US Count Votes.org seeking precinct level data

[L. Coyote]

This via e-mail:

"The PhD level 2 dozen or so statisticians on our email list for statisticians have come to the conclusion that to prove anything validly, we need to have precinct level data.

If you can begin collecting precinct level data from each county's election officials in each state in its orginal form, that is what we need now to do our study.

It does cost money and can be a pain in the neck, but we urgently need volunteers to begin collecting this data for us.

Best,
--
Kathy Dopp
US Count Votes
http://USCountVotes.org

[StephanieMarie]

1. Go to the site and donate

What a great cause! And they're going to use the data to create a nationwide database by precinct that can be used in future elections to highlight possible "irregularities". Awesome!

[glitch]

2. Precinct level data for WA State King County right here

http://www.metrokc.gov/elections/canvass/2004/index.htm go for the 146 mb zip file.

Just King County - it's a start.

[L. Coyote]

3. USCountVotes has created individual state email lists for people who...

More from Kathy Dopp:

"USCountVotes has created individual state email lists for people who want to coordinate and discuss the study of any particular state's election results.

Please feel free to subscribe and ask your friends and associates who are working on a particular state to subscribe and freely use these list which are there totally for the use of persons like yourself who are working on analyzing a particular state's election results.

http://uscountvotes.org/index.php?option=com_content&task=view&id=61&Itemid=70

Also, please ask your friends, family, and associates to join you in donating, in order to speed our project's technical implementation along.

http://uscountvotes.org/index.php?option=com_content&task=view&id=13&Itemid=40

Thank you again.

[L. Coyote]

10. Wrong domains for the state email lists .. now corrected .. re-subscribe

The individual state mail lists (whose domains all end with .net not the .org) may have been wrong when you subscribed. They are under development, so it is understandable.

Please (re-)subscribe to the state email lists on:

http://uscountvotes.org/index.php?option=com_content&task=view&id=24&Itemid=71

[skids]

4. Where do they want them sent?

I have some data, but they've given me no clue as to where to send it.

[Carolab]

5. Go to their website

and find the contact link

[skids]

7. I kinda doubt.

They want me to paste data into that web textbox.

Their FAQ says:

 Hang on to the orginal documents until USCountVotes' public archival 
system is announced via USCountVotes' email announcement list.  We will 
need historical data of all types as well, from prior elections, so 
please try to obtain those as well.  

I am on that list, and haven't gotten any email saying it's time yet.

So my advise to everyone is, check rasgroup.com to see if data for your county has been posted, if not, collect data where you can get it, and post it to rasgroup.com. When USCountvotes is ready, we can post it.

I'll try to get to posting what I have sometime this weekend.

(btw, I like the new USCountVotes webpage.)

[Lil]

11. Skids, if you don't mind, which state/county do you have?

Thanks.
Lil

[skids]

13. Several Ohio counties.

Not all the data is very complete -- some of the BOEs don't seem to care to break out overvotes, some lump all non-votes and write-ins into one category, and some just massage the "ballots cast" numbers so there is no overvote to see. Most don't break out absentees, and when they do, they don't break them out by precinct, though if you have the down-ticket data they are usually broken out by senate/rep district in those cases.

Most of the data is only 2004 presidential, but for some of it I've hit the web to get primary 2004 and some down-ticket races.

I also have some extracted results from sundry other county webpages all over the U.S.

As far as OH, what I have is this:

Pickaway, Geauga, Stark, Hancock, Ottowa, Cuyahoga (2 sets), Perry, Sandusky, Butler, and the correct data for Trumbel's absentee discrepancy (I emailed Werner and he helped me correct a few minor transcription errors.)

I may be getting Greene by hardcopy in the mail soon, but it will take a while to OCR/proofread it.

[Lil]

15. Thanks, Skids.

I passed on the email address listed on your website to one of the researchers involved in uscountvotes.org. It might be faster right than communication via their website. Thanks again.

[sunshinekathy]

41. Hi from USCountVotes

You've done a good job of recapping some of the efforts going on.

