Is Diebold's Voting Machines One of the Greatest Threats to Democracy?

[russ_watters]

The entire list of complaints regarding the 2004 election can be found https://voteprotect.org/index.php?display=EIRMapNation&tab=ALL". In Ohio, there were 4,166 irregularities reported. Among them:

-1192 Registration related incidents
-548 Long line reports
-500 Polling place inquiry incidents
-340 Absentee ballot related incidents
-280 Machine problems
-263 Votor intimidation incidents
[etc]

 

[pattylou]

Thank you for the link (2 posts up), Russ. There is regular blame being directed at the various election officials, in this report - blame that indicates non-machine-hack type problems favored Bush. That is a separate topic, but it is useful for establishing the take home message of the investigation.

It is also clear that some machines are considered, as a result of this investigation, to be unsuitable for elections:

While there is no reliable evidence of actual fraud in the use of these machines in Ohio in 2004, our expert advises that DRE (touchscreen) machines are not sufficiently safeguarded against fraud and are less usable for the broad population of voters than earlier simpler technologies; and that existing standards and practices for certification are insufficient to ensure th security requirements of DRE (touchscreen) systems.

It bears pointing out that (1) this report was published before the Hursti demonstration and thus no specific questions about the memory card hack (etc) were asked and (2) the hypothetical hack would have only needed to swing the vote by about 2% (they knew it was a close race). A swing this small may not be evident in an analysis that shows merely:

The tendency to vote for Kerry in 2004 was the same as the tendency to vote for the Democratic candidate for governor in 2002. That the pattern of voting for Kerry is so similar to the pattern of voting for the Democratic candidate for governor in 2002 is, in the opinion of the team's political science experts, strong evidence against the claim that widespread fraud systematically misallocated votes from Kerry to Bush.

In other words, I am not sure if the "pattern" they are describing would clearly show a 2% alteration. I assume they know what they are talking about (them being mathematicians and all!)(EDIT! They're not mathematicians, they're political scientists - they are giving their "best opinion." Oy vey!) but I want to see the analysis in more detail, and I'll appreciate your feedback when I look at the data. If their positive conclusion (as you stated it, that no fraud occurred) rests solely on this 'pattern' that they describe, and if there is scatter in the data such that the similarity of the pattern in voting in 2002 vs 2004 has some error in it, then it seems that the conclusion (positively concluded no fraud) may not be as cut and dried as it sounds in the two paragraphs on page 11 of the summary you linked.

I have found the full report (thanks again for the summary). Your point that people in both parties are interested in getting to the bottom of this, is well made. (I would, however, suggest that Howard Dean and others would be unwilling to be too vocal about machine hacks (either to avoid being labelled a conpiracy nut or to avoid revealing corruption in their party as well). )

As I have said previously in this thread, there is corruption in both parties. I am glad to see the democratic party investigating this, but it is not enough to convince me that Diebold did not fix a few votes.

off to look at the full report. It's 204 pages long. Don't wait up.

http://a9.g.akamai.net/7/9/8082/v001/www.democrats.org/pdfs/ohvrireport/fullreport.pdf

 

[russ_watters]

Your tone has reverted back to something like condescending and patronising.

My tone is just not letting you wiggle out of making a point by playing word games. It is disingenuous to argue against word-choice. What you should be doing is correcting my word choice if it doesn't fit what you really think (edit: after much pressure, you have now, mostly, done that). I want to know what the correct characterization is. And once you give it, I phrase my responses using your words to try to avoid future non-responses. If you say (for example) that you wish to approach this issue scientifically, I will hold you to that and you should expect to have it pointed out when you fail to do so.

 

[Informal Logic]

What are you guys talking about? THERE WERE investigations done. The Democratic Party itself did one that concluded there is "strong evidence [based on statistical analysis] against the claim that widespread fraud systematically misallocated votes from Kerry to Bush." http://www.johnkerry.com/features/count/Section.pdf
Guys, the fact that no one on the national level has been charged with fraud is not because no one is investigating, it is because public entities doing the investigations - even those with partisan biases toward the Democratic party - have positively concluded that systematic fraud did not occur.
Not even the democrats in Congress have pressed for opening criminal investigations (one exception: the Conyers report - but he got very little support from his Democratic peers and because of that, the issue died). Is it because they are wusses? Maybe, but more likely they just plain don't think they'd find anything.

I do not have time right now to read the entire document, but it appears that study was conducted only in regard to Ohio and in regard to whether "every eligible voter can vote and every vote is counted." I would need to read it fully to know if it was determined that votes were counted for the correct candidate, which would seem difficult if not impossible without a printed ticket for e-votes.

 

[pattylou]

I do not have time right now to read the entire document, but it appears that study was conducted only in regard to Ohio and in regard to whether "every eligible voter can vote and every vote is counted." I would need to read it fully to know if it was determined that votes were counted for the correct candidate, which would seem difficult if not impossible without a printed ticket for e-votes.

It is a brief summary (12 pages) and the parts relating to possible electronic vote fraud are even briefer (2 paragraphs) and are essentially entirely stated here:

Quote:
While there is no reliable evidence of actual fraud in the use of these machines in Ohio in 2004, our expert advises that DRE (touchscreen) machines are not sufficiently safeguarded against fraud and are less usable for the broad population of voters than earlier simpler technologies; and that existing standards and practices for certification are insufficient to ensure th security requirements of DRE (touchscreen) systems.

The tendency to vote for Kerry in 2004 was the same as the tendency to vote for the Democratic candidate for governor in 2002. That the pattern of voting for Kerry is so similar to the pattern of voting for the Democratic candidate for governor in 2002 is, in the opinion of the team's political science experts, strong evidence against the claim that widespread fraud systematically misallocated votes from Kerry to Bush.

IOW, according to the summary that Russ linked, political scientists hold the opinion that the patterns of votes for Kerry in 2004 are similar enough to the patterns for democratic votes in the governor's race 2002, that they consider misallocation of votes from kerry to bush unlikely.

They are not statisticians (they are political science experts), and the "conclusion" is explicitly stated to be an opinion.

Quick perusal of the full report http://a9.g.akamai.net/7/9/8082/v001/www.democrats.org/pdfs/ohvrireport/fullreport.pdf shows "best fit" lines among points with such incredible scatter (see pages 147, 148, 149, for example) that any discussion of "pattern" of voting leading to any "conclusion" of certainty one way or the other ----

Well it seems a bit extreme to *this* non-mathematician.

Please take a minute to look at the lines (best fit I presume) made on those pages, and tell me what your impression is, particularly if you have some good education in statistics.

-Patty

(I'm not trying to diss the party, this just looks like a red flag for bad data analysis to me. I don't even see error bars. Perhaps they're included in the captions, and as I said I've only perused it so far.)

 

[loseyourname]

I was referring to what you yourself write, which is usually your own opinion editorial, and not quotes for sources with a few of your own comments in regard to the quote. That's fine, but please don't criticize me for providing sources, which I do on a regular basis -- and this is the second time you've made a derogatory remark about using Wikipedia as a source. Wikipedia is considered reliable, but nonetheless it is by no means the only source I've referenced.

I didn't criticize you for using wikipedia as a source this time. I said that I was glad you chose an article that was not red-flagged. Granted, I was being sarcastic, and reiterating my previous criticism for when you did use the flagged article. Wikipedia might generally be considered a good source, but even you must admit that using an article that says right at the top of it that it has inaccurate content is not a good idea. Townsend was ripped to pieces when he linked to an article with inaccurate content on Fidel Castro.