Reply #148: I hope to... [View All]

[euler]

...answer some of your questions.

You say: "Does this sentence mean that there's no reason to believe the exit polls?"

I say: No. It means that, given the design of this exit poll, it's not possible to use the exit poll numbers to prove, or disprove fraud. The exit poll is still valid for the purpose it is intended.

You say: "are they saying that exit polls do not do what they are said to do?"

I say: Exit polls do what they are designed to do and no more. Please consider this quote from the report:

"some analysts inappropriately attempt to use current exit poll results to investigate whether the results in a locale show... that fraud might be involved in an election. A certain form of exit poll could be used for this purpose, but again the designs would have to be different. To validate results in specific precincts or from particular machines, the designs would have to incorporate larger numbers of interviews with voters leaving the polls for precision. And the stratification strategy would also need to be different, focusing on a combination of machine types and geography, for example, including a larger number of precincts at the first stage."

If you want a exit poll that can be used to verify the actual vote count, then as the authors say, "the designs would have to incorporate larger numbers of interviews with voters leaving the polls for precision. And the stratification strategy would also need to be different"

In other words, an exit poll that can be used to verify the actual vote count requires that it be conducted with a methodology that can deliver such results. This particular exit poll was NOT conducted that way; therefore, it can't be used to question the vote count.

You say: "The results of the exit polls in Germany are so accurate that they are used to determine the winners until the vote can be counted by hand"

I say: This question helps me make my argument. Exit polls in Germany are nothing like exit polls in the United States. Go here to see the difference:

http://www.mysterypollster.com/main/2004/12/what_about_thos.html

The German exit CAN be used to verify the actual vote count because they conduct their exit poll in such a way to make election verification possible (they make sure their respondents are a true random sample). If the US conducted exit polls like the Germans do, we could verify the actual vote count too.

You say: "When they qualify their remark by saying 'as currently designed and administered in the US,' do they mean that polls have not yet developed to the level where they can be used with confidence? Or is it just in the US where the problem lies?"

I say: No. What they mean is that the customer who paid for the exit poll (MSM) did not request an exit poll that can be used to verify the election. MSM, as they always have, asked for an exit poll that would enable them to make really cool statements about demographics. Such an exit poll costs about 10 million dollars. Had they asked for the additional requirement that the exit poll provide them with the means to verify the actual vote count, they would have had to pay about 10 million more, because a exit poll like that is hard to implement. It's all about cost. In Europe someone pays the price for an exit poll that can be used to verify the election. In the US, No one does - yet.