I'm working full-time on this effort and so is USCountVotes' sysadmin, but there is a lot of work to do. I'm as frustrated as anyone else is with the rate of progress which could have been much quicker with the help of financial resources or political capital.

We're a few days away from testing our data archival system (which is only the first step of this process). Yet as soon as it's tested we'll have to take it down for a few days to mail the server out to its high-speed location in California.

I've had to dip into my own meager savings to pay our now full-time (as of this week only) system administrator and programmer through the end of January with Bruce Bodell, USCountVotes' VP.

See http://uscountvotes.org/ and About Us to see board members.

There is a lot going on behind the scenes right now that will mean success. Please ask people to sign up on the individual state lists and coordinate with each other so that people avoid duplicating efforts at collecting data:

http://uscountvotes.org/fairelection/statesubscribe.html


Best,

Kathy Dopp
http://uscountvotes.org

[adolfo]

6. Florida here!!!

I've been advising this route for a while, it's about time. This is the only way we can know which areas to target. How do you guys think the Rethugs know where to strike?

Florida precinct data (some have unofficial results):


http://www.recountflorida.com/ivrs_county.php

[L. Coyote]

8. kick

[Lil]

16. Adolfo, I sent you an email message from one of the researchers. . .

. . .from uscountvotes.org.

Since communication from their website may be a problem right now, you can contact him directly if you wish to. We didn't see an email address on your website. Thanks for your work.

[NVMojo]

9. requesting precinct level data is how my county found 271 votes

left on 3 cartridges from 3 machines and the votes had not been counted! Lack of cross referencing is what happened. And this was on paper trailed Sequoias.

[Lil]

17. " Lack of cross referencing. . was on paper trailed Sequoias"

Isn't that interesting. Cross referencing with poll books?

[L. Coyote]

12. Another e-mail from US Count Votes .. 50 people need to re-subscribe

US Count Votes is getting traffic. Rec'd via e-mail:

Over 50 people today tried to subscribe to individual state volunteer
email lists and could not do so because I made a mistake on
http://uscountvotes.org

Please subscribe to the individual state email lists again:

http://uscountvotes.org/index.php?option=com_content&task=view&id=61&Itemid=70

or

http://uscountvotes.org/index.php?option=com_content&task=view&id=24&Itemid=71

The individual state email lists are for your use to coordinate your efforts to achieve integrity of elections within any state. Anyone who is leading an or helping with any effort to collect and analyze data or obtain recounts within any state, please use that state's list to help you and others to sign up so you can more easily communicate and coordinate re states you are working on.

You may use these lists any way you wish.

If there is any problem develops with spammers, etc. on any of these state lists, please let me know and I can add some of you as moderators.

If anyone would like to assume, or has already assumed, a leadership role in a particular state's effort, I would be happy to post your name, email, or nonpartison web site URL next to that state's mail list.

For now, every member on a state email list can post emails to it.

http://uscountvotes.org/index.php?option=com_content&task=view&id=24&Itemid=71


Best regards,

--
Kathy Dopp
US Count Votes
http://USCountVotes.org

[rumpel]

14. Geo started a whole thread on this for CA

ironically CA has passed a State constitutional amendment this past election. Public's right to access "Public Records, Open Meetings" Measures 59. Therefore, Geo thought it shoule be free.

In any event for Los Angeles County link

http://lavote.net/svc/posttally0024/bigframe.htm
and the first DU thread is

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=91254&mesg_id=91254&page=

Part II is

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=91254&mesg_id=131075&page=

[Lil]

18. Thanks, rumpel

[jfern]

19. Here's some

Complete 2004 Tompkins County, New York results:
http://owasco.co.broome.ny.us/boe/2004/general%20election%20results%202004.htm

Complete 2000-2004 Alameda County, California results:
http://www.co.alameda.ca.us/rov/sov.htm

San Francico data for the last 10 years:
http://www.sfgov.org/site/election_index.asp?id=5877

Here's the worst precinct in all of San Francisco:

PCT 2734 Kerry 310 Bush 223 Others 8 - only 57% for Kerry

[sunshinekathy]

45. Need by vote type breakout & Thanks for CA results links

Hi. I linked to them for now from:

http://uscountvotes.org/fairelection/statedata.html

The ny url wouldn't load for me.

