Does anyone know information on Election Science Institute, formerly

[mod mom]

Votewatch? If OH Republican Damschroeder hired this group(ESI is providing their own funding for the analysis), you have to wonder.

Questions:

Who is funding this group?

Does anyone have background information on Steve Hertzberg?

Why did Damschroeder agree to this analysis?

[KaliTracy]

1. Rebecca Mercuri PhD has done much to push for Voter Verified Paper Ballots

in wake of the 2000 elections.


Rebecca Mercuri
Rebecca Mercuri is among the foremost experts on electronic voting. She has been interested in electronic voting since 1989. She is presently a Fellow at Harvard University's Radcliffe Institute for Advanced Study (http://www.radcliffe.edu) where her research focuses on transparency and trust issues in computational systems.

She is well known for having popularized the idea of using voter-verified paper ballots (often referred to as the "Mercuri Method"), whereby an electronic voting machine prints a paper ballot for the voter to verify and deposit in order to cast their vote.

http://www.answers.com/topic/rebecca-mercuri (provides several links)

perhaps someone knows more though.

2001 Testimony Presented to the U.S. House of Representatives Committee on Science
Subcommittee on Environment, Technology, & Standards
Tuesday, May 22, 2001, Room 2318, Rayburn House Office Building

"Good Morning. I am Dr. Rebecca Mercuri of Lawrenceville, New Jersey, an Assistant Professor of Computer Science at Bryn Mawr College in Pennsylvania, and President of Notable Software, Inc. (a New Jersey computer consulting firm). My testimony today represents my own opinions and not those of my employers or any professional organizations with which I am affiliated. Thank you for the opportunity to address your Committee on this important matter.

"For the last decade, I have investigated voting systems, with particular emphasis on electronic equipment (hardware and software) used to collect and tabulate ballots. Through this research, I have identified numerous flaws inherent to the application of computer technology to the democratic process of elections. These flaws are both technologically and sociologically based, so a quick (or even long-term) fix is not readily apparent. For example, present and proposed computer-based solutions are not able to resolve (and in some cases even increase) the likelihood of vote-selling, coersion, monitoring, disenfranchisement, and fraud in the election process.


Some of the problematic issues with electronic balloting and tabulation systems are as follows:

"Fully electronic systems do not provide any way that the voter (or election officials) can truly verify that the ballot cast corresponds to that being recorded, transmitted, or tabulated. Any programmer can write code that displays one thing on a screen, records something else, and prints yet another result. There is no known way to ensure that this is not happening inside of a voting system.

"Electronic balloting and tabulation makes the tasks performed by poll workers, challengers, and election officials purely procedural, and removes any opportunity to perform bipartisan checks. Any computerized election process is thus entrusted to the small group of individuals who program, construct and maintain the machines. The risk that these systems may be compromised is present whether the computers are reading punched cards or optical scanned sheets, or are kiosk-style or Internet balloting systems.

more: http://www.house.gov/science/full/may22/mercuri.htm

Post Election Analysis
by Dr. Rebecca Mercuri Election Science Institute (October 2003)
http://www.votewatch.us/forums/general/680938898239



edit: clarity

[Bill Bored]

2. http://www.votewatch.us/ does nt

[mod mom]

3. I found what the web site said, but was looking for personal knowledge.

A website is the information that the organization wishes people to know. If any election reformer, especially left coaster has info, please post.

I found this on Steve Hertzberg-it was listed on his bio from his website for Election Science Institute:

"Mr. Hertzberg spent the first several years of his career as a civilian within the US Department of Defense.  While serving as a Project Manager and Test Director for highly visible military development programs, Mr. Hertzberg received the U.S. Army’s Civilian Special Act Award."

Also for what it's worth, it appears from his bebsite that the funding comes from 2 groups:

1. Aguirre International-Dr Edward Aguirre was appointed Us Commissioner of Education by Gerald Ford in 1976. After leaving this position he founded Aguirre International.

2. CommonCause Education Fund.

Does this raise any questions?


Thanks.

[Wilms]

5. Links, please. I'm still lost. n/t

[mod mom]

7. link

http://www.votewatch.us/

Elections Board Announces Examination of Election Process

Goal, in the wake of the November 2004 presidential election, is to set new standards for election process


COLUMBUS, Ohio – The Franklin County Board of Elections announced today that it was launching an independent examination into the election process used in last November’s presidential election. The aim is to conduct a thorough review of the Board’s strengths and weaknesses in the wake of the much-scrutinized Bush-Kerry presidential contest in Ohio, and enact changes guaranteeing maximum transparency and accountability in future elections.

Franklin County, home to the state capital, Columbus, has already conducted an internal assessment of areas in need of improvement – such as adequate distribution of voting machines and avoidance of long lines at polling stations – and made the results available to the public last month. But Matthew Damschroder, Director of the County Board of Elections, said he wanted an external, fully independent review as well to make sure nothing was overlooked.

