

* WQAM --- 560 AM- Miami, FL (3.27 miles) -Naples -sports
Phone: [phone]
Fax: [phone]

* WIOD --- 610 AM- Miami, FL (5.76 miles) - Clear Ch.
News Radio
Phone: [phone]
Fax: [phone]

* WSBR --- 740 AM- Boca Raton, FL (39.06 miles)
Money Talk Radio
Phone: [phone]
Fax: [phone]

* WAXY --- 790 AM- South Miami, FL (27.62 miles)-variety
Phone: [phone]
Fax: [phone]

* WHSR --- 980 AM- Pompano Beach, FL (32.32 miles) -internat’l
Phone: [phone]
Fax: [phone]

* WJNO -- 1040 AM- Boynton Beach, FL (48.42 miles) -Clear Ch.
News talk
Phone: [phone]
Fax: [phone]

* WJNO -- 1040 AM- Boynton Beach, FL (CP) (48.42 miles)-Clear Ch
News talk
Phone: [phone]
Fax: [phone]

* WVCG -- 1080 AM- Coral Gables, FL (22.03 miles)-tropical music
Phone: [phone]
Fax: [phone]

* WAVS -- 1170 AM- Davie, FL (21.05 miles) -tropical
Phone: [phone]
Fax: [phone]

* WLQY -- 1320 AM- Hollywood, FL (18.57 miles) -ethnic
Phone: [phone]
Fax: [phone]

* WKAT -- 1360 AM- North Miami, FL (3.27 miles) -classical
Phone: [phone]
Fax: [phone]

* WFTL -- 1400 AM- Fort Lauderdale, FL (26.31 miles)-talk
Phone: [phone]
Fax: [phone]

* WWNN -- 1470 AM- Pompano Beach, FL (28.10 miles) -health talk
Phone: [phone]
Fax: [phone]

* WMBM -- 1490 AM- Miami Beach, FL (3.59 miles)-gospel
Phone: [phone]
Fax: [phone]

* WSRF -- 1580 AM- Fort Lauderdale, FL (21.38 miles) -tropical
Phone: [phone]
Fax: [phone]

______________
SPANISH
____________

* WCMQ -- 1700 AM- Miami Springs, FL (13.74 miles)
**SPANISH
Phone: [phone]
Fax: [phone]

* WWFE --- 670 AM- Miami, FL (18.80 miles)
**SPANISH
Phone: [phone]
Fax: [phone]

* WAQI --- 710 AM- Miami, FL (17.72 miles)
**SPANISH
Phone: [phone]
Fax: [phone]

* WQBA -- 1140 AM- Miami, FL (18.21 miles)
**SPANISH (Cuban)
Phone: [phone]
Fax: [phone]

* WNMA -- 1210 AM- Miami Springs, FL (13.70 miles)
**SPANISH
Phone: [phone]
Fax: [phone]

* WSUA -- 1260 AM- Miami, FL (14.17 miles)
**SPANISH
Phone: [phone]
Fax: [phone]

* WOIR -- 1430 AM- Homestead, FL (29.96 miles)
**SPANISH
Phone: [phone]
Fax: [phone]

* WOCN -- 1450 AM- Miami, FL (4.58 miles)
**SPANISH

* WXDJ -- 95.7 FM- North Miami Beach, FL (0.33 miles)
**SPANISH - El Zol ‘95
Phone: [phone]
Fax: [phone]

* WCMQ -- 92.3 FM- Hialeah, FL (0.33 miles) - SPANISH clasica ‘92
Phone: [phone]
Fax: [phone]

* WRMA - 106.7 FM- Fort Lauderdale, FL (15.19 miles)
**SPANISH - romantic music
Phone: [phone]
Fax: [phone]

-------

Liberal Talk Radio In Florida!

Spread the word. Tell your friends to listen in. Call the station every Saturday and give them your supportive comments (239-732-9369). Call The Guy James Show live on the air (239-530-1660).

Saturdays 3-6 pm

http://www.theguyjamesshow.com/

America's #1 Democratic Talk Show! Broadcasting from behind enemy lines on WCNZ 1660 Naples, Florida! World Wide audio streaming live Saturdays, from 3-6 PM EST. Call us on the air! 239 530-1660.

