Many factors affect voter behavior. at which factor is the cartoonist poking fun?

Kim Mills: Although it is not yet November, the 2020 elections have effectively begun. Election officials are seeing a surge of early in mail and ballots around the country. And lots of other voters will head to the polls on November 3rd to vote in person. Many Americans see this as the most consequential election in recent American history, amid the backdrop of a pandemic, a season of racial unrest, a suddenly opened Supreme Court seat and months of natural disasters fueled by climate change. What will shape voters decisions this year? What are the psychological forces at play when people decide to vote and whom to vote for. Welcome to Speaking of Psychology, the flagship podcast of the American Psychological Association that explores the connections between psychological science and everyday life. I'm Kim Mills.

Mills: Our guest today is Dr. Jon Krosnick, a social psychologist who directs the political psychology research group at Stanford University. He has spent 30 years studying American's political attitudes, what drives them to turn out for elections and how their political attitudes shape their voting choices. In recent years, he's had a particular interest in climate change attitudes, which Americans say is one of the most critical issues we are facing. Welcome to Speaking of Psychology, Dr. Krosnick.

Dr. Jon Krosnick: Thank you so much for having me. It's a pleasure to be here.

Mills: Right now polls are showing Joe Biden ahead of Donald Trump in many battleground states. You've done research on polling and survey methodology. How seriously should voters take these polls right now?

Krosnick: Well, it's a wonderful question and it's one that I'm very glad that the country is interested in. Unfortunately, a memory that many people have about the polls is that in 2016, many people went to bed the night before the election believing from the poll aggregators FiveThirtyEight.com, the Upshot at the New York Times and others that Hillary Clinton had a probability of winning that election somewhere between the high 60s and the 90s. And when in fact she didn't win, of course it was a historic moment for the country, but it was also an important moment for social science because the sense in which the polls are broken and perhaps hopelessly broken was certainly an appealing interpretation to many people.

Krosnick: But I'm happy to say that that is wrong, that the polls... It's important to differentiate two kinds of surveys, the scientific ones and the not so scientific ones. Scientific surveys involve truly random samples of people and that methodology, done with what's called random digit dialing where we generate random numbers to call all possible landlines and cellphones with known probabilities. That method has produced startlingly accurate pre-election predictions for decades. And in fact, that method worked beautifully in 2016. The polls done during the last week before the election, either national polls or state polls, predicted the outcomes for President Trump and for Secretary Clinton with about 1 percentage point of error, just about exactly right.

Krosnick: The unfortunate fact is that there were many, many more non-scientific polls that involved not random samples and instead samples of people who either volunteered to do surveys regularly for money or samples of people who were scooped up by a method called river sampling. So these are people who are doing something on the internet and we, for example, want to read a newspaper article and a box pops up that says before you can read the article you need to answer this question, who are you going to vote for on Tuesday? There's nothing scientific about those samples. There's nothing random about those samples. Unfortunately, companies are selling those samples as if they are scientific, and those are the samples that were way off with errors as big as 16 percentage points off, not surprising given that there was no science behind it.

Krosnick: The answer is that happily I think many news organizations learned that lesson. And for many of us in 2020, we are watching the polls. More often the ones getting high visibility, not always but the ones getting high visibility so far, are the scientific random sample polls. And unfortunately for your listeners the challenge is when you see a poll, how can you decide whether it's one that you should trust or not? And of course I know the survey organizations and so when I see one I know what methods they use, and so I can tell and I would like it if the news media would tell everybody when they report on a survey whether it's a truly random sample or not. And if you care, take a minute and click some links to follow through to find a methodology write up. And if you find the description is a random digit dial by telephone, by human interviewers to landlines and cellphones, that's the critical factor.

Krosnick: And so the good news is by careful sampling we can still learn, and I hope before this election, that there will be not only lots of scientific national polls, but lots of scientific state polls so that we all have a good sense of what to expect when the votes are counted. And as I say, there is a track record that goes back many decades showing we should have strong expectations based on scientific polls.

Mills: So it's not necessarily a brand or a name that the public should be looking for, although that may be part of it, like whether you look at a Gallup poll or National Opinion Research Center out of the University of Chicago, for example. But that really educated consumers should be looking at the methodology, that that's the trick?

