Polling season ready to begin … plus or minus margin of error

By Michael Falcone
Daily Bruin Senior Staff

The gap between the winner and loser of the 2000 presidential
election is likely to be one very small number ““ a stark
contrast to the thousands of polling numbers that inundate the
public via the news media during election season.

In the weeks before the November election, a barrage of polls
will bombard the American people, and while many polls claim to
have their finger on the pulse of the electorate, few are actually
as accurate as they say.

According to a statement co-authored by The Gallup
Organization’s current Editor in Chief, Frank Newport, most
Americans think a poll of 1,500 to 2,000 respondents “cannot
represent the views of all Americans.”

But Gallup and a host of other historically reputable polling
organizations say well-conducted, scientific polls are accurate
predictors of public opinion. Even so, a great deal of crucial
information is often left out when news organizations report poll
data, and this leaves some wondering whether the media should be
doing a better job of explaining exactly what all these numbers
mean.

“There is too much preoccupation with who’s ahead
and who’s behind and not enough reporting of trends in the
electorate,” said Michael Traugott, the former president of
the American Association for Public Opinion Research and a
professor at the University of Michigan, Ann Arbor.

A poll’s comprehensiveness depends in part on the detail
and scope of the questions asked as well as the amount of
background information collected about the respondents. The better
a poll is, the better the chance that the kinds of trends which
Traugott is referring to will be brought to light.

One recent example is the apparent gender gap between Al Gore
and George W. Bush supporters, which was reported in a recent
CNN/USA Today/Gallup poll. The poll, which showed a higher
percentage of women favoring Gore and a higher percentage of men
supporting Bush, did receive media coverage, but other demographic
trends may go unreported.

Traugott said it is the so-called “pseudo polls”
that give polling a bad name and he advised voters to beware of
them. These poorly conducted polls include self-selected polls as
well as “push”polls.

The Internet is a perfect breeding ground for self-selected
polls where with a few clicks of a mouse people can submit their
opinion on a particular poll question posted on a Web site and the
site will update their percentages based on the respondents. In
such a poll there is no way to control for bias in the respondents,
and it is possible that the same person can submit a poll response
hundreds of times.

“Push” polls, conducted over the phone, are not
polls at all, but instead a method of trying to sway voters to
support a particular candidate or issue using the guise of polling.
In February the AAPOR called for a nationwide elimination of the
use of push polls, but Traugott said many political campaigns still
use the push poll tactic.

In addition to pseudo-polls, some polling critics argue that
more polls may be contributing to the nationwide problem of low
voter turnout. If a pre-election poll shows one candidate ahead of
another, some segments of the public may believe the election is
already decided and there is no need to vote.

Though Traugott said polling could possibly diminish the value
some citizens attach to their vote, in other instances polls may
actually help increase the number of people that turn out on
election day.

“If the polls show that the race is too close to call, it
may stimulate voter turnout,” he said.

In a publication issued by the National Council on Public Polls,
Sheldon R. Gawiser and G. Evans Witt, the cofounders of the
Associated Press/NBC News Poll, said the real challenge for both
the journalists who report poll data and the public who listens to
them is to ask the right questions about the numbers.

Poll financing, the number and method of selecting interviewees,
those left out of the interview or out of the data reporting, the
way the actual poll or interview was conducted, the kinds of survey
questions that were asked, their wording, and the order in which
they were asked are all valid areas of skepticism when interpreting
poll results according to Gawiser and Witt.

Of the scientific polls out there Gallup is the most well-known,
but polls conducted by major daily newspapers like The New York
Times and Washington Post, often in conjunction with television
networks, also offer an important second or third opinion to
Gallup’s numbers.

Election season and polling season are now nearly synonymous,
and with new poll results coming out every day, the public’s
constant task is to find truth in the numbers.