The new investigation is supported by the Franklin County Democrat and Republican parties and will be conducted by Election Science Institute (www.electionscience.org), formerly known as Votewatch (www.votewatch.us), a non-partisan San Francisco-based election monitoring group. Election Science Institute will not be spending taxpayer dollars, but will finance its examination with locally raised private funds. The organization will not be under contract to the County Board of Elections, but will report to a separate independent commission. Its findings will also be made public.

"We want an honest and critical review of our election operation as well as recommendations for improvements," Damschroder said. "We want this analysis to be above reproach and not simply another government funded study of a governmental operation."

[Wilms]

4. I'm lost.

Damschroeder, I guess is a state senator.

What did he hire VoteWatch for?

[mod mom]

6. Damschroeder is the Franklin County OH (Columbus) Republican BOE Director.

Why would he want Votewatch/Election Science Institute to analyze the election? This is not an INTEGRITY issue, the thugs don't have it in them. By the way this group is self funding their work.

[stevenstevensteven]

8. Hello, let me introduce myself...

I'm Steven Hertzberg, I've been a member of DU for over 3 years and have been quietly working in the election reform arena for approximately the same period of time. A friend of mine pointed me to this discussion this evening, and I'd like to make myself available to answer any questions that you may have.

First, Franklin County did NOT hire ESI. ESI filed a lawsuit (Writ of Mandamus) against Franklin County in an effort to obtain public records relating to the Nov 2004 Presidential election. As our efforts to secure information moved forward, we started a conversation with Matt Damschroeder and over time we both realized that our mutual goals could be met by collaborating to secure election transparency. We worked very hard to form this collaboration, and we are now excited that we may both have the opportunity to create an election model that can form the basis for other counties nationwide.

Second, I have personally provided a substantial amount of funding toward Votewatch and ESI over the past three years. Additionally, we have received funding from several high net worth individuals and a couple of foundations. While I have agreed to release the identity of any new donors who fund the Franklin County effort, I have not secured this release from our current donors. I am currently trying to secure the approval of our existing donors to release their names, but I do not have their approval yet. I am pleased to provide that information as it becomes available. However, it will ultimately be published with our organization's tax forms, as we are a 501c3 project.

Third, I'm Steven Hertzberg, what questions do you have about my background? Check out www.hertzberg.org for a photo and a brief bio. If you have questions after you read that page, then please provide them here.

Fourth, I think you should ask Matt personally why he agreed to this analysis. But my opinion is that he understood that we need to determine what really went right and what really went wrong during the last election - a serious scientific and objective analysis is needed. ESI has worked extremely hard to solicit credible scientists who place their objectivity and reputations above partisanship. We are all trying to raise confidence in our electoral process prior to the next big election, and I believe that this type of collaboration is the best way for us to realize this goal.

Thanks for your questions. I don't typically visit DU any longer, but I will try to check back once per day to see if you have any additional questions.

Steven

P.S. Matt Damschroeder is the Director of the Franklin County, Ohio Board of Elections.

[Melissa G]

9. Looks good! Like your web page and your quotes

I'm personally grateful to folks making a difference on this issue..
Steve's Quotes



“The means are the ends in the making” - Gandhi
and...

SUCCESSFUL PEOPLE GENERALLY HAVE more errors to their credit, and often bigger ones, than unsuccessful people. They view these in the same way that scientists view failed experiments: not as moral setbacks but as the necessary concomitants of discovery. While plodders see failure as a demon, achievers see it more as a void, oppressive perhaps but not intimidating, and capable of redemption by the first success that comes along. They know, however, that success, no matter how much praised or how well rewarded, will open up new challenges, new risks of failure.

Robert Grudin, Time and the Art of Living

[meganmonkey]

10. Thanks for introducing yourself!

Please keep us updated on the process, if you find time. As you can tell, we are obsessively interested in this!

:hi:

[stevenstevensteven]

11. Thanks meganmonkey

Dear meganmonkey,

I will certainly do my best to keep you up to date. You can also always register on our site to join our mailing list (www.electionscience.org).

Our first piece of work will be to analyze the wait times experienced by some voters in Franklin County and to see if we can develop a better methodology for avoiding long lines during future elections.

Steven

[meganmonkey]

12. I'm checking out the site

and I signed up for the mailing list. Thanks for the info!

:)

[garybeck]

13. Steven, thank you, and I have a question or two

I really appreciate your openness. When your funding sources are fully disclosed I believe we can set to rest any doubts that people may have. I hope you can understand people's hesitancy regarding the circumstances.

I have a few questions. From your message, you said that the collaboration began with a lawsuit.