Call us on the air!
Guy James Show Live (Saturdays 3:15 PM - 6 PM)
239-530-1660

EMAIL: [e-mail]
The Guy James Show
12870 Trade Way Four # 108
Bonita Springs FL 34135

WCNZ Radio News 1660
Naples, Florida
239-732-9369

The Guy James Yahoo Group
http://groups.yahoo.com/group/GuyJames/


-------

SOME OTHER SUGGESTIONS (sorry, don't have contacts)

Jim Lampley - sports writer/commentator - very good on election fraud/reform! (famous)

The New Yorker

The Nation

Vanity Fair

Christopher Hitchens

Keith Olberman

The Lone Star Iconoclast (very, very kickass newpaper in Bush home town, Crawford, TX)

Knight-Ridder (!!!!!!)

Agence France Presse

Reuters

BBC News (--they seem to have reporter/video units all over US, and they get around)

Al-Jazeera (--might like to compare Bush democracy rhetoric re Iraq with reality in U.S.)

International Herald Tribune

Scoop.com (New Zealand)

Air America, of course. (Randi Rhodes, Mike Malloy, Janine Garofola)

Some of the usual suspects, maybe--New York Times (warmongers but maybe would like to save their soul), Washington Post (ugh), Philadelphia Inquirer (used to be good on investigations), Los Angeles Times (pretty good investigative reports on some things but very snotty on election fraud/reform), ABCNBCCBSCNNFOXTIMEWARNER (it would be convenient if they had just one generic blackhole into which to deposit vital info about our democracy).

[Peace Patriot]

6. Re: all those music radio stations listed above for Miami...

Some may do "Public Service Announcements" (PSA's - usually 15 or 30 seconds) between songs or in transitions. Free publicity for public events, for non-profits, community groups, etc.

Some may do brief news reports. Give them something catchy and short. And, who knows? - some of the DJ's may like the rebel subject.

(And some of the stations are 24/7 news, news-talk, variety or public stations. If you have to be selective, fax/call those first.)

[Peace Patriot]

7. MORE MEDIA: A very activist student group at Florida State...

[e-mail]

Ask their advice about contacting university media. Invite students to the press conference.

[eomer]

8. South Florida print media

Peace Patriot, here are a couple of additions for your list.

Robert Steinback, columnist at the Miami Herald, leans toward the conservative side but has recently written a column advocating paper ballots:
http://www.grandforks.com/mld/miamiherald/news/columnists/robert_steinback/11377424.htm

Here is his email:
[e-mail]

Also, the Palm Beach New Times ran an article a few months ago, I believe about Clint Curtis. Here are their website and email:
www.newtimesbpb.com
[e-mail]

And the Miami New Times is another possibility:
www.miaminewtimes.com
[e-mail]

[Febble]

12. I think you'll find

that "Kerry response bias" translates as "reluctant Bush responder".

A response bias is when a greater proportion of one group than another group respond to a survey.

The question now is "how small"?

[Land Shark]

20. The best propaganda has a kernel of truth to it n/t

[autorank]

25. I second that remark. Enough of the quibbling. Time for SOLIDARITY

AND ACTION.

[Peace Patriot]

34. Thanks, Febble! I appreciate the clarification (I really didn't get that

...it was just rBr). Speaking of that, have you seen my posts on the recent news story out of Waynesville, NC? It's highly relevant to "rBr."

Last week, 30 members of a Baptist Church were thrown out of their congregation for refusing to pledge support to Bush. I think that's our "reluctant responder" --or the most likely picture of one in the U.S. of A. today--a voter in a hotbed of Christian rightwingery and Bush-worship (true of many Republican areas these days), who couldn't take this hypocritical crap any more, decided to vote for Kerry, felt protected by the secret ballot, but would have been very reluctant to admit such a vote to a strange pollster especially in a public area that might be only blocks from his/her church, in a precinct where the congregation votes--for fear of serious repercussions. This voter has been living in a social and political atmosphere of intimidation and bullying by Bush supporters, which just boiled over in the expulsion of non-Bush supporting members. Imagine what it must have felt like to that voter, and his/her 30 compadres, back on Nov. 2, 2004, with Jesus W. Bush's Second Coming on the line.

