![]() | This page is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Just ran across this, we should add some of it to the page somehow, sorry it's so big: http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=45003&mesg_id=45003
(full text removed. Copyright status indicates this should not be copied here verbatim)
I have this image:
from the Associated Press, published 2 November in an article on 'The Louisville Channel' web site. Are there PD objections to using the image, since it was published in mainstream press? File:2004longvoterlines.jpg
I'm puzzled by the meaning of the term "effectively disenfranchised." What does this signify? How does one go about "effectively disenfranchising" voters? Long lines are a problem but as I understand it most states require polling places to stay open until all voters who were in line at the closing time, have voted. The word "effectively" implies equivalence, something that we don't have here. Rhobite 20:48, Nov 12, 2004 (UTC)
It is also hypothesized that long lines at urban polling places would negatively effect turnout for Kerry voters. Kevin implied that this is an irrefutable logical proof and deleted the word "hypothesized," but there are a few hidden premises in that argument. This is someone's opinion, and it shouldn't be presented as fact. Rhobite 20:52, Nov 12, 2004 (UTC)
- *it is more likely that more people left the line throughout the day than were in the line when the polls close. lower voter turnout numbers correlate with counties with a higher rate of machine shortages and long lines. + - *a implies b implies c (not "is hypothesized") see logic. Kevin Baas | talk 22:56, 2004 Nov 12 (UTC)
Please help me understand, why were a number of new content and source just deleted by Rhobite in the main page? -- RyanFreisling
I made a correction. The 'dixiecrat' phenomenon is not part of the Caltech study. It is a valid issue, and belongs in the document, but not in that section. -- RyanFreisling @ 11:07, 13 Nov 2004 (UTC)
(deleted for brevity: outdated)
Just ran across this, we should add some of it to the page somehow, sorry it's so big: http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=45003&mesg_id=45003
(full text removed. Copyright status indicates this should not be copied here verbatim)
Machine shortages in precincts with democratic voters, charts: [1] Kevin Baas | talk 20:52, 2004 Nov 20 (UTC)
'Stinking Evidence' of Possible Election Fraud Found in Florida by Thom Hartmann noosphere 22:33, 2004 Nov 20 (UTC)
KB, I think we should have a discussion of what main categories this page should be broken up into before we vote on each one... I'm interested in some of your proposals but I think we should draft a complete organizational breakup instead. -- kizzle 12:35, Nov 21, 2004 (UTC)
This is actually a good idea, the detailed sections on voting machines can possibly be moved to the electronic voting article, including most if not all of the expert testimony. Though many of the issues do relate to non electronic voting machines. Criticisms of Diebold and brief mentions of the potential for fraud from lack of paper trail or auditability etc are relevant and should be left in this article. Zen Master 23:32, 11 Nov 2004 (UTC)
What do others think?
James proposed that we split the article into 2, one for data irregularities, one for all other election controversies, what do people think and exactly where should the dividing line be? I guess almost all of the voting machine info should stay in the irregularities article as it's related, but maybe not. There may be some overlap between the articles. Zen Master 05:49, 15 Nov 2004 (UTC)
I added the current New Mexico incident (a single rural county has missed the provisional ballot count deadline). I could not find evidence of it occurring before. Do others agree that validates the situation as a noteworthy Election irregularity of public record? Thanks. -- RyanFreisling @ 08:51, 15 Nov 2004 (UTC)
The article is getting too big, too redundant and the organization unwieldy, I propose we create a new wikipedia article to the effect of "Criticisms of voting machines" or "Criticisms of Diebold voting machines" or "Criticisms of electonic voting machines" or all of the above. With the titles to be decided by whoever actually creates those pages. Note: those articles would be general/historical criticisms and not cover the 2004 election, that stuff would still belong in this article. Zen Master 23:13, 13 Nov 2004 (UTC)
(dormant? related to split/edit discussion?)
we have successfully added 175KB in 3 days to a talk page, which is 7 times the guideline max for an article itself... this discussion is getting us nowhere, as any valid discussion is within minutes buried by new additions. we now have what looks like at least 20 active editors working on this page. Obviously the proposal for organization which I made wasn't taken too seriously, but some level of organization needs to be applied to this page, there are simply too many cooks in the kitchen for us to keep posting randomly in the manner we have been doing, IMHO. -- kizzle 03:29, Nov 13, 2004 (UTC)
What I actually meant was this talk page, there needs to be some organization to it, as the information being added/debated is tremendous. We don't need to use the templates, but we need to separate this talk page into "proposed passages" "current passages" and "other", or something like that, and make sure we quote verbatim the place that needs discussion, cause keeping up with the talk on this page is highly time-intensive. --
kizzle 09:16, Nov 15, 2004 (UTC)
Kevin:
Let's say we are weighting the exit poll data based on the actual vote counts (by the way, the allegation that this happens has not been sourced). But let's assume. When you weight a variable, you assign it an importance. This importance is arbitrary - it is up to the user of the data to assign weights to variables. Let's assume the exit poll data in this article are correct:
Now, let's take the actual Ohio results, according to Wikipedia, dropping third party candidates:
We have our data, and we can "weight" the exit poll results. Please keep in mind these are all "mathematical facts," as you put it.