Is the lawsuit a matter of the public record? If so, can do you have the case number?

How would you characterize the nature of your collaboration? It seems like it began as an investigation into 2004 irregularities, but now from your wording it appears it is more focused on creating a model for future elections. Can you summarize the purpose of your project?

Do you think the project will include hand counting of the ballots in Ohio to check accuracy of ballot counters and tabulators?

Sorry for so many questions but this is very interesting and I appreciate your openness.

Regards,
Gary Beckwith

------------------------------------
the solar bus
ELECTION JUSTICE CENTER
your home for updated information on the fight for democracy in America
http://election.solarbus.org
------------------------------------

[stevenstevensteven]

14. garybeck, I'll try to answer your questions below

Thanks for the interest. Below are my replies:

1. "When your funding sources are fully disclosed I believe we can set to rest any doubts that people may have." With regard to our funding sources, I really hope that we are able to find an equal mix of republicans and democrats. Also, please take a look at our advisory board, as this should give you a better idea of who we are. Additionally, we're setting up a bi-partisan oversight committee in Ohio, names to be released soon.

2. "Is the lawsuit a matter of the public record? If so, can do you have the case number?" The lawsuit was filed in December 2004 by Votewatch against several Ohio county BOE's and the Ohio Secretary of state. I don't have the case number handy, as we've really let the lawyers handle this. But if you want it, I'll get it and post for you. It all should be in the public record.

3. "It seems like it began as an investigation into 2004 irregularities, but now from your wording it appears it is more focused on creating a model for future elections. Can you summarize the purpose of your project?" The investigation into the 2004 election continues, as we are finalizing several reports that we plan to make public over the next several months. This effort has not been interrupted by the Franklin County relationship. With respect to the purpose of our Franklin County project, it is to build a set of best practices in one county that allow for better transparency, accountability and audit-ability in our election system. Our goal, as applauded by the Election Assistance Commission, is to create a model that may be replicated across Ohio, and perhaps the USA.

4. "Do you think the project will include hand counting of the ballots in Ohio to check accuracy of ballot counters and tabulators?" Good question, and one I don't have an answer for yet. Your suggested work scope is not included in our initial scope of work as we laid-out in our proposal. However, I am sure there will be a number of projects that we will embark upon that have yet to be defined.

Hope this helps!

Steven

[garybeck]

15. Thanks Steven, I believe #4 is the most important question

With all due respect, I don't see how anyone can investigate what happened in 2004 without manually counting some of the ballots and looking inside the hardware, software, audit logs, of the machines.

If you are not planning on checking the accuracy of the vote counting, can you let us know what you are checking? You mention a proposal about the investigation. Is this available?

Suggestion - you say there will be more projects that have yet to be defined. I think I can speak for many when I say, if you do any more projects, can you please focus on the vote counting, the ballots, the hardware, the software, and the machines? We already know a lot about the long lines, allocation of machines, etc. We want to know what happened with the vote counting and if it was done properly.

Every day that goes by before this is addressed, the longer the ballots sit there in a somewhat vulnerable position.


I hope that you or someone has the wherewithal to do what millions of people in this country want done. We want to know if those ballots were counted right. We want to know about those stickers on the Kerry ballots. we want to know why the Triad people were fixing the recount. Was it an innocent attempt at saving money or would actually counting the ballots reveal something?

Thanks

[stevenstevensteven]

16. Reply to questions

Gary Beck,

First, let me say that vote counting has always been this organization's highest priority. We are the only organization to have requested and analyzed the vote counting audit trail in numerous counties across Ohio.

Second, in your request you say, "if you do any more projects, can you please focus on the vote counting, the ballots, the hardware, the software, and the machines? We already know a lot about the long lines, allocation of machines, etc. We want to know what happened with the vote counting and if it was done properly." Gary, I'm not sure you fully appreciate the scope of the vote counting effort a large county like Franklin undertakes as part of the canvass process. To review all of the paperwork associated with the canvass is an extensive and extremely tedious assignment. Vote counting does not just involve the ballots, hardware and software. It involves hundreds of reports that create an extensive audit trail that must be reviewed.

You also say, in your request, that "We already know a lot about the long lines, allocation of machines, etc." Actually, we don't yet fully understand the facts around this issue because it has never been 'objectively' studied. What we have is rumor and innuendo, some information, but virtually no data or analysis. We plan to do this because it needs to be done in order to determine what really happened and to solve it.

Third, we've completed a comprehensive audit of the election result in dozens of Ohio counties. So, our effort in Franklin will focus on ensuring that future elections are counted accurately and that Ohioans can have complete confidence in the Franklin county election result. With this focus, our resources and our time will be centered on improving the system so that questions regarding accuracy will be answered before they are mounted. So, I can certainly see, as part of our effort to secure transparency, a focus on solutions that will facilitate full disclosure of the data stored on the memory cartridges in the Danaher machines, or better chain of custody mechanisms for the memory cartridges, or more comprehensive L&A testing. These are all approaches that will help determine inaccuracies in future elections.

