Articles >

Is polling a matter of statistics or sociology or what?

Published by The Straits Times, Singapore on 2004-10-18

IN Mr John Zogby's mind, there's little question that the electorate in the United States is divided into what he calls "two warring nations," not just one favouring President George W. Bush in the upcoming November 2 presidential election, and the other siding with his Democratic challenger, Senator John F. Kerry of Massachusetts. The increasingly sharpening divisions are over ideology - conservative versus liberal - and over the very core values of the American system, which has long prided itself on tolerance and magnanimity.
"Each side feels that if the other wins, it would be the end of the Republic," Mr Zogby told The Straits during an overnight visit to Singapore sponsored by Reuters. "Whoever the next president is, he is going to have to be, first and foremost, a healer - or at least someone to stops this widening, this deepening of ideological divisions in the US. The ugly thing is, this meanness has spilled over from the halls of Congress to Main Street, USA."
What does it matter what Mr Zogby thinks of the American Republic's predicament? It's not just because he's widely considered to be one of the country's pre-eminent pollsters, whose accuracy was the envy of more established psepsologists; it's more because he's studied and taught history and the culture of societies.
It's not so much the arithmetic behind the statistics in polls but the sociology of respondents that matters, he said. That means it's the personal environment of the respondent that ultimately determines his or her political choices.
"Societies need to have sustained civility in their political discourse," he said. "Remember the election of 1800?"
His interviewer did not.
"Well, let me tell you," the bespectacled Mr Zogby said. "It was a bitter election, and Thomas Jefferson was finally declared the winner. His first exhortation to Congress was, 'We're all Federalists, we're all Republicans [the Democratic Party hadn't been founded then] so let us proceed to bond, to heal our wounds."
That's Mr Zogby's way of saying that President Bush has been "the most divisive president in modern American history."
"Republicans and Democrats aren't talking to each other across the aisle," he said. "In Congress, they used to swim together, play tennis and racquetball, they used to party together - and how. But now? Now they don't even know one another's names. The civility has gone."
It has gone, in his view, because President Bush deliberately chose to ignore the "creative centre" that traditionally energized American politics. From the very start of his term in January 2001, Mr Bush resolved that he would go as far right as the political structure would tolerate. Unfortunately, the very tolerance that has signified magnanimity permitted Mr Bush to take his policies to unprecedented locations rightward.
"The danger of ignoring the creative centre is that you then cannot appeal to both sides in the political and sociological game," Mr Zogby said. "That's why the next president has to be like Thomas Jefferson. The American political system simply cannot take any more of this push to the right."
Mr Zogby - he's clearly flattered when he's called Professor Zogby - belongs to an industry that, as elections and public-opinion sampling go, is relatively new. In their 1988 book, Polls and Surveys: Understanding What They Tell Us, Professors Norman M. Bradburn and Seymour Sudman say that the presidential election of 1936 brought the new "science" of polling to prominence when three independent polls (by Crossley, Gallup, and Elmo Roper) predicted Franklin D. Roosevelt's victory over Alf Landon.

They point out that these polls stood in stark contrast to the prediction of the Literary Digest that Roosevelt would receive only 40.9 percent of the vote. As in its widely publicized reports about public opinion over the preceding decade, the Digest rested its 1936 prediction on a tally of ballots returned from millions that had been mailed out across the country. That Crossley, Gallup, and Roper had the audacity to base predictions on relatively small samples (compared to the 2 million ballots on which the Digest based its claim) were itself newsworthy, according to Mr Bradburn and Mr Sudman.

"But when their projections were borne out by the election returns, the validity of modern polling had been established. The point had been made that the way a sample is drawn is more important than its size. Bias toward the affluent inherent in the lists from which the Digest had drawn names--telephone subscribers and owners of automobiles--could not be offset by large numbers," they say. They also point out that in 1848, the Belgian mathematician, Mr Adolphe Quetelet, advanced the idea of the "average man" by arguing that the concept of the normal distribution of observations around a mean could be applied to analysis of society as well as to the physical world.

Polling has come a long way since then. Globally, it's now a US$4 billion industry annually. It employs nearly 100,000 people and has five distinct
sectors: (1) The Academic and Nonprofit Sector; (2) The Private
Sector; (3) The Mass Media Sector; (4) The Government Sector; and (5)
The In-House Sector. In 1939, Rensis Likert developed polling for the US Department of Agriculture, the first such operation within an agency of government. Polls by Hadley Cantril provided President Franklin D. Roosevelt with valuable information on American public opinion throughout World War II.

The Office of Public Opinion Research he established in 1940 at Princeton University conducted research into the methodology of polling and became a central archive for polling data. In 1941, the National Opinion Research Center, now at the University of Chicago, was established.

In short, everyone polls nowadays - corporations conduct polls in order to determine how to launch new brands or reinvigorate old ones. Politicians poll in order to focus their campaigning funds on specific regions and communities. Companies even poll their own employees to assess corporate morale and productivity.

All this means that Mr Zogby is a pretty busy man - and certainly a wealthy one. His 20-year-old firm, Zogby International, which based in Utica, New York, conducts polls in several dozen countries. Of Arab descent, Mr Zogby is especially interested in Middle East issues, and is currently undertaking a series of polls in Iraq about issues such as the viability of establishing a genuinely democratic state there.
But isn't polling still a somewhat uncertain science after all these years? There's little question that after the 2000 presidential election in the United States - when the key state of Florida was first awarded to Democrat Al Gore, then to his Republican challenger Mr Bush polling took a huge black eye. It suffered a solid blow to the solar plexus this past May when virtually every major poll in India predicted that the ruling United Democratic Alliance led by the right-wing Hindu fundamentalist Bharatiya Janata Party would easily be re-elected.
And, of course, there's the mother of all polling disasters: in 1948 almost every American pollster predicted that President Harry S. Truman would be defeated by Governor Thomas E. Dewey of New York; Mr Truman, of Missouri, won by almost five percentage points.
"Polling can be uncertain only if you rely on statistics alone," is Mr Zogby's riposte. "That's why an effective pollster has to rely on culture, history and sociology. I repeat - polling is the study of human behaviour, not simply a sampling of people's preferences."

That may well explain his success. His big test will come on November 2. In Singapore last Friday, he flatly predicted that Mr Kerry would become the 44th President of the United States.

And what if he's wrong?

Mr Zogby looked at his questioner with some irritation. But it could have been the fatigue on his face after a 24-hour flight from New York.

Pranay Gupte,
Senior Writer and Global-Affairs Columnist

© Copyright 2003 - 2008, - by Fluid Design