Let's place a 10% weight on the actual vote results. We can calculate a "corrected" exit poll percentage like so: Corrected = (W * ActualVote) + ((1 - W) * ExitPoll). For a 10% weight on Kerry's results: (.1 * .4875) + (.9 * .5206) = .5173, or 51.73%. Obviously this doesn't support your assertion that weighted results "matched the actual vote count" or "supported the conclusion." Similarly, a weighting of .25 or even .5 is still not enough to push Kerry's exit poll result below 50%. For your statement to be true, they would need to place more importance on the polling result than the actual exit poll, in order to come up with their published exit poll number. The only way for the results to match "perfectly," as you say, is to throw out the exit poll results and simply use the actual vote.
Also, in the article, please back up your statement about how they weighted the results. Rhobite 23:37, Nov 11, 2004 (UTC)
Your mathematics are correct, and so are your conclusions.
If I do recall correctly, I think I put a citation or two in the article somewhere. Here's a great source for information on exit polling: [7]
Here's a source corroborating what I said: [8] "...the weighting of exit polls to match actual results is not new, but a standard procedure used since the early days of exit polls. Second, the weighting to actual returns does not occur all at once but continuously, precinct by precinct, over the course of election night. The exit pollsters weight their sample to match incoming actual results for each sampled precinct as actual returns become available..."'' Kevin Baas | talk 19:34, 2004 Nov 12 (UTC)
(dormant?) The map of the Ohio counties showing colors to indicate "electronic voting machine incidents reported to the EIRS" has Cuyahoga County, OH colored Orange. Cuyahoga County, OH used a punch card system, not electronic voting machines. Copies of the ballots for all districts are available in .pdf format from the Cuyahoga County Board of Elections website. http://www.cuyahoga.oh.us/BOE/ballots/ballots.htm
Ahhh... Up for deletion... Go figure Cyberia23 08:32, 13 Nov 2004 (UTC)
Once again, Netaholic has deleted it outright, opting for a subtractive approach instead of a dialogue here. I reverted, with corroborating sources. The issue may have an explanation, but the controversial event (change in the data on CNN) took place, and the Internet's resolving/debunking it is an important part of the story. -- RyanFreisling @ 23:05, 14 Nov 2004 (UTC)
In the image in the voting machine problems section with 2 maps of Florida & 2 of Ohio, the yellow counties in the Ohio map only have 1 incident reported a piece, and they are either really really petty, or aren't actually machine problems. I think it's misleading. I'm tempted to make a new image w/those counties grey instead of yellow. What do people think about this?
--I think the whole image needs to be removed all together. There is absolutely no reason for pointing out that such counties are traditionally Democrat other than for suggesting conspiracy theories. First, Democrat counties often times are urban areas, which tend to get electronic voting machines, while sparsely populated rural Republican counties do not. So pointing out at least one electronic voting error happened in a county while also pointing out that county leans Democratic fuels this insinuation. What a proper map should do, is point out which counties had electronic voting, point out which counties reported problems, and mention the severity of these problems.
But that's not what we have. We have a full state county map that does not show which counties have electronic voting, we have Democratic explanations thrown in for insinuation, an explanation that does not explain the severity of the problems, and we don't have any references to back any of this up to begin with!
The image and caption fails in every way of being fair, meaningful, and objective. For that reason, it should be removed or replaced by an updated version that fixes every problem listed.
-- Non-US folk may not know the normal Dem/Rep split of the US, so the map might be useful. CS Miller 17:00, Nov 16, 2004 (UTC)
(dormant?) Minus votes from the 12 or so states with data irregularities how many votes did bush and kerry each receieve? Is there evidence of smaller degree of fraud/irregularity in the other 38 states? Maybe I will perform that calculation myself. Zen Master 22:34, 13 Nov 2004 (UTC)
using the 12:22 data for all states (ommiting the 4 no-data states), popular vote is bush:50.43%, kerry: 48.56% Kevin Baas | talk 23:06, 2004 Nov 13 (UTC)
Remember I have my excel file available. It has per state data from the vote count and three different exit poll sources (taken at different times). 9and if anyone wants the maps i've in vector format (i.e. you can change the colors of the states easily w/the right software), i can give them.) Kevin Baas | talk 19:40, 2004 Nov 14 (UTC)
Rhodite, made POV corrections. -- RyanFreisling @ 22:04, 14 Nov 2004 (UTC)
(dormant / move news items to new article?)
Anon User '66.30.181.242' has just performed edits that are POV. Modifiers like 'few', 'some', etc. are inappropriate.
The other edit, changing the quote about diebold from a direct quote to hearsay, is factually untrue. Link provided. Thoughts?
Note: I took "It should be noted however that these same counties historically have voted for the Republican presidential nominee" out of the Issues section, as it has no factual corroboration in the piece. Anyone have any data so we can put that back in? If it's a fact, it's important. -- RyanFreisling @ 03:27, 16 Nov 2004 (UTC)
if everyone is ok with the new format above, i'm going to begin deleting duplicate sections that got moved upwards (will not delete info that isn't up there already)... this should help out tremendously instead of having to scroll the entire page just to understand what is going on, and should help prevent good discussions from being buried in the talk page. -- kizzle 20:38, Nov 16, 2004 (UTC)
It is suggested that:
Can someone verify these and if so add them into the article, in appropriate places. FT2 21:41, Nov 17, 2004 (UTC)
In addition the article is getting too long, and much of it is just "backing reference information". Whilst this information is necessary for now, since there are multiple perceptions and uncertainty regarding much of the detailed information needed to review this controversy, and what information is sourced and verified versus rumour, I am concerned that the sheer volume is detracting from an encyclopaedic approach. I'd like to suggest with the archiving of much of the talk page, and the end of the VfD issue, that collaborators give serious consideration to siphoning a lot of the backing information into a separate article, "2004 U.S. presidential election controversy (summary of sources)".