Four, the election process as an extensive and complicated system, and the voting machines that you refer to are only one component in this vast system. The problem with those that favor paper trails on the vote recording equipment is that they are focused on a single risk point within the system. There are many other nodes within the system where vote counts may be affected, whether the system uses paper or not. Hence, there is more than one way to secure the accuracy of the vote count than to secure the vote recording and vote tabulating machines themselves. If the processes around the machines are sufficiently designed, then audit-ability, transparency and vote verifiability can be achieved. So, we can look to other areas to determine accuracy and we can also improve the system dramatically without altering the machines.

Finally, We will continue to review machine audit logs, as we have been conducting in numerous counties across Ohio. As far as the stickers on the Kerry ballots or the Triad tabulators, we won't be looking at either of those in Franklin county as Franklin county does not use paper ballots or Triad tabulators (we are quite familiar with the Triad systems, as we have folks on our team that reviewed this system when they were first used in the 1980's). Additionally, let me be clear that this is not a witch hunt to placate Kerry supporters. Our organization is non-partisan, and we work extremely hard to keep it this way. Our goal is to improve the election system for everyone.

P.S. The proposal that I referred to can be found on our web site: www.electionscience.org

[KaliTracy]

17. Steven cubed

(sorry I couldn't help it... ) :)

I live in Butler county -- if you have looked at any machine allocation problems in that county I just wanted to volunteer to say that my neighborhood didn't have allocation problems and in fact, considering the problems other areas had (including Warren county) my district probably had more machines than recommended, though I'm only guessing. It is for this reason (my perceived excess amount of machines in my area .vs precincts that didn't have machines) that I got involved at DU.

I also question the huge number of precincts that have "merged" or disappeared which started in the late 90s but "started" for the 2000 election.

I've got some numbers surrounding this if you are at all interested (don't want to clutter up this forum with all of the info I've posted before). But just for reference:

In the 2000 election, Ohio lost 926 precincts from 1998 (there was no 1999 precinct data).

In the 2004 election, Ohio lost 786 precincts from 2000.

The reasons given for cutting back on precincts in 2000 were "low voter turnout" numbers, and for 2004, "low voter turnout" numbers and expectation of moving to electronic voting (they merged precincts in anticipation of this, but didn't "unmerge them" when they decided not to go with the machines).

From 1992 on -- there have been increases in voter registrations, but decreases of over 500 precincts at a time.

**From 1992 to 2004 there has been an increase of voters by
1,437,734

** From 1992 to 2004 there has been a decrease of precincts by 2,372 -- the most not occurring this year, but in 2000!

I posit (but have no proof, only a "gut feeling") that the "low turnout numbers" in 2000 were because of similar problems that occurred in 2004 (long lines, not enough machines, confusion, etc.), and that people got fed up and left instead of staying in long lines to vote. It is probably impossible to get accurate turn-out information from 5 years ago (people voting .vs people who showed up but didn't vote).

Feel free to contact me for the numbers I used.


edit: clarity

[garybeck]

18. Steven, some more questions

Thanks for your reply, I am unclear about a few things you said:

About voting machine allocation:
Actually, we don't yet fully understand the facts around this issue because it has never been 'objectively' studied. What we have is rumor and innuendo, some information, but virtually no data or analysis. We plan to do this because it needs to be done in order to determine what really happened and to solve it.

Steven have you seen the data that Dr. Richard Hayes Phillips and other statisticians have collected on this? I think it hardly qualifies as rumor or innuendo. There is hard data and quite a bit of analysis on this. For example:







This is all real data, isn't it?

I'm not sure how to take your witch hunt comment. Can you clarify? Are you saying that anyone who questions the integrity of the Ohio election and wants to investigate the stickers on the ballots and things like that is on a witch hunt? There are sworn affidavits of many people who saw the stickers covering up Kerry votes. We also know the recount was done illegally; the hand counts were not selected randomly, and recount observers were not given access to the recount that is provided to them by law. I would think that anyone who cares about our democracy would want to know the truth about these things, and it would be our duty to investigate them regardless of what side we're on. Can you clarify the witch hunt comment?

I'm not trying to be critical, I'm just asking for clarification.

Thanks
Gary

[stevenstevensteven]

20. response

1. Can you please send me the links to the reports that you cite below, they are not showing up on the discusison board...

"Steven have you seen the data that Dr. Richard Hayes Phillips and other statisticians have collected on this? I think it hardly qualifies as rumor or innuendo. There is hard data and quite a bit of analysis on this. For example:"

2. "I'm not sure how to take your witch hunt comment. Can you clarify?"

