For whom the poll tolls

Did you notice how many commentators, after each of the debates, would end an evaluation of what they had just seen with a caveat like this: "But we will have to wait a few days until the polls determine how the debate actually affected the race between Bush and Gore."

I've never known a presidential campaign when the polls were as much a part of the race itself - when they have been so clearly influential in shaping the media coverage and thus the public perception of what is going on.

Early on, well before the conventions when the polls were finding George W. Bush some 20 points ahead of Al Gore, the stories from the media weighed heavily in favor of Mr. Bush merely by repeating again and again that Bush had an immense lead. There's nothing more favorable than stories that say a candidate just might be unbeatable because he is so far ahead.

This all changed at the Democratic convention. The polls found a Gore upsurge and then a Gore lead. And the press followed with what could only be called "Gore stories" since it depicted day after day the vice president's climb in overtaking Bush and how he did it.

And now Bush is getting the better of the coverage as the polls have shown him drawing up even with Mr. Gore and then going ahead.

Oh, certainly, the media were doing their own research, talking to a lot of people, gaining their own slant on things - or at least trying to do so.

But the media tend merely to corroborate what they are learning from the polls. Indeed, it is often their own pollsters, like those used by The New York Times and The Washington Post, that reporters will rely on as their indicators of how the race is going.

Reporters know their own weakness. They know they have to base their stories on relatively few interviews. So they naturally find a poll (like a recent Washington Post-ABC poll, which had been based on some 1,500 interviews) quite persuasive as they put together their next stories on who is ahead.

Most reporters that I know are very diligent and do a lot of digging. And the best of them have this capability of picking up the "feel" of voter intentions. But I can understand why reporters will turn to the polls for guidance, particularly those they believe to be reliable, produced by experienced and highly regarded pollsters.

I find myself turning again and again to pollster John Zogby for insight on the New York and national races, with no apologies. I'm not out and around the country these days, making my own judgments from interviews. So Mr. Zogby and several other pollsters have been invaluable to me.

But here I am, pumping up the polls when I still have great questions about them.

Polls do intrude on elections. Polls showing a candidate ahead will encourage that candidate's workers and stimulate his supporters to contribute money. Such a poll at the time of the election will also encourage a candidate's backers to get out and vote. On the other hand, the vote will tend to fall off for a candidate who is shown by polls to be behind - particularly if he appears well behind.

Polls also can be wrong. Their interviewers simply may be asking the wrong questions. Or voters may not be revealing how they really feel.

During the Minnesota Democratic presidential primary of 1956, when Adlai Stevenson was supposed to demolish Estes Kefauver - according to the Minnesota poll, which had been in business for years and had never, repeat never, been wrong. Well, guess who won? Senator Kefauver did that sneaky deed in an upset that surprised the nation at the time.

To Minnesota's farmers, Mr. Stevenson had looked and sounded like a city slicker. And the polls somehow hadn't picked that up.

Then there are the dishonest pollsters (probably of yesterday, or so I hope) who weren't above asking the kinds of questions that resulted in answers that would please their clients.

And there are the pollsters who even today will push too hard for definite views from voters on how they will vote and, thereby, persuade uncertain voters to say they are for one candidate or the other when they really aren't.

Oh, those pesky pollsters! We can't live with them - and it seems we can't live without them.

(c) Copyright 2000. The Christian Science Publishing Society