Krosnick: Yeah. And luckily there's only a little bit of the methodology to look for. This pretty much differentiates the pros from not. The random sampling, random digit dial, RDD it's called, telephone, human interviews, landline, cellphones. And I can tell you that there are certain organizations. You mentioned Gallup. Gallup stopped doing pre-election polls. I'm not aware that they're doing them now. But the Pew Research Center is doing them. They're very reliable. ABC News is another example of a very reliable organization. And there are others as well. Quinnipiac University is another one that partners, I think, maybe with the Washington Post. That's another one using random digit dial. And there are other others as well.

Mills: You published a review study called Why do people vote? And one thing you found is that door to door canvassing is one of the most effective ways to increase voter turnout whereas phone calls seem to make no difference. Can you talk about that research and how that might be playing out this year?

Krosnick: Yes. Really many people think of Don Green, who is a political scientist at Columbia University, as a really important contributor to our understanding of voter mobilization because for many decades in America, it's been big business at the time of elections for candidates to pay for television advertising, to pay for radio advertising, to mail out postcards and letters to voters in the hopes of influencing their behavior. And yet up until Don started to study all of this, we had very little evidence; and Don himself was one of the first people to test the impact of television advertising of candidates and to show really tiny effects of ads that dissipate so quickly they can't possibly influence election day.

Krosnick: And so, in a sense his message was tremendous amounts of money are being spent on ads that have no real measurable impact. And others as well, Lynn Vavreck at UCLA is another one who has done some work in this area and showed the same thing. It's really very difficult to find any evidence that television advertising, which is quite expensive for candidates, has any consequences. Don then turned to this question of voter mobilization and the logic really comes from the following, that for a long time candidates thought that whether you are going to vote or not is pretty much a fixed attribute of you and it has nothing to do with them. So either you're a voter or you're not. And the real question is, can they convert you to vote for them instead of for their opponent?

Krosnick: And in recent years, the professionals working in this area became much more enlightened by realizing that actually it's quite hard to convince people to move from one side to the other in terms of candidates, but it is much easier to activate a person who probably will vote but might not vote, into actually voting. So imagine that you have kind of a 70% background probability of voting, that giving you a bit of a nudge can make a significant difference. And if before I give you that nudge, I know that if I can nudge you into voting, you'll vote for me, then that's a good thing.

Krosnick: And so the Obama campaign is widely credited in 2008 with really paying attention to Don's research and realizing that it made sense on a scientific and an empirical basis to deploy resources to try to figure out who would vote for Obama if they vote and to figure out who to nudge to have the most impact on them. And what Don's field experiments showed was that mailing out postcards to people has very little if any effect, making phone calls to people has very little if any effect, but having the right conversation on the doorstep actually produced quite measurable effects, increasing turnout. So for psychologists, we ask the question why. What is it that happens during that face to face conversation that causes an increase in turnout?

Krosnick: And in fact, we've known for quite awhile from Tony Greenwald, who's at the University of Washington and his research were way ahead of the curve. He realized that it was possible, during a conversation like that with a person, to simply ask one question; on election day, are you going to vote? And asking that question is enough of a nudge to actually alter behavior. And I'll tell you about the studies that he did. What he did was to randomly assign households that were listed in the state of Washington official government records as being registered to vote. He had undergraduates at the University of Washington, I think, make phone calls to those randomly selected households for which he could get phone numbers. And each household was randomly assigned to get one of the following two quick conversations.

Krosnick: Hi, I am an undergraduate at the University of Washington survey in the psychology department, it's just one question today. Are you going to vote in the election on Tuesday? Or who do you think is going to win the Seattle Seahawks football game on Sunday? And so each person's randomly assigned either to get that vote question or to get the football question. And the evidence that Tony provided from a series of studies is that simply asking that question, are you going to vote, actually increases turnout because there are people who kind of lean in the direction of voting. And when they hear themselves say, "Yeah, I'm going to vote," it actually changes their self concept and creates a little bit more crystallization of a behavioral intention. So that same logic I think makes sense and applies on doorsteps.

