Monday, January 31, 2005

TRUTH LEFT OUT: MSM Fail to Report an Important Exit Poll Study By Prominent Experts That Invalidates Mitofsky's Explanation

U.S. votergate media blackout


I have found the first comment made by Mitofsky since the damning report by refuting his conclusions.


It's really short.

Here it is:

The Edison/Mitofsky report was not investigating election fraud. We were looking at how to guarantee the accuracy of exit poll data.
- from - February 2, 2005

(more blank space for dramatic effect)

I really don't know what to say. That comment says it all. Either Mitofsky is:

1. Completely off his rocker


2. Being forced to say these things, which defies all logic and reasoning.

I'll just state the obvious and be as brief as Mitofsky.

How the hell can you guarantee the accuracy of exit poll data if you don't even consider if there was counting error or not????!!!!!

How the hell can you guarantee the accuracy of the exit poll data if your conclusions contradict the data????!!!!

And what's more -- "the integrity of the election process" (what is about) does not necessarily equate to "vote fraud".

Or does Mitofsky know something more?



Word is slowly, and I mean slowly, getting around about this important report. An article by Judy Bertelson at the Berkeley Daily Planet has made a very astute observation about the Mitofsky / Edison report. She states:
...Furthermore, the Edison/Mitofsky report elides from hypothesis to assertion of fact, without the benefit of confirming data.

Edison/Mitofsky are quoted as saying, “While we cannot measure the completion rate by Democratic and Republican voters, hypothetical completion rates of 56 percent among Kerry voters and 50 percent among Bush voters overall would account for the entire Within Precinct Error that we observed in 2004.”

(Within Precinct Error is defined as “an average of the difference between the percentage margin between the leading candidates in the exit poll and the actual vote for all sample precincts in a state.”)

However, this hypothesis is treated as fact on page four of Edison and Mitofsky’s Executive Summary, “It is difficult to pinpoint precisely the reasons that, in general Kerry voters were more likely to participate in the exit polls than Bush voters.” The hypothesis has morphed into asserted reality.
Are we to believe that a highly respected researcher, such as Mitofsky, has made such an egregious error in logic by mistake? What is the "probability" of that?!



I waited a few days to see if any of the MSM would pick this story up. Surprise, surprise -- NOTHING! This report by prominent statiticians and mathematicians across the U.S. has concluded that Mitofsky is simply WRONG with his own exit poll analyses. Based on the numbers provided by Mitofsky in the "investigative" report entitled "Evaluation of Edison/Mitofsky Election System 2004", the professors and researchers report that Mitofsky has no evidence to support his claim that the exit poll participation rate was skewed in Kerry's favour. In fact, there is evidence to suggest that the exact opposite was true.

As Newsclip Autopsy has mentioned multiple times before, there are only 2 explanations that could account for the exit poll discrepancy:

1. Significant Systematic Exit Poll Bias

2. Counting Error (benign or malicious)

According to Mitofsky's research, there does seem to be a few variables which points to number 1. However, as I've noted below, the discrepancy still exists when the ideal conditions are met. Unfortunately, the report does not address this particular point. It is my belief that this systematic bias was NOT significant enough to account for the overall discrepancy. I have emailed Dr. Freeman (one of the contributers of the report) about this particular query, but I have received no response as of yet.

Nonetheless, this new evaluation of the Mitofsky data makes it clear that FURTHER ANALYSES OF THE RAW DATA is needed in order to address this issue and others. So far, none of the media outlets will allow this information to be released.

As for "counting error" of the actual votes, statistician Josh Mitteldorf explains that:
Now we have statistical evidence that these reports were the tip of a national iceberg. The hypothesis that the discrepancy between the exit polls and election results is due to errors in the official election tally is a coherent theory that must be explored.
(from Scoop - January 31, 2005)
Furthermore, the report goes on to say that the hand counting of paper ballots showed NO STATISTICAL DISCREPANCY with the exit polls, as was the case with the counting and tabulating with the other machine methods. Mitofsky DID emphasize in his investigation that there was no error difference between e-voting and voting using paper ballots, but he did not compare "hand counting" against "machine counting". Why he did not do this is odd -- considering he had the data for comparison. Afterall, this difference was of major importance during the Ohio recount in December when only 3% of the votes were actually recounted by hand. There were multiple instances when the the hand recount did not match the machine count. By state law, a full hand recount should have taken place. However, with orders given by SOS Ken Blackwell, the machines were replaced with new ones and a full hand recount did NOT take place.

This report proves that Mitofsky DID NOT do his homework properly and HAS NOT adequately explained why a significant discrepancy exists between the exit polls and the official results of the 2004 U.S. Presidential election. It also begs other basic questions not queried by Mitofsky -- supposedly "the father" of exit polling. The report is an extremely important piece of work done by highly respected professional and credible mathematicians and statiticians.

When are we going to see this critical analyses in mainstream media???!!!

Indeed, when is the mainstream media even going to release the raw data so that further objective analyses may continue???!!!

Image Hosted by
Location: Vancouver, British Columbia, Canada