I think this the most likely candidate for Reluctant Responder, circa 2004.

Now think about the bullying majority who threw the non-Bush supporters out of the church. Shy? Reluctant? Fearful? Nope, they would be getting rewards for voting for Bush, and for being very vocal about it. They would be proud of it. They would want to announce it to the world--as they have now, in their notoriety over this incident. They are living in Bush's World. What do they have to be afraid of? They watch Bush-can-do-no-wrong, propagandistic news on TV and hear it on radio. (You can hardly get away from it, here.) Repressive, intimidating, self-righteous people who want to shut others up, who think God elected Bush, and they have a direct line to God.

What this ugly little incident tells us is that there was significant dissent in Bush strongholds, that the dissent was controversial, that Kerry voters had reason to fear repercussions, and that the Bush supporters were not above retaliation and punishment.--all of which supports the opposite of "rBr."

I suspect that 'rBr' is a phantom, and that one of the really important things to look for is what happened to the votes for Kerry in Republican areas such as Waynesville, by people who had good, solid, palpable reasons to shut their mouths after they voted.

Such votes obviously may not have left much trace in the exit polls, but possibly can be inferred from other statistics, for instance, odd discrepancies between Kerry's totals and those of other candidates (like the unknown, underfunded judge candidate in Ohio who seemingly got far more votes than Kerry, or this strange thing in California with Kerry's vote being significantly lower than liberal Democratic senate candidate Barbara Boxer only in the rightwing counties--the difference between the Boxer and Kerry votes in these counties possibly pointing to yet more Republican dissenters from the Bush paradigm in Republican-controlled areas (and also pointing to their votes for Kerry having been stolen).

If Edison-Mitofsky are perfectly happy with the Bush paradigm themselves, or feel no personal impacts from it because of their class or financial position, perhaps their minds would automatically turn to something like 'rBr" for an explanation (if they're not outright lying), because they haven't a clue what it must have been like being a poor Baptist anti-Bush dissenter in a bigoted, repressive, fanatical pro-Bush subculture. E/M's intuitions wouldn't lead them there. And of course such dissenters would further put the lie to the official tally, which E/M assumes to be correct (unreasonably, it's seems to me).

'rBr' makes little intuitive sense to me, from my own experience of the Bush subculture, and the last three years in the U.S. have further struck me as highly unusual in recent American history for the open, public and aggressive statement of intolerant "know-nothing-ism," the tendency of Bushites to ridicule and belittle any dissent from their views, and their not being the least shy about it. They have shown in many ways their belief that others don't have a right to dissent (Waynesville being a startling example of this, and the Bush Republicans' treatment of Democrats in Congress being another), and so, to me, the notion of a shy Bushite seems quite unreal. But his silent dissenting wife or employee or true, gentle-hearted Christian church member, afraid to voice their dissent to a stranger for fear of who it might get back to, seems very real, indeed.

[Sawyer]

35. "the notion of a shy Bushite seems quite unreal"

1. You know you're getting off track when you start believing your own propaganda. Trust me - there are all kinds of people who voted for Bush, some are standoffish, some are outgoing, some are social, some are anti-social, some are shy, some are gregarious.

2. RBR does not mean "shy Bush voter" literally. It just means that Bush voters refused to answer exit polls at a slightly higher rate than Kerry voters. Both types of voters refused to answer at a very high (close to 50%) rate.

3. The reasons for non-responding could be various. Some would be actually "shy". Some would be in a hurry. Some would distrust anyone trying to find out how they voted. Some would be too arrogant/obnoxious to stop. Some would mistake the NEP interviewer for some kind of solicitation. Etc. etc.

[MarkusQ]

19. A small bias

What does "small Kerry exit poll response" mean? Small WHAT (voter response to exit pollsters? low level of response from Kerry voters? ???). And then what does, "small Kerry exit poll response BIAS" mean? (--bias toward or against what? bias with what result? ???)

It seems crystal clear to me. A small response bias means that (as many others have suggested) there was a bias in the response to exit pollers (e.g., that Kerry supporters were more likely to answer honestly about who they voted for) but that this effect was small.

