Ohio SOS shenanigans in NEP report

[mgr]

I was reviewing the latest USC report and re-reveiwing the NEP report, when I noticed a section that I had overlooked on pp 70-71.

http://exit-poll.net/election-night/EvaluationJan192005.pdf

"The most important state was Ohio in which we were forced to file suit against the Secretary of State of Ohio the day before the election. Five days before election day the Ohio Secretary of State told county election officials to keep all exit poll interviewers at
the 100-foot electioneering distance. He had previously stated in writing that exit polling was not electioneering and that no Ohio statute regulated the distance from the polling place for conducting exit polls. During the summer the Ohio director of elections had
assured us there was no change in the Secretary of State’s position.

Our suit was filed the day before the election. We were successful in overturning the Secretary of State’s ruling, but the court ruling did not occur until 10:30 PM on the night before the election. Although we were able to contact all of our interviewers before the polls opened, many Ohio election officials at our polling places did not know of this ruling when the polls opened and many of our interviewers in Ohio were delayed in starting their interviews until the local election official was informed of this ruling. The last local election official did not permit our interviewer to begin work until close to 5PM."

I am in the process of presenting a thread that lays out a methodology that will permit anyone to rehabilitate a state's exit poll to where the MOE is reduced, and comparisons to the actual vote could indicate fraud. This section in the NEP report is critical to anyone who is already undertaking that effort.

Mike

[qwghlmian]

1. About that methodology for accurate exit polls

Exit polls as they are conducted currently in the United States cost somewhere around $10M (see http://www.washingtonpost.com/ac2/wp-dyn?pagename=article&node=&contentId=A51692-2003Jan13¬Found=true about halfway down the page) and are not designed to predict the outcome of the election or to verify the non-fraudulency of the election.
(see http://election04.ssrc.org/research/InterimReport122204.pdf)


A German-style exit poll that would be suitable to detecting fraud would cost about $80 to $160 million - judging by the methodology used and comparing the sizes of Germany and US. (see http://www.mysterypollster.com/main/2004/12/what_about_thos.html for the description of differences between the two types of exit polls) and would still not be as accurate, due to the US high non-response rates and the fact that German Census bureau does official demographic surveys inside polling places and nothing like that exists in the US.

Here is what is needed to change US exit polls to the much accurate German model:

* Increase sample size by the factor of 10 to 12
* Increase the number of interviewers by the factor of 20, and increase their training
* Change laws in the US that would allow exit pollsters to be present at the immediate exit from the polling place, and not some distance away
* Do exit polling through the whole election day, currently there are gaps in the middle and in the end
* Somehow increase response rates from the current below 50% to 80-85%
* Have the government conduct demographic survey of voters INSIDE the polling place

Some of this takes money. Some needs legislation.

[tandem5]

2. television advertising might be another helpful component

Familiarize the public with the uniforms and any other identifying markers the pollster will be wearing. Additionally use the airtime to show what type of questions they will be asking and exactly how the interview will be conducted. Set it up so it’s made clear in the ads that the sole purpose of this survey is to verify the integrity of the vote.

[mgr]

3. Can you save that for when I post it, please?

What I was talking about is (a coming attraction) applying an ANOVA along with an F test that will segregate the precincts to those that accurately sample the precinct population so that when aggregated into larger numbers the MOE does not have to reflect the bias that occurred with the precinct level sampling.

I'm trying right now to see if one could combine the LA Times data with the NEP data.

I wanted to post this in hopes it may spur discussion without my shepherding it along. Oh well, let the border collie start:

It suggests that the Ohio exit poll may have been intentionally bullocked up by the SOS, in two ways, locating the pollster 100 feet from the precinct, and or delaying the start of individual precinct exit polls. What it should suggest is that the WPE for Ohio may be so great as you might not be able to overcome its effect, but that other states may not have as great a bias effect, since outlier effects disproportionately distort the mean from the median (skew it).

It also can be the basis for a legal complaint against Blackwell, since if the NEP and the previous incarnations were allowed to be in the precinct, and they changed the rules, an stronger argument can be made regarding his intent, for the very reason that an exit poll is to meant to address fraud.

Mike

[mgr]

4. Reply 2 Excellent post

Don't get my first post wrong.

Mike

[mgr]

5. One minor quibble

Mitofski claimed that the exit poll could be used for such a purpose as detecting Fraud, however he did not elaborate. The ANOVA is the usual tool to assess whether smaller samples may be representative of a larger population or not. My opinion is that it is only with the state exit polls that we can assess if any fraud took place, not the national poll.

The variablility that characterized WPE appears to have a geographic component that may not apply to every state(the discussion on exit polling place distance suggests this), so we might be able to resurrect an exit poll in Florida with a smaller MOE than the 6.5 overall. Whether this will be sufficient to bracket possible election fraud remains to be seen.

Mike