Stuff that could go in there = most of everything that is "source material re-repesented" as opposed to "encyclopediac summary of the issues and information known at this time".
The main article would then reference this sources article for much of its substantiation and detail.
Comments? FT2 17:54, Nov 17, 2004 (UTC)
I could do a good cleanuup job on it that way, and make the main article really work without losing citations and quotes. But I'd want at least something of a consensus to do so, not just 1 or 2 voices, and I'd want to check with others as well. FT2 01:12, Nov 18, 2004 (UTC)
A lot is as you say to do with these differences - but a lot isn't. Some is evidence of security issues from specialists and hearings, some is internal memos of manufacturers knowledge of fraudulent claims, some is industry report summaries, some is political and corporate information relevnat to the matter, some is raw data sources used in maps, or tables, or non-core links, or news references. I think a single split between "summary" and "backing material" could benefit the main article. I'd be minded to title the other article something like "2004 U.S. presidential election controversy/Sources" or "/Supporting material" FT2 21:50, Nov 18, 2004 (UTC)
I think we should devise an overview article like JML suggested which then links to the current article broken into chunks. I'm sure we can come up with 3-5 main topics which merit daughter articles in themselves, whether it mirrors the top-level categories of the page itself or some other format. But definetely not shorten or revise the article, in such heavily scrutnized and "dismissed" claims that we have here, the more detail and direct quoting from sourcing the stronger it will hold up to such counter-claims. -- kizzle 08:01, Nov 19, 2004 (UTC)
Someone seems to have quoted large blocks of independent analysis into the article, that is ok with me but we need to be clear to the reader NPOV wise when there are headers like "Conclusion" in the article, those are not wikipedia's conclusions but are conclusions of the third party research. A header titled "Conclusion" should not be thought of as a wikipedia header in this case, it should be treated the same as the rest of quoted third party analysis text. Also, I propose that we have at least one citation link per quoted analysis paragraph so it's further clear that this is not a wikipedia conclusion, citations can be repeated. What do people think? Zen Master 22:01, 17 Nov 2004 (UTC)
I removed the following from the exit poll discrepancy section. CalTech/MIT have identified these are no different from normal, challenging this are analyses which contend CalTech/MIT made methodological errors, including comparing not native exit polls, but exit polls adjusted to match the official figures, and it is therefor no surprise that they found no evidence of significant difference. See below for CalTech/MIT's analysis.
I'm not sure of the facts being asserted, so I'm not sure. But at the very least, that sentence needs a restructuring. It's POV, makes characterization and conclusion and is grammatically damaged. Anyone want to take a crack at it, so we can get this on the article page? -- RyanFreisling @ 04:12, 18 Nov 2004 (UTC)
Its not difficult. This is what the two views say, in simple terms:
There is also substantial comment that the CalTech report was flawed and (perhaps) politically inclined on various websites:
Whilst some of the above are obviously personal POV, the key points appear to be well sourced and capable of verification, namely:
Ive added the key ones to the article to firm up the "methodological errors".
FT2 18:44, Nov 18, 2004 (UTC)
nice. -- kizzle 21:53, Nov 18, 2004 (UTC)
It looks like the 'partisan ties' section is now glutted with duplicative content, pasted in front of previous content that covered the same points. There's good stuff there, but lots of redundancy. I'm going to take a few passes at it. -- RyanFreisling @ 00:52, 19 Nov 2004 (UTC)
Had a skim and a history check. Looks pretty good. FT2 03:17, Nov 19, 2004 (UTC)
Several days ago I suggested creation of an overall summary article, and I proposed specific text for a lead section that would give an idea of what I had in mind. I've now created a first draft of that article at 2004 U.S. election voting controversies.
I think it’s very important that we try to keep the new summary article free of the problems plaguing this one (the older one). The mass of detail here makes this article hard to read and even harder to edit. Specific problems, IMO, include:
The new article is not a blog, or a newsfeed, or a Wikisource document, or a directory of links. For example, it summarizes the whole EVM issue in a couple paragraphs, with a link back to this article. The same treatment is appropriate for other issues. I suggest that anyone who wants to pile up a list of (for example) every voter registration dispute anywhere in the country in 2004 should create a separate article for that purpose and link to it. I warn everyone that, although I’ve largely given up trying to edit this (original) article, I’m going to be aggressive in editing the new one to prevent the kind of bloat we’ve seen here. JamesMLane 19:55, 21 Nov 2004 (UTC)
I understand, but having considered both I disagree. What someone said way up above seems the key to me: in the course of time, it will become clear what is gereally known and what is generally not known of the factual material. Right now a lot of the information is at some kind of "grass roots" level, by that I mean that its collected in the web pages of many minor newspapers and organisations, off-beam court cases, emails not reported on any american daily newspaper's front page. So for now, the background information although its big, is needed. I dont think many articles are needed.
At a pinch if you wanted to split voting machine issues from exit poll controversies... but they are different in character, at heart if you read both sections, both are about evidence... the only difference is one is exit poll statistical evidence (indirect evidence), one is manufacturers and other documentary evidence (direct evidence). I'm not convinced it needs splitting multiple ways just yet - but if it was, what would the subject matter of each sub-article be? Each section has potential to grow. I dont think we've seen the last of this yet... FT2 18:57, Nov 22, 2004 (UTC)
I notice this article still discusses the E-touch versus opscan machine voting difference in Florida. The view that it's explained by the Dixiecrat effect is mentioned with references to NYT and Wired News articles but isn't given a lot of weight. I'd just like to add that, after seeing the initial reports, I spent a boatload of time looking at the Florida by-county numbers and I concluded (sadly) that the Dixiecrat effect does indeed account for the differences. There are three points I'd like to make about it.