The meaning of "significant shifting of votes cast for Kerry to Bush" seems clear, but what does "a combination of" mean? How are you combining these things, and what does it mean that "the patterns in E/M" can "only be reproduced" by this combination?

Again, this is pretty simple. There were two problems, a big one (vote stealing) and a small one (Bush voters refusing to give honest answers to pollsters). It takes a combination of the two to explain the observed data--you can't just say that it was one or the other that did it.<P>

Try recasting it in an other situation and see if it helps:

I've been slaving away to understand the occasional funny smell in the second floor conference room and create a new automatic-sniffer that has allowed us to distinguish the odor of spoiled food versus the odor of serious plumbing problems. The paper that I'll be releasing in a few days clearly shows that the smell can only be reproduced by a combination of small amount of leftover food in the trash can and significant stench from the toilet in the bathroom overflowwing.

The paper is short and readable, full of graphical charts. All the technical stuff has been pulled into the Appendices which are much longer than the actual paper, for those who want to check or replicate our work.

There will also be a downloadable simulation kit that anyone can download, check our formulas, and play with to see what various amounts of spoiled food or amounts of poop on the bathroom floor smell like, and see how close they can come to replicating the smell. (Must do quite a bit of poop, plus small amount of leftovers, it looks like). All the programs will be made publicly available for our programmed simulation as well.

I hope that helps.

--MarkusQ

[Febble]

22. Right.

So how do we distinguish between a small response rate differential and a large one?

What would a large one look like? How would fraud look different?

We are getting there. Maybe.

[MarkusQ]

36. Error signatures

If some group of people is statistically less likely to answer the pollsters honestly (and instead either lie or refuse to answer) the reporting error will be proportionate to number of people in that group in the sample. Unless you are supposing that someone rigged the election specifically in proportion to the number of people in each precinct that were in this group, the statistical profiles will be different. If the rigging was done by something crude (such as taking voting machines away from precincts where your opponent is strong and giving them to precincts in which you are) the difference should be quite clear.

--MarkusQ

[helderheid]

31. Great media links!

Kathy, Clarity here.

My experience with dealing with the press is that if they get a fax, they'll read it. If they get an email, it will see the delete button.

Do you have a way to fax out? If not, contact me and I can plug these faxes into my contact manager and fax it for you.

[sunshinekathy]

40. Thank you, and I didn't see this in time darnit.

Peace Patriot,

I am so sorry I missed this until now.

We released the report this a.m.

http://uscountvotes.org/ucvAnalysis/US/exit-polls/USCV_exit_poll_simulations.pdf

(first 11 pages are the report - the rest are appendices)

Thank you so much for posting all this pertinant info. I stayed up almost all night for the last two nights and worked around the clock to get the paper ready at the last possible minute this a.m. for Ron Baiman and Pete Pekarski (sp?), and I just sent out a press email notification at about 3 a.m. I just didn't have time to do anything but get the collaborative paper done.

Would you possibly be willing to join a new mail list for our pr team that we could rely on once in a while for some pr help?

If so, email me at [e-mail]

In fact, I haven't even updated our web site yet, and my personal life is a mess,..

I just didn't find time to do any press notification other than emails which is not very effective.

I'm sorry bout that.

Best,

Kathy

[Amaryllis]

43. I second the request; please explain the points PP asks about.

[Peace Patriot]

9. DEFINITELY GO HERE AND CHECK THIS OUT!

And DEFINITELY CONTACT Jim Lampley (sports writer). If you haven't been aware of his three recent columns on election fraud '04, you need to get aware. Send the new report to him. Invite him to the press conference (don't know where he's located). Get his help in publicizing your work, and fundraising. This guy is GREAT!

http://www.huffingtonpost.com/theblog/archive/2005/05/to-byron-york-and-other-o.html

Sports = fairness = transparent elections! This is the key to America's heart!

[eomer]

10. Kathy, do you have a flier...

or anything that could be used as one?

If so, I would be willing to print and post some around town, particularly at UM, FIU and Miami-Dade College.

PS, I guess I should check the schedules first. I'm not sure whether those campuses are in finals week, in between semesters or what. Regardless, I can post info somewhere around town.