Things always take a little longer than expected, but we're making lots of progress.

You can help by calling these counties, making friends with the county election officials and their staff and obtaining a by voting type breakout of the data for us so that we know how many ballots were cast via

provisional early
provisional election day
absentee
early
election day

and what vote counting method was used for each by precinct or county.


Best,

Kathy
http://uscountvotes.org

[Lil]

20. Just a thought - what about putting a brief description in subject line -

of what you have.

It might be easier to sort them out especially if we start getting more responses?? (IMHO)

[Lil]

21. If you have the precinct voting records for your voting precinct. . .

. . .post your State/County/Precinct in the subject line and I can get your email from your profile and pass that on to the researchers at:

http://USCountVotes.org

Thanks for your help

[Niche]

22. So many DUers have info. Please guys come back and help.

Don't know where all old time DUers are but can you email privately. So much data used to be posted now there's nothing. Please come out of the deep underground DUers and help.

[Lil]

23. Thanks, Niche. - also - if any of you can contact your local office . . .

. . . and get PRECINCT results it could very important. Sometimes only a selected few precincts are chosen for cheating. The county count may not raise red flags.

[Lil]

24. Precinct vote tally inf. welcomed here. Thanks. n/t

[Carolab]

25. Kick

because THIS IS IMPORTANT

[L. Coyote]

26. kick this n/t

[minvis]

27. Will be getting Clermont County, OH the first week of January

This will be precinct by precinct count as well as precinct shared locations and the ballot order for each precinct.

[Lil]

29. Thanks, minvis.

Are you working with US Count Votes.org? If not, and if you have trouble getting your data to a researcher there, let me know via email or private message in my profile. I don't know if the website is still having problems in that regard or not.

[minvis]

30. US Count Votes.org

No, I'm not working with US Count Votes.org, so when I receive the information, I'll e-mail you so you can send it on to them.

[Lil]

31. Great! Thanks very much for your help, minvis.

[helderheid]

28. Is there any chance the moderators might pin this?

[L. Coyote]

32. kick n/t

[Lil]

33. Another thought, if anyone reading this thread NEEDS specific data...

. . . from a precinct, I think it would be appropriate to ask for it here?? Someone on the ground might be able to snatch it for them. :-)

[L. Coyote]

34. Another E-Mail from US Count Votes ... Go to domain to help ..

Go to their domain and make contact if you can contribute anything. Get on the mailing list or join a state list.

Here's their latest mailing:

23 Dec 2004
From: "Thomas G. Knight" <[e-mail]>

US Count Votes Gains Staff, Readies For Archival Kickoff
December 23, 2004

USCountVotes - a new volunteer scientific project to investigate the
accuracy of elections in America - formed in December 2004 to objectively
investigate the accuracy of elections in every part of America. Our goal
is to provide statistically-valid methods to help detect any and all
precincts with probable vote-counting errors.
http://uscountvotes.org/fairelection/missionstatement.html

New staff announcement:
Just one week after launching our new web site, US Count Votes has
committed itself to hire its first full-time paid staff person. Thomas
Knight has been in the IT industry for over 9 years, specializing in
software development and system administration. He is currently wrapping
up the programming and testing needed to prepare for the launch of US
Count Votes' election data archival system - the first step in our
data-gathering and analyzing mission.

Submit your local media contacts:
In order to ensure that USCountVotes public archive and statistical
analyses gets full public exposure when we announce its launch, Thomas
Knight has programmed a "media collector" online form. Please help us to
build a comprehensive national media contact list, covering every state in
the US, by submitting the names and contact information of newspapers,
broadcast media outlets and independent journalists that you know at:
http://uscountvotes.org/fairelection/mediacollector.html

Donations update:
We are very grateful to our early financial supporters, who've allowed us
to purchase a 150 GB database server/archival engine and make our rapid
progress to date possible. We acknowledge all contributors who wish to
make their names public at
http://uscountvotes.org/fairelection/contributors.html - donate today and
add your name to our list of supporters. We encourage everyone to help
ensure the fairness and accuracy of voting in America by making a
donation: http://uscountvotes.org/fairelection/donate.html
(USCountVotes is registered as a Utah non-profit and will file for
tax-exempt status with the IRS in 2005.)

Volunteer opportunities for fundraisers:
If you are experienced in fundraising or grant writing we need your
advice, and whatever amount of time you can spare, to ensure that we
secure the financial resources we need to be successful (and justify
Thomas Knight's "leap of faith" in agreeing to work full-time for our
newly-formed non-profit organization!) We need his continued employment
in order to fully implement our database and analysis. Please email our
VP, Bruce O'Dell (http://uscountvotes.org/fairelection/boardmembers.html)
at [e-mail] or join our fund-raising email group by sending
an email to [e-mail]

Volunteer opportunities for data collection:
US Count Votes plans to begin accepting data documents for its archive
from its civic volunteers before the end of the year. Volunteers in each
state are needed who can collect precinct-level data for every
jurisdiction in the US, starting with the 2004 elections. We're
encouraging you to sign up for our state volunteer mailing list now, so
that you may team up and coordinate with other volunteers, and we can
notify you just as soon as we're ready to go.
http://uscountvotes.org/fairelection/statesubscribe.html

About USCountVotes:
USCountVotes will create a single database containing precinct-level
election results for the entire United States. All data and analysis will
be published and there will be free public access to the data, via the
Internet, that will allow anyone to verify the results. This data will
then be analyzed by statisticians.

If the database and analytic tools can be put in place by the national
election in November 2006, for the first time in American history, it
could be possible for candidates to be reliably warned of indications of
machine or human-caused vote count errors, in time to challenge the
results. With a sound scientific approach and methodology, it may be
possible for US Count Votes project staff to serve as expert witnesses, or
to help develop statistical evidence in support of legal filings.

Please visit USCountVotes online at http://www.uscountvotes.org for
additional information.

Contact: Bruce O'Dell at 612-309-1330 [e-mail]

--
Thomas G. Knight
System Administrator
http://www.uscountvotes.org
[e-mail]

[Lil]

35. kick

[L. Coyote]

36. kick n/t

[PeterPan]

37. Is there a basic format determined?

Since every county in every state seems to have a different way of accounting for themselves it would be useful to know exactly what they are looking for.

Also I would warn very strongly to distinguish between unofficial numbers that may be on an SOS website and the cetified numbers found on the official canvass reports. Much of the analysis I see on DU - excellent and insightful as it is - has been based on unofficial numbers. I was able to get the New Mexico numbers in a massive Access file directly from the Secretary of State's office - I don't know how easy that will be in other states but it is essential that serious analysis be based on certified numbers.

Great project!

[L. Coyote]

44. To start, the format of the official canvass report -- I think ...

I can't speak for them, but I think the official data is the best place to start. They are also putting up links to data sites.

As you can see below, Kathy Dopp of their organization is following this thread too.

[rumpel]

38. I have LA County precinct data, but only on the presidential race

Unfortunately, the way they have posted the sheets, are only on a per page view basis. Therefore, I have created pdf files for each 130 or so pages. In hindsight, after trying out various ways to import the data into excel, I found out, I might have as well saved it in text format.
I can e-mail the pdf files in batches, as they amount to a large file as a whole. I found it comes back undeliverable in the past. (not to you)
Please tell me if pdf is good enough at this time.

[rumpel]

39. btw. CA State still has not posted the final on the website

In order to prepare the official Statement of the Vote for the November 2, 2004 Presidential Election, these pages are no longer being updated. The official final results will be posted after they are certified on or before December 11,2004.

Therefore, what I have is the latest publically obtainable data.

[helderheid]

40. kick

because Kathy Dopp is one hard working woman (and I know - we both live in Utah and have worked on keeping the evote machines out of Utah).

