I did try to take a look at the report, as I’m always interested in things like this — but the pdf file is broken at this point.

Drop a message on the board when it’s working again — I’d like to read the report in depth (as it’s always, *interesting* to see how the popular press interprets things like this…!)

This talks about our methodology, shows the full questionnaire, etc. We are preparing Wave 2 – so any comments, advice or critiques you might have would be most useful.

And non-responses are only mentioned as an aside to the research report (e.g., 1000 surveys were mailed, of which 400 were returned), because non-responses are exactly that, non-responses.

I fail to see what your point is re: asking someone’s political affiliation. Asking questions that determine an audience’s demographic makeup is an absolutely accepted and standard practice in the polling and research world.

Also, regarding your question about people that just “walked away,” I would be willing to bet that the poll was done via the phone, with perhaps an online component, as well. If people don’t complete a survey because they “walk away,” hang up the phone, etc., they’re not included in the sample. Why WOULD they be?

I even appreciate your feedback Mr Inspector – I’m sure you could go “on and on” with your list of serious flaws. It is easy to do so when you have no idea what you are talking about. You are making enormous assumptions about the survey that are simply not correct. The survey was not conducted in the presence of a pollster. It was mailed to respondents to complete. We know the numbers for people who returned and did not return the questionnaire and did not have higher non-response from a particular political group, etc. And I’m sorry if a pollster has hurt you in some way.

If any of you have any questions or comments feel free to write me at any time or post here. I’ll respond in a few days when I return from a conference. It will be at least entertaining to see how much I have annoyed “Mr Inspector.”

Thanks again for your interest and keep up the good work!

Any attempts to justify this with the old, “we’re trying to group our participants”, or “we are gathering data about specific groups” merely points out how un-scientific the results are. The study should be totally blind, the questions should be designed to be as non-ambiguous as possible, and conducted without a pollster in order to provide any usable data. And still, whomever compiles said data, may impart a bias to said results intentionally or not.

Hence, the very not scientific nature of the poll/survey.

Where in your numbers do you account for the people who merely gave your pollsters a dirty look and walked away without a word? How did you classify those according to political affiliation? I can go on with a list of serious or fatal (to the survey) flaws.

No doubt about it, we are still the Planet of the Apes