On Edit: are there sessions that will be open to the public at the AAPOR event? That's what I was thinking of in terms of doing local postings. If there are not, then I'm not sure local postings would be the way to go.

[sunshinekathy]

42. Thank you, and I didn't see this in time darnit.

Argh. Thank you.

I "really" need help. I'm getting further and further behind. That is an EXCELLENT Idea.

Thank you for that offer and I wish I had time and been organized enough to create a flier for you.

I need to find someone that can come live in the extra room (w/ own bath and nice mountain views) in my home and be paid a pittance ($400/month) to help me run US Count Votes. I need five of me, or funding. I can't keep up and my personal life is a mess.

Do you have any suggestions for how I could best find someone who may be willing to help run US Count Votes like that?


Best,

Kathy

[Febble]

11. Just bear in mind

that all simulated data (including mine) are exactly that - simulated data.

If we want to know whether fraud occurred we need real data. All I can do, and all USCV can do, is to devise scenarios that might produce the data points in the E-M report.

We do not know what the actual data looks like. All I have argued, and continue to argue, is that there are many forms that data could take, and still give the data points in the report, and that there is no way we can rule out randomly distributed bias (whether in count or poll).

I believe USCV are going to argue that the data points we have are consistent only with a pattern of bias that is greatest in Bush strongholds and least in Kerry strongholds, and that this is the signature of fraud. If I am wrong, please correct me, but this seems to be what is implied in the USCV paper.

Whether or not such a pattern is indeed the signature of fraud, the one and only way of answering the question as to whether that pattern really exists in the data is, not by simulations, but by asking it of the data itself.

Warren Mitofsky is going to be presenting at AAPOR. Why doesn't Ron Baiman ask him whether there is, in fact, a significant correlation between bias and vote-count margin in the data? And whether that correlation indicates greater bias (in the direction of Kerry over-estimates) in higher Bush-count precincts? If the answer is yes, the USCV model is upheld. If the answer is no, then mine is. Actually I win both ways, because all I am saying is that we can't tell without seeing the actual data.

And it still leaves us with the question as to whether either pattern implies fraud.

But now that USCV has such a clear hypothesis has become so clear, AAPOR would seem like a good time to pose the question to the man himself, no?

If that slope is there, I will be only too happy to concede.

[pgh_dem]

13. I'm still confused why E/M gets the benefit of the doubt...

What conditions exist that gives E/M any trust with regard to answers they might give at a conference?

Rather than ask a question about their data, why not just ask "Will you release your data to a responsible third party, which will review your data to see if it actually comports with the rBr hypothesis?"

Honestly, what is stopping Warren Mitofsky from just lying about what the data says? Why on earth would anyone trust a word out of his mouth? At best, we would be relying on the word of an incompetent, who somehow missed these hundreds of thousands of shy bush voters (There would have to be at least that many in order to account for 'randomly' having so many in the sample, right?). At worst, we would be trusting that someone complicit in a fraudulent cover-up would just volunteer evidence that contradicted his thesis.

Related: Is there any indication from Zogby, other pre-election pollsters that indicate a high rate of registered republicans refusing to be polled as to presidential preference? If so, why would E/M's model not have taken this into account? If not, why would so many repubs voice their support for * over the telephone, but refuse to state it to an exit pollster?

I think the last fallback position of E/M has already been voiced by some other poll-theorists on DU. That is: anecdotal evidence (my mom knows somebody who..) that *democrats* (read 'security moms') were either avoiding or refusing to answer pollsters that they had actually voted for *, because they didn't want to 'change horses' in the middle of a war. (Edit to ease back on conclusion) Kerry exit polls could be inflated in this way (cross-party voting under-represented in polls), and even more signficantly by dems lying about voting for Kerry when they really voted for * (causing double effect of impression of vote for K that really went to *). However, none of the anecdotes I've read have suggested that. I'm sure folks will get around to it.

Again, these fallback positions are just so much hypothesizing/number-shuffling without access to the actual data. E/M should not be allowed to dance around conclusions, offering unsubstantiated support for any and every theory that avoids the possibility of inaccurate vote counting.

[Febble]

14. Well if Mitofsky is going to lie about the data

at the conference, how do you know he hasn't lied in the report?

