In a now-famous recent exchange on Good Morning America, ABC reporter Martha Raddatz asked Vice President Dick Cheney about the fact that "two-thirds of the American people say [the war in Iraq] is not worth fighting." The vice president said, "So?" and Ms. Raddatz asked, "So you don't care what the American people think?" Mr. Cheney responded that "you cannot be blown off course by the fluctuations in the public opinion polls."
Six days later, Mr. Cheney reasserted that his answer was about "polls," adding, "The point I wanted to make, and I would make again ... is the president of the United States ... can't make decisions based on public opinion polls. ... He cannot make judgments based upon what the polls say."
Irrespective of one's position on the Iraq war, the vice president is correct about the proper use of polls insofar as making policy decisions is concerned. The Declaration of Independence doesn't demand "a decent respect to the opinions of mankind" as measured by polls. There is a difference between what the American people think and what public opinion polls say they think.
To the dubious credit of pollsters and their sponsors, the omnipresence of polls and polling has convinced people that public opinion polling results and public opinion are the same.
How do they differ? Let us count the ways:
1. Polls imply that opinions are neatly formed, clear and without ambiguity. That's because so many people's opinions with so many variations are summarized into the same - often overly simplistic - categories, such as those who "think we should pull out of Iraq immediately."
You cannot get the nuances of opinions through such phrases. What if people who support an immediate pullout were asked, "Should the United States pull out immediately if Gen. David Petraeus argues that al-Qaida in Iraq is being dealt a serious blow that would be reversed if we do so?" Would the results be the same?
2. Polls usually imply that people hold opinions with equal intensity; they do not. Some polls ask how strongly people hold their opinions, but most do not, and even fewer polls publicly report such results. The New Hampshire Democratic primary polling debacle, of which pollster Neil Newhouse said, "We just didn't see it coming," may have been a result of unmeasured, low-intensity supporters of Sen. Barack Obama going at the last minute for Sen. Hillary Clinton. That, of course, may not explain the other Democratic primary polling errors just in the last few months that exceeded the margin of error.
3. Polls are often reported without the history of opinion-changing that preceded the current polling. In 2003, according to ReliablePolitics.com, about 75 percent of the public supported the war in Iraq. A Washington Post-ABC News Poll in December 2004 found that "a strong majority of Americans, 58 percent, support keeping military forces in Iraq until 'civil order is restored.'"
In a representative democracy, we elect people to voice our opinions, and we reject plebiscitary democracy in which all issues are put to a popular vote. We support this sort of representation because the public cannot know all of the issues and nuances of issues, nor can it know the long-term effects of political actions. Equally important, public opinion polls cannot accurately measure the complexities of opinion-holding, including the depth of opinions, the strength of opinions and how immutable they are.
When Vice President Cheney was told the most recent polls show that the public thinks the war "is not worth fighting," he probably should not have responded with the arrogant-sounding, "So?"
It might have been better to articulate the multitude of reasons public opinion polling should not dictate ongoing war policy.
Richard E. Vatz is a professor of rhetoric and communication at Towson University. His e-mail is firstname.lastname@example.org.