First, county size turns out to be a rather bad proxy for Dixiecrat population in a county. The opscan counties are not just smaller; they're also generally in the northern half of the state, while the E-touch counties are generally clustered in the southern part. Florida is highly inhomogeneous, and is stratified by latitude.
Second, I looked at the 2000 election results in Florida, and, broken down into "2004 E-touch" and "2004 op-scan" counties, they show exactly the same pattern as the 2004 results.
Finally, voting in Louisiana in 2000 showed the same strange patterns, which lends weight to the claim that it's conservative southern Democrats who are breaking the curve here. Interestingly, though Louisiana went solidly for Bush in both elections, there was not one parish in Louisiana in which more than 50% of the voters were registered as Republicans, and there was just one parish in which Republican voters constituted more than 40%. In contrast, there were a number of counties with ~70% Democrats.
If anyone's interested, details of this are on http://www.physicsinsights.org/elec04.html
-- Salaw 13:48, 22 Nov 2004 (UTC)
"Dick Morris, a career pollster (Republican), "
Dick Morris is most famous for his work under Bill Clinton, a Democrat.
While I understand that his beliefs now are more in-line with those of Republicans, is it fair to label him a "Republican?" It seems to me as if that label there is meant to disqualify his opinion, which violates NPOV. RNJBOND
<><>Here's a quick note, on subtle efforts to persuade: The parenthetical word "(Republican)" was evidently used, not to discredit what Dick Morris said, but instead to invest the statement attributed to Morris with some "against-interest" credibility. The technique here seems to have been to point out that, even though Morris was a supporter of Bush and an opponent of the Democrats (and would thus naturally have a partisan interest in arguing that the vote counts were accurate), Morris "admitted" that the discrepancy between the exit polling and the vote count was "incredible," because (in Morris' own view) exit pollling is almost always very accurate. Of course, it surely misrepresents what Morris actually was saying, to suggest that he meant to cast doubt on the vote counts in 2004. Instead, he seems to have been blasting the unprecedented incompetence of the exit-polling (apparently this was actually what he intended to say). As a general statement, I believe it would not be inaccurate to characterize Dick Morris as a Republican in recent years; Morris had a famous falling-out with the Clintons and has written many columns (and even some books) attacking the Clintons as dishonest, etc.
So now we also have the People For the American Way Foundation taking legal action in Ohio. What are all of the organizations? Should we make a list and put it in the article? Kevin Baas | talk 01:23, 2004 Nov 28 (UTC)
Kevin Baas created three new articles to break up this mass of material -- 2004 U.S. presidential election controversy, voting machines; 2004 U.S. presidential election controversy, exit polls; and 2004 U.S. presidential election controversy, vote suppression. Since then, however, no one has edited those articles. Instead, people keep piling everything into this article. I opined several days ago that it was far too long, and it's only grown since then. Is there some purpose being served by creating a huge, unreadable article? JamesMLane 05:54, 29 Nov 2004 (UTC)
Why is this 2004_U.S._presidential_election_controversies second article (which does not point here, or to other subpages) necessary? Thanks. -- RyanFreisling @ 18:17, 3 Dec 2004 (UTC)
May I say that I'm pleased with Z-M's new split of the lede on this entry. It was getting a bit unwieldy. Baylink 22:07, 30 Nov 2004 (UTC)
i added the Boston Globe's graphic and source. Good stuff!
Because of a decision by The Ohio Supreme cournt, for a recount in Ohio to happen before the deadline, Kerry, not Cobb or Badnarik, must file for it to be expediated. Kevin Baas | talk 20:59, 2004 Dec 1 (UTC)
Well, the counties have all certified their results, but, as I'm sure we all predicted, Ken is taking his sweet time certifying the state results. Kevin Baas | talk 00:38, 2004 Dec 2 (UTC)
This suit has to succeed: [13] [14]
I have a theory why 1/3 (8099/24472) of the provisionals were thrown out in Cuyahoga County:
Voter turnout on avg. in Ohio ws about 70%. 70% of 10,000 is 7,000. That leaves 1099, less than 10% of 24472.
This is why I believe that the election being overturned is contingent on the above law suit succeeding. Kevin Baas | talk 19:53, 2004 Dec 2 (UTC)
An excellent summary of the irregularities under active investigation can be seen in the House Democrats' letterto Kenneth Blackwell, sent Dec 2. [16]. Might do well for us to put some stakes in the ground around those specific questions and possible answers to come from OH Sec. of State Blackwell. -- RyanFreisling @ 07:14, 3 Dec 2004 (UTC)
How did you find those links? Is this significant enough to put on the current events page? Kevin Baas | talk 22:59, 2004 Dec 3 (UTC)
I have one question about the recount...if electronic machines have no paper trail, how exactly is a recount performed? --
kizzle 23:13, Dec 3, 2004 (UTC)
If stuff under new has been added please add "- Added" to the title", same with any current passages that have been removed, "- Removed" in the title.
