Get PJ Media on your Apple

The PJ Tatler

Reflections on the Colorado Polls

What to make of a polling company that doesn't release polls.

by
Charlie Martin

Bio

September 11, 2013 - 3:15 pm

People whose memory extends back to 2012 may recall I wrote a good bit about polls then. I noted at the time that the polls necessarily would have to be wrong one way or the other because they both showed very high Republican voter identification, registration, and enthusiasm, and yet showed Obama leading.

Sadly, I guessed wrong about which side of the polls was wrong.  I’m still not sure I understand why, although we have now seen many convictions for voter fraud (and many more reports of apparent voter fraud that were not prosecuted), the clear interference by the Adminsitration with anti-Administration political groups, and of course the repeated assertions in the legacy media that the only possible reason someone might vote against Obama was racism.  I’m led to the essentially untestable hypothesis that the election hinged on a very active voter suppression effort that kept a whole lot of those “very enthusiastic” GOP voters home.

Of course, what goes around comes around (no this isn’t a Buddhism column) and it turns out the Public Policy Polling got their own surprise in the Colorado recall elections:

We did a poll last weekend in Colorado Senate District 3 and found that voters intended to recall Angela Giron by a 12 point margin, 54/42. In a district that Barack Obama won by almost 20 points I figured there was no way that could be right and made a rare decision not to release the poll. It turns out we should have had more faith in our numbers because she was indeed recalled by 12 points.

What’s interesting about our poll is that it didn’t find the gun control measures that drove the recall election to be that unpopular. Expanded background checks for gun buyers had 68/27 support among voters in the district, reflecting the overwhelming popularity for that we’ve found across the country. And voters were evenly divided on the law limiting high capacity ammunition magazines to 15 bullets, with 47% supporting and 47% opposing it. If voters were really making their recall votes based on those two laws, that doesn’t point to recalling Giron by a 12 point margin.

We did find on the poll though that voters in the district had a favorable opinion of the NRA by a 53/33 margin. And I think when you see the final results what that indicates is they just did a good job of turning the election more broadly into do you support gun rights or are you opposed to them. If voters made their decision based on the actual pretty unobtrusive  laws that Giron helped get passed, she likely would have survived. But the NRA won the messaging game and turned it into something bigger than it was- even if that wasn’t true- and Giron paid the price.

But what’s interesting here, really, is two parts of this mea (non) culpa: first of all, the admission that he didn’t release the poll that was showing a big margin for recall — in fact, the recall was at the outside of his margin of error on the high side — because he “didn’t believe the results.” I’ll note, with a certain pleasant schadenfreude, that when I didn’t believe a poll, I said so and said why, and took my lumps later.

It’s a little hard to not observe, however, that the effect of a poll showing the results that far off would very likely have been to suppress the pro-Giron turnout. If you report polls where your side is winning, and suppress polls where your side is losing, one might just wonder whether you are an objective and trustworthy source.

The post doesn’t do a lot to reassure us, either. I mean where he says:

We did find on the poll though that voters in the district had a favorable opinion of the NRA by a 53/33 margin. And I think when you see the final results what that indicates is they just did a good job of turning the election more broadly into do you support gun rights or are you opposed to them. If voters made their decision based on the actual pretty unobtrusive  laws that Giron helped get passed, she likely would have survived. But the NRA won the messaging game and turned it into something bigger than it was- even if that wasn’t true- and Giron paid the price.

This looks to me like flat-out denial. To some extent he’s correct: as Dave Kopel points out, it was about something more than the “pretty unobtrusive” laws (which we in Colorado did find pretty damned intrusive, thank you very much). But at last count, the pro-recall groups were outspent about 8 to 1 by the anti-recall groups. Hell, the NRA was nearly matched by Mike Bloomberg alone, out of his own pockets.

So remember this. The next time a conservative finds a poll hard to credit, it might just be for good reason.

Charlie Martin writes on science, health, culture and technology for PJ Media. Follow his 13 week diet and exercise experiment on Facebook and at PJ Lifestyle

Comments are closed.

All Comments   (1)
All Comments   (1)
Sort: Newest Oldest Top Rated
The curious thing is the pollster cannot distinguish between the talking points about the laws and the actual laws themselves. The focus-grouped, poll-tested, Bloomberg-funded catch phrases -- "expanded background checks" and "limiting high capacity magazines" -- test well, but the actual laws as implemented were unpopular.

Also curious is that they are SIMULTANEOUSLY saying it was entirely their decision not to release the results AND that as a private polling company, they cannot release all the polls they conduct. That's a complete non-sequitor, as if this were a privately funded poll, then it was the commissioner of the poll who would have decided about its release. If it's a self-funded poll, then you have to ask a) was this the best use of company funds, and b) was the decision not to release in the best interests of the company?
45 weeks ago
45 weeks ago Link To Comment
View All