Because I can't ever completely lay off the political posts...

[c9]

Expected vote tallies are based on the percentage of registered Republicans in a county multiplied by the voter turnout in that county. In Florida during the 2004 US Presidential election, counties that used touch-screen voting systems recorded an average increase over the expected Republican vote tally of 27.93%. That is, if 10,000 Republican votes were expected to vote, the results showed 12,930 Republican votes. The Democrat vote showed an average increase over expected tallies of 23.80%. Not a huge difference, when you're playing with such focused numbers.

In counties that used optical scan ballots, which are considered more precise than other methods by many, the differences were a little different. Democrat turnout averaged a drop of 22.29%. Republican turnout averaged an increase of 134.79%. That's not a typo.

One county had an increase of 712% over the expected Republican vote. 26 counties showed more than a doubling of the Republican vote, 11 a tripling. The top five were 297.40%, 358.10%, 433.20%, 459.30% and 712.30%.

The county showing 712% has 4075 registered voters, of whom less than ten percent registered as Republican. 88.3% registered as Democratic. But the final vote tallies show nearly 2000 votes for Bush, and only 1070 for Kerry.

To be honest, something is not right there. But I don't know if anybody will ever figure out what's going on.

That's why I'm pretty happy leaving computers out of elections here in Canada. The efficiency hit seems to be worth the trouble.

Comments

[ex-halfwitte432.livejournal.com]

nikkinewsnet has a good article today about why the statistical anomalies have occurred.

[c9.livejournal.com]

Hmmm. SOme good information there, though I found it didn't directly address some of the stuff I'm seeing. It's easy to find conspiracies, of course. Just a shame.