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 #30: OK [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
Febble Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Feb-09-07 01:46 PM
Response to Reply #29
30. OK
The TOTAL number of 2000 voters (including first-timers) who did not vote (DNV2k) comprised 17% of the total 2004 vote.

I presume that you mean that the TOTAL number of voters in 2004 who did not vote in 2000 was 17% of the total 2004 vote.

How do you know? From the NEP "past vote" data? Do you assume that people correctly report having voted in 2000, just as you assume that people correctly report who they voted for?

TIA, all your analyses assume that exit poll sampling is unbiased, and that people correctly report what they did four years ago (e.g. whether they voted; who they voted for). While it is possible that these two assumptions are correct, there is no reason to believe that there are, and a fair bit of reason to believe they are not.

Contrary to your repeated assertions, I for one have never assumed that the vote count is correct. Even when I established that there was absolutely no correlation between increase in Bush's vote share and precinct-level redshift, I kept on analysing and reanalysing the data to see if there was any way that widespread fraud could both account for the even a substantial proportion of the total redshift and be consistent with that result. I have not succeeded. It is not for want of trying. Which is why I conclude that the exit poll data are NOT evidence of massive fraud, and, if anything, contra-indicate vote-switching fraud on a scale of millions.

Here is a plot of that analysis:

On the X axis is the precinct level discrepancy expressed in standard errors (positive values=redshift). On the Y axis is Bush's increase in voteshare in percentage points (negative values=decrease in voteshare).

You can see clearly, as you would expect, that the vertical zero line is to the left of centre, indicating that although many precincts had exit poll results that were well outside the MoE, substantially more of these were in the "redshifted" direction (count redder than poll) than in the "blueshifted "direction (count bluer than poll). Also note that the horizontal zero line is below centre, indicating that while in many precincts Bush's vote-share dropped, his vote-share rose in rather more.

Now, as you and many others have pointed out, vote-shifting fraud would tend to show up as an exit poll discrepancy - indeed what you and others allege is that vote-shifting fraud was responsible for the exit poll discrepancy in 2004. Note also that the exit poll discrepancy in 2000 was much smaller than in 2004.

OK. If vote-shifting fraud in 2004 occurred on a large scale, it would show up as an overall redshift, such as the one we see in this plot. It would also show up, as you surely must agree, as an overall improvement in Bush's vote share (you are, after all, alleging that Bush's increased vote share came from stolen votes, not real votes), again, such as we see in this plot. However, if Bush's increase in vote-share AND redshift both arose from the same cause (vote-shifting fraud), the two are likely to be positively correlated.

And they aren't. The upper confidence limit of the slope of the regression line is very low. This strongly suggests that the two phenomena (redshift; Bush's increase in vote-share) have not the same cause (fraud) but different causes (participation bias; increase in Bush's actual vote-share).

How else would you explain the complete absence of any discernable positive correlation between swing to Bush and redshift to Bush if both had the same substantial cause, namely, fraud? And how do you square this finding with your insistence that the NEP polls are a perfect random sample and that people accurately report their votes at the previous election?

So to address your question: given that IMO the exit poll data contra-indicate massive vote-switching fraud, then one possibility is that some of Bush's extra votes came from Gore voters, and voters who did not vote in 2000, and that of these, a proportion incorrectly reported having also voted for Bush in 2000. There is good evidence that this happens. I see no reason to think that it didn't happen in 2004, especially in the light of my findings that large scale vote-switching fraud is unlikely.

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