Best Of: Are We Really Living In Separate Worlds?

It’s been an incredibly divisive year, and we’re constantly told we’re more politically divided than ever. But, as our team takes some time with their families for the holidays, we want to re-share a more hopeful conversation with you that sheds some new light on these seemingly unbridgeable divides in our country.

We hope you enjoy it, and we’ll be bringing you brand new episodes after the holiday.

Transcript:

Anthony Fowler:

Hello, listeners, and Happy Holidays. 2020 has been an incredibly divisive year, and we're constantly told that we're more politically divided than ever. But as we take a short holiday break, we want to reshare a more hopeful conversation with you that shed some new light on these seemingly unbridgeable divides in our country. We hope you enjoy, and we'll be bringing you brand new episodes after the holiday. Thank you for listening

William Howell:

I am Will Howell.

Wioletta Dziuda:

I'm Wiola Dziuda.

Anthony Fowler:

I'm Anthony Fowler. And this is Not Another Politics podcast.

William Howell:

So, we've got to dive headlong into concerns about polarization, no? These vast differences between Democrats and Republicans. There's a notion that they live in separate worlds and they perceive very different facts about the state of the world such that there isn't any commonality that can bridge differences even by reference to basic facts. We could certainly imagine this being true today, no, when you think about evaluations of the effect that COVID-19 is having on the country.

Wioletta Dziuda:

And I think we have some early evidence that this is indeed happening. So for example, I just saw this poll yesterday, of course I'm cooped up in my office, I browse the internet a lot. So people were asked whether they think that COVID is not worse than seasonal flu. And surprisingly about 20% of Democrats said yes, it's not worse than seasonal flow, which was a little bit strange to me, but I think what's even more glaring is that 56% of Republicans said so. There seem to be a huge gap in beliefs about how deadly the virus is between Democrats and Republicans.

Anthony Fowler:

And you can see the concerns. I mean somebody might conduct that poll and write a headline like, "Partisanship is literally killing us." I mean Republicans don't seem to realize that this is a serious problem and they're not social distancing and they're not washing their hands and all of that and they're going to die because of partisan misinformation. They somehow think this isn't serious as it really is.

Wioletta Dziuda:

So it seems to me that there are two questions. One is why do people report holding very different beliefs about something that's actually relatively easy to check? And the second question is what are the consequences of that?

Anthony Fowler:

All right, so Will, you talk to somebody who has an interesting study on this question that can tell us whether or not partisanship really is killing us or not, or at least give us some evidence that's relevant to that debate.

William Howell:

Sure. I talked to Greg Huber who is a terrific scholar at Yale University. He's also the chair of the political science department there and he did a series of experiments that are trying to evaluate just how much Democrats and Republicans really do disagree about factual matters and what consequences they have for American politics. So why don't we give that a listen?

Greg Huber:

I would say the beginning motivation for this is an understanding of what we do when we ask people surveys. And so when we ask people a survey question, the first thing we know is that people will answer even if they don't actually have an opinion on things, we know people will answer a survey. The second thing that, at least we intuited at the very beginning is that when people answer a survey question, they're not just answering the question you ask them. They're answering the question they want to answer, they're trying to express an opinion often. So, one can think of an analogy where you go to a restaurant and you have a wonderful meal and the service is terrible. And the next day I call you and say how was the food? And you say it was terrible. And you're not telling me what you thought of the food. You are telling me what you thought of the service.

So, one way to think about public opinion data, public opinion data for example, about how the economy is doing is people aren't telling you what they think of the economy, they're telling you what they think of the president when they're answering that item. I think the third thing people are doing when they answer questions that relates to the notion that people will answer a question just because you asked it, they're taking a guess and often they're taking a wild guess but a guess without any consequences. So if I'm answering a survey item and the survey item gives me a chance to say something I'd like to say about my party or about a candidate, and I don't really know the truth, there's really not much of a cost to me of just saying the thing that I think is most expressively desirable about my side.

So, one of the things we did in this study is we tried to create in one condition, an objective benchmark, some force that would cause people to literally lay on a table the cards as they thought they really were. And we did that by giving people money to get answers right. What's the unemployment rate? A Democrat might say under a Democratic president, "Oh, it's 2%." And a Republican might say, "Oh, it's 10%." And the truth might be something like 4% or 5% so they're both going to their corners, defending their president. But now let's say I say something like, "Well, if you get it right, I'll give you a dollar." Well, what happens then? And the first thing we find is that there's a fair bit of convergence between the two parties towards a common answer, not always the truth, but away from those partisan corners when we put money on the table.

And why that's important is that all we're doing there is adding a benefit to being accurate, but we think that a lot of partisan responding may not mask knowing partisan responding but may in fact mask knowing ignorance. And people may understand that they don't actually know what the right answer is, that people would say almost anything because they didn't know what the truth was. And so if I give you money to get it right and you don't have any reason to have a good idea about what the truth is, you won't generally answer any differently because you're literally guessing.

And so, the other intervention that we did was to pay people just like they used to use in the old-fashioned SAT to say, "I don't know." You got a point for getting it right. We'll give you a dollar if you get the answer correct, we'll give you 10 cents if you say, "I don't know." And what was shocking is that more than half of survey respondents on basic questions about the economy said, "I don't know." So, if the partisans differed in their beliefs about the truth, we wouldn't think putting money on the table would have any real effect on the answers that they would give.

William Howell:

And so these results in some ways are encouraging. That is they're encouraging in that if what you worry about our state of democracy is imperiled because Democrats or Republicans don't just disagree about policy, it's that they understand the world to be radically different. Even when you point them towards facts, wherein if they are objectively evaluating those facts, they should offer common responses.

Greg Huber:

I think that's right. At core it suggests that people walk around with something like two sets of books in their head. They have a set of books that when they're cheering on their team, they express those attitudes and opinions, but if they were actually going to make consequential decisions about their lives, about their investments, about high stakes things, they would behave differently. And I find that encouraging. I've often thought one of the more interesting things to do would be to ask whether or not Democrats and Republicans really react differently, radically differently, in their stock portfolio investment after election outcomes.

And I think the reason we've never seen any evidence of that is that they don't. Even though highly educated wealthy Democrats probably think Republicans are worse for the economy than at least in survey data they don't change how they behave in a relatively sophisticated investor fashion when an election outcome goes against them.

William Howell:

Right, after spending months talking about how everything's going to go to hell in a hand basket, if the opposition assumes the mantle of the presidency, they don't then pull all their money out of the stock market.

Greg Huber:

Exactly. And in this vein, I can talk at length about the wonders of the mid-80s Chicago Cubs, my baseball team of my youth, and about how Bobby Dernier was the best center fielder of that decade. And I wouldn't put money on it, but I really feel that way and I'm happy to talk about it. I know what I'm doing. I'm expressing my affinity for the mid-80s Chicago Cubs, but it's not something where I really believe deep inside that's the truth. I understand that distinction. I would say that the other side of this is actually quite deeply pessimistic. And that is that we are very lazy in how we use survey data. We assume that if we ask someone a question, the answer that they give us reflects their actual factual beliefs about the world.

So, if you want to write an explosive journal article or an explosive newspaper headline, you can ask questions like, "Did the president do X or Y?" And partisans will pretty reliably end up in their own corners. And you can say, "Look, we're all coming apart. We can't even agree on these basic facts." But I don't think we should take survey data so seriously. And so that requires us to be more deeply skeptical.

William Howell:

Is that what you're doing, taking it less seriously? Or is what you're trying to do is take it more seriously and parse it, as to say a variety of things are being communicated in the answer that you've offered. There is learning, there is something meaningful in recognizing that the responses differ to factual questions among the two parties. But the thing to conclude is not that they in fact see different worlds. It's that there are expressive benefits associated with cheering on your preferred politician.

Greg Huber:

I would say that's exactly how I think of it. And when I say less seriously, I would say that people don't take their survey answers as seriously as we as survey researchers do. They understand what they're doing, that it's often expressive and yet we have a tendency to worry that the world is falling apart because people say crazy things on surveys.

William Howell:

So let me ask you this. In thinking about whether or not the world is falling apart and as on the basis of the answers that are offered, might there be some instances where this performative quality that's being propelled by these expressive benefits might actually be deeply problematic. And that we should nonetheless worry about the state of American discourse and people talking past one another in ways that can be really costly. And we ought to have an even dimmer view because people are saying things about the state of economy that they know are not in fact true when pressed, but then nonetheless feel free to come out in very public forums, not to root for your favorite Cubs player from the 80s but to talk about things of real lasting import.