Krosnick: And so that turns out to be a technique that was used in 2008 and in subsequent elections. And as you know, 2016 allowed it to happen, 2020 is not so much. The idea of going around and knocking on doors and talking face-to-face with people is unimaginable for most. And so that's one of so many things that makes this a remarkably different direction from what we've been used to. And I think many people who want to make a difference are volunteering to make phone calls and it may make the phone callers feel better to make the phone calls. But at least Don's evidence is that not seeing somebody, not talking to somebody who you can connect with on the doorstep and answering that question freely, it at least in his studies didn't work. But we should remember in Tony Greenwald's studies, it did work to call people on the phone and simply ask that question. So, maybe we shouldn't completely abandon that notion in 2020.

Mills: Given that the research shows that advertising and postcards and some of these phone calls are not very effective, why do campaigns continue to spend money on them?

Krosnick: Such a wonderful question.

Mills: Seems logical to me.

Krosnick: I'm sorry to tell you there's so many instances in which I am amazed that business spends money in ways that don't work, that government does things that the public doesn't want it to do. One potential answer to your question is that they simply aren't aware of the research. That the research is being done, it's being published in academic journals. It's certainly known in some circles, but maybe not in their circles. I'll tell you a story though that gave me a little bit of insight in general into this.

Krosnick: I was lucky to be able to collaborate with some folks at the University of Chicago at their public policy school in sponsoring a series of workshops, wonderful workshops, where they brought together political psychologists, political scientists together with campaign professionals. And when we set this up, we thought really on the academic side that we had a lot to learn in this situation. And what we wanted to learn was how professionals whose careers live and die based upon their insights into how voters think could tell us who, let's face it, we're academics, we sit at our desks, we write papers for each other. Our careers don't live and die on the accuracy of our scientific assessments.

Krosnick: We thought for sure there would be a lot that we could learn from them, and that would enhance our scholarship and our literatures and our understanding of voters and elections. And what we were amazed to learn was when we brought these folks together, and essentially what we said was come to Chicago, we'll tell you what we academics think we know, and then you tell us how we're wrong and what you know.

Krosnick: So when we got together and made the presentations, the professionals said, "Thank you so much. We really appreciate all of this insight. We don't think this way, we had no idea that there was such literature. This is really helpful. And frankly, we have no theories of our own." And we said, "Well, how can you not have theory? I mean, how do you know what to do?" And they said, "Well, we just do what we always did." That there are traditions and making ads of certain types, and we just keep right on doing it because we don't see any alternatives." And so when we asked them, "For example, when you make a television ad, do you test it to see the impact that it has the way a scientist would?" "Oh no. Definitely we don't do that." "Why not?" "Oh, well, because we don't have time and we don't have money."

Krosnick: So when we, academics, evaluate the impact of ads, what we find is most of them have no impact. A few of them have impact in the intended way and some of them actually backfire. And so if I were a candidate and I were paying for those ads, I would certainly ask that they be tested, especially because testing isn't all that expensive. And it's of course exactly the bread and butter for psychologists. But it's just not something that's routinely done, but that in many fields people operate in making decisions based more on instinct, I think, than on data when there are actually are scientific data that are reliable and available for them.

Mills: Is there a particular type of ad that is effective?

Krosnick: Well, interesting you should ask. As you know, there has been an attraction in political advertising over the last 30 years to what are called negative or attack ads. The idea is don't spend time promoting your candidate, spend time attacking your opponent and trying to discredit that person. There actually is a logic in psychology in our literature for that, it's called negativity effect. In general, the notion from lots of different types of research is that I can tell you five good things about a person. But if I tell you one bad thing about that person, it influences your impression of the person disproportionately.

Krosnick: And so you might imagine that it makes sense to focus attention on discrediting. The problem is that so many of these attack ads require kind of making up stories that are sort of puffed up to be scary and are presented with what I call shark fin music and dark images and a voice, an ominous sounding voice that I think we've all gotten pretty tired of them.

Mills: The one in Courtney ad, for example.

Krosnick: That is one exactly, one of them. And so the sense in which Americans tune out the ads is one issue. And so our group at Stanford has actually started to pursue an alternative which is to explore respectful, informational, educational ads. What I mean by that is that for example we have looked at the economic track records of various political parties. And so you can imagine that if you were to picture a graph that along the bottom has years from 1940 to the present, and then over on the left hand side is the unemployment rate in America. We looked at the patterns of unemployment up and down over that time period under Democratic presidents and Republican presidents.