I will be very brief as I'm short on time. What I am saying here is that most if not all of the conclusions that I have seen are not supported by sufficient evidence. Further, many of the investigations that I have seen have either not evaluated all of the data available, or they have omitted certain facts, or the authors do not fully understand the intricacies of elections. In short, there is strong evidence to support the conclusion that many of the reports, circulating the Internet, are part of a larger agenda, which is to demonstrate that the election was somehow stolen (these are not my words, but represents the sentiment that I have observed by many following this issue).

ESI's goal is to maintain total objectivity with respect to these incredibly important issues. We will not publish findings that do not fully account for all the facts. We will publish reports that articulate all of the potential findings that the data supports. Hence, we do not have any agenda other than to find out what is deficient about our election system and to work as hard as we can to mitigate these deficiencies.

Specifically, you say: "Are you saying that anyone who questions the integrity of the Ohio election and wants to investigate the stickers on the ballots and things like that is on a witch hunt?" My response: are you working on determining how voters for Bush may have been disenfranchised, or how votes for Bush may have been lost or recorded incorrectly? If you are not, it seems to me that your agenda is not election reform, but to support a particular election outcome.

You also say, "we also know the recount was done illegally; the hand counts were not selected randomly." My response: While ESI and the American Statistical Association offered to help every county in Ohio with their precinct selection for the recount on a pro-bono basis, the random selection was not done "illegally", as you state. The law in Ohio does not define the term "random" and thereby gives the election officials a tremendous amount of flexibility here. However, I agree that the recount procedures need to be re-evaluated and improved.

Overall, I have found that there are a tremendous amount of detail one needs to understand in order to be objective when evaluating the election process. To conduct outside investigations, immediately publish these results and their findings to the Internet, without giving due process to understanding the election system, is short sighted and unproductive.

[garybeck]

22. answers, and here are the URLs

Yes, I have looked into disenfranchisement of Bush voters. I've spent hours in the EIRS system and found that the vast majority of reports and complaints favor Bush. I can't quantify it but others have. It's not just the quantity, either. Some of the reports are really disturbing. Like people being asked who they were voting for and told they had to wait longer if they answered Kerry. It's not just one, there's a definite pattern.

It should also be noted, there are several Republicans who have looked at this and concluded there is valid reason for concern. Jim March and Chuck Herrin are two that come to mind.

Here is a listing of complaints just for Franklin county.

http://www.flcv.com/franklin.html

I have found that no matter what system I look in, I find more complaints were registered by people trying to vote for Kerry. So yes, I have looked into both sides.

here are the URLs of the images that wouldn't display:







(I'm sorry I have no idea why these won't view. I will import them on to my server and try to reference them there). The article with all the images is here:

http://www.freepress.org/departments/display/19/2004/900 )

I don't see how these tables can be called heresay or rumor in any way. They look like raw data imported into a graph to me.

I don't see where you have responded to the stickers on the ballots. This has been sworn to by many witnesses in sworn affadavits and there is photographic evidence. It is not heresay. Even the Election Official had to admit publicly that they were there eventually, and as I understand it, it is illegal for the ballots to be altered in any way, by anyone, including the election officials. Am I on a witch hunt if I want to know what happened with those stickers? Or maybe I just want to know the truth about how they got there.

I'm not sure how to take your remark about the randomness of the hand recounts. I don't think the word "random" has to be defined. The intent is clear enough - no one is supposed to know ahead of time which precincts are going to be hand counted, so they choose them randomly.

If you view the video taped interview of the Triad employee, it is clear, he asked which ones were selected and he was given the information he requested. I don't see how this can be defined as random in any stretch of the imagination. He also admitted on tape that he instructed the election officials to use cheat sheets and ignore the numbers they got on the recount.

I've talked with ovservers who were supposed to be able to view the recount. They tell me that when they showed up to observe, they were just shown the results and they never got to observe anything.

This is not heresay or rumor, it's documented, much of it is on video tape, and it's real. And I don't think it's a witch hunt to demand an investigation into these things.

I have to admit, I have concerns about your report, because from your comments you seem overly concerned about being non-partisan, to the point that you would not want to come to any conclusions that would upset either side. What would you do if your investigation showed that one side was up to funny business? Wouldn't you be concerned about the funding you are receiving from that side? Do you think that bi-partisan investigations can be tainted because they avoid confrontation from either side? Aren't they predisposed to make both sides happy, to the same extent that one-sided investigations would be predisposed to find evidence that supports its claim?

Personally I favor an independent investigations over bi-partisan. Funding of any sort should not be the issue, whether it's from one side, the other, or both.

