For the first time in my life I received a telephone call from a polling firm asking me about my political preferences. I have been surveyed many times about my favourite toothpaste or radio station but never have I ever been called for a political poll.
It was an automated poll. The system asked me questions and I pressed buttons on my phone to answer.
It was a short poll, only nine questions, of which, five were the standard demographic questions.
The poll asked one question each on whether I thought the country and Ontario were heading in the right or wrong direction.
Then it asked me who I would support if an election were held in Canada and Ontario.
Interesting that they asked the direction questions first. The answers to those questions could influence how someone answers the subsequent party support questions.
Having designed surveys and polls in the past my choice would have been to put the party support questions first. It would make analysis of the estimates from those questions much simpler and reliable. Done the other way I cannot really determine the level of influence the direction questions would have on the party support questions so any analysis of the party support estimates would be much less reliable.
Not that it matters any more as I have been out of that business for almost a decade but I still have an interest in how these companies design their polls.
It was an interesting 3 minutes.
1 comment:
Hi Ottlib, I have never ever been polled in my 67 years and it makes me wonder why that is so and do the polling keeep a record of who and how they answer the questions and their answers indicate to who they will poll next depending on the answers they want to receive.
Whats your fellings on this?
Manipulative questions by polsters are common and they do not present clear unbiased questions. At least that's the findings I see when I hear of a lot of those polled on being targeted time and time again. It almost seems that they are always targeted by what party they support.
Post a Comment