J. Ann Selzer

Jun 14, 2019  | 27 min  | Ep 4640 | Podcast



An unprecedented field of more than 20 candidates plus shifting rules for the Iowa Caucuses creates an uncertain world for political pollsters. We sit down with Iowa's Ann Selzer on this edition of Iowa Press.


Funding for Iowa Press was provided by Friends, the Iowa Public Television Foundation. The Associated General Contractors of Iowa, the public's partner in building Iowa's highway, bridge and municipal utility infrastructure. I'm a dad. I am a mom. I'm a kid. I'm a kid at heart. I'm a banker. I'm an Iowa banker. No matter who you are, there is an Iowa banker who is ready to help you get where you want to go. Iowa bankers, allowing you to discover the genuine difference of Iowa banks.     


For decades Iowa Press has brought you politicians and newsmakers from across Iowa and beyond. Now celebrating more than 40 years of broadcast excellence on statewide Iowa Public Television, this is the Friday, June 14 edition of Iowa Press. Here is David Yepsen.

Yepsen: It's difficult to quantify the shifting landscape of political polling in recent years. Increased mobile phone use, a lower percentage of calls being answered and plenty of skepticism about polling in the general public. And that is before the 2020 democratic presidential field grew to more than 20 candidates and Iowa Caucus planners laid out new rules. Well, to sift through the current status of this field is Pollster Ann Selzer and she joins us now at the Iowa Press table. Ann, welcome back.

Selzer: It's always a pleasure to be here, David.

Yepsen: Good to see you again.

Selzer: Thank you.

Yepsen: Across the table, James Lynch writes for the Gazette in Cedar Rapids and Kay Henderson is News Director for Radio Iowa.

Henderson: Ann, let's walk through the poll. And for the benefit of our viewers, you don't have to take notes at home, we have a graphic to show you these numbers. But it appears there is a top tier. Who is there and what does that tell us about this race?

Selzer: Well, one of our goals was to see if we could in fact segment this field because it had been a rosy picture for Joe Biden in our last poll and every other piece of data was good news for Bernie Sanders and that was it. Then there was the rest of the field. So we now have a legitimate top tier with four candidates getting into double digits and that is Joe Biden, again, Bernie Sanders, again, but they are joined by Elizabeth Warren and newcomer Pete Buttigieg.

Henderson: And Harris is sort of in the mix?

Selzer: Harris is sort of in the mix and that is where we embedded some new questions to, again, give us some help. We wanted to take a look not only traditionally their first choice and second choice, but who are people actively considering. And I'm sure you know when you're out covering these events people talk about who is on their list, so who are the seven, plus or minus two, candidates that people are actively considering. And Kamala Harris who gets 7% as a first choice, I think that number is right, and does better in terms of having good second choice numbers and when you add in her actively considering, more than half of likely in person caucus goers say that she is on their list. That's a very strong showing.

Henderson: So we knew Biden and Sanders had support here heading in based on the previous data. You mentioned Warren and Buttigieg. How significant was their move in terms of history here, in terms of the Iowa Poll history?

Selzer: Well, as I said, Joe Biden strong still, the top of the list. What I think is important to understand before we get too far down the road is that underneath the surface Joe Biden's lead is a bit shaky. That is, we wanted to take a look at enthusiasm for people's first choice candidate and he had a lower proportion of his supporters describing themselves as extremely enthusiastic. Then when we put the other three of the top tier combined and took a look it was a difference of 29% for Joe Biden, 43% for the other group. So there is an enthusiasm deficit. Bernie Sanders also has something that we have known to take a look at in terms of strength of support. He is more popular with people who say they will probably attend the caucus rather that definitely attend the caucus. We always kind of look at that other group as well maybe they'll attend and maybe not. You don't want stronger support with softer caucus goers. So that's there. Elizabeth Warren's numbers are really quite good. When we look at that footprint, that is first choice, second choice or actively considering, she is even with Joe Biden. So she makes up for his lead through these other measurements. And then Pete Buttigieg just sort of arrived and he has got a very strong favorability rating and that is art of what put him in that top tier.

Yepsen: James?

Lynch: The Democratic National Committee will be hosting debates later this month and to determine eligibility they are using polling data to see which candidates will get in. And given how early we are in the process, given the margin of error in some of the polls, is that a fair measuring stick to use to decide who to include in this process?

Selzer: I understand why they want to do it. As a pollster I don't like it because the difference between somebody getting slightly less than 1%, so maybe it's .4% and somebody getting .6%, there's a lot of error there and somebody's potential future rests on whether they appear on that debate stage or not and I don't like to be part of that.