I believe the problem here is that just about everyone is partisan in some way. Most people involved in this entire question voted for someone. That's why I think it's interesting that the third party candidate Cobb, who has little if anything to gain regardless of what is discovered, is leading the way into calling for an investigation into these facts.

I will review your report when it comes out, as objectively as I can.

Regards,
Gary

[stevenstevensteven]

24. very quick response to gary

Gary: Yes, I have looked into disenfranchisement of Bush voters. I've spent hours in the EIRS system and found that the vast majority of reports and complaints favor Bush. I can't quantify it but others have. It's not just the quantity, either. Some of the reports are really disturbing. Like people being asked who they were voting for and told they had to wait longer if they answered Kerry. It's not just one, there's a definite pattern.


My Response 1: The EIRS system is not statistically significant. They did not setup their model with statistical significance nor objectivity in mind. While this is certainly interesting information, the information is not delivered in context with respect to the greater system that was in use on election day. One cannot extrapolite these findings across the entire USA and draw conclusions.

My Response 2: One cannot say that "there's a definite pattern." unless you can prove it scientifically. There are a myriad of statisical complexities to deal with here. Additional objective research is needed, which I strongly support.

Gary: I have to admit, I have concerns about your report, because from your comments you seem overly concerned about being non-partisan, to the point that you would not want to come to any conclusions that would upset either side. What would you do if your investigation showed that one side was up to funny business? Wouldn't you be concerned about the funding you are receiving from that side? Do you think that bi-partisan investigations can be tainted because they avoid confrontation from either side? Aren't they predisposed to make both sides happy, to the same extent that one-sided investigations would be predisposed to find evidence that supports its claim?

My Response: ESI is practicing science, that means we do not publish findings without the evidence and data to support them. Partisanship is not part of our model.

Overall, I do not want to use this forum for a debate. That was not my intent when I said I would answer questions, as I was referring to our work in Franklin county. I will allow our reports to address the specific issues that you raise.

Steven.

P.S. I still can't read the URL's that you've posted.

[eomer]

29. "They did not setup their model with statistical significance..."

Sorry, Steven, to go on another rant but there are just so many fallacies in your posts that I cannot help myself.

Steven said: "They did not setup their model with statistical significance nor objectivity in mind."

Reply:

This particular fallacy sounds suspiciously similar to a talking point I've heard repeatedly, often in reference to the exit poll. The talking point goes like this: "Such and such a system was not designed to support the statistical conclusions you are drawing from it."

Let me refute this fallacy by way of an example. Let's say that the government collects the physical characteristics of every single member of the armed forces. Let's say these characteristics include both a DNA sample and eye color. The data is collected for the purpose of identifying remains.

Clearly this system is not "designed" for the purpose of any statistical analysis; it is "designed" to be applied one soldier at a time. Does that mean we cannot apply statistical analysis to the data and form conclusions? Absolutely not. There are many conclusions we can draw from this data using statistics. We may very well be able to prove a correlation between blue eyes and a particular gene with statistical significance. Even though the design of the system had no goal whatsoever with regard to statistical analysis.

So it is totally a fallacy to claim that statistical intent must be designed into a system or else you cannot apply statistics to it. This is true of both the EIRS and the exit poll. Both are sets of data that have been collected according to a known protocol. A practically endless array of questions can be analyzed by applying statistics to the data.

Now I will grant you in advance of your making it the point that there may be errors in the data. But you should be careful claiming that errors in the data mean you cannot use statistics. I am certain that your coming reports will involve taking data that have errors and then applying statistics anyway.

Here is a specific question you may want to analyze: "Does the EIRS data preclude with statistical significance the premise that visible vote switching on DREs between Kerry and Bush was unbiased?" That is, the premise that in those cases where a switch occurred, the underlying probability of a switch from Kerry to Bush was .5 and the probability of a switch from Bush to Kerry was .5 and the variance from those rates seen in the data can be explained by mere random variation within some stated confidence interval.

[eomer]

31. EIRS reports show a definite pattern of Kerry votes switching to Bush

Steven said: One cannot say that "there's a definite pattern." unless you can prove it scientifically.

Reply: I looked at the EIRS data for S. Florida where I live. I included Miami-Dade, Broward and Palm Beach Counties, all of which used touchscreen DREs. I counted the reports of a vote switching from Kerry to Bush and the reports of a vote switching from Bush to Kerry. Here are the totals:

Voter pushed Kerry, vote switched to Bush: 42 reports
Voter pushed Bush, vote switched to Kerry: 2 reports

I don't need to use any science to say that there is a definite pattern here. We can debate what the cause of the pattern is but the conclusion that there is a pattern is beyond debate.

I invite you, Steven, to agree that there is a pattern.