[sunshinekathy]

42. Yea I never sleep

I "am" hard working. Thanks.

We've actually managed to get Utah to drop its plans to select DRE voting machines of ES&S and Diebold and recently were lucky enough to receive word that Utah's Election Director has just announced that she will quit at the end of December. She had irrationally continued to push for DREs even after many letters from Utah computer scientists advised against purchasing DREs, and even after she was forced to stop the selection process that was supposed to result in a final contract being awarded to a DRE vendor in December.

Well, I've become an old insomniac who isn't willing to let democratic elections die if I can do something about it, and I believe I can.

Some good things seem to be happening. It looks like the Social Science Community may join our effort to push for funding for a thorough study of all the election results and a few people in New Mexico have done a wonderful job gathering precinct-level vote-type level data that should help begin the process.


Merry Xmas or whatever holiday you celebrate or don't!

Best,

Kathy Dopp
http://uscountvotes.org

[jkd]

48. Thank you, Kathy

Hi Kathy, I'm also a Utah resident next door in Heber City. Thanks for your efforts to stop the DRE's here.

I speak with the Wasatch County Clerk, Brent Titcom, as often as I can about the purchase of a system to replace the punch-cards. He informs me that he'll probably have to go along with what the state decides. I let him know that I would hold him personally responsible if the system chosen by our county didn't have a paper trail and wasn't easily re countable. He said that although ES&S and Diebold had been considered, he thought that some form of opti-scan system was now being decided upon. He didn't have any details. When I suggested that they just vote a paper ballot and recruit enough people to count them, he vehemently disagreed. He said that would be far less accurate than any system they could chose. I believe some of these county clerks would like machines that just selected the winner without having to go through the process of an election. Perhaps that is what we already have.
I wrote a letter to the editor in the Wasatch Wave the week prior to the election on the impending crisis with electronic paperless voting. I read the Park Record, but you probably don't read the Wave. Why would you? I have a hard time accepting something that I can't see. I had a few well wishers, but most people here now probably consider me a fringe kook. Sorry, this should be a PM message, but I don't have enough posts on DU to do that.

I really respect your friend, L. Coyote. I learned a lot from the posts. If people of that caliber are working with you, I expect great things. I will sign-up, but I don't know if I'll be much help. I'm not much good at math or statistics, and I don't have excess bucks. What I can do, I will.

[sunshinekathy]

43. USCV Needs Precinct Data by Vote Type

Hello Everyone,

It is heartening to see that people are willing to work hard to restore democratic elections in America.

FYI, I just spoke with Warren who has been collecting precinct and vote type level data in NM by working with many friendly staff persons who've helped him obtain all the detail he needed. We'll try to post some of the New Mexico data on http://uscountvotes.org within a few days and analyses as soon as it's been peer-reviewed.

It turns out that it is critical to break down the precinct level election results by these types:

provisional - early
early
provisional - election day
election day
absentee

because Warren has found out that in New Mexico there are many phantom votes (2,087 state-wide). i.e. there may be 200 absentee votes cast but 500 counted, hence 300 "phantom" votes in a particular county or precinct. So essentially there is vote padding going on that is covered up or undetected because precinct level or county level reports included these phantom votes in with their "undervotes" so that the phantom votes cancel out some of the undervotes making it seem like there are fewer undervotes in a county.

Warren and others working on the New Mexico recount have taken these numbers directly from the sources there, and he feels that all states will show similar phantom votes if we obtain the detailed information by type of vote, as well as by precinct.

Please sign up for our state email lists and use them to communicate and coordinate data gathering for any states:

http://uscountvotes.org/fairelection/statesubscribe.html

We have to finish some programming, test it, then ship our server to our co-location in California before we announce, hopefully by the end of December, that we're ready to begin archiving the public election documents and web pages.

Best,

Kathy Dopp

http://uscountvotes.org

[helderheid]

46. Wave!

Glad to see you here Kathy! :)

[rumpel]

47. Project CA volunteers hope you continue here.

:hi: