>Apologies if this has been posted before, but the premise that 70% of
>those asked to be polled refused remains an interesting caveat to the
>poll results thus far.
>
> - Deb
>
> October 2003
>The Public Opinion Polling Fraud
>from Z Magazine <http://www.Zmag.org>
>by Marc Sapir and Mickey Huff
>http://www.retropoll.org/polling_fraud_z-mag.htm
Sapir himself uses some pretty questionable techniques in his Retropoll. I think we went over them here.
But the Pew study I posted the other day <http://people-press.org/reports/display.php3?ReportID=211> says that the refusal rate appears to have little effect on accuracy. The sample still match national demographics, and predict elections rather accurately.
Doug