Krosnick: What we found was a remarkable pattern whereby under one party, unemployment almost always has gone down and under the other party, unemployment has almost always gone up. And Americans are thought to be very importantly influenced in their voting by economics and they're thought to be influenced by unemployment in particular. And so we did a very simple experiment where we exposed some people to an informational ad that showed them this graph and explained it to them. The treatment group saw the unemployment ad but control group instead saw an ad for some consumer product that was unrelated. And what we found was seeing the ad improved that... Seeing that one ad statistically significantly improved attitudes toward the party that presided over reductions in unemployment, increased confidence in that party's ability not only to manage unemployment but to manage the economy, increased identification with that party. More people called themselves members of that party. And people actually said they were more likely to vote for candidates from that party after seeing the ad.

Krosnick: And so this is an example of a very different kind of ad that I think provides, respectfully for people, factual information. Obviously you now know that in the world we live in, there's so much uncertainty about what actually is factual or not. That presenting this information because the unemployment rate is measured by the US Census Bureau and it's been measured and is on the record under all of those presidents, that the Census Bureau of course was operating under the approval of those presidencies, that it's reasonable to present that as factually correct information. And so by presenting that information to people and heard, what we saw was people became better informed about the economic track records of the parties and that in fact shaped their attitudes and their voting intentions strikingly. So I guess I would say that's an example of something that works.

Mills: Campaigns obviously don't know that.

Krosnick: They don't, because they'd [inaudible 00:20:36]. That's odd.

Mills: Let's change gears for a minute. I want to talk about the influence of fear on voters. Now, one campaign, I guess I should say it's the Trump campaign, is trying to motivate people to do things like vote in order to preserve the suburban way of life to end what they say is left wing violence in some cities, maybe even to vote twice because the system is rigged. The Democrats are saying things like another four years for Trump will lead to the end of the Affordable Care Act, Roe v. Wade, continued unrest. So both of these parties and campaigns are pushing messages of fear. How compelling are such messages as a motivator to get people to vote and to vote for their guy?

Krosnick: A great question. You've of course accurately described what's happening these days. And I think there are two ways to think about the messages that you have described. You might say that a message saying, and in fact Joe Biden did say the other night during the debate that President Trump is seeking to overturn all of Obamacare and that the appointment of a new Supreme Court justice before the end of this term for the president could well provide the needed votes in order to accomplish that. And so for Joe Biden to say that, to say, "Hey America. Listen, just bear in mind, if this justice gets appointed, most likely the Supreme Court will declare Obamacare unconstitutional and the entire program will be eliminated."

Krosnick: You can imagine portraying that as a message meant to induce fear, but you can also think of that very same message as an educational message, right? That many people, I think including President Trump, would say, "Yes, that's exactly right. I do want to overturn Obamacare and I want to rely on the Supreme Court to do that. And increasing the number of Republican leaning justices on the Supreme Court will enhance my ability to accomplish that." So I think the president actually wouldn't disagree with any of that. And the fact that Joe Biden mentioned it you could say is a helpful step in educating voters about the policy choices that they face.

Krosnick: And yet as you say, for people who have healthcare right now through Obamacare and who feel that they have benefited from a program that they might consider affordable, that they might have serious preexisting conditions, they might be worried that their coverage for preexisting conditions would disappear. That for them it is a fear inducing message, it is scary. And what we've known for many decades in psychology is that fear can be a motivator up to a point. That fear can activate people to pay attention and to learn better and to be more motivated to take action.

Krosnick: But fear is not helpful if you don't have a strategy available to you to effectively feel that you're able to reduce that fear and reduce the likelihood of the undesirable outcome happening. And the problem I think right now for America is that as so many people are saying, we're in this unprecedented moment in history when the president of United States has said, "I don't think I can trust the voting process." And when that message comes from elected officials, in addition to the president, who are charged with voting, who are charged with managing the process of mail and ballots being counted in a timely fashion, in a way that normally we would take for granted.

Krosnick: The postal service is kind of irrelevant to this. Of course they do their job and there's nothing to worry about. Now Americans have been told there is something to worry about involving simply the delivery of those mail and ballots. That there are so many aspects of the voting process itself that contributes, I think in the minds of Americans at the moment, to a lack of confidence that they can actually solve this problem. In other words, if they want Obamacare to stay in action, what can they do? Well, the answer is you can vote except is it clear, can you vote? Is your vote actually going to get counted?