The entire USCV position is based on means and medians provided in the report.

If Mitofsky is part of some grand conspiracy, it's a bit late for him to be lying now.

And if he was lying before why should we trust any of the data that USCV are basing their inferences on?

[pgh_dem]

16. I think my point was that he did lie in the report...

...when they suggested that rBr could account for the deviation.

I think my accusation is supported by the possiblity that the released data was intentionally vague, and that initial criticism of E/M's job was based specifically on deviations that could only be legitimately addressed through access to the real data.

In other words, my impression is that the entire USCV position is that the means and medians provided in the report do not, in and of themselves, support the rBr hypothesis, independent of precinct-identified data (anybody have a short-hand for that, i'm always unsure of how to concisely refer to the 'stuff we don't know, that E/M does'.)

I dunno, does that address your questions? Just saying he wouldn't be starting to lie now, just continuing to lie...and as long as he doesn't ever have to tip his cards, he can bluff around the rBr hypothesis indefinitely.

ps I'm glad you haven't left DU. I do believe others who have tangled with TIA in these discussions have been disingenuous, but I (for one) am convinced of your integrity.

(edit to add tinfoil hat grouphug)

[Febble]

17. Thanks, I appreciate the hug

I wouldn't call a hypothesis a lie, but let that go.

The question is whether Mitofsky would make up data. Frankly, the thought of a respected academic and social scientist making up data is completely unlikely. Academics will argue vociferously about interpretation of data, and may be biased in their interpretation of data, but the data is absolutely sacrosanct. Data fabrication is a hanging offence in academia, and the chance of someone of Mitofsky's standing indulging in it seems beyond belief. I know, I know, I'll get that tinfoil hat on in a moment....

But the real point is, if he's been making up data then the E-M report is worthless, and so, by implication, are the inferences made by USCV from that data.

And if he's not making up data, just, in your view, misinterpreting it, there can be no reason to doubt his answer to a simple data question. You don't even need to say why you are asking.

The question is, simply:

Is bias correlated with vote-count margin?

If it is, then USCV may want to interpret it as fraud. Others may devise alternative hypotheses based on response rate differentials.

If it is not, then everyone needs to think again.

But if USCV is so sure that the data presented in the E-M report are incompatible with a regression line of zero through the whole data set, which is what I am arguing is perfectly possible, then there can be no reason not to ask.

He may not tell us, but then, on the other hand, he may.

People (including me) have been clamouring for the data, or alternatively for specific hypotheses to be tested against the data, and here, finally, at AAPOR, we have a clear hypothesis and an opportunity to have it tested.

Go for it, I say.

[pgh_dem]

18. regarding hypothesis vs lie

All I'm saying is that the hypothesis could be considered a lie of omission. I'm not talking about E/M insisting on the rBr hypothesis as fact (although I can't figure why, if they have all the raw data, they can't prove this hypothesis to the satisfaction of critics). I'm suggesting that it would be a lie to avoid stating that their data is at least as consistent with counting error causing deviation as with their hypothesis of shy * voters, for which their substantiation has been intentionally vague.

As respected academics and social scientists, is their adoption of a null hypothesis of a clean election warranted, given widespread reports of machine error, vote suppression, and acknowledged vulnerability to vote count manipulation? Aren't they assuming a fact not in evidence, simply because they reject 'unproven' charges, (which would signficantly affect their results) that haven't actually been investigated? Further, if they are withholding information which can refute or support their hypothesis, why would they be trusted to give an honest answer to any questions?

To take a different approach, say the vote count was in fact in error, but was (as I think you've posited) due to small quantities of vote-switching over a much wider range of precincts (* and K strongholds alike). The answer to your questions could be 'No, the bias is not correlated with vote-count margin', and still be misleading as to what the raw data *in their possession* shows. So why ask a question with an obvious out?

I think eight months later, there has been more than enough foot-dragging on the part of every party that has access to illuminating information (a few that leap to mind: Blackwell on recount procedures, Damschroeder on machine allocation, FBI on Warren County 'terror lockdown'), that it is time to stop pussy-footing around and just ask the direct question which may provide a definite answer.