Lynch: Are there, is polling valid though to say if somebody doesn't have 1%, I think somebody off the street might say, why even waste time on them? I think what you're saying is that maybe the difference between 1% and less than 1% is pretty minute.

Selzer: Well, it's a very lower bar for one. What I'm happy about is they added some other measures. So you had to have a fundraising minimum, you had to have some other things, the Iowa Poll also looks at whether you have visited Iowa. I think it's proper for there to be parameters so that they're not flooded with people who just want to be on the debate stage. The last time it was polls alone, which again, I have just an innate discomfort for taking that tack. But there's got to be some way when you have this many candidates to say, well, who are the people who have a legitimate chance and are taking it seriously?

Yepsen: And despite all the criticisms of polls, this one struck me as being a good one because you had all the candidates and the race has been going on for a while. You also made adjustments in your methodology for this virtual caucus where people are going to participate. So is it fair, participate by remote computer or phone, is it fair to say this is a good baseline poll for what is about to come in the next few months?

Selzer: Well, that's exactly what we set out to do, David, which is as soon as we read the rules, and they're not yet final but the state party seems confident that this is the way they're going to go forward, I and the poll committee and our partners at CNN put all of our heads together because I didn't want to be the only one worrying about this because it is a significant wrinkle in the way things are going to go. And I wanted an early gauge of what it was going to be like on our last poll when we're not going to have much time between when we finish interviewing and when we're releasing data and to see what is this virtual caucus, how many people think that would be the way they would want to participate. So we answered some questions and we certainly had a good trial heat, if you will, trial run of how that's going to go.

Yepsen: I want to go back to Joe Biden. How firm is his "lead"? I look at that and people are talking about he's a frontrunner, 24%, and to me that says 75% of the likely democratic caucus goers want somebody else. Is that really all that strong?

Selzer: Well, I think that's my takeaway is that while Joe Biden is nominally the frontrunner, if I were looking at the full details of this poll I would be concerned that he's got a lot of shoring up to do with Iowa likely caucus goers.

Henderson: If you look at some of the past Iowa Polls that have been taken at about this time of this cycle, I believe John Edwards was in the lead in 2007, he wound up finishing second. Scott Walker on the republican side was leading at this point in 2015 and by September he wasn't even a candidate. What sort of pile of grains of salt should we be using when we look at these numbers given the fluidity that we see in Iowa historically?

Selzer: Well, Kay, as you well know these candidates are about to spend many, many weeks and months and millions of dollars to change these numbers. So as I commonly say when people say, well do you expect these numbers to hold, if I expected these numbers to hold I would stop polling. The whole idea of a race is that there are things that change with it. We want to take a measurement and see where things are now. Many things have happened since our March poll and this poll reflects significant changes. But imagine trying to cover this field of candidates without a poll that would guide you into who is getting traction and who is not. And that is one of the other aims that we had for this poll was to really kind of show who is the bottom of this list? Who is failing to get any sort of recognition still in Iowa?

Henderson: Have you ever polled and had a candidate get zero before? Bill de Blasio and the Mayor of a Florida community got zero.

Selzer: Literally zero people. But here's the thing, they have almost two dozen options here. There's only a hundred percentage points to go around. So it's hard in that big a field. It's realistic for people to get literally zero.

Yepsen: James?

Lynch: So is this really a four person race, not a 24 person race when you look at it? You talked about the top tier of candidates and who is at the bottom. So should we be ignoring 20 of these people?

Selzer: Well, I Think the four we've talked about plus I think you have to take a look at Kamala Harris, again because of the size of her footprint matching Joe Biden's, that's an accomplishment. It's just not showing up in first place votes. I think you have to look at Beto O'Rourke and Cory Booker who are the only other candidates where a majority say that they feel favorable toward the candidate. So that's step one is for people to know you and like you. So they are poised to be there and they could be the candidates that came from out of nowhere. In previous caucuses we've had so much experience with candidates starting slow and then having a fast burn at the end.

Lynch: But isn't it true that generally the people who are leading in the early stages of the race are still there in the end and the people at 1%, has anybody ever risen from 1% to a caucus winner?

Selzer: The first time we polled ahead of 2016 Bernie Sanders got 3% and he nearly won on caucus night. It was less than one delegate equivalent. So that is my reminder and that chart is emblazoned in my head, that people can come from nowhere.

Lynch: So there's hope for the one percenters?

Selzer: Yes, let's say it that way, well said.

Lynch: Or should some of these people be dropping out at this point?