Krosnick: I think it's reasonable for many Americans to say, "I don't have the confidence in that now that I had two years ago or four years ago or eight years ago." So unfortunately the solution at hand or what might be proposed as a solution for fear, that is vote. If you want to make sure the president stays the president, vote. If you want to make sure the president's out, vote. And the problem is even thinking about that action induces this greater sense of fear and uncertainty.

Krosnick: We've talked about this election as one that may have the highest levels of turnout that we've seen in America in a long time. And yet it's also possible that if this keeps up for another month, that people may become depressed and inactivated and to feel incapable. And for people in psychology, you might remember the notion of learned helplessness that was demonstrated years ago in laboratory studies. If you put an animal in a situation where they get an electric shock and they learn that they can't avoid it, they just kind of sit down in the corner and take it because they've lost their sense of being able to control and get out of the pain. And there is a very real possibility that learned helplessness may kick in with some Americans and thereby undermine their turnout.

Mills: It would seem at this point that there are very few voters who don't know which of the two presidential candidates they support. Now, I've seen between 3% and 11% undecided right now. Is this an unusually low number of undecideds? And how difficult is it to change people's minds at this point, do you think?

Krosnick: Well, you are accurately describing results of surveys that ask people, have you decided or not? And I'm going to surprise you by telling you I don't think that's a sensible question to ask in a survey because if somebody says, "Yep, I've made up my mind," and then they find out that there's some news that Joe Biden or Donald Trump is caught at home with a knife in the chest of someone who they've just murdered, then of course people are going to rethink their decision. So when somebody says, "I've made up my mind," what they're really saying is, "I can't imagine information that I would get between now and election day that would lead me to change the way I'm leaning at the moment."

Krosnick: So the real issue here is, are they imagining realistically? And my guess is the answer is most of us can't be because we could not have known that the New York Times last week would publish Donald Trump's tax returns and we cannot know various other things that are undoubtedly going to happen between now and election day. So what we would really like to do is to tell people, "Okay, you don't know this yet, but these are the kinds of things that are going to happen between now and election day. For example, there's going to be a debate. During the debate, there's going to be lots of yelling between the candidates. They're going to be interrupting each other and so on."

Krosnick: The viewership was not terribly high for that debate, but the news coverage of it has been substantial and that's just one example of something that I think most people would have said we've never seen this at this level in presidential debates before. And so people guessing about are the debates going to change my mind, well, they're thinking the debates are normal debates the way we're used to.

Krosnick: So, given that people can't know all the events that are going to unfold, we really can't expect them to guess whether they're going to change their minds. And if you'll allow me, this is a wonderful moment to mention other landmark research in psychology done especially by Tim Wilson at the University of Virginia and Dick Nisbett from the University of Michigan back in the '70s actually pulling together a huge literature in psychology showing that when you ask people questions like that, how do you make decisions? Why did you decide to buy this car instead of that car? Why did you marry this person instead of somebody else? That in general we don't have access to the working of our brains enough to be able to accurately describe the answers to those questions. And in fact, instead what we do is we make up stories. We say, "Well, why would somebody marry a person like that? I guess here's a good reason."

Krosnick: So in this case asking people to guess will information, that they can't even anticipate, change their minds is asking them what I call a Nisbett Wilson style introspective question that we're actually not prepared to answer. So, forgive me for saying the premise of your question is exactly right. Those are the kinds of little tiny numbers of undecided voters, at least according to their own claims and surveys. But my guess is that the numbers of undecided voters are considerably higher than that, particularly because I think we know that there is plenty of ambivalence.

Krosnick: There are people who are leaning in particular directions but they are also on both sides of the aisle disappointed with their candidates to some degree. That there are Republicans who say, in the presidential race for example, "I lean toward President Trump, but I wish he would stop doing X, Y, and Z." And others on the Democratic side who say, "I've always been a Democrat but I wish Joe would do X, Y, Z." And so in situations like that, where voters are feeling ambivalence, they are susceptible to being nudged. So I think there is probably more nudgability out there than those surveys might suggest.

Mills: Just to change gears again. I want to talk about some of your other research because I know that in recent years you've been looking at climate change attitudes and you released a survey just the summer, I believe, where you found that voters are increasingly focused on climate change and that the number of voters who say the issue is personally important to them is at an all time high. Can you tell us more about what you found in that survey and what it means?