[Febble]

21. I agree

I don't think a correlation one way or the other would prove or disprove fraud.

But Kathy and USCV seem fairly keen to refute my claim that there may not be a correlation. If fraud doesn't depends on their being a correlation, why go to such lengths to refute me?

And what every schoolboy knows about correlations is that they cannot prove causality. Actually, like most things that every school boy knows this isn't strictly true, but for present purposes it is.

E-M does not cite rBr as fact. They cite it as a hypothesis that could explain the data. This is all a hypothesis can do.

Actually they do more than that. Because they find that error was greater (in the direction of over-estimating Kerry's counted vote, and I'm not going to be drawn into an argument here about whether that is because the vote as wrong or the poll was wrong, I'm just trying to be clear as to the direction of error) in places where sampling protocol was more likely to be compromised, it is actually a reasonable inference that departure from sampling protocol had something to do with the discrepancy.

Sampling bias is a major headache in any survey, and although it can be minimised with good sampling technique and high response rates, it will never go away. You can never be sure that the people who refuse to take part in a survey are the same kind of people (as we say in statistics "are drawn from the same population") as the people who agree to take part. They usually are not. So the worse your technique, the worse your sampling bias is likely to be. This is well-known, and there are vast undergrad textbooks on the subject.

So, if you have evidence of differential non-response (and forget the possibility of fraud for a moment) the first thing you test for is: was it greater when various things conspired to make sampling protocol less than idea. And it was. So differential non-response starts to be supported.

The problem with testing fraud hypotheses is we don't have any. We don't know what fraud looks like. Where it occurred. Contrary to some allegations, it the discrepancy was not greater in swing states. USCV thinks it was greater in Bush strongholds, and that this is evidence of fraud. I think that a) it was not necessarily greater in Bush strongholds, and that b) like you, even if it was, it would not rule out fraud.

Frankly I think the whole exit poll business is unlikely to tell us anything. Correlations, and all this analysis is correlational, don't allow firm inferences regarding causality. What inferences they do allow, so far, tends to support the hypothesis that poor sampling protocol allowed differential non-response to be exacerbated in some precincts, which in turn supports the hypothesis that it was a factor.

As I say, it would be surprising if it was not.

And check out my paper on Damschroeder, if you haven't already:

http://electionarchive.org/index.php?option=com_content&task=view&id=65&Itemid=63

[Amaryllis]

44. Mitofsky is COMPLICIT in election fraud! have you checked out Votescam?

http://www.votescam.com/chapters.php

[dalloway]

15. NOT FRIDAY!

The conventional wisdom I constantly hear around these woods is that Friday is a TERRIBLE DAY to put out any important news--that it totally gets lost and gets very little attention. I am not a media person, so please, those with more media experience jump in here.

Too much advance notice though and we may end up with another national terror alert.

[Febble]

23. I'd wait until we hear what Mitofsky has to say on Saturday

if I were you.

Maybe he's proved it was all fraud after all. There must be some reason for that session being moved to prime time.... :)

[TruthIsAll]

32. Fat chance. n/t

/

[elehhhhna]

26. Yep. Publish that on Friday and they'll have all weekend to

prepare a pine-cone dropping air-o-plane to buzz the Capital on Monday.

[Chi]

24. Use the conference to challenge Mitofsky, that might get the Publicity.

[kimpossible]

27. "Vote Embezzlement" is a great term!

Much catchier than "election fraud". And it gives people a way to talk about the problem and still save face about using the F word (fraud).

Let's start saying Vote Embezzlement everywhere!

[EvolvedChimp]

28. Scary Stuff

How long did this whole procedure take you?

[CantGetFooledAgain]

29. Kathy: thank you for all of your efforts

I have a question that I have not yet received an answer for, maybe you can help.

If the "raw" exit polling data are still being witheld from the general public, what reasons are given for this action? Is there any precedent for this?

I cannot think of a single, legitimate reason, at this late date, well after the inauguration, that all of the raw data would not be released and fully available for all to run their analyses.

Under what pretense is E/M concealing this information? And, why hasn't it been leaked by any of the media outlets that used it?

[Sawyer]

30. The reasons given are two:

One is that this exit poll was paid for by E-M's clients, the price was not small, and they would have to authorize the release of that data, since they own it.