Last thing we need to do, let us aim to archive all entries that already have been added both in New Passages and Possible Passages for Inclusion within the next few days, if you see something that is already on the page, please help out by putting " - Added" to the end of the title of the section. -- kizzle 21:50, Nov 18, 2004 (UTC)
Anon IPs that have mildly trolled this talk page: 68.107.102.129 05:36, 14 Dec 2004 (UTC) [unsigned]
![]() | This page is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Just ran across this, we should add some of it to the page somehow, sorry it's so big: http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=45003&mesg_id=45003
(full text removed. Copyright status indicates this should not be copied here verbatim)
I have this image:
from the Associated Press, published 2 November in an article on 'The Louisville Channel' web site. Are there PD objections to using the image, since it was published in mainstream press? File:2004longvoterlines.jpg
I'm puzzled by the meaning of the term "effectively disenfranchised." What does this signify? How does one go about "effectively disenfranchising" voters? Long lines are a problem but as I understand it most states require polling places to stay open until all voters who were in line at the closing time, have voted. The word "effectively" implies equivalence, something that we don't have here. Rhobite 20:48, Nov 12, 2004 (UTC)
It is also hypothesized that long lines at urban polling places would negatively effect turnout for Kerry voters. Kevin implied that this is an irrefutable logical proof and deleted the word "hypothesized," but there are a few hidden premises in that argument. This is someone's opinion, and it shouldn't be presented as fact. Rhobite 20:52, Nov 12, 2004 (UTC)
- *it is more likely that more people left the line throughout the day than were in the line when the polls close. lower voter turnout numbers correlate with counties with a higher rate of machine shortages and long lines. + - *a implies b implies c (not "is hypothesized") see logic. Kevin Baas | talk 22:56, 2004 Nov 12 (UTC)
Please help me understand, why were a number of new content and source just deleted by Rhobite in the main page? -- RyanFreisling
I made a correction. The 'dixiecrat' phenomenon is not part of the Caltech study. It is a valid issue, and belongs in the document, but not in that section. -- RyanFreisling @ 11:07, 13 Nov 2004 (UTC)
(deleted for brevity: outdated)
Just ran across this, we should add some of it to the page somehow, sorry it's so big: http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=45003&mesg_id=45003
(full text removed. Copyright status indicates this should not be copied here verbatim)
Machine shortages in precincts with democratic voters, charts: [1] Kevin Baas | talk 20:52, 2004 Nov 20 (UTC)
'Stinking Evidence' of Possible Election Fraud Found in Florida by Thom Hartmann noosphere 22:33, 2004 Nov 20 (UTC)
KB, I think we should have a discussion of what main categories this page should be broken up into before we vote on each one... I'm interested in some of your proposals but I think we should draft a complete organizational breakup instead. -- kizzle 12:35, Nov 21, 2004 (UTC)
This is actually a good idea, the detailed sections on voting machines can possibly be moved to the electronic voting article, including most if not all of the expert testimony. Though many of the issues do relate to non electronic voting machines. Criticisms of Diebold and brief mentions of the potential for fraud from lack of paper trail or auditability etc are relevant and should be left in this article. Zen Master 23:32, 11 Nov 2004 (UTC)
What do others think?
James proposed that we split the article into 2, one for data irregularities, one for all other election controversies, what do people think and exactly where should the dividing line be? I guess almost all of the voting machine info should stay in the irregularities article as it's related, but maybe not. There may be some overlap between the articles. Zen Master 05:49, 15 Nov 2004 (UTC)
I added the current New Mexico incident (a single rural county has missed the provisional ballot count deadline). I could not find evidence of it occurring before. Do others agree that validates the situation as a noteworthy Election irregularity of public record? Thanks. -- RyanFreisling @ 08:51, 15 Nov 2004 (UTC)
The article is getting too big, too redundant and the organization unwieldy, I propose we create a new wikipedia article to the effect of "Criticisms of voting machines" or "Criticisms of Diebold voting machines" or "Criticisms of electonic voting machines" or all of the above. With the titles to be decided by whoever actually creates those pages. Note: those articles would be general/historical criticisms and not cover the 2004 election, that stuff would still belong in this article. Zen Master 23:13, 13 Nov 2004 (UTC)
(dormant? related to split/edit discussion?)
we have successfully added 175KB in 3 days to a talk page, which is 7 times the guideline max for an article itself... this discussion is getting us nowhere, as any valid discussion is within minutes buried by new additions. we now have what looks like at least 20 active editors working on this page. Obviously the proposal for organization which I made wasn't taken too seriously, but some level of organization needs to be applied to this page, there are simply too many cooks in the kitchen for us to keep posting randomly in the manner we have been doing, IMHO. -- kizzle 03:29, Nov 13, 2004 (UTC)
What I actually meant was this talk page, there needs to be some organization to it, as the information being added/debated is tremendous. We don't need to use the templates, but we need to separate this talk page into "proposed passages" "current passages" and "other", or something like that, and make sure we quote verbatim the place that needs discussion, cause keeping up with the talk on this page is highly time-intensive. --
kizzle 09:16, Nov 15, 2004 (UTC)
Kevin:
Let's say we are weighting the exit poll data based on the actual vote counts (by the way, the allegation that this happens has not been sourced). But let's assume. When you weight a variable, you assign it an importance. This importance is arbitrary - it is up to the user of the data to assign weights to variables. Let's assume the exit poll data in this article are correct:
Now, let's take the actual Ohio results, according to Wikipedia, dropping third party candidates:
We have our data, and we can "weight" the exit poll results. Please keep in mind these are all "mathematical facts," as you put it.
