[muawiyah]

Sorry, "they" aren't correcting for that particular bias ~ they like it fine.

However, if you push on too thin a sample with too great a bias, and ignore the fact you just disqualified 91% of your random sample calls, you are just making a sham of random sample techniques and theory.

This is called Augury ~ it keeps the goobers entertained ~ it should not be used to support one or the other arguments in the making of public policy.

[JCBreckenridge]

That’s a variation of about 3 percent in a sample of 1200 people. Just simple math.

[muawiyah]

It's a variation of 3% only if the special interests answer the phone at the same frequency as the ordinary people ~ but they don't ~ particularly if they are tipped off about the survey.

If the special interests ALWAYS answer the phone and the ordinary people only answer 9% of the time, the special interests effectively create 11X the response rate for themselves than that created by the ordinary people.

That 3% becomes 11X 3%, or 33%.