Polls That Make Your Head Explode

Back in the early 60s, the then-Canadian Prime Minister John Diefenbacker (who was a real Prairie populist) was asked what he thought about polls. "Polls are for dogs," was Dief the Chief's terse reply.

These days, we watch polls the way ancient sages consulted chicken bones and goat entrails — we're all just trying to hedge our bets. This year, however, polls have done more to confuse the future outcome than reveal it. Hillary Clinton's totally unexpected victory in New Hampshire is the prime exhibit. Pollsters (who seem to be outnumbered only by Elvis impersonators these days) defend their profession vociferously, but they still don't seem to be any better at clearing away the fog from important issues.

For instance ... there are two major media polls today, one from USAToday/Gallup and one from CBS News/New York Times.

The USAToday/Gallup poll shows Sen. Hillary Clinton with a seven-point lead nationally over her rival Sen. Barack Obama. And the survey shows that Obama has been deeply injured by his relationship with Rev. Jeremiah Wright.

But if you don't like that, then don't worry. The CBS/New York Times poll shows Obama ahead by 12 points nationally and says that he's moving beyond the Wright controversy.


Were these two groups of pollsters surveying the same country? They were done over the same time period — last Thursday through Saturday — with about the same number of respondents. It's like the USAToday/Gallup people just happened to find every white rural American woman over 60 who feels uneasy voting for a black, while the CBS/New York Times poll was only calling people under 45, who live in urban neighborhoods and who have been to college.

I'm sure there is some logical explanation for this that pollsters can provide; maybe this kind of result only happens on a Cinco de Mayo that falls on a Monday in a year with an eight in it. Something, you know, scientific.



Please keep your community civil. All comments must follow the NPR.org Community rules and terms of use, and will be moderated prior to posting. NPR reserves the right to use the comments we receive, in whole or in part, and to use the commenter's name and location, in any medium. See also the Terms of Use, Privacy Policy and Community FAQ.

I take offense with the way you use the words "a black." A black what? I don't think if Barack was white you would phrase it as "a white." It's so Fox Newsy.

Sent by Margaret | 12:40 PM | 5-5-2008

I'm glad you included the reference to "chicken bones and goat entrails", because that is the way I look upon the media's collective dependence on polls.

And as you pointed out, the surrounding circumstances can (and do) influence the results. Sample size, regional differences, time-of-day, all change the outcome.

Sent by Harold | 1:16 PM | 5-5-2008

NPR thanks our sponsors

Become an NPR sponsor

Support comes from