John: Does polling itself affect events? .... If it [the poll] has any impact at all, that impact is tempered by context, what comes before and after it, our opinion, if we have one concerning the polling organization and whether the results touch is one of our hot-button issues, and, of course, the spin put on the results by those who comment on them.
My first thought of this came from Iraq polls. The context of an "ongoing issue" poll, like an Iraq poll, is that it is a feature onto itself. It can't be used as filler the way a poll about, say, women's health can be inserted in a domestic segment and later show up on a network TV "magazine.". It is a poll that is also a news item. "The news is that 56, 47, or 38 percent of Americans support the war." It is opinion as news.
Its context is that it is the weekly poll you're going to see on almost all flavors of cable TV. It'll be shown on NBC, CBS, ABC, MSNBC, CNN, and FOX. It will be discussed on Hardball and Meet the Press. It will appear in a small box on the front page right in the offerings from many newspaper chains. Turn on morning TV to check the weather and you'll see the poll shown full-screen with bar graphics. When your dog-eared copy of The Nation shows up in your mailbox, it'll have that poll figure mentioned in an editorial. Then Newsweek will show up and it'll have the poll embedded as a graphic on the magazine cover.
The poll is more or less the same question asked again and again. It is compared and trended, updated every week. It's its own show, like a highly successful segment of another show that is about to be spun off. News pundits will cite the "ongoing issue" poll without even attributing it to Zogby, Harris, or Josephine Schmoe. It becomes a generic fact.
If polling does alter poll results, one would imagine that this kind of "ongoing issue" poll would be prime to investigate. An ongoing issue poll is never in danger of becoming just white noise. It's the most in-your-face kind of poll result; it's a continuing episode (ongoing poll); and it has wide emotional appeal.
John: The right way to frame the problem is "How much? And by what mechanisms?" The answer may then teach us something we didn't assume at the start.
So maybe find a way first to isolate the process? Some double-blind experiment with the pollsters being polled? Take the political issue out of the poll, but leave something equally emotional in the poll?
Seems that if you couldn't first isolate the process, assuming it exists, you wouldn't be able to discuss how it interacts with the other contexts and spins that you've mentioned.
------------------------------------------------------------------ To change your Lit-Ideas settings (subscribe/unsub, vacation on/off, digest on/off), visit www.andreas.com/faq-lit-ideas.html