Frederica Freyberg:

The attorney general’s race, along with the governor’s race, has been carefully tracked these past few months by the Marquette University Law School Poll. This week the results from the final poll were released. Poll director, Charles Franklin, joins us now. Charles, thanks very much for doing so.

Charles Franklin:

Good to be here.

Frederica Freyberg:

I know it’s been a long season for you and you’re working with a little bit of a scratchy throat, but I’m sure we’ll get through this. Let’s go right away to the results of your final poll among likely voters in the governor’s race. And what we see here, as we've read since Wednesday is that Walker sits at 50%, with Burke at 43%, an undecided of 3%. Were you surprised by your own results?

Charles Franklin:

I’m always surprised by the results. I’m not good at guessing what they’ll be. So three weeks ago it was a tie and now to a seven-point Walker lead. I think to put it in perspective, our polling since Labor Day has shown a Walker by three and by five and a tie, and now by seven. But in July and August we had Burke ahead by one and two. So there’s been some variation. But it’s driven by variation in terms of who’s going to turn out to vote. When you look at the registered voter numbers, you got a tighter race.

Frederica Freyberg:

Let’s look at that registered voter number. As you say, it's tighter with Walker at 46% and Burke at 45%. What else is happening other than just the enthusiasm to certainly vote among Republicans?

Charles Franklin:

Right. Well, this time Republicans are reporting that they will vote at the same levels as they did for the recall, which is very high. Democrats are actually up about five points in reported turnout, but independents down a little bit. But there’s been that shifting in enthusiasm. But I think the way that that enthusiasm and the way that groups swing, like independents swing a little more to Walker, or men swinging pretty strongly for Walker, those things come and shift in the likely voter numbers more than they shift in the registered voters. So if there’s ever an example about why turnout matters, that gap between a one-point registered voter and a seven-point likely voter is a good example of that.

Frederica Freyberg:

So what we’re also saying in all of that, though, is that come election day it may not necessarily be a seven-point spread.

Charles Franklin:

You know, our poll is our best estimate of where the vote will be based on the interviews we did last weekend. And so that’s our estimate of where it is. But, you know, the nice thing is, at this point, it’s not in our hands. It’s in the hands of the voters. The voters that go to the polls between now and election day are the people that are going to decide where this race is.

Frederica Freyberg:

How can polling results change voting behavior, especially so close to the election?

Charles Franklin:

Actually, there’s very little evidence that there’s much of a change. When you look at the difference between the polls and the actual outcomes across hundreds or thousands of races nationwide, the errors above are equal to the errors below. And so it doesn’t look like there’s any systemic bias. But you do get rapidly-changing differences. If you look at the last five or six polls in the state, they’re all within a one-point margin. Our poll stands out as different from the last other five or six polls. So listeners should, viewers should take that into account and think about it. But the bottom line is don’t vote because of what our poll says. Vote because of what you want to vote for.

Frederica Freyberg:

Or don’t not vote because of what the poll says.

Charles Franklin:

That, too. Absolutely not.

Frederica Freyberg:

So another number we want to look at is the attorney general’s race. This one, this one, among likely voters remains really tight, it seems to me, 43% for Brad Schimel, 39% for Susan Happ. What’s up with that very large undecided number?

Charles Franklin:

Voters really have not come to know either of these candidates, which is a shame. To their credit, both candidates were willing to do three debates with varying formats. But we are still seeing nearly two-thirds of voters who don’t know either of these candidates’ names, don’t recognize them. So the race remains tight. It’s a dead tie among registered voters, just a four-point Schimel advantage with likelies. But this is a race that I always thought voters would focus on and learn more about, but, again, as of last weekend’s interviews, still enormous lack of knowledge.

Frederica Freyberg:

So does it end up for them that whoever wins the governor’s race probably wins the attorney general’s race, kind of that straight line?

Charles Franklin:

There’s certainly the chance for straight ticket voting. We’ve seen more of that. And when voters don’t know much about the candidates, the one thing they know from the ballot is the name and they know the party. And so that’s an easy rule of thumb to follow. But remember, Jim Doyle won and so did Republican JB Van Hollen. So they don’t have to move in lock step, but they often do.

Frederica Freyberg:

Let’s take a look at your independent voter respondents. Among likely voters, 52% for Walker, 37% for Burke. And then let’s just flip over to the registered voters as well, and that’s closer, 46% for Walker, 40% for Burke. But just this question, and it’s because I’m not a pollster, why are we even looking at registered voter tallies?

Charles Franklin:

Sure. Excuse me. The registered voters give us a sense of what the full electorate believes, and the comparison between registered and likely is exactly showing you what difference turnout makes. So that’s the clear comparison here. If everybody voted, we’d look more like the registered. If just the people right now who say they’re sure they’re going to vote, then we get the likely result. So it’s for that comparison to understand how important turnout is and what difference it makes.

Frederica Freyberg:

As far as the independent voters go, are these voters who genuinely swing their votes between parties and candidates

Charles Franklin:

They are clearly less connected to a party that people that call themselves partisans. They’re also less likely to call themselves liberals or conservatives. They’re more likely to call themselves moderates. But they’re also somewhat less likely to follow politics and less involved in politics. So it’s a mix of things. But it means that they’re not as anchored to a party or a candidate as they might be. They’re more likely to vote based on performance and their views of how the performance of an incumbent governor, for example, looks to them.

Frederica Freyberg:

We just have about half a minute left, so I don’t have time to spin through all the issue graphics and numbers, but you describe gaps in issues among these candidates as enormous. Give us an example.

Charles Franklin:

The really striking thing is that out of about a dozen or more issues, about half of them in the majority leans in a Democratic direction and half lean in a Republican direction. But when we look at Burke voters and what they want on issues and Walker voters and what they want, there’s a minimum of a 30-point gap in preference on that. The Act 10 question is the most clear. There's only a 43% to 50% split on Act 10. But among Walker voters only 9% would restore collective bargaining, among Burke voters 79%. That's the single biggest gap. That’s a 70-point difference between the two coalitions.

Frederica Freyberg:

Very big polarization. Charles Franklin, may you find your voice again, and thank you for all your help over this season.

Charles Franklin:

Thank you. It’s been my pleasure.