09.18.2012 1

Can you trust the polls?

Marist PollBy Rick Manning — In my Norco High School geometry class, my teacher Mr. Reid, used to teach us how to evaluate equations with the undergirding principle that in geometry there cannot be contradictions between rules.  If two sides of a four sided figure were longer than the others, it could not be a square, and so on.

Over the past months, the public has been barraged by polling that shows the incumbent president slightly in the lead with bigger leads in swing states.

This is not altogether shocking in that Obama’s campaign has spent heavily on television and other advertising during the normally quiet months of summer, while the Romney campaign has seemed content to send me at least one letter asking for money every day.  Now, Romney has a cash-on-hand advantage that should begin to show up on the airwaves and in targeted communications to those few unlucky souls who have been deemed to be undecided or persuadable voters.

But what about these polls?  Are they accurate or are they just more spin?

An interview by radio host Hugh Hewitt with the head of the Marist polling company proved instructive in answering this question.

Marist is polling in Ohio and around the country in something that is released as the Marist/NBC/Wall Street Journal poll.  They are showing Obama with a seven point lead in Ohio, which is a significantly larger gap than other pollsters.

Hewitt, who is from Ohio, points out to the pollster that in 2008, exit polls showed that 8 percent more Democrats came to the polls than Republicans, and Obama won by 7 percent in the battleground state.

Yet, the Marist/NBC/Wall Street Journal poll, which shows Obama with an identical 7 percent lead in 2012, includes 10 percent more Democrat voters than Republican voters to get the result.

For this voter turnout model to be even remotely accurate, you would have to believe that Obama, who couldn’t fill a stadium for his 2012 inaugural speech, is actually even more wildly popular today than he was four years ago.

When confronted with this obvious point, the pollster ducked the question going into an explanation involving voters who only have cellular telephones (a strong Obama vote) that simply revealed that he had dramatically oversampled Democrats in the portion of the poll conducted with people who have land lines.  At the end of the interview, which is linked above, the pollster actually predicts an extraordinary 8 percent turnout advantage for Obama in Ohio.

The Marist/NBC/WSJ poll would just be a sampling outlier, if you did not see the same oversampling phenomena repeated again and again by others.

At the end of the day, pollsters make predictions using a set of assumptions about who is likely going to turn out to vote, and voter intensity is one of the key determiners on those turnout models.

In 2008, Democrats were extremely motivated to elect a young, handsome, African American who offered soaring rhetoric about America’s future.

Today, that man is president, and he speaks to half-full arenas as his message is no longer new, and his promise of lifting America has been stunted by economic realities that he never confronted while playing the Ivory Tower community organizer.

The lofty message has been exchanged for a series of attack ads that pretend that this president hasn’t had a full term to get it right, and that’s why the economy is still in disarray.

These attacks from Obama’s campaign reflect the kind of desperate grasping that one sees when a politician is 10 points down and has nothing to lose.  We don’t see a Reaganesque “Morning in America” type of campaign that looks forward to a hopeful future from Obama. 

Instead, we get Obama surrogate attacks that accuse Romney of everything from pet abuse to murder.  The exact types of over the top attacks that a campaign engages in when their own internal polling shows they are a hairbreadth from losing.

When we watch Obama campaign events in 2012, we see empty union halls and stadium cancellations.  Long gone are the days of eager, adoring eyes of a bunch of naïve college students.

Hardly validation that enthusiasm for re-electing this president is running higher now than it did when he was a fresh, new commodity emerging from Styrofoam carved Roman columns.

If my high school geometry teacher were to look at the Obama campaign and the polling assumptions that voter turnout among the Democratic base will be higher in 2012 than in 2008, I suspect he would shout, “Aha, contradiction” as he used to do in class.

A word to the wise, when it comes to polling that you hear or read about, if it tells you something that directly contradicts what your eyes see, check its underlying assumptions and trust your eyes.

Rick Manning is the Director of Communications for Americans for Limited Government.

Copyright © 2008-2020 Americans for Limited Government