The other reason is confidentiality. If E-M releases the raw data, individual responders may be identified from it by inference, and they were promised complete anonymity when they responded.

[CantGetFooledAgain]

33. Those aren't bad reasons, but...

I can't believe that there isn't some arrangement that could be reached, given the potential of this data to either prove or put to rest the various suspicions surrounding this election.

[sunshinekathy]

38. New USCV Study Posted - Shows Pattern of Bias vs. Vote Theft

For Immediate Release - Press Conference Today

2004 Presidential Election: Hypotheses of Fraud Remain Credible; New Scientific Study Released
http://uscountvotes.org/ucvAnalysis/US/exit-polls/USCV_exit_poll_simulations.pdf


Contact: Kathy Dopp, US Count Votes, President

[phone] [e-mail] electionarchive.org

The persistence of credible hypotheses of election fraud, six months after the election, underscores the fragility of the U.S. electoral system. US Count Votes continues its systematic statistical study of the discrepancy between the Edison-Mitofsky exit polls and November's reported presidential election results.

Miami, FL. – Ron Baiman, Ph.D. of US Count Votes and the Institute of Government and Public Affairs of Chicago, will release the new results at the meeting of the American Association of Political Opinion Researchers today, Saturday at a 2:15 p.m. Press Conference in the Hotel Fontainebleau Hilton Resort lobby, 4441 Collins Avenue, Miami Beach, FL 33140.

Peter Pekarsky, the lead attorney being sued by Ohio Secretary of State Kenneth Blackwell, will be available to talk about what has happened in Ohio, and why the election needs to be investigated.

Mitofsky, of the Edison/Mitofsky group who released an analysis on January 19th of their November 2nd exit poll that had predicted a strong win for Kerry, will also be in attendance at the AAPOR conference.

The National Election Data Archive (NEDA) today has released a new report, demonstrating that data from the Edison/Mitofsky analysis is consistent with the hypothesis of a corrupted vote count, and inconsistent with the competing idea that Bush voters were under-sampled in the poll. Using numerical modeling techniques to simulate the effect of polling bias, NEDA scientists are able to reproduce signature patterns in the Edison/Mitofsky data by incorporating a general shift in the official vote tally in the model.

Most telling is the fact that the highest participation rates and the peak disparity between poll and official returns both occurred in precincts where Bush made his strongest showing. This feature of the data is inconsistent with the Edison/Mitofsky assumption that polling
bias was responsible for the gap.

For the complete report, see
http://uscountvotes.org/ucvAnalysis/US/exit-polls/USCV_exit_poll_simulations.pdf

This paper follows an earlier study released on March 31, 2005, by a group of statisticians for the National Election Data Archive Project, Analysis of the 2004 Presidential Election Exit Poll Discrepancies.

[pauldp]

39. Kick! This is Huge! n/t

[Febble]

41. Not necessarily.....

See my post on Kathy's new thread.

Lizzie

[sunshinekathy]

45. Yes it IS - Febble is working with Mitofsky

Here is the more recent thread and press release:

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=369091&mesg_id=369091


Please take the time to print and read the 11 page US Count Votes' paper with lots of pictures.

http://uscountvotes.org/ucvAnalysis/US/exit-polls/USCV_exit_poll_simulations.pdf

If you want to see how simple Liddle's equations were to derive from the work that US Count Votes did, read appendix C and D.

She was in our team for a while, but the rest of our team, when we derive a few equations, write them up in an easy to understand way, and send them to the group for inclusion in our collaborative work, not rush to publish them independently as though we had singly done all the work to obtain them ourselves.

Febble, or Elizabeth Liddle has been working directly with Mitofsky, as Ron Baiman and I figured out, and Liddle openly admitted on another thread, so perhaps her assertions come from that direction.

Please take the time to read the momentous work of US Count Votes. It should be the shot fired round the world.

Try using our spreadsheet simulator for yourself. We'll be making improvements on the vote shift simulator soon.

Kathy Dopp
http://electionarchive.org

p.s. Please put copies of the USCV paper into the hands of your elected officians so we can educate them not to concede by the Nov 06 election.