[garybeck]

32. the URLs are here; and please read this about randomness

regarding randomness of recounts, I don't think it could be any clearer than Mr. Blackwell's own Lucas County investigation states, and I quote:

"The requirements for the selection of a random recount sample were clearly laid out in Directive 2004-58. Instead of following these guidelines, Ms. Hicks-Hudson ordered a staff member to purposefully
select certain precincts that would easily balance during a hand recount."

For the full report see:

http://serform2.sos.state.oh.us/sos/elections/lucas/LucasCountyInvestigationReport.pdf

This is not about people questioning the definition of the word "random" by any stretch of the imagination.

Here is the URL for the images that weren't showing up:

http://www.freepress.org/departments/display/19/2004/900

Gary

[eomer]

25. Definition of the phrase "must randomly select"

In Directive 2004-58, the SOS issued an Outline of Recount Procedure. This directive includes the following step:

d) The board must randomly select whole precincts whose total equals at least 3% of the total vote. These precincts’ ballots must be manually counted.

I would submit to you that the phrase "must randomly select" is clear and unambiguous on its face.

The clear meaning of this phrase is that any purposeful selection is precluded. Only a selection that is not purposeful and therefore gives each precinct the same chance of selection as any other is random.

To support my claim, here is the definition of random from Merriam-Webster Online:

1 a : lacking a definite plan, purpose, or pattern b : made, done, or chosen at random <read random passages from the book>
2 a : relating to, having, or being elements or events with definite probability of occurrence <random processes> b : being or relating to a set or to an element of a set each of whose elements has equal probability of occurrence <a random sample>; also : characterized by procedures designed to obtain such sets or elements <random sampling>

Please let me know how this word can be construed to mean selecting according to precinct size, party predominance, urban vs. suburban vs. rural, and so on. All of these selections are purposeful and by definition not random. If these selections are random then all selections are random and the word is left with no meaning whatsoever.

You have chosen to claim objectivity for your part and impugn that of others who are not present to defend themselves. Ironically this is a sweeping generalization that is lacking in specificity or supporting evidence. May I suggest that you back up a step and talk to me about each conclusion separately and on its merits rather than painting them all with the same broad brush. (Let's call this approach "objectivity".)

As you can probably tell I am so far not convinced of your objectivity. One of your first assertions requires a twisting of the meaning of words that could be called Orwellian. I will try to keep an open mind as I evaluate your coming reports. But I will also promise to let you know if any of your conclusions are not supported by sufficient evidence or are lacking in objectivity.

[emlev]

26. More on "random"

While I personally agree that the meaning of "random" is clearer than stevenstevensteven suggests, apparently it is not clear in Ohio election law. I think that what's relevant here is not Merriam-Webster's definition but the legal definition in the state of Ohio.

The Green Party recount site lists among other changes necessary to ensure fair elections in Ohio in the future:

"Clear up the vagueness in Ohio Election Law that allows each county to interpret words such as "random" recount in their own way."

http://www.votecobb.org/recount/ohio_reports/

Does anyone know how to look up the legal definition applicable in the state of Ohio?

[eomer]

28. "Ohio Election Law is very clear on this point:"

Another quote from the same page of the Green Party website:

Ohio Election Law is very clear on this point:

"The board must randomly select whole precincts whose total equals at least 3% of the total vote, and must conduct a manual count."

The Green Party contradicts themselves within this web page, saying in one place that the wording is very clear on this point and in another place that it is vague. I think their intent was to say that the phrase is very clear but that they recommend perfecting the wording to make it inescapably clear and eliminate any perceived wiggle room.

(By the way, no slur is intended toward the Green Party when I say they contradict themselves. They have done a tremendous service in their work on the election and I particularly appreciate them making all the recount info public.)

As far as the "legal definition applicable in the state of Ohio", it is not typical to define a term at that broad level. If a term is defined in some law or regulation it is usually (should I say always?) defined for the purposes of some specified scope. A definition of a term "randomly" within some environmental law would typically not be intended to have any effect on the use of the word "randomly" in the context of some election law, for example. I've looked over the election statutes and do not find any definition.

If someone finds a definition that is legally applicable in this context then I'll rethink the question. In the meantime I believe you have to interpret the directive by looking to the English language and what the words mean as they are commonly used.

And doing that, I just don't see any ambiguity in the phrase "must randomly select".

Please let me know a definition of the word "randomly" that permits all the variations that occurred in the recount but still has some meaning. To have some meaning it will have to somehow constrain or affect the variations that are permitted. But remember that your definition has to have enough flexibility to permit all the variations that occurred. I think you will find this an impossible task. I think you will find that the variations that occurred were not "random" in any imaginable definition of the word.

edit: spelling

[Bill Bored]

34. I think the Greens were just being diplomatic in their second statement.

Unless there's a court case we don't know about that showed there really was some ambiguity in the Ohio law about what "randomly" means.