Let's place a 10% weight on the actual vote results. We can calculate a "corrected" exit poll percentage like so: Corrected = (W * ActualVote) + ((1 - W) * ExitPoll). For a 10% weight on Kerry's results: (.1 * .4875) + (.9 * .5206) = .5173, or 51.73%. Obviously this doesn't support your assertion that weighted results "matched the actual vote count" or "supported the conclusion." Similarly, a weighting of .25 or even .5 is still not enough to push Kerry's exit poll result below 50%. For your statement to be true, they would need to place more importance on the polling result than the actual exit poll, in order to come up with their published exit poll number. The only way for the results to match "perfectly," as you say, is to throw out the exit poll results and simply use the actual vote.
Also, in the article, please back up your statement about how they weighted the results. Rhobite 23:37, Nov 11, 2004 (UTC)
Your mathematics are correct, and so are your conclusions.
If I do recall correctly, I think I put a citation or two in the article somewhere. Here's a great source for information on exit polling: [7]
Here's a source corroborating what I said: [8] "...the weighting of exit polls to match actual results is not new, but a standard procedure used since the early days of exit polls. Second, the weighting to actual returns does not occur all at once but continuously, precinct by precinct, over the course of election night. The exit pollsters weight their sample to match incoming actual results for each sampled precinct as actual returns become available..."'' Kevin Baas | talk 19:34, 2004 Nov 12 (UTC)
(dormant?) The map of the Ohio counties showing colors to indicate "electronic voting machine incidents reported to the EIRS" has Cuyahoga County, OH colored Orange. Cuyahoga County, OH used a punch card system, not electronic voting machines. Copies of the ballots for all districts are available in .pdf format from the Cuyahoga County Board of Elections website. http://www.cuyahoga.oh.us/BOE/ballots/ballots.htm
Ahhh... Up for deletion... Go figure Cyberia23 08:32, 13 Nov 2004 (UTC)
Once again, Netaholic has deleted it outright, opting for a subtractive approach instead of a dialogue here. I reverted, with corroborating sources. The issue may have an explanation, but the controversial event (change in the data on CNN) took place, and the Internet's resolving/debunking it is an important part of the story. -- RyanFreisling @ 23:05, 14 Nov 2004 (UTC)
In the image in the voting machine problems section with 2 maps of Florida & 2 of Ohio, the yellow counties in the Ohio map only have 1 incident reported a piece, and they are either really really petty, or aren't actually machine problems. I think it's misleading. I'm tempted to make a new image w/those counties grey instead of yellow. What do people think about this?
--I think the whole image needs to be removed all together. There is absolutely no reason for pointing out that such counties are traditionally Democrat other than for suggesting conspiracy theories. First, Democrat counties often times are urban areas, which tend to get electronic voting machines, while sparsely populated rural Republican counties do not. So pointing out at least one electronic voting error happened in a county while also pointing out that county leans Democratic fuels this insinuation. What a proper map should do, is point out which counties had electronic voting, point out which counties reported problems, and mention the severity of these problems.
But that's not what we have. We have a full state county map that does not show which counties have electronic voting, we have Democratic explanations thrown in for insinuation, an explanation that does not explain the severity of the problems, and we don't have any references to back any of this up to begin with!
The image and caption fails in every way of being fair, meaningful, and objective. For that reason, it should be removed or replaced by an updated version that fixes every problem listed.
-- Non-US folk may not know the normal Dem/Rep split of the US, so the map might be useful. CS Miller 17:00, Nov 16, 2004 (UTC)
(dormant?) Minus votes from the 12 or so states with data irregularities how many votes did bush and kerry each receieve? Is there evidence of smaller degree of fraud/irregularity in the other 38 states? Maybe I will perform that calculation myself. Zen Master 22:34, 13 Nov 2004 (UTC)
using the 12:22 data for all states (ommiting the 4 no-data states), popular vote is bush:50.43%, kerry: 48.56% Kevin Baas | talk 23:06, 2004 Nov 13 (UTC)
Remember I have my excel file available. It has per state data from the vote count and three different exit poll sources (taken at different times). 9and if anyone wants the maps i've in vector format (i.e. you can change the colors of the states easily w/the right software), i can give them.) Kevin Baas | talk 19:40, 2004 Nov 14 (UTC)
Rhodite, made POV corrections. -- RyanFreisling @ 22:04, 14 Nov 2004 (UTC)
(dormant / move news items to new article?)
Anon User '66.30.181.242' has just performed edits that are POV. Modifiers like 'few', 'some', etc. are inappropriate.
The other edit, changing the quote about diebold from a direct quote to hearsay, is factually untrue. Link provided. Thoughts?
Note: I took "It should be noted however that these same counties historically have voted for the Republican presidential nominee" out of the Issues section, as it has no factual corroboration in the piece. Anyone have any data so we can put that back in? If it's a fact, it's important. -- RyanFreisling @ 03:27, 16 Nov 2004 (UTC)
if everyone is ok with the new format above, i'm going to begin deleting duplicate sections that got moved upwards (will not delete info that isn't up there already)... this should help out tremendously instead of having to scroll the entire page just to understand what is going on, and should help prevent good discussions from being buried in the talk page. -- kizzle 20:38, Nov 16, 2004 (UTC)
It is suggested that:
Can someone verify these and if so add them into the article, in appropriate places. FT2 21:41, Nov 17, 2004 (UTC)
In addition the article is getting too long, and much of it is just "backing reference information". Whilst this information is necessary for now, since there are multiple perceptions and uncertainty regarding much of the detailed information needed to review this controversy, and what information is sourced and verified versus rumour, I am concerned that the sheer volume is detracting from an encyclopaedic approach. I'd like to suggest with the archiving of much of the talk page, and the end of the VfD issue, that collaborators give serious consideration to siphoning a lot of the backing information into a separate article, "2004 U.S. presidential election controversy (summary of sources)".
