News
Garber Announces Advisory Committee for Harvard Law School Dean Search
News
First Harvard Prize Book in Kosovo Established by Harvard Alumni
News
Ryan Murdock ’25 Remembered as Dedicated Advocate and Caring Friend
News
Harvard Faculty Appeal Temporary Suspensions From Widener Library
News
Man Who Managed Clients for High-End Cambridge Brothel Network Pleads Guilty
With the election less than a week away, pundits, journalists and others are turning to the polls to venture their guess about who will be the next president. But the answer is not so clear.
Polls have been increasingly volatile. And during the same day different polls have often reported drastically different results.
In the crucial state of Florida, for instance, yesterday's polls showed great disparities. The Zogby/Reuters/MSNBC poll had Vice President Al Gore '69 leading by 11 points in the state, while the Los Angeles Times Poll had Gov. George W. Bush ahead by 4 points.
In this past week alone, the national Gallup/CNN/USA Today poll had Bush up by 13 points, down by 1 point, up by 11 points and now up by 3 points.
Experts attribute the disparities to the large number of polling agencies, each of which uses different techniques and voter samples. The basic methods used by the agencies are standard, but there are variables: what questions are asked, when the poll is conducted and how many people are surveyed.
Both David J. Bender, a veteran political analyst and former contributing editor for George, and Frank Newport, editor-in-chief of the Gallup polling agency, say that there are more polls this year than ever before.
While the record number seems to be a logical explanation for the wide discrepancies between polling agencies, it still does not explain the volatility of individual polling firms over short periods of time.
Maxine Isaacs, lecturer in public policy at the Kennedy School of Government (KSG), says pollsters and news agencies have been careless and irresponsible this year in processing poll data.
She says the most serious problem with polling agencies is small sample sizes.
"It is inadequate to say that a small group of citizens can represent the views of the entire population," she says.
What she calls inaccurate polling gives a false impression of dramatic shifts, she says, whereas the voting population is actually relatively stable.
Bender says that today's polls do not accurately reflect the state of mind of the voting public.
"It is a misstatement to say that the volatile polls show that the public itself is that volatile," he says.
But Gallup's Newport defends his profession, saying the volatility is not generated by the polls but by the voters.
"I think the polls do reflect what's going on; we design the polls to do this," he says. "There is a lot of short-term volatility in response to recent events, like debates and speeches."
"I wouldn't be in the polling business if I didn't believe that a randomly selected group of people accurately reflected public opinion," he adds.
But both Bender and Dan Balz, chief political correspondent for The Washington Post, name Gallup as the most widely fluctuating polling source this year. Bender says that other polling agencies, such as Zogby and ABC News, have been relatively stable in their polls.
Thomas E. Patterson, Bradlee professor of government and the press at the KSG, says he thinks the polls are accurate: a representative picture of an electorate that hasn't made up its mind.
"The Republican convention gave Bush a huge boost, and Gore got a boost from the Democratic convention too," he says. "Both candidates have had a lot of trouble with the independent base, and that's where the volatility lies."
Patterson said that the difference between polls is not really volatility but merely a discrepancy in methods of polling.
In addition, he says that polling agencies often admit their results are significantly inaccurate about 5 percent of the time.
"By chance, 1 out of 20 polls will be off the mark, and we make a big deal out of that one poll," he says.
In the end, some ask, does all of this talk of technique and accuracy make a difference?
Balz and Isaacs say that they think polls have little affect on the voting public.
"Only the media and political elite are preoccupied with what's going to happen; voters really aren't interested," Isaacs says.
Newport, however, contends that his line of work does interest the public, whether or not it actually affects how people vote.
Zogby and ABC News could not be reached for comment.
Want to keep up with breaking news? Subscribe to our email newsletter.