[Bill Bored]

27. More about "random"

Steven,

The evidence that the precincts were not selected at random, as I understand it, was that larger precincts were systematically excluded from the selection process to avoid exceeding 3% of the vote count by a large margin. Since the statute says "at least 3%", this should not have been an issue at all and there was no need to skew the precinct size distribution non-uniformly, i.e., non-randomly. In a random selection process, by any definition of the word that I am aware of, all precincts would have had an equal chance of being selected, which was not the case.

What we also know from the record is that Mr. Blackwell and others in the state, thought the recount request was frivolous, not having come from the Kerry campaign, and that the cost of ~$120,000 was considered to be too low to cover the counties' expenses, due to inflation since the original statue became law over 20 years ago. So I believe some effort was made to minimize the cost of the hand count by coming as close to the 3% target as possible by not selecting precincts randomly.

If one knew in advance that this would occur, he might be inclined to perpetrate election fraud in the larger precincts, knowing they would never be hand counted. On the other hand, maybe they were just trying to save the taxpayers a few bucks.

Either way, unless there is evidence to the contrary, it does not seem as if the law was followed as the precincts were not selected randomly by any reasonable definition of that term. This is no way to engender confidence in the electoral process.

Good luck with your investigations.

[garybeck]

33. Lucas county investigation reveals NON-random selection

"The requirements for the selection of a random recount sample were clearly laid out in Directive 2004-58. Instead of following these guidelines, Ms. Hicks-Hudson ordered a staff member to purposefully
select certain precincts that would easily balance during a hand recount."


http://serform2.sos.state.oh.us/sos/elections/lucas/LucasCountyInvestigationReport.pdf

[eomer]

19. My questions

Steven, thanks for offering to answer our questions. Here are mine.

You said, "Third, we've completed a comprehensive audit of the election result in dozens of Ohio counties."

From your website: "Election Science Institute is a nonprofit nonpartisan organization of citizen volunteers, statisticians, lawyers technologists, journalists and election officials who monitor public elections in the U.S.A., analyze patterns, and make their findings public prior to the certification of the election."

Have you published the results of auditing these dozens of Ohio counties? If not, when do you plan to do so?

I went to your website hoping to find some published results. I found that there is not much there and most of it dates back to December or earlier. One of the most recent reports concludes that "Votewatch Finds Ohio Vote not Currently Auditable, Due to Issues of Timeliness, Completeness, and Quality of Documents from County Election Officials."

There seems to be some conflict between your statement that you have audited dozens of counties and your earlier one that the election was not auditable. Was there a turn around in availability and quality of data or do you stand by your earlier conclusion?

Do you plan on pushing more information out to the public?

Thanks in advance for your answers.

[stevenstevensteven]

21. response to eomer

1. "Do you plan on pushing more information out to the public?"

Yes, absolutely!

While our goal was to have our findings out prior to election certification, we've learned that the scope of work we undertook was far more extensive and time consuming than we ever could have imagined... it's the first time it has ever been done.

We are working now on consolidating our findings into several "bite sized" reports and publishing them as they are ready.

2. "There seems to be some conflict between your statement that you have audited dozens of counties and your earlier one that the election was not auditable. Was there a turn around in availability and quality of data or do you stand by your earlier conclusion?"

One has to conduct an audit in order to determine its audit-ability. So no, there wasn't a specific turn around, it is just that the further we got into the process the better we were able to understand the audit-ability of the public records. I think we've done a good job quantifying what we have, and that will be reflected by one of our first reports, which is in peer review now.

[eomer]

23. Thanks.

I look forward to your reports. I'm sure you will find everyone here is eager to read and evaluate your findings.

[Chi]

30. Not a single Ohio county, out of 49, gave them what they asked for...

At least up to date of certification.
The question that comes to mind at this point is...
How many counties have now complied 100%...Five months later?

http://www.votewatch.us/Members/StevenHertzberg/report.2004-12-15.2641756478/report_contents_file/

And if the answer is still (to date) pathetic...why make excuses for them...

"Election departments across Ohio faced many public records requests in the wake of Election Day and were
understandably occupied with certifying election results."

[mod mom]

35. kick. nt

[mod mom]

36. Steven, The data Richard Hayes Phillips used to demonstrate problems in

the allocation of machines came from the Franklin County (OH) BOE spreadsheet that broke down by precinct: 2000 Voter registration, 2000 Voting machines, 2000 Official Turnout, 2000 Voters per machine, 2004 Voter Registration, 2004 Active Voters, 2004 voting Machines, 2004 Unofficial turnout, 2004 Voters per Machine, Voter Reg % Change, Machine Change, Turnout % Change, Voters Per Machine % Change. This is not speculated data but internal numbers from the BOE. Why do you question his results?