Stuff that could go in there = most of everything that is "source material re-repesented" as opposed to "encyclopediac summary of the issues and information known at this time".
The main article would then reference this sources article for much of its substantiation and detail.
Comments? FT2 17:54, Nov 17, 2004 (UTC)
I could do a good cleanuup job on it that way, and make the main article really work without losing citations and quotes. But I'd want at least something of a consensus to do so, not just 1 or 2 voices, and I'd want to check with others as well. FT2 01:12, Nov 18, 2004 (UTC)
A lot is as you say to do with these differences - but a lot isn't. Some is evidence of security issues from specialists and hearings, some is internal memos of manufacturers knowledge of fraudulent claims, some is industry report summaries, some is political and corporate information relevnat to the matter, some is raw data sources used in maps, or tables, or non-core links, or news references. I think a single split between "summary" and "backing material" could benefit the main article. I'd be minded to title the other article something like "2004 U.S. presidential election controversy/Sources" or "/Supporting material" FT2 21:50, Nov 18, 2004 (UTC)
I think we should devise an overview article like JML suggested which then links to the current article broken into chunks. I'm sure we can come up with 3-5 main topics which merit daughter articles in themselves, whether it mirrors the top-level categories of the page itself or some other format. But definetely not shorten or revise the article, in such heavily scrutnized and "dismissed" claims that we have here, the more detail and direct quoting from sourcing the stronger it will hold up to such counter-claims. -- kizzle 08:01, Nov 19, 2004 (UTC)
Someone seems to have quoted large blocks of independent analysis into the article, that is ok with me but we need to be clear to the reader NPOV wise when there are headers like "Conclusion" in the article, those are not wikipedia's conclusions but are conclusions of the third party research. A header titled "Conclusion" should not be thought of as a wikipedia header in this case, it should be treated the same as the rest of quoted third party analysis text. Also, I propose that we have at least one citation link per quoted analysis paragraph so it's further clear that this is not a wikipedia conclusion, citations can be repeated. What do people think? Zen Master 22:01, 17 Nov 2004 (UTC)
I removed the following from the exit poll discrepancy section. CalTech/MIT have identified these are no different from normal, challenging this are analyses which contend CalTech/MIT made methodological errors, including comparing not native exit polls, but exit polls adjusted to match the official figures, and it is therefor no surprise that they found no evidence of significant difference. See below for CalTech/MIT's analysis.
I'm not sure of the facts being asserted, so I'm not sure. But at the very least, that sentence needs a restructuring. It's POV, makes characterization and conclusion and is grammatically damaged. Anyone want to take a crack at it, so we can get this on the article page? -- RyanFreisling @ 04:12, 18 Nov 2004 (UTC)
Its not difficult. This is what the two views say, in simple terms:
There is also substantial comment that the CalTech report was flawed and (perhaps) politically inclined on various websites:
Whilst some of the above are obviously personal POV, the key points appear to be well sourced and capable of verification, namely:
Ive added the key ones to the article to firm up the "methodological errors".
FT2 18:44, Nov 18, 2004 (UTC)
nice. -- kizzle 21:53, Nov 18, 2004 (UTC)
It looks like the 'partisan ties' section is now glutted with duplicative content, pasted in front of previous content that covered the same points. There's good stuff there, but lots of redundancy. I'm going to take a few passes at it. -- RyanFreisling @ 00:52, 19 Nov 2004 (UTC)
Had a skim and a history check. Looks pretty good. FT2 03:17, Nov 19, 2004 (UTC)
Several days ago I suggested creation of an overall summary article, and I proposed specific text for a lead section that would give an idea of what I had in mind. I've now created a first draft of that article at 2004 U.S. election voting controversies.
I think it’s very important that we try to keep the new summary article free of the problems plaguing this one (the older one). The mass of detail here makes this article hard to read and even harder to edit. Specific problems, IMO, include:
The new article is not a blog, or a newsfeed, or a Wikisource document, or a directory of links. For example, it summarizes the whole EVM issue in a couple paragraphs, with a link back to this article. The same treatment is appropriate for other issues. I suggest that anyone who wants to pile up a list of (for example) every voter registration dispute anywhere in the country in 2004 should create a separate article for that purpose and link to it. I warn everyone that, although I’ve largely given up trying to edit this (original) article, I’m going to be aggressive in editing the new one to prevent the kind of bloat we’ve seen here. JamesMLane 19:55, 21 Nov 2004 (UTC)
I understand, but having considered both I disagree. What someone said way up above seems the key to me: in the course of time, it will become clear what is gereally known and what is generally not known of the factual material. Right now a lot of the information is at some kind of "grass roots" level, by that I mean that its collected in the web pages of many minor newspapers and organisations, off-beam court cases, emails not reported on any american daily newspaper's front page. So for now, the background information although its big, is needed. I dont think many articles are needed.
At a pinch if you wanted to split voting machine issues from exit poll controversies... but they are different in character, at heart if you read both sections, both are about evidence... the only difference is one is exit poll statistical evidence (indirect evidence), one is manufacturers and other documentary evidence (direct evidence). I'm not convinced it needs splitting multiple ways just yet - but if it was, what would the subject matter of each sub-article be? Each section has potential to grow. I dont think we've seen the last of this yet... FT2 18:57, Nov 22, 2004 (UTC)
I notice this article still discusses the E-touch versus opscan machine voting difference in Florida. The view that it's explained by the Dixiecrat effect is mentioned with references to NYT and Wired News articles but isn't given a lot of weight. I'd just like to add that, after seeing the initial reports, I spent a boatload of time looking at the Florida by-county numbers and I concluded (sadly) that the Dixiecrat effect does indeed account for the differences. There are three points I'd like to make about it.
