After offering my take on the first presidential debate, it was interesting to see most of those presenting opposing comments based their response on the results of polls conducted after the debate, particularly one by CNN (not their “Perception Impression” gadget, but actual polls.)
I do not mean to change anyone’s opinion at this point. It is clear that in regards to that debate and perhaps others to follow, if you’re an Obama supporter, your guy won. If you’re a McCain supporter, your guy won. As Iranians, we have always been politically stubborn and perhaps we are stubborn Iranian-Americans now.
However, the nature of polling is still at question and for those who do not question it, often citing some magical “scientific” reason behind the numbers, just ask yourself one question; if the polls were such an exact science, why would campaigns conduct their own separate, rigorous, frequent and very expensive polling instead of just looking at the CNN results?
The answer has much to do with the same “science” behind the polls, for as everything so scientific also leaves an opening to be scientifically manipulated, and it often is. And the reason I turned off my television right after the debate to write my opinion was to not hear those same polls as well as the spin-masters’ version of what took place.
There are many ways to manipulate the outcome of a poll. The most common are of course;
- What questions were asked? In one of the campaigns I worked for, we got two separate results when asking whom “do you feel WON the debate” versus “do you feel CAME ON TOP at the debate”. The difference was close to a huge10-point margin when all else was equal and nobody could explain why. To this date I can’t understand why people responded so differently to essentially the same question.
- In what order were the questions asked? It is known that you can “lead” a subject by just choosing the order you ask the questions. If you get 4 positive answers in a row, the 5th is most likely a positive also, for example. Or if you set up the interviews so your most crucial question (D) is always asked before (E), (G) and (Z) to get the result you want.
- How random were the interviewees? Again from my personal experience, I’ve seen media choose random phone-in interviews to conduct a very critical poll. What they didn’t take into account was that according to latest stats, 58% of renters in California don’t have landlines anymore and this was a poll that directly affected renters on a proposition initiative in CA. Needless to say, they had a list that excluded cell numbers, leaving the most important folks out of their poll. They were calling areas that had high renters, but mostly talking to their landlords that had 2-6 units for rent in, living in the same complex!
I am not suggesting CNN wanted to manipulate the results (although it would be naive to suggest any media is without a bias these days) or they are close-minded to such issues. But they are a news organization, filling their time and I am not going to let them tell me what I think. Poll or no poll, from what I watched, Obama lost round one. Sorry if he’s your candidate and hope he prepares better for round two.
This post first published on my blog at www.eyeranian.net deals with traditional and accepted methods of polling and not what is used at most focus groups today with a small sample of people and a gadgets to indicate based their immediate responses.