News Articles

Polls, polls, polls

ALEXANDRIA, La. (BP)–“Opinion polls,” someone once said, “measure the public’s satisfaction with its ignorance.” While that assessment might seem somewhat cynical, it is probably not too far from the truth. Which is why I have never understood the media’s obsession with polls as a primary source of news.

Let me draw an analogy to football. I am an avid fan of NCAA Division III football. This level of NCAA football does not offer athletic scholarships. Hence, the players at D-III are student-athletes in the purest sense, which is one of the reasons I like it.

What does NCAA D-III football have to do with opinion polls? Unlike NCAA Division IA — now known as the Football Bowl Subdivision — D-III has a playoff system to determine its national champion. And while there are a couple of D-III “Top 25” polls that are conducted each week, in the grand scheme of things they are meaningless.

Those who participate in the poll are fans of D-III and are very knowledgeable of the teams they vote on. One is even a poll of D-III coaches. I respect the voters’ opinions. However, when it comes down to determining who the best team is, the polls don’t really matter.

When it comes to the current political polls — whether they be on candidates or initiatives — they might have a bit more relevance than a college football poll, but not much.

While political polls are usually in the ballpark when it comes to predicting the outcome of elections, there are historical examples where they got it wrong.

The most famous example of polls gone wrong is the 1948 presidential campaign that featured President Harry Truman, a Democrat, and Republican challenger Thomas E. Dewey. The three major pollsters of the day predicted a Dewey victory. The Crossley and Gallup polls had Dewey winning by five points. The Roper poll predicted Truman would be trounced by 15 points.

Elmo Roper was so confident of a Dewey victory that he had his company cease polling the presidential campaign. “My whole inclination is to predict the election of Thomas E. Dewey by a heavy margin and devote my time and efforts to other things.”

The polls were wrong, very wrong. Not only did Truman defeat Dewey, he did so by more than four percentage points. That represents a nearly 10-point swing from the most conservative poll numbers and nearly 20 points from the Roper poll.

While it is true that polling has become more sophisticated in the 60 years since the Dewey debacle, there are still elements of subjectivity to every poll that can cause them to be less than accurate.

One element that can skew a poll is the very people that are polled. At least some of them change their minds after the polls have been conducted. Sometimes, as in the case of the 1948 election, many of them change their minds.

“The so-called science of poll-taking is not a science at all but mere necromancy. People are unpredictable by nature,” observed American author E.B. White. “And although you can take a nation’s pulse, you can’t be sure that the nation has not just run up a flight of stairs.”

The timing of a poll, the questions that are asked, how a question is asked and even the order in which questions are posed, all have a bearing on the outcome of a particular poll.

Since pollsters don’t always make public the aforementioned details, all polls should be taken with a grain of salt.

“The techniques behind the numbers [in a poll] are a complex blend of probability studies, statistics, and sampling techniques that can easily and ruinously be thrown off target by a seemingly minor change and questionnaires that can be skewed by a word that hints at an expected answer,” wrote Pulitzer Prize winning reporter Walter R. Mears in the forward of “A Journalist’s Guide to Public Opinion Polls.” He added, “In the age of polls, all those sins happen, all the time.”

A poll is nothing more than a snap-shot of what a segment of society, which pollsters hope is representative of the population at large, is thinking at that moment. It is only good at the time it is taken.

Polls properly interpreted can have some value for political candidates. However, the average voter is not privy to the kinds of detailed information that make a poll relevant. And I would put most journalists in the same boat.

“We are all ignorant,” said Will Rogers, “just on different subjects.” When it comes to political polls, the American humorist’s observation is more than accurate — most of us are just ignorant. As such, whether your candidate is ahead or behind in the polls, you would be wise to ignore them.
Kelly Boggs is a weekly columnist for Baptist Press and editor of the Baptist Message (www.baptistmessage.com), newsjournal of the Louisiana Baptist Convention.

    About the Author

  • Kelly Boggs