>Jim Westrich wrote:
>>Could I add a smallish critique to your list? One of the things
>>I notice in looking at random digit dialing surveys/polls is that
>>women answer the phone by a 2 to 1 margin (after all, answering
>>the phone is housework and women still do a disproportionate share
>>of housework). Now, I am sure they reweight by gender but that
>>would definitely increase the margin of error for men (by
>>weighting the responses they do get more). There seems to me to
>>be a small bias toward Kerry in that (the men who actually do answer
>>the phone would be more likely Kerry supporters).
>
>If the "wrong" kind of person answers the phone, they often ask for
>the "right" kind. I'd ask in more detail, but I'm sorta in the
>doghouse with the pollster world for my Gallup lese majeste.
>
>Doug
I can speak from experience that at least one of (and probably all of) the reputable pollsters do something to reduce the chance of bias in terms of who answers the phone. When the Quinnipiac Poll contacted my house before the Massachusetts presidential primary, the pollster asked to speak with the adult in the house whose birthday was next to occur. While such a question might not eliminate any gender bias, I suspect that the pollsters are very likely to get the correct person this way.
--tim francis-wright "You know, the very powerful and the very stupid have one thing in common. They don't alter their views to fit the facts; they alter the facts to fit their views, which can be uncomfortable if you happen to be one of the facts that needs altering." -- The Doctor, Face of Evil.