Politicians, Voters Rely Too Much on Polls

Polls are overwhelming the political process.

Politicians and the public have come to rely heavily on opinion surveys.  Our belief in them has increased just when we should have become more skeptical.

Two stories in the past few days prove the point.

President Obama has reportedly been uncertain about reacting to the civil war in Syria, because he reads the polls as telling him that the American people are opposed to any more involvement in the Middle East.

And, in Maine, the potential entry of Congressman Mike Michaud into next year’s race for governor has led to a spate of reports citing a poll being used by pundits rushing to handicap the contest.

There’s much wrong with reliance on polls in both cases.

Because of his reluctance to act, Obama has seen the situation in Syria deteriorate to the point that it may pose danger to American interests. Iran seems to be gaining a foothold in Syria, and Lebanese terrorists are now fighting in support of the Syrian government.

When Obama’s “red line” against the use of chemical weapons there was crossed, he delayed acting, making the line look like little more than an empty threat.

Reportedly, former President Bill Clinton finally got on his case, telling him that a leader is elected to do what it right for his country not what the polls say a largely uninformed public thinks.

In other words, it is the leader’s job to shape public opinion by defining policy and explaining it to the people.  Could Obama change poll results if he took action with respect to Syria rather than avoiding the leading role much of the world demands of him?

Meanwhile, by taking no action, well short of sending troops, the United States may be strengthening Iran, a country whose nuclear program could be a threat to the U.S., and terrorist groups that could pose an increased danger to the U.S. and its allies.

Under pressure from Clinton and ally Great Britain, Obama may be willing to move out from the shadow of the polls.  Just how far he will go on his own remains to be seen.

In the Maine election picture, some of the worst defects of polling have already appeared.  A recent poll pitted GOP Gov. Paul LePage against independent Eliot Cutler and an unnamed Democrat.

This poll came a year-and-a-half before the election, with no campaign having taken place, and, almost laughably, with one candidate’s identity unknown.

It was hardly a surprise that the phantom Democrat fared poorly, proving that it is tough to beat somebody with nobody.

It proved only that some people are what’s called “yellow dog Democrats” – people so loyal it is said they would even vote for a yellow dog if it were the party’s nominee.

The poor showing by the political nobody should have no influence on Michaud as he makes up his mind about running.  In fact, he almost certainly has already decided to run.

Polls keep playing a prominent role, because the media likes them.  When polls showed the GOP candidate gaining in the upcoming Massachusetts special election for the U.S. Senate, some news reports excitedly hyped the race as a toss-up.  In fact, all signs show Democrat Edward Markey winning next Tuesday.

While polling can be useful, it needs to be handled with care.  Polls today have enough obvious problems to raise question about how much leaders or the public should rely on them.

Many polls have been shown to have built-in biases.  Some, intentionally or not, often produce results favorable to one party or another.  Of course, a few are intentionally meant to produce the false impression of objectivity while favoring a candidate or cause.

Pollsters face a challenge to reach cell phone users on telephone surveys.  To the degree they miss a part of the population, their accuracy becomes questionable.

Many people, selected randomly, refuse to be interviewed. And some people lie to interviewers.  Both the abstainers and the liars undermine the value of surveys.

Even more important is timing.  Public opinion today is obviously not the same as it will be on Election Day 2014.  In 2010, support for the Democratic candidate for governor was much higher in the polls only a few days before the election than it was when people voted.  The false impression may have affected the election outcome.  

Polls have some value, of course, but these concerns suggest we should accord them less importance.  And leaders could try to influence opinion rather than merely trying to follow what the polls show.