Tuesday, April 05, 2005

Exit Polls Revisited - Likelihood of Fraud is High

A very significant report was just completed by a bunch of PhD statisticians on behalf of the non-profit organization US Count Votes. Jerry Policoff, over on Daily Kos, wrote an even more condensed and clear summary.

Here are the most important facts:

  • The exit polls predicted a Kerry win by 3%. Computer counts showed Bush winning by 2.5%. Not only is that a 5.5% swing, that is the largest difference in exit poll history.

  • The pollster, Edison/Mitofsky, validated their process by selecting from actual computer data using the same decisions made in selecting voters. When that count was done, the results were only off by 0.3%, and biased in the opposite direction to the 'error' in the exit polls.

  • The margin of error in this rigorous, validated scientific study was 1.3%. 5.5% is more than 4 times that figure and therefore demands an explanation.

  • The given explanation by the group conducting the exit polls is that Bush voters were less likely to state their preference than Kerry voters. The pollster gave no evidence supporting this assertion. In other words, it was just a guess. But for that to be true, more people should have responded to the pollsters in democratic precincts than in republican precincts. However, the data shows the reverse is true. People were more likely to talk to pollsters in republican precincts. So if anything, the error should have overestimated Bush's lead.

  • The exit polls asked about both the Presidential and Senate races where applicable. The exit polls correctly predicted the Senate races, but not the Presidential races.

  • As the report states:
    This difference between the accuracy of the Senate and presidential exit poll is puzzling. Historic data as well as the exit polls themselves indicate that the ticket-splitting rate is low. It is reasonable to expect that the same voters who voted for Kerry were also the mainstay of support for the Democratic candidates in the Senate.
  • These simple facts caused these PhD staticians to conclude the following:
    The many anecdotal reports of voting irregularities create a context in which the possibility that the overall vote count was substantially corrupted must be taken seriously. The hypothesis that the discrepancy between the exit polls and election results is due to errors in the official election tally remains a coherent theory.
In other words, we need to now face the facts: The vote count as presented simply cannot be correct. The numbers are significantly skewed beyond mathematical error towards Bush. If we don't press to find out what happened, we're giving someone permission to steal our vote in the future.

Our vote is the most precious asset we possess as citizens in this country. People have died so we could avoid taxation without representation.

So what can we do? Much.

1. Please join one of the many groups working to protect our vote. Give them time or money or both.

2. Write your Congressperson and ask them a) to seriously and comprehensively investigate the discrepancies between the exit polls and the counts, (preferably using some of the same, honest people who wrote the report, as opposed to a "Warren Commission" guaranteed to produce a preordained conclusion), and b) ask him or her to support Rush Holt's bill H.R. 550, the "Voter Confidence and Increased Accessibility Act of 2005."

3. Find out what your Secretary of State is doing about electronic voting in your state. Ask to see their plan.

Please get busy. There's no time to lose. It takes a year to put anything in motion, and the 2006 election is therefore right around the corner.

This is really serious. If we give up our vote, we are no longer a democracy. It's ours to save or give away. Demand that it count.


Post a Comment

<< Home