Pre-election advice on Exit Polls from Edison Media !!!

[k8conant]

http://www.edisonresearch.com/home/archives/2004/10/what_to_watch_f.html

• What to Watch for on Election Night
  
  October 28, 2004
  by Larry Rosin, President
  
  6) How do they know?
  
  While the networks are promising to be a bit more transparent on this point, most Election coverage viewers are still mystified when the networks say things like “With 2% of the precincts reporting, we are projecting that will win.”
  
  How do they do that?
  
  This is where Edison Media Research comes in. Working together with Mitofsky International, we are providing the networks with two streams of data, Exit Polls and the Quick Count.
  
  In most states, the outcome of the Presidential election will be “callable” from the information in our Exit Polls, and these are the “top of the hour” calls.
  
  In other states, the Exit Polls will imply that one candidate or the other is very likely to win, but there is not enough information to confidently make the call. Often, the second stream of information, the “Quick Count” can lend greater assurance. We will have people stationed at thousands of voting locations across the country. They will be calling in the results from those sample precincts as soon as the votes are counted. Our system will add this new information to the Exit Poll information, and often this is the push that makes the outcome clear.
  
  Finally, the Associated Press will be counting all the votes. This final stream of information will be added to our system to eventually (we hope) allow a winner to be called, or at least to show that a race will go to a recount or to further discussion.
  
  This article is being written on Wednesday, October 27, six days before the election. As of today, no one has any idea who will win. Election Night promises to make for extremely exciting television or radio.

I had never heard of the "Quick Count" before. Apparently those are actual results from certain precincts. I wonder which ones they use. They're adding that to the Exit Poll data.

:shrug: :kick:

[dargondogon]

1. Write to polling association, ask for Mitofsky's head

Tired of secrecy about exit polls? Turns out secrecy violates pollsters' code of ethics.

Republican pollster Frank Luntz got smacked down by the American Association for Public Opinion Research for failing to disclose his methodology.

Perhaps we should complain to AAPOR about Warren Mitofsky.

Time and again I've read about researchers complaining they can't get the raw data or clear explanation of methodology from Edison/Mitofsky's National Election Pool. If true, that appears to be a violation of AAPOR's Code of Professional Ethics and Practices.

From AAPOR's news release:
"In particular, the AAPOR inquiry focused on Luntz's reporting, prior to the November elections in 1994, that his research showed at least 60 percent of the public favored each of the elements in the GOP "Contract with America." When later asked to provide some basic facts about this research, Luntz refused.

"AAPOR holds that researchers must disclose, or make available for public disclosure, the wording of questions and other basic methodological details when poll findings are made public. This disclosure is important so that claims made on the basis of opinion research findings can be independently evaluated. Section III of the AAPOR Code states: "Good professional practice imposes the obligation upon all public opinion researchers to include, in any report of research results, or to make available when that report
is released, certain essential information about how the research was conducted."

Mitofsky is a bigwig among AAPOR's members. Perhaps it's time they asked him to live up to the organization's code of ethics.

[Fortunato]

4. I might be wrong but...

...I think the actual surveys used have actually appeared on the web, and that's all that is required by the law you state here (unless there is more detail you are not stating).

In other words, Mitofsky is not required to release the actual counts of the polls. he is simply required to release the methods and questions he used in the poll, along with any generalized results, so that other experts can evaluate whether the methodology/questions he used was sufficient and theoretically accurate. He has done that, I think I recall.

But Mitofsky still owns the actual data he collected and is permitted to release it to whomever he sees fit (i.e., those who pay enough money). America runs on the capital system.

[Carolab]

2. I forwarded this to Arnebeck

I think they need to see this too. Very odd. Good find.

[k8conant]

3. I hope so! n/t