Playing Hide and Seek With Polls

America loves democracy, where the mandates of government are determined by the consent of the governed. The public's temperature is gauged regularly - even daily - by a media obsessed with polling data on absolutely everything.

But are these media surveys intended to document public opinion - or affect it?

Take the war. At the outset of verbal hostilities, the media's pollsters asked the generic question: Do you favor war in Iraq? Maybe they felt the question too broad, or the number in agreement too high. Whatever the reason, the pollsters felt the need to get more specific. Do you favor war - if the U.S. goes around the United Nations? Do you favor war - even if ground troops are used? As the numbers curved lower, pollsters even asked if the public would still favor war if American ground troops suffered large casualties.

Polls are a good way to measure how the country is absorbing politics; how the country is responding to the statements of our political leaders; how the public reacts to sabre-rattling by our enemies; and what we think of our "allies" at the UN. But the media are not just taking the country's temperature. They attempt to manipulate public opinion by touting results gleaned from sometimes loaded questions. What's even more fascinating is how the media selectively report their own poll results.

In the pre-war buildup, ABC News found it necessary to report, as a formal news story, the polls showing support slipping for the White House. I'll buy that. But if that's true, how does ABC explain its decision not to tell viewers - as a formal news story - when their own poll numbers revealed growing support for the White House case?

On January 21, Peter Jennings reported: "An ABC News/Washington Post poll finds that public support for attacking Iraq has declined somewhat: 57 percent of Americans now support U.S. military action to get rid of Saddam Hussein. It was 62 percent in mid-December, and as high as 78 percent a few months after the 9-11 attacks."

The ABC news judgment: the administration's lost support is news.

A week later, hours before the State of the Union address, Jennings mentioned - mentioned - the number had risen four points back to 61 percent figure on his newscast. But he did not describe it as an increase. And for good measure, he buried it under a stack of more liberal-pleasing numbers: "An ABC News poll for this occasion finds that 64 percent of Americans believe the UN weapons inspectors should be given a few more months to do their jobs, 61 percent support attacking Iraq eventually. But only 44 percent of American support a war if the UN does not approve."

The ABC news judgment: the administration's growing support was not news.

Do you doubt me? The night after the President spoke to Congress, ABC's pollsters found the number favoring war rose to 63 percent, but Peter Jennings only said "we conducted a poll when he was finished and we found that people had not changed their minds in significant numbers."

So, in a nutshell, this is ABC's idea of news judgment: support for war dropping from 62 to 57 percent was worth a story isolating that fact. Rising back to 61 percent was not reported as a jump, and buried in one sentence of a larger story about other matters. Another bump to 63 percent was dismissed as no change.

Poor ABC News. It's hard to imagine how frustrating it must be for their reporters and their Canadian anchorman to report day after day why Bush is moving too rashly, why he isn't sophisticated enough for the French, and why it's wrong to use those gauche words like "evil" to describe the Iraqi regime - and then see the polls go against them anyway. The numbers climbed after the State of the Union, jumped after Colin Powell laid out Iraqi deceit before the UN, and rose as people saw the strangeness of peace marches and live UN deliberations on TV.

By March 10, Jennings reported how a new ABC News poll found that "61 percent believe support from the UN Security Council is not necessary to attack." But he did not explain the number's significance: support had surged 17 points, up from 44 percent in January. The results were even worse for Jennings and his pro-UN cheerleading. ABC's "Poll Vault" Web site report showed the number for those who feel UN authorization is not necessary jumped to an almost-stratospheric 71 percent "if allies participate."

Jennings and his ABC team have crusaded relentlessly against this war, and the public rejected them. Any day now, Jennings may again find the American people "threw a two-year-old temper tantrum." That's how he reacted after the 1994 Republican triumph - the last time democracy let him down.