Krosnick: Absolutely, thank you for asking. My group that's based at Stanford has been doing research since the mid-90s actually tracking American public opinion on climate change. I think we've now done about 24 national surveys over that time period all using those scientific methods we talked about earlier, random digit dialing by human interviewers to landlines and cellphones. When we started the research, to be honest, I started to study this not because I was interested in the topic, but because I was approached by the Electric Power Research Institute asking whether I might be interested in taking a grant to do some survey studies in this area on this topic of global warming. And I said, "What's that?" I actually didn't know what it was.

Krosnick: And so I agreed to do the work and got addicted to the topic because from the very start, from the late 1990s when we were doing national surveys, what we found was that a huge majority of Americans at that point were better informed about the issue than I was and actually believed the earth had been warming over the last hundred years, that it's been caused importantly by human activity, that it's a threat and that it should be dealt with by government. And so though that finding of a large and sometimes huge majorities expressing those points of view were unusual back then in American politics and they're even more unusual in the context of our image of the country as being so polarized, and that just became very interesting to me.

Krosnick: And what I can tell you that we found over the intervening 20 years since those first surveys was amazing stability of these opinions, that Hurricane Katrina happened and Hurricane Sandy happened and Al Gore's two movies came out and wildfires were devastating in parts of the country and so on. Lots of events happened of extreme weather. Hundreds of millions of dollars spent on advertising to try to convince Americans they should be concerned about this issue. And those huge majorities have remained remarkably stable.

Krosnick: And yet when we were gearing up for our survey now in 2020, we actually stretched out the interviewing period for 80 days during the COVID crisis because what we wanted to see was, first of all, were opinions any different now than they had been before? But as the crisis evolved, what would we see in perhaps changes in people's opinions? And the reason this is an interesting hypothesis is because of Abraham Maslow's notion that many will remember from the introductory psychology class being his hierarchy of needs. And what Maslow proposed quite reasonably is kind of a, you can think of a pyramid shape where at the bottom are the most important needs that we all have to satisfy first in order to get through the day.

Krosnick: We need to have a place of shelter and safety. We need to have food to eat. And so those very basic needs of safety and security Maslow said need to get satisfied before any individual has the luxury of worrying about higher order need satisfaction. And so for example, higher order need might be making meaningful friendships with other people. And another one might be promoting the environment, protecting the environment and so on. That worrying about the extinction of a butterfly specie halfway around the world is something that is what economists might call a luxury good, something that you can only afford if the basics of your life are taken care of.

Krosnick: And that's a hypothesis that a number scholars have taken seriously and some have published papers saying this is true and in particular saying that there's a tension between public support for environmental protection and public concern about the economy, as if there is an inevitable trade off there. Either you attach importance to the economy thriving or you attach importance to the environment, but you can't do both. So if that hypothesis is right, this year of 2020, as horrible as that has been in terms of the health crisis and the economic crisis, it does offer us as researchers an opportunity to test this notion. It's really what you can think of as an interrupted time series design where by no fault of our own as the researchers were handed this opportunity to compare measurements of climate change opinions before the crisis to those during and after the crisis.

Krosnick: And I'm thought it was maybe possible that Americans would say, "This is really an unprecedented double whammy. I don't feel safe physically due to the virus. I don't feel safe economically due to the crash of the stock market at the beginning of the crisis and due to the surge of unemployment and the threats to so many small businesses. I've got no mental resources here to worry about the environment." So, it was a great shock to me when we repeated our survey in 2020 and we found absolutely no evidence of Americans turning away from this issue. That the percentages of Americans saying today that the earth has been warming over the last hundred years, that that warming is due to human activity, that government should work aggressively to address that in the future, we see absolutely no evidence of those numbers dropping in any meaningful way.

Krosnick: And the headline that you mentioned a moment ago was in fact the headline that the New York Times chose to emphasize when they covered our survey results in its first release. And the headline that you mentioned and that the Times picked up on initially has to do with what political scientists call the issue public. And so it turns out that we're used to thinking in surveys about the general public in America, and you might think about the voting public in America. But what we've learned is it's also important to think about little groups of people who are passionate about particular policy issues. So there's a gun control issue public, there's an abortion issue public and lots of issues that have been on the radar screen and on public debate for a long time that are successful at attracting the passions of small fractions of Americans.

Krosnick: These are people who wake up every morning, open their eyes, look across the pillow and say, "Good morning gun control. Another day, another opportunity for me to do something about you." So that we really are thinking that way, they're married to the issue. It comes about not easily because marrying an issue means making a big commitment. That commitment tends to last a long time and it tends to not only make folks who are passionate about an issue likely to write letters to express their points of view to elected officials and to publications and on blogs, get to give money to lobbying groups to attend protests. And most importantly, to vote based on the issue. They're listening very carefully to everything that gets said on this issue that they're married to and they learn a lot and they develop strong preferences and they use those preferences to make voting decisions.

Krosnick: And so just as there are issue publics to those other issues, there's a climate change issue public as well. And the big headline is that that climate change issue public has been what I would call a typical magnitude four issue public on many other issues, five to 12 percentage point range. And yet in this instance the issue public has surged to an unprecedented level up to 25%. This is one of the largest issue publics we have ever seen and this is the percent of people in the country who say this issue is extremely important to me personally.

Krosnick: And for most issues, the issue public is divided about equally between people who are sort of on the liberal side of the issue and the conservative side of the issue. So for example, about as many passionate pro-gun controllers as there are anti-gun controllers in the issue public, and what that means is it's a problem for candidates because a candidate who says I'm for strong gun control laws, they will alienate about as many people as they woo.

Krosnick: But on climate change, it's not like that. It's actually remarkable. This is the only issue we've ever seen where more than 90% of the passionate people are on what you might call the green side of the issue. These are people who believe that climate change has been happening, is a threat, and they support government action to address it. And there's a little group of them, less than 10% of them who are passionate and skeptical. And those folks certainly make plenty of noise. They express their points of view in various different ways, but they are vastly outnumbered by the passionate green folks on climate.

Krosnick: And what that means is that when a candidate is trying to decide what to talk about, how to message their constituents, how to present themselves at rallies, what positions to take publicly, climate teams offers a wonderful opportunity because it turns out that all candidates, Republicans, Democrats, anybody else, who chooses to take a green position is going to win more voters than they will lose. And because every issue public is small, the candidates can only assemble a coalition of supporters enough to win an election by pulling together a bunch of different issue publics and taking positions on various issues to attract enough voters to support them.

Krosnick: And here what we've learned is that more than 50 million Americans, that's more than a quarter of the adults in the country, are prepared to vote based on this issue when the candidate expresses an enthusiastic green position. So for Democrats and Republicans, it's an opportunity to gain some votes by talking about it.

Mills: Those are amazing numbers and I wasn't aware that really they skewed in that direction and that in other cases candidates can cancel themselves out by taking a position. So it seem like this would be a no brainer and yet we still find candidates who deny that global warming and climate change exists. It's kind of shocking, they need to read your study clearly.

Krosnick: Well, actually there's a backstory to that that will help [inaudible 00:41:47] and that is that a woman named Coral Davenport who is a writer for the New York Times did an investigative report some years ago to try to understand why it is that about 60% of Democratic candidates tend to take green positions and almost no Republicans talk about climate change. And the argument that she made was from her investigative reporting that the messages delivered to candidates and their campaign staff by fossil fuel lobbyists is, "We'd love to support your campaign, happy to give you money. But if you talk about climate change, we won't be doing that. So feel free to make your decisions about what you want to talk about, but we'll be happy to support you as long as you don't talk about climate change."

Krosnick: And from the point of view of the fossil fuel industry in light of the survey results I've told you, then that is kind of a desirable thing that best not to talk about this issue on which Americans would like to see change. And what's interesting is put yourself in the position of the candidate. A potential contributor to your campaign says, "Okay, you've got a choice here. Either you take my money or you talk about climate change." And you think to yourself, "Okay, I know Krosnick research. I know I could win some votes if I talk about climate change, but maybe if I don't talk about climate change and I take the money from this contributor, I can spend it on more television ads and I can win votes that way."

Mills: So here we come full circle.

Krosnick: Exactly. And so what they didn't do was they didn't read Lynn Vavreck's research and they didn't read Don Greens research that would have told them "No, no. Don't do that. Don't spend the money on advertising."