Selzer: Well, another aim of this poll was to see if the likely Iowa caucus goers were satisfied to have this many candidates or if there was any sort of wish for some people to drop out. 77% say they wish at least several would drop out and about 1 in 4 said they wish most would drop out. Now most is a dozen candidates. So I think there sort of is a wish that the field would start winnowing sooner than it normally does.

Yepsen: I want to go back to the list of candidates and talk about Mayor Pete, Pete Buttigieg. First openly gay candidates for the presidency, would be our first openly gay president. Does that seem to bother caucus goers? Is it a plus or a minus for him?

Selzer: We asked about a series of traits, is that more of an advantage or more of a disadvantage? They certainly weren't seeing it as an advantage and more were seeing it as a disadvantage, but many were saying it doesn't really matter. And I'm going to say something that I can't make the link in the data directly but people, we had a finding that about 2 in 3 say it's more important to them to have a candidate with a strong chance to defeat Donald Trump more than that they share their stands on particular issues. So that says electability. But we gave them a set of seven traits you look for in a candidate, tell us one or two that's important, because we figured electability would be the first thing they would say. So we gave them, okay we wanted to see what their second option would be, electability in fact came in fourth. First came in integrity and second came in intelligence, third was leadership. And so they want to win, they want to win, but this is their preferred arsenal is integrity and intelligence and that may be what Pete Buttigieg sort of is exuding in a way that other candidates could emulate perhaps and in a way that is getting some traction there.

Yepsen: What about Biden's age? Is there a conflict here between people who think he is too old and people who are looking for a winner?

Selzer: Right. Well, we took a look, again, we asked whether having a candidate who would be over the age of 70 is an advantage or disadvantage and there were a substantial number who say that's a disadvantage. There was not a single Biden supporter who said it would be an advantage. So you might sort of in theory say they've got the most wisdom, gained from the most experience, that's an advantage. Not one of Biden's supporters was willing to say that.

Lynch: So, this is sort of a process question, but as a pollster how do you poll for 24 options? And how do you keep people on the phone long enough to run through that list of candidates? Do they hang up before you get halfway through?

Selzer: We have a special relationship here between the Iowa Poll and caucus goers and so once they're willing to participate in the poll we do not get significant drop-offs. We do introduce all of the candidates through what we call the feeling thermometer, rate your feelings toward each of these candidates, and then we give them the first choice and the second choice and then we go through again to ask them about whether they are actively considering. That is two-thirds of the poll. And so we were able this time because we weren't polling on the republican side to devote some resources to getting more questions. But keep in mind, these people are interested in what is going on and what other people think. And so when you've got a population invested in learning what we're asking about they tend to hang in there.

Lynch: Does that skew the results that they are so interested and they're willing to hang in there, that that might not be reflective of all Iowa democrats?

Selzer: We don't mean this to be reflective of all Iowa democrats. We mean it to be reflective of the people who are going to show up on caucus night. And so they, we have a screening procedure that qualifies them as that. And there will be things that change the shape of the race and so some people who thought they were going to go won't and some people who didn't think they were going to go will and hopefully we've laid down this benchmark now to track that and see changes in that going forward.

Henderson: You mentioned that you asked a question about age, you asked a question about abortion and some of the other issues that are roiling through the electorate. Did any of those responses surprise you because the age thing seemed to have some sort of interesting cross tabs, if you will.

Selzer: I think, we asked about key issues and whether you considered them a must have in order to support the candidate. There were three where a majority said they were must have and one was a woman's right to choose abortion and nearly as strong was climate change and seeing it as the greatest threat to humanity. It was a strongly worded position and yet very strong there. And then the third was restoring the ban on assault style weapons, that those three were there. And some of the other issues that you hear talked about a lot, Medicare for all, the Green New Deal, others didn't hit that mark. So I do think there was a fair amount of learning about issues.

Henderson: You, though, asked about age and you have three candidates above the age of 70 if my calculations are correct. And then you got an interesting result, right?

Selzer: Help me out, Kay. Where are you --

Henderson: So you had people say age is kind of an issue for me and then you had an equal number of people say but I like the 70 year olds, right?

Selzer: We had a strong showing by three 70 year olds. I think the question is sort of saying is there some advantage to it or does it give you cause for concern. Any candidate that ends up with the nomination is going heads up against a candidate who is also over the age of 70 so that may sort of level that playing field.

Lynch: Shifting gears real quickly for a moment, we're coming up on the 2020 census and being in the polling business I just wanted to ask you your thoughts about asking a citizenship question, whether that is going to change the outcome and what impact it might have on the outcome?

