Exit Polls

When I voted last week, I answered an exit poll outside the door. The company was Yacoubian Research, and they told me they were hired by Fox News. I am curious about that data. Where is it? Who has it? What do they think it says?

I know that the media outlets use exit polling data to improve the accuracy of their extrapolations on election night. Some returns plus the polls allow them to make very accurate calls early in the evening so that we can all go to bed knowing who won and still make it to work on time the next day...Remember 2000?

I also know that campaigns sometimes use this kind of information. I am not drawing any conclusions, but as I researched this question, I discovered that Yacoubian Research is a company from Memphis. It is named for pollster Berge Yacoubian who has advised many political candidates in his career, including Bob Clement (Memphis Flyer, February 7, 2002). I also discovered that Bob Clement's campaign has used Yacoubian in this campaign (now, I wish I could produce this evidence, but I cannot get the City Paper links to work. Google Clement and Yacoubian, and you will see the links...I just can't get them to display).

Here were the questions:

1. Whom did you vote for today?
2. For whom would you vote in a Clement-Dean run-off?

That's it. I am happy to be asked and answer these questions, and I did, but when the interview was over, I got to ask my questions. The pollster, who was very nice, by the way, said she was with Yacoubian Research and that this company was hired by Fox News.

This is not a blog about specualtion and inuendo. So, let me be clear: I could care less who really hired the group. I am willing to take them at their word, and I expect the Fox affiliate in Nashville will use the data tonight. I am interested in this poll as one who spends alot of time looking at them.

What is curious about this poll is that it only asked these two questions. The second question could easily be designed to elicit certain perceptions about the race. This practice is called push polling. I hope that the Fox affiliate will produce a story tonight in which they analyze this whole concept of defensive voting as it played out in this race. I can see a direction for interpreting the data from these two questions in order to extrapolate how much Briley and Dean affected each other and whether anti-Clement sentiment really played a role in selecting one candidate or another. We'll see. That could be interesting stuff.

I will only say this: If I were the editor in charge at Fox, I would try to use a company with a more plausibly independent face than Yacoubian which has obviously worked for one of the candidates in this election. I don't ever want this blog confused with a cynical conspiracy theory joint. All media whether institutional or blogospheric make selection decisions about what to include and what to exclude. I just want people to remember that so they will investigate further and make up their own minds. At Fox they like to say that they report, and we decide. I am just including a little more reporting that might be useful in deciding.

Leave a comment if you are polled today.