Polls, politician, and the Gulf war

National Review, May 13, 1991 by Humphrey Taylor

POLITICIANS who read the polls (and which ones do not?) to help them decide which policies to support, beware. While well-conducted polls are the only reliable way to measure current public opinion, they are unreliable predictors of future opinion for two reasons. First, public opinion sometimes changes dramatically; today's polls measure only what people are thinking now. Second, the public is not good at predicting how it will react to future or hypothetical events.

This lesson has been strongly reinforced by the Gulf War. No doubt many members of Congress who voted, on January 12, against authorization of the war did so for honorable reasons. But some surely did so with one eye on the polls.

Seven different national polling organizations which conducted surveys just before January 15 all reported that less than 50 per cent (between 44 and 49 per cent) favored immediate military action, while a similar or slightly higher proportion favored allowing more time for sanctions to work. Immediately after the air war began, opinion changed dramatically. The polls reported between 68 and 84 per cent of the public approving the decision to go to war, while those who favored waiting had dropped to between 13 and 26 per cent.

Just before the ground war began, public opinion clearly preferred a continuation of the air war to an invasion. For example, a CBS/New York Times poll reported that a massive 79 per cent majority wanted the coalition to "continue mainly bombing from the air," while only 11 per cent favored "beginning the ground war soon." But 11 days later another CBS/New York Times poll reported a huge 75 to 19 per cent majority believed that it was "right to start the ground war" when it was started. Events, not the public's previous opinions, shaped public opinion. In politics, as in much else, nothing succeeds like success.

Longtime students of the polls, here and abroad, will not find this volatility surprising. A very similar pattern of polling data was reported in Britain before, during, and after the Falklands War. Before the first shots were fired, as British forces sailed south, polls showed most people believing that Mrs. Thatcher would not be justified in using force if even very few British lives were lost. In the event, over 250 were killed, including the sailors burnt to death in the floating inferno of the frigate Sheffield. But, far from flinching, public opinion quickly became overwhelmingly supportive of the military option.

Mrs. Thatcher's lack of interest in the polls while the British prepared to recapture the Falklands was totally in character. Not even her worst enemies accused her of pandering to public opinion. For the first eight of her 11 years as prime minister she showed that doing the right thing, or what she saw as the right thing, even if it was initially unpopular, was not just good policy but also good election-winning politics.

Commercial marketing researchers have known for years not to rely on consumers' own expectations of what they will do or think in the future. In the 1950s George Katona, at the University of Michigan, demonstrated that many people who say they will buy something will not do so, while many others who say they won't buy it, will. This is equally true when it comes to politics.

None of this means that politicians should, or could afford to, ignore public opinion. Mrs. Thatcher was finally undone by her stubborn commitment to her very unpopular poll tax. Success in politics--that is, re-election--depends in large measure on how the people perceive your record, on whether or not they think you did right by them. Selling your policies and your record to the public is much easier if you understand what the public knows, thinks, feels, and wants. But the key to re-election is to distinguish between what the public thinks is right today and what it will think when the next election comes around. That can mean persuading the public to change its mind. If politicians are now more aware of this difference--the difference between leadership and followership--that is one more benefit of the Gulf War.

Humphrey Taylor is President of Louis Harris and Associates, Inc.