Selzer: I'm a member of the American Association of Public Opinion Researchers, which is the trade association for pollsters. Our motto is hey, if we want your opinion, we'll ask for it. And that organization has taken a very strong stand about not including the citizenship question and the reason being is we want census data to be as complete as it can possibly be. The Census Bureau ran tests to see what would happen with that question included and with it not included and there's no question that there is a drop off of participation. Our numbers are helped along by comparing them to census data and making adjustments where necessary and if that data is fundamentally flawed we all lose. So that is the positon of the Association, I'm a member in good standing of that association and I support that position.

Henderson: Back to the caucuses, on caucus night democrats will announce delegate equivalents, caucus math, and then they will also announce raw votes. So as a pollster how do you decide the value of the second and third choice question that you make given the fact that when those raw votes come out people are going to say the person who had the most votes won the thing?

Selzer: We've had endless discussions about exactly this. This is the first time in our poll we've ever paid attention to doing something that would look like delegate equivalents. And that is because this virtual caucus is going to be made to look, they're going to represent 10% of the delegate equivalents on caucus night. And we figured out a mathematical way to make that happen with our data.

Henderson: Better you than me.

Selzer: It wasn't obvious, let's just say it took some work. So what it came down to is that the first number out, I'm told, will be the delegate equivalents. And so to sort of not have something that looks like that and to wait and wait and wait and wait, which could be hours before the raw totals are released for the initial preference for the people who caucused on caucus night and for the people who participated virtually, we're going to have numbers that match up with all of this. Well, we're not going to have numbers that match up with the final preference, that's not something we can measure, we can only measure how they tell us they intend to cast their preference.

Yepsen: This is confusing for everybody and I'm sure a lot of our viewers are scratching their head because it's new. First of all, you have this virtual caucus, which is an equivalent to an absentee vote, it's designed to allow more people to participate. People will sign up and then they will go online and vote. And you looked at who those people are. How did they differ from the people who are telling you they're going to be at the caucus?

Selzer: Yeah, that's a very good question. I want to make clear that it's not like an absentee vote in that you don't do it on your own time, you go to a session and it will be run like a caucus. So there will be the message from the party chair, it will be a time commitment, it will just be one that you register for and get locked in.

Yepsen: A lot of people don't understand virtual caucus, they understand the concept of absentee in that you do not have to be at the --

Selzer: That's right, that you do not have to be at a particular place. You do have to be at a certain time. But your question was about --

Yepsen: Differences between virtual caucus goers and --

Selzer: So they are younger, they are less experienced, more than 50% say this would be their first caucus. They are less likely to hold a college degree. So they are demographically different. In their choice for who was their first choice they were more Biden than the people who said they were going to show up on caucus night. So among the caucus night people he got 23%, the virtual people it was 33%. Now, they only count for 10% so when we married them together that turned into the 24% that we mentioned before.

Yepsen: Is releasing this raw body count, just the initial preferences of everybody as they walk in, which has not been done before, is the party running the risk of really causing confusion here because, first of all, people don't understand it, but secondly, if they release what are the estimates of how many delegates a candidate won, the delegate equivalents, and then come back later and show a different set of numbers in that raw body count, is that going to damage the caucuses?

Selzer: Well, I think it's something that is for them to worry about. It is I think possible that there will be many people claiming bragging rights to having won the Iowa Caucus but using a different number because of all of the ways that they're going to be reporting it out.

Yepsen: We've got less than 30 seconds. How big a caucus turnout will there be?

Selzer: Well, David, I have a crystal ball collection but so far none of them has revealed that to me. I think with this many candidates, with the enthusiasm from the midterms, with the interest in defeating President Trump, I think it will be historically big. How big? I don't know.

Yepsen: Ann, thank you for being with us again, always a pleasure.

Selzer: My pleasure.

Yepsen: And thank you for joining us. We'll be back next week for another edition of Iowa Press at our regular airtimes, 7:30 Friday night, Noon on Sunday and anytime at IPTV.org. So for all of us here at Iowa Public Television, I'm David Yepsen. Thanks for joining us today.


Funding for Iowa Press was provided by Friends, the Iowa Public Television Foundation. The Associated General Contractors of Iowa, the public's partner in building Iowa's highway, bridge and municipal utility infrastructure. I'm a dad. I am a mom. I'm a kid. I'm a kid at heart. I'm a banker. I'm an Iowa banker. No matter who you are, there is an Iowa banker who is ready to help you get where you want to go. Iowa bankers, allowing you to discover the genuine difference of Iowa banks.     

More from this show

Iowa Bankers Association
Associated General Contractors of Iowa