Polling for politics

By Jody Tanji

There's little question that popular media have a disproportionate amount of influence on the slice of the electorate that is lazy, gullible, apathetic, or – wonder of wonders -- still doesn't know which candidate they support. Many of these voters, in deciding whether or not they are going to expend the calories required to actually get off the sofa, pick up their car keys, and get themselves to their polling place on Nov. 2 are looking for justification, any justification, to skip it.

How about, say, a seven-word headline or a 20-second sound bite? Suppose that headline or sound bite was about recent poll results in swing states. Suppose those poll results purported to show that one of the candidates in this historically close race was pulling away from the other by a margin significant enough to allow the members of this voting bloc to say to themselves, "Well, it looks like Candidate X is going to win anyway, so I'm not going to bother."? And that would be it.

But what if those poll results were substantially different from results of all the other polls conducted in the same states, with approximately the same number of participants, over the same time periods, with the same margin for error? And what if this "outlier" poll was conducted by an affiliation of three of the most-recognized, powerful media and public opinion names of our time: CNN/USA Today/Gallup? Wouldn't you wonder about that? And wouldn't you wonder, further, why there's been no real reporting on this in the major metropolitan dailies?

Follow this link for an excellent analysis done by people who still seem to understand what reporting is. No need to be a conspiracy theorist to smell something.

http://www.emergingdemocraticmajorityweblog.com/donkeyrising/

For the polls in question:

http://www.cnn.com/ELECTION/2004/special/polls/index.html

Jody Tanji is a former employee of the Minneapolis Star Tribune now working in business. She is a keen observer of politics and American society.