Does that poll really reflect the headline?

Posted by: Jason Zacher on Tuesday, October 21, 2014

It’s October and an election year. So along with the changing leaves and Oktoberfest beer in the supermarkets, we’re also being bombarded with political polls.

Before my Chamber life, I was the Communications Director for the House Republican Caucus in Columbia. My life revolved around polls during election years. Sometimes they were spot on, and sometimes the polls were garbage. Here are a few things I learned:

Look at Polls, not a Poll. One poll means very little. Many people might remember a few outlier polls in 2008 that had Republican John McCain inside the margin of error against then-Senator Barack Obama. Polls just a few days on either side had Obama comfortably ahead. But those single polls sure gained a lot of headlines. If you want to track polls over time, the website Real Clear Politics does a great job. You can even find polling for our governor’s race, our U.S. Senator races here and here, and perhaps our Congressional races. Many pros will cite the “RCP average” as a snapshot of the race.

Polls show a moment in a changing sea. But polls are just that, a snapshot. Take a picture of the ocean. Then go back and look at it the next day. Or the next week. It’ll never look exactly that way again. The same thing goes with a poll. Don’t get too excited, or concerned, about a single poll. A headline, geopolitical event, economic event, or just about anything, can dramatically change a poll from day to day. Good pollsters can even out many of the variables, but remember that if the pollster had called 1,000 different people on the same night the results might be dramatically different.

Be sure the poll is of “likely voters” not “registered voters.” Many polls that are reported in the news are of registered voters. But in a mid-term election like 2014, perhaps fewer than 50 percent of registered voters will actually vote. A likely voter is someone who has an established voting history. A pollster will try to sample likely voters based on someone’s election history, but that doesn’t always mean they’ll vote this year. On the flip side, by not polling people who have never voted, you can miss big demographic trends, such as with younger, pro-President Obama voters in 2008 or TEA Party voters in 2010. Which leads us to…

People lie to pollsters. Sampling a poll is increasingly difficult. Cell phones, the Do Not Call List, and other generational changes mean getting valid phone numbers is hard. They lie about things that are not socially acceptable (“Do you plan to vote this November?”) and about things that could be embarrassing. There is also a distrust of pollsters by some conservatives, who believe it’s the liberal media doing the poll, and older-generation minorities in the South, who vividly remember a day when you just didn’t give someone your opinion over the phone.

Try to find the questions. Many online news articles will link to the text of a poll. Take a look at the questions. And look at the questions that lead up to the “head-to-head” question. If you ask a series of questions about the problems that occurred during Governor Smith’s time in office and then ask for whom they’re voting, people will lean to the opponent. If you lead into the head-to-head question with a series of questions about how great things have been, people will lean toward the governor.

Sometimes, a head-to-head number might be released after a candidate did a large amount of “message testing” – where they look at the response to certain issues. Those can really skew numbers because the respondent just received an equivalent of $1 million in positive TV for the candidate.

The moral of the story is to take a poll with a grain of salt. Remember that many voters literally don’t make their minds up until the final days of a campaign, so positive numbers now can swing dramatically in the final few weeks. That’s why you, as a business leader, need to get involved early and often with the pro-business candidate of your choice. I wrote a lot about that in September: here, here, and here.

As always, if you’re a member of the Upstate Chamber Coalition, feel free to shoot me an email and I’ll be happy to answer any questions you have about public policy and elections.

Comments

0 comments on "Does that poll really reflect the headline?"

Leave a Comment

Back to top