First, county size turns out to be a rather bad proxy for Dixiecrat population in a county. The opscan counties are not just smaller; they're also generally in the northern half of the state, while the E-touch counties are generally clustered in the southern part. Florida is highly inhomogeneous, and is stratified by latitude.
Second, I looked at the 2000 election results in Florida, and, broken down into "2004 E-touch" and "2004 op-scan" counties, they show exactly the same pattern as the 2004 results.
Finally, voting in Louisiana in 2000 showed the same strange patterns, which lends weight to the claim that it's conservative southern Democrats who are breaking the curve here. Interestingly, though Louisiana went solidly for Bush in both elections, there was not one parish in Louisiana in which more than 50% of the voters were registered as Republicans, and there was just one parish in which Republican voters constituted more than 40%. In contrast, there were a number of counties with ~70% Democrats.
If anyone's interested, details of this are on http://www.physicsinsights.org/elec04.html
-- Salaw 13:48, 22 Nov 2004 (UTC)
"Dick Morris, a career pollster (Republican), "
Dick Morris is most famous for his work under Bill Clinton, a Democrat.
While I understand that his beliefs now are more in-line with those of Republicans, is it fair to label him a "Republican?" It seems to me as if that label there is meant to disqualify his opinion, which violates NPOV. RNJBOND
<><>Here's a quick note, on subtle efforts to persuade: The parenthetical word "(Republican)" was evidently used, not to discredit what Dick Morris said, but instead to invest the statement attributed to Morris with some "against-interest" credibility. The technique here seems to have been to point out that, even though Morris was a supporter of Bush and an opponent of the Democrats (and would thus naturally have a partisan interest in arguing that the vote counts were accurate), Morris "admitted" that the discrepancy between the exit polling and the vote count was "incredible," because (in Morris' own view) exit pollling is almost always very accurate. Of course, it surely misrepresents what Morris actually was saying, to suggest that he meant to cast doubt on the vote counts in 2004. Instead, he seems to have been blasting the unprecedented incompetence of the exit-polling (apparently this was actually what he intended to say). As a general statement, I believe it would not be inaccurate to characterize Dick Morris as a Republican in recent years; Morris had a famous falling-out with the Clintons and has written many columns (and even some books) attacking the Clintons as dishonest, etc.
So now we also have the People For the American Way Foundation taking legal action in Ohio. What are all of the organizations? Should we make a list and put it in the article? Kevin Baas | talk 01:23, 2004 Nov 28 (UTC)
Kevin Baas created three new articles to break up this mass of material -- 2004 U.S. presidential election controversy, voting machines; 2004 U.S. presidential election controversy, exit polls; and 2004 U.S. presidential election controversy, vote suppression. Since then, however, no one has edited those articles. Instead, people keep piling everything into this article. I opined several days ago that it was far too long, and it's only grown since then. Is there some purpose being served by creating a huge, unreadable article? JamesMLane 05:54, 29 Nov 2004 (UTC)
Why is this 2004_U.S._presidential_election_controversies second article (which does not point here, or to other subpages) necessary? Thanks. -- RyanFreisling @ 18:17, 3 Dec 2004 (UTC)
May I say that I'm pleased with Z-M's new split of the lede on this entry. It was getting a bit unwieldy. Baylink 22:07, 30 Nov 2004 (UTC)
i added the Boston Globe's graphic and source. Good stuff!
Because of a decision by The Ohio Supreme cournt, for a recount in Ohio to happen before the deadline, Kerry, not Cobb or Badnarik, must file for it to be expediated. Kevin Baas | talk 20:59, 2004 Dec 1 (UTC)
Well, the counties have all certified their results, but, as I'm sure we all predicted, Ken is taking his sweet time certifying the state results. Kevin Baas | talk 00:38, 2004 Dec 2 (UTC)
This suit has to succeed: [13] [14]
I have a theory why 1/3 (8099/24472) of the provisionals were thrown out in Cuyahoga County:
Voter turnout on avg. in Ohio ws about 70%. 70% of 10,000 is 7,000. That leaves 1099, less than 10% of 24472.
This is why I believe that the election being overturned is contingent on the above law suit succeeding. Kevin Baas | talk 19:53, 2004 Dec 2 (UTC)
An excellent summary of the irregularities under active investigation can be seen in the House Democrats' letterto Kenneth Blackwell, sent Dec 2. [16]. Might do well for us to put some stakes in the ground around those specific questions and possible answers to come from OH Sec. of State Blackwell. -- RyanFreisling @ 07:14, 3 Dec 2004 (UTC)
How did you find those links? Is this significant enough to put on the current events page? Kevin Baas | talk 22:59, 2004 Dec 3 (UTC)
I have one question about the recount...if electronic machines have no paper trail, how exactly is a recount performed? --
kizzle 23:13, Dec 3, 2004 (UTC)
If stuff under new has been added please add "- Added" to the title", same with any current passages that have been removed, "- Removed" in the title.
Last thing we need to do, let us aim to archive all entries that already have been added both in New Passages and Possible Passages for Inclusion within the next few days, if you see something that is already on the page, please help out by putting " - Added" to the end of the title of the section. -- kizzle 21:50, Nov 18, 2004 (UTC)
Anon IPs that have mildly trolled this talk page: 68.107.102.129 05:36, 14 Dec 2004 (UTC) [unsigned]