HolyCoast: Gallup vs. Zogby
Follow RickMoore on Twitter

Monday, November 01, 2004

Gallup vs. Zogby

Jonah Goldberg on National Review Online made this point:
It's pretty amazing that CNN/USAT/Gallup and Zogby could conduct polls almost simultaneously in Wisconsin and come up with such diametrically opposed findings. Gallup (10/27-30) Bush +8 (52-44). Zogby (10/28-10/31) Kerry +7 (44-51). That's a fifteen point difference, which is just massive.

Here's a pretty good analysis of the differences between the Gallup Poll and the Zogby Poll, also courtesy of National Review Online.
Gallup: This poll is truly the gold standard in presidential polling. They leave their partisan identification unweighted. This lends itself to swings in partisan identification, but it can enable it to pick up the shifting partisanship of the electorate. Gallup also has gobs and gobs of demographic statistics available to its subscribers. They have a demographic analysis that is perfectly in line with industry standards. They are the best poll in the business. Bar none. In particular, they have a good reputation for filtering unlikely voters out of their samples.

Zogby: Drummond seems to think that Zogby simply fell ass-backward into his present notoriety. Zogby predicted 1996 accurately and came close in 2000. However, his track record is generally lousy. Zogby has a devastating methodological problem -- his polls are not conducted randomly. He uses regional quotas. He is the only one of all these polls (that make any methodology known) that employs this technique. Furthermore, Zogby does not conform his data to industry standards. Rather, he uses -- among other sources -- his previous data. Neither of the standard-setters in the polling industry accept this practice. Zogby also weights his polls to strictly conform with the 2000 partisan turnout results -- a result that BC04 has spent tens of millions of dollars to change. Zogby, a supposedly independent pollster, also called the race for Kerry...IN THE SPRING! Zogby is also one of the chief practitioners of the internet poll, which is really quite unacceptable. Because I have (unfortunately) used Zogby here at this site, I will quote the very estimable Drummand at length (consider it part of my pennance): "All in all, Zogby’s habit of confusing his personal opinion with data-driven conclusions, his admitted practice of manipulating the respondent pool and his demographic weights, by standards not accepted anywhere else, along with mixing Internet polls with telephone interview results, forces me to reject his polls as unacceptable; they simply cannot be verified, and I strongly warn the reader that there is no established benchmark for the Zogby reports, even using previous Zogby polls, because he has changed his practices from his own history."


No comments: