Wednesday, October 3, 2012

New Arizona Poll Samples (Even) More Heavily From GOP Voters

(UPDATE, 10/4/12, 11:15AM: Correcting High Ground's name and adding comments from High Ground VP Paul Bentz.)

(UPDATE, 10/4/12, 11:40am: Updated to include PPP's response to inquiry that it does NOT weight its polls for party affiliation.)

The new Public Policy Polling ("PPP") Arizona poll might help disprove the recent (national) Republican meme that mainstream polling outfits are overstating President Obama's support because they are under-sampling Republicans.  Or, it might indicate that the GOP teeth-gnashing is having an effect on (some) pollsters.

(UPDATE:  PPP's Tom Jensen told Arizona's Politics that it does not weight its polls for party, indicating that it was not reacting to conservative criticisms of recent polling.)

Of course, the third option is probably more likely: that polling pools often vary in their party affiliations, because pollsters do not control for that because party self-identification is a varying element that often is at odds with how people are actually registered.  (Personal experience confirms this: my mother often told people she was a Democrat, and was very surprised when I told her that she was actually registered as an "R".)

Current registration numbers show that 36% of Arizona voters are registered as Republicans, 30% as Democrats and 33% as something else.  Yet, the new PPP poll has a pool consisting of 46% self-identified "R's", 29% "D's", and 25% "I's/others".

That is not only up from PPP's last Arizona poll three weeks ago (44-36-21), but it is even greater (for Republicans) than the percentages that Republican consultants at High Ground thought would be appropriate, when they bemoaned the national polling pools.  (They thought Republican voters should make up 41-44% of any Arizona polling pool.)

(UPDATE:  Paul Bentz, VP at High Ground, responded to Arizona's Politics inquiry by noting that PPP's poll "is interesting in that it does indeed oversample Republicans. Its not outside the realm
of reason in its sampling, but it emulates the 2010 "tea party" turn out - not the 2008 and 2004 turnout."

He notes that the PPP poll, as well as High Ground's, confirm that the race is very close and that "it will
be critical for someone to grab the momentum in the next week coming up on Early Ballots.")

PPP confirmed to Arizona's Politics that "we do not weight our polls for party."  As PPP does not do that, it is interesting that some 10% of the random pool would seem to identify as Republicans even though they may be registered differently.  That is outside the 4% margin of error, and would seem to indicate that Arizonans are - currently - even more GOP-philic than the registration figures indicate.

We welcome your comments about this post. Or, if you have something unrelated on your mind, please e-mail to info-at-arizonaspolitics-dot-com or call 602-799-7025. Thanks.

No comments: