Electronic Vote Switching from Kerry to Bush - 9:1 swing state ratio

[Time for change]

This is a cross post to a thread I just posted on GDP:

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=132x2020121

In May of this year, I did an analysis of the U.S. Electronic Incident Reporting System (EIRS) database (for the 2004 presidential election), in which I showed that incidents of "electronic vote switching" that favored Bush outnumbered incidents that favored Kerry by a ratio of greater than 12:1 – there were 87 incidents that favored Bush, and 7 that favored Kerry.

My conclusions were:
1. The probability of such a lopsided ratio in favor of Bush was extremely unlikely
2. Although it was possible that these results could be explained by a reporting bias, whereby Kerry voters were more likely to report incidents than Bush voters, it seemed highly unlikely that this could result in such a large ratio.
3. Therefore, election fraud (i.e., purposeful programming of the computers to switch votes from Kerry to Bush) would seem to be by far the most likely explanation for these findings.

Recently, mgr has suggested that another way to test these conclusions would be to compare the rate of vote switches in swing states vs. non-swing states. It seems probable that Republicans would target election fraud in states that were likely to have influence on who won the election. Therefore, if the rate is much greater in swing states than in non-swing states, that would provide confirmatory evidence that bias was a highly unlikely explanation for the lopsided ratio of vote switches in favor of Bush.

Go to above link for a description of this study.