You are viewing an obsolete version of the DU website which is no longer supported by the Administrators. Visit The New DU.
Democratic Underground Latest Greatest Lobby Journals Search Options Help Login

Reply #9: that's a claim, and your own evidence undermines it [View All]

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
Home » Discuss » Topic Forums » Election Reform Donate to DU
OnTheOtherHand Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-04-09 04:44 PM
Response to Reply #8
9. that's a claim, and your own evidence undermines it
Again, you posted evidence that the gap between lever and paper votes in New York was (to use your hyperbolic word) "uniform" in 2000, 2004, and 2008. If that suggests anything, it suggests that the level of miscount on lever machines was similar (whether high or low) in those years, and therefore, that at least some of the exit polls were wrong.

(I think this is far from the best evidence that the exit polls have been wrong, but hey, it's your evidence, so I'm trying to work with it. I think we all noticed that provisional voters aren't exactly a random sample of the electorate. Didn't we?)

For a moment, you tried to solve this problem by arguing that the gap wasn't uniform at all -- that, properly construed, it was narrower in 2000. That didn't work, so now apparently you are back to trying to change the subject.

The EVIDENCE shows that the late votes closely MATCHED the exit polls.

That is incorrect. Your table actually shows that the closest "MATCH()" between late votes and exit polls was in 2004.

That EVIDENCE was CONFIRMED by the STRONG CORRELATION between the exit polls and the late votes.

That inference is incorrect, and rather jarringly so. Using your data, I was able to replicate your finding that for the 50 states plus DC, the correlation between late vote and exit poll share is 0.72. I was also able to calculate that the correlation between initial vote and exit poll share is 0.976. By your reasoning, surely, this is very strong evidence that the initial vote totals are correct and some of the late votes have been corrupted. I can't imagine how to construe it as evidence that the late vote count is more accurate than the initial vote count.

Limiting the analysis to your top 20 states narrows the gap but doesn't fundamentally help your cause. Again, I replicate your correlation between late vote and exit poll share, 0.92; the correlation between initial vote and exit poll share is 0.984. This might be construed as evidence that some of the late votes have been corrupted -- or that some late-vote counts are more representative (or differently biased) than others.

I further reckoned, using your data, that the correlation between WPE and "change" from initial to final vote is 0.095, p = 0.51 -- exquisitely non-significant. (In the top 20 states, it's actually negative and non-significant.) This result constitutes strong prima facie evidence that these two quantities do not measure the same underlying variable (fraud or anything else).

Maybe we could at least check whether the late votes are more strongly correlated with the initial votes or the exit poll returns? Certainly we could. In the top 20 states, the correlations are 0.948 and 0.920, respectively, which could be construed as weak evidence that the initial votes are more accurate than the exit poll results. Across all 50 states plus DC, the correlations are 0.726 and 0.721, basically indistinguishable (although the difference is in the same direction). I don't think any of these results amount to much, but they surely don't support your argument.

Since these correlations are reckoned from your own data, it's hard to fathom how you could have failed to compute them, or failed to report them, whichever is the case. At this point, you might want to consider asking readers to forget all about the late vote counts.


I don't claim that the official vote count was strictly correct, but I do believe that the exit polls were wrong (i.e., that the error exceeded the statistical "margin of error"). As a separate issue, I also believe that a full recount (if possible) would confirm that Bush received more votes. None of this amounts to claiming a fair election. It's hard to believe that you're incapable of making these distinctions.

Do you think your position is flattering to Democrats? I don't. But, more to the point, I don't think it's true. If you want me to change my mind, first you will need better arguments.
Printer Friendly | Permalink | Reply | Top

Home » Discuss » Topic Forums » Election Reform Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002
Software has been extensively modified by the DU administrators

Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

Got a message for Democratic Underground? Click here to send us a message.

© 2001 - 2011 Democratic Underground, LLC