A note about that poll showing the race tied

A recent Emerson College poll showed the Kentucky governor’s race as a tie between Andy Beshear and Daniel Coleman. Immediately, everyone’s hair was on fire.

Well, I’m here to provide a little fire extinguisher.

Polling Kentucky is hard – we all know that. And, it’s gotten harder over the past few years, as more and more people have abandoned their landline, and refuse to answer calls on their cell phones if they don’t know the caller.

And if you want to really analyze the impact and validity of a poll, you can’t stop with the top line results. You have to dig into the cross-tabs to see what the top line is based on. And that’s what I did with this poll.

First of all, let’s look at how the poll was taken. Here’s the various ways Emerson got to 1,000 respondents:

From their website: “Data was collected by contacting a list of landlines via Interactive Voice Response (IVR) and emails provided by Aristotle, cell phones via SMS-to-web, along with an online panel of voters provided by Centiment.”

So, over half the responses came via landline phones. To me, that immediately makes the results less reliable. In today’s world, who has landline phones? Older adults. And how do older adults tend to vote? Conservative.

What ages, then, make up the poll respondents?

Over half the respondents are over 50. Emerson would say that that matches turnout results, and that’s probably correct. But I’m not sure that past turnout is necessarily a good gauge of what is going to happen in this election.

In short, I think the latest Emerson poll is an outlier. I think the makeup of the respondents, with such a large reliance on landline phone respondents, is misleading. And I think Beshear has more support than this poll is showing.

But guess what? We’ll know for sure in just a few days. So forget all the polls, and go vote, no matter what.

--30--