Monday, November 29, 2004
Vote Fraud: Rampant but Not Enough to Throw the Election
Today I investigated more closely the votes in Florida's counties. I used the data I referred to in yesterday's blog. I included the Bush and Democratic 2004 and 2000 votes, and computed for each county the percent that each of Bush and the Democrat won between each other (that is, I excluded the minor candidates). I also included the attribute of "Machine", whose values were Computer (for the new computerized touch-screen voting) and Op-Scan. I then computed for each county the amount by which Bush and the Democrat gained between 2004 and 2000 (in percent). I ordered the data by descending Bush gain and found that in the upper half, where the Bush gains were greatest, there were only three "Computer"s and the rest were "Op-Scan"s. In the lower half (smallest Bush gains) there were ten "Computer"s and the rest "Op-Scan"s. To me this tells me there was vote fraud. I would have to make a statistical analysis to see the probability of a 10-3 split, but I think it is low. Something put Bush votes into the Op-Scan machines.
But did this fraud throw Florida and hence the election? To check that out, I made a new table. For Bush I took the 2004 vote if the county is a Computer county; for Op-Scan counties, I took Bush's 2000 votes times the total vote (Bush and Democrat) for 2004 divided by the total vote for 2000. Thus I took the 2000 split, adjusted to make it comparable to 2004 figures. I did the same for the Democrat, yielding a vote for Kerry. I then totaled the votes and came up with Bush 3,713,029 votes, and Kerry 3,510,944 votes. This is a victory for Bush by 51 to 49 percent. The margin shrunk, but not enough to make Kerry the winner. I therefore conclude that although there was fraud in this election, it was not enough to throw Florida and the Presidency to Kerry.
Therefore, I think that this whole vote fraud thing about the 2004 election should be ignored. It's a conspiracy theory; it is hopes pinned somehow on these little irregularities that they might upturn the election. Maybe someone can show me how the figures show somehow that Kerry should have won Florida, Ohio, or the election, but right now I think the vote fraud theory has been debunked.
But did this fraud throw Florida and hence the election? To check that out, I made a new table. For Bush I took the 2004 vote if the county is a Computer county; for Op-Scan counties, I took Bush's 2000 votes times the total vote (Bush and Democrat) for 2004 divided by the total vote for 2000. Thus I took the 2000 split, adjusted to make it comparable to 2004 figures. I did the same for the Democrat, yielding a vote for Kerry. I then totaled the votes and came up with Bush 3,713,029 votes, and Kerry 3,510,944 votes. This is a victory for Bush by 51 to 49 percent. The margin shrunk, but not enough to make Kerry the winner. I therefore conclude that although there was fraud in this election, it was not enough to throw Florida and the Presidency to Kerry.
Therefore, I think that this whole vote fraud thing about the 2004 election should be ignored. It's a conspiracy theory; it is hopes pinned somehow on these little irregularities that they might upturn the election. Maybe someone can show me how the figures show somehow that Kerry should have won Florida, Ohio, or the election, but right now I think the vote fraud theory has been debunked.
