Are the Pollsters Confused, or Is It the Voters?

September 27, 1992|By EVERETT CARLL LADD

The polls are again in the spotlight and under criticism as the United States enters the stretch run of the 1992 campaign. One especially insistent question: How do we explain a situation in which one survey shows Bill Clinton leading George Bush by 15 percentage points, while another -- completed just a day later by an equally able research organization -- puts the Clinton lead at only 6 points? Such divergent results are in fact common.

What's more, discrepancies in the presidential trial heats are small compared with what one often finds on other less publicized questions. For example, in a survey taken Sept. 9-12. CBS News and the New York Times sought to determine whether voters were really inclined this year to turn against incumbent members of Congress. They asked: If an election were being held today, "would you vote to re-elect your Representative, or would you vote for someone else?" The results were enough to take incumbents' blood pressure down 20 points. Forty-eight percent said they would re-elect, while just 35 percent were inclined to back a challenger.

On the other hand, an NBC News/Wall Street Journal poll of 12-15 Sept. came to the precise opposite conclusion. It asked the same basic question, although with a slight- ly different wording: ". . . do you feel your Representative deserves to be re-elected, or do you think it is time to give a new person a chance?" Only 31 percent -- unprecedentedly low -- would back their own incumbent, while 56 percent would give the challenger a chance.

These two sets of results on the likely fate of congressional incumbents are entirely incompatible, of course. Their existence means polls don't give us a clue as to whether the public's obvious dissatisfaction with the performance of Congress is likely to result in sustained anti-incumbent voting.

Still, this really is a case where we shouldn't shoot the messenger. The "fault," if that's what one wants to call it, lies not in our polls but in ourselves. That is, there's just so much softness, ambivalence and uncertainty out there in the electorate. The chief source of the polls' variability over the summer -- and the main reason we shouldn't lean very heavily on their trial heat numbers as predictors of the November 3 decision -- involves the indecision of many voters.

We have been so conditioned to believe that numbers are real. A race where Clinton leads by 15 points is pretty much over, isn't it? A race where the incumbent president is down just 6 points and closing is a real contest, isn't it? So which is it? We find it hard to accept that it's neither. If polls come up with different numbers, we believe the answer must lie in their contrasting methods. That is, if science yields contradictory results, we must find the source in contrasting scientific approaches.

So, there's been speculation this year that the divergent results may stem in part from the fact that some survey organizations report their findings of Bush-Clinton strength on the responses of all registered voters, while others base their findings on a subset of registered voters who have been determined to be likely to vote.

Forget it. The differences we have been seeing have almost nothing to do with such methodological distinctions.

We need to keep our eye on the ball. It should be an axiom of survey research that one of the hardest things to measure is an opinion which really hasn't formed or a decision not yet reached. A huge slice of the U.S. electorate finds itself unusually cross-pressured this year and hasn't yet begun to resolve its conflicting impulses. For these voters, preferences of the moment are so lightly held that they are easily altered by relatively minor happenings, and may even move for no substantial reason at all.

People not anchored by clear partisan ties or philosophic commitments have always been distinctively "up for grabs" in presidential contests. Over the last several decades, their ranks have steadily expanded, as party loyalties have weakened across much of the electorate. Never before have so many American voters been so unanchored -- and thus so hard to locate reliably in pre-election trial heats.

Into this evolving electoral setting, the specific conditions of the 1992 campaign present pollsters with a special, added challenge. An unusually large segment of the electorate now feels itself tugged in opposite directions. After 12 straight years of Republican control of the presidency, and with the economy stuck in third gear, the impulse is strong to give the other side a go at it. "Time for a change" sentiment is the great threat to Bush's re-election.

Tugging many voters the other way, though, are strong, continuing doubts about the Democratic standard bearer. The "character issue" dogs Mr. Clinton's campaign now six weeks before the balloting much as it did early in the primary season

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.