[lit-ideas] Re: interaction of polls and public opinion

  • From: Eric Yost <eyost1132@xxxxxxxxxxxxx>
  • To: lit-ideas@xxxxxxxxxxxxx
  • Date: Sat, 11 Nov 2006 12:54:07 -0500

>>Does this mean that you're saying that what is traditionally taught about probabilities and stats isn't

relevant to politics?

I was just curious about the effect polling itself has on polls. It doesn't have to be about Iraq. Say you're watching a TV poll that says 30 percent of the US wants to convert to Islam as a way of ending the war on terror. That 30 percent number is encouraging, gives you as an individual permission to hold that view; after all, 30 percent of the polled group believes that.

Anyway they keep polling the same question. Every night there's a new poll on the same question. Over time, the 30 percent number starts to edge upward.

As the polling percentage moves upward, that increasingly legitimizes the minority opinion polled. I mean, they were at 30 percent two months ago, now they're at 43 percent. And every night they compare the results of the poll conducted one or two months ago with the current poll.

And every night the TV shows the poll and the new results, keeping the issue foremost in your mind. You note the change in polled opinion, and it gives you increasing permission to adopt what was once a true minority opinion. Convert to Islam? Maybe it won't be too wretched.

So when pollsters talk about polls having a statistical deviation of such-and-such, they seem to be missing the real impact of polls as shapers of public opinion rather than mirrors of it.

To change your Lit-Ideas settings (subscribe/unsub, vacation on/off,
digest on/off), visit www.andreas.com/faq-lit-ideas.html

Other related posts: