[muawiyah]

The problem isn't the methodology, nor is there a model being applied to the data ~ it's the sampled population ~ you call 10,000 people and you will end up with about 910 old enough and willing enough to answer a few questions.

With CALLER ID in general use what this means is the ONLY people responding to the poll are those who want to respond. That is NO LONGER A RANDOM SAMPLE!

My judgment is this isn't a reliable or meaningful poll. Ignore it. You can ignore all the others too. They are all like that according to PEW.

[Mr Rogers]

I can’t remember the last time I bothered with responding to a call about politics. If there is any sign it is a “poll” or from a candidate, I either don’t pick up, or hang up. I have no idea what the political breakdown is of people who answer telephone polls, but I’d bet a lot of Republicans act like I do.