Should the press poll? Information is our job


As Super Tuesday bears down on us like an electoral tsunami of pontificating, pandering, posturing and prognosticating, the pundits have kicked into overdrive.

A piece in The Wall Street Journal this past week, for example, predicted the presidential primary results will likely cause "red faces among pollsters -- again."

The story described a couple of examples of polls that appeared to miss the mark and suggested that a glut of first-time voters and voters who changed their minds at the last minute might be responsible for the muffed calls by many respected and usually spot-on pollsters.

Among the examples listed were two from Nevada, one claiming that polls overestimated the support for John Edwards in the Democratic caucuses here by a factor of four or five and the other suggesting that Mitt Romney's capture of a majority in the GOP straw poll "shredded predictions."

Those comments brought to mind something said by the man who conducts political polls for the Review-Journal.

"Statistics can be like guns -- dangerous if you don't know how to handle them," commented Brad Coker, managing director of Mason-Dixon Polling & Research in Washington, D.C.

What most people failed to consider, including several local columnists and bloggers, when comparing the poll results for statewide voters and the final percentage of Democratic delegates selected to attend county conventions, is that it took two ballots to determine those delegates.

Yes, the Review-Journal's poll taken on the Monday, Tuesday and Wednesday before the Saturday caucus showed Edwards garnering 14 percent of those polled. He picked up only 4 percent of the delegates, at first blush well outside the 5-point margin of error. But in many precincts, Edwards failed to get the required 15 percent support on the first ballot, requiring his supporters to move to another candidate.

In fact, Coker pointed out that a poll conducted by Edison/Mitiofsky Network Entrance Poll of Democrats entering the caucus sites election morning revealed 11 percent supporting Edwards.

As for both Hillary Clinton getting 6 points more than the poll predicted and Barack Obama getting 7 points more, simply divvy up the 11 percent undecided in the poll as well as those who supported Edwards on the first ballot.

Our poll had Romney winning by double-digits over all comers, with 11 percent of Republicans undecided, but we had Ron Paul finishing fifth instead of second. On the other hand, a poll for the Reno Gazette-Journal had John McCain winning in Nevada and Romney in third.

With pollsters, as with journalists, over time the reader will be able to discern who is more trustworthy, honest and accurate. None has a perfect crystal ball.

No matter how accurate or inaccurate polls might be, there are some who argue we should not be doing them at all, that they somehow distort the election process by creating a bandwagon effect in which people either abandon their candidate because his or her poll numbers are lagging or jump on to the bandwagon of the apparent winner.

But Frank Newport, editor in chief of The Gallup Poll, argues in his book "Polling Matters" that polling is nothing more than a virtual town hall meeting. "Polling simply distills the opinions of all of one's neighbors in an accurate fashion -- an expanded version of what the voter would find out by asking neighbors in the community where they stand on the issues of the day. The idea of town meetings wherein people get together to discuss their opinions is essentially as old as America itself."

Even in the media there are some respected journalists who balk at polling, fearing it overly influences the overcome of elections.

Kenneth F. Warren, president of the Warren Poll, in his book "In Defense of Public Opinion Polling," quotes former Philadelphia Inquirer Editor Robert Rosenthal as saying "newspapers have enough problems, let alone having to deal with the problems of being accused of influencing the outcome of political races."

The problem with that line of thinking is that all information -- be it truthful or not, unbiased or not, balanced or not -- influences people's decisions. Newspapers are in the business for providing information that ultimately influences the decisions of readers.

Knowing what your fellow citizens are thinking is a part of that.

Thomas Mitchell is editor of the Review-Journal and writes about the role of the press and access to public information. He may be contacted at 383-0261 or via e-mail at tmitchell@reviewjournal.com.

 

Rules for posting comments

Comments posted below are from readers. In no way do they represent the view of Stephens Media LLC or this newspaper. This is a public forum. Read our guidelines for posting. If you believe that a commenter has not followed these guidelines, please click the FLAG icon next to the comment.