Mills: Right. Take the votes.

Krosnick: Takes the votes. And so, yes. So there is indeed... Thank you for appreciating the humor that yeah, that's what's going on.

Mills: One last and highly unfair question. What's going to happen on November 3rd?

Krosnick: Oh Lord. Well, certainly I'm the last person that you should pay me when it comes to making predictions. But I'll tell you this, in my opinion. I think to answer a question like that, we have to use a series of considerations. The first one is that high quality scientific polling has shown Joe Biden with healthy leads in many important states for him and neck and neck races in some other important states for him such as Florida. And so in a sense, first of all Biden appears to be winning the popular vote in the country as a whole. That means nothing as far as who ends up in the White House in January.

Krosnick: So what really matters is the state polls and what we need is, as a country I think, as many scientific surveys before election day as possible to create a shared understanding on a scientific basis of what will happen the next day and at least what should happen the next day or the next weeks, depending upon how long it takes to count those votes. Bearing in mind, of course, that what surveys can measure for us is what people say they plan to do, or exit polls that are asking people what did they do, at best can measure what people think they did.

Krosnick: And as you know, there are lots of reasons why a ballot can be invalidated according to the law. So for on paper ballot, for example, if somebody were to check Donald Trump's name at the top of the ballot and then also write by hand Donald Trump's name on the other line at the bottom of a ballot that offers that opportunity, instead of that candidate clearly being endorsed by that voter, that ballot is invalidated because it appears that according to the law the person has voted for two different people. So the surveys can at best measure what people want to do or what they think they did and there may be some slippage between that and a legitimate government count of those ballots according to legal prescriptions.

Krosnick: But still at the same time I think the first consideration to answer your question has to be that the polls at the moment, that I have confidence in, are issuing a signal that at the moment if the election were held now that Biden would have healthy leads in many places and then would be neck and neck in other places that matter and that could end up going in his direction. On the other hand, we know that as I mentioned earlier, other unpredictable events will happen between now and election day and they may well be significant, and I can't know what that's going to be. So that's a reason why I can't make a prediction.

Krosnick: And lastly, as we know in this unprecedented world that we're living in, there is apparently a strong evidence from intelligence community that the Russians were knocking at the doors of dozens of states' election bureaus working to try to get past blockades to access the electronic records of the votes and perhaps change them. As I understand it, the conclusion as of now is that no votes were actually changed as a result of that. But there's certainly a sense in which in some communities the ability to detect and prevent those kinds of electronic intrusions has not necessarily improved over the last four years.

Krosnick: And so that's another element that I can't personally anticipate and that may well be consequential. Of course, remember it need be small. It doesn't have to be a big nudge. It just in a case where the race is close, just a few votes can be pushed one way or another. And HBO in particular has a couple of very overwhelming documentaries for those who are interested in watching them that describe the flaws in the electronic voting technology that are used. And so if you're looking for something else to do tonight, go look for those documentaries, one very new and one older, that will make the hair on the back of your neck stand on end.

Krosnick: And these are all challenges for us as a country at the moment, but they are also challenges for me in trying to answer your question to make a prediction about what's going to happen next month. So the answer is I have no prediction.

Mills: We'll just have to wait.

Krosnick: Yeah, I hope we are all left standing up when it's all over.

Mills: Amen to that. Well, thank you so much for joining us today Dr. Krosnick. It's been really fascinating talking with you.

Krosnick: Well, it's a great pleasure. I thank you and I thank the American Psychological Association for doing your work. We as psychologists toil in the fields but it's events like this that provide us a privilege of being able to disseminate our work and I hope that your listeners have found it valuable and I hope we, as a field of psychology, continue to thrive for many years to come.

Mills: I am sure they will and we will. Thank you. You can find previous episodes of Speaking of Psychology on our website at www.speakingofpsychology.org or wherever you get your podcasts. If you have comments or ideas for future podcasts, email us at , that speakingofpsychology, all one word, @apa.org. Speaking of Psychology is produced by Lea Winerman. Our sound editor is Chris Condayan. Thank you for listening. For the American Psychological Association, I'm Kim Mills.

What are the 3 main factors that affect voter turnout?

What factors affect voting behavior quizlet?

Which three types of factors influence the decisions of voters at the polls quizlet?

What factors that affect voting behavior?