Greg Huber:

I think partially it depends on what you want to use survey data to do. So if I want to predict who someone is going to vote for, it may be the case that their assessment of whether or not they endorse a particular view about how the economy is doing is powerfully predictive and it might be powerfully predictive for the very reasons that you just articulate, which is those people who are willing to stand up and say things that make their team look good are on average going to vote for their team no matter what. And that would be a concern if what you thought was that in a massive electorate, you'd want people to be looking at the state of the real economy and responding to it. And I think that's a very real concern.

William Howell:

It isn't just the action that will follow. It's that their willingness to say it unto itself is potentially costly to the extent that you worry about the capacity of us to engage in open and serious minded deliberation.

Greg Huber:

Well, I guess the question is when you talk with your friends and family or at work with people who may differ from you, politically not in the academic setting, but in other work settings, do people behave more like they behave in surveys or do they behave more like they behave when they run into people at little league practice where they try to be civil and try to talk about things in real terms?

And I think of as a survey as a one-sided expression, it's the least deliberative forum that you can imagine. Work on deliberation, much of which is done by political scientists, often finds that when you put people in settings that encourage deliberation, it takes place. The other thing I wanted to mention is that as political scientists, we often use surveys not just to predict elections, but to understand them. And so for example, we say, well, people voted this way because of their views about the economy but if I know that my view of the economy reflects who I want to vote for, then it's not correct to say that I voted for my candidate because of their view of the economy if my view of the economy is a statement about who I want to vote for. It's circular.

William Howell:

Exactly, exactly. There's not much learning about the status of facts and how they inform one's willingness to vote one way or another.

Greg Huber:

We will see that in the months and weeks after every election. And it's not just academics. The day of an election, there are exit polls and you see a correspondence between, "Well, do you think X and who did you vote for?" And that's all the commentary that takes place. Much of it before the election outcome has been announced that is used to interpret the election outcome. So, partially the narrower view here is that when academics are using survey data to make claims about cause and effect and journalists do it also, we should perhaps be more cautious in pushing on the survey data a little bit more to ask whether or not it's real.

Anthony Fowler:

Well, I'll just start by saying I like all the studies that we discuss on this podcast. I really like this study. This is a really creative, innovative, clever thing to have done. Political science and social science in general, lots of fields have been doing big surveys for many, many decades and we have been putting so much stock in those survey results.

And Greg points out it through his study and his coauthors point out that maybe we shouldn't put as much faith in these surveys as we really do, which does challenge a lot of our existing survey research where that's essentially the format. We ask people a political question in a long political survey. They have no stakes whatsoever. There's no incentive for them to give a true or false belief. And probably there are a lot of people who actively misreport their beliefs because why not? And there are probably lots of other people who really don't know but they give a guess anyway and they present it as if it's a confident guess.

William Howell:

Nothing more reliably makes Anthony go cross side than claims that voters are dumb and irrational like that. We can count on that. If you feel like he's having a good day and you want to ruin it, you just go out and say that and this is a point in favor of offering a corrective to that kind of view. I have continued concerns though about what it says about the state of our American politics for a slightly different reason. I don't really care what voters believe in their heart of hearts in the same way, I don't really care what Trump's true emotional state is or what he thinks facts really are. What I care is what they actually do express what they actually do say. And so—

Anthony Fowler:

Do you care about what they say in a survey, Will, or—

William Howell:

I don't care about per se what they say in a survey. That was quite a line, but to the extent that what they say in a survey is emblematic of what they would say in public discourse generally. I mean, I'm not rooting for the industry of survey design to grow because I care about the people who work in this space, but I think that to the extent that it encapsulates not the content of what people truly believe, but it encapsulates what they're likely to express in public and maybe downstream how they're likely to act, then we're on to something important and so I do think it's consequential that Democrats and Republicans in the context of surveys as I suspect they would in the context of conversations that they would have in public and with their friends say things that would deviate from what they know to be the truth if they're pressed.

Wioletta Dziuda:

And why are you worried about that? Why should we be worried about people having strong circular discussions over the dinner table that are not based in their true beliefs?

William Howell:

Doesn't that question answer itself? I mean, don't we want to have a public discourse that says people are expressing what they truly believe, what they know to be the truth and that they are engaging with one another in ways that are responsible and they're not just cheerleading. That cheerleading is really problematic. If we're going to have hard debates about how we meet things, what do we do about this pandemic, right? There are all kinds of really hard questions to be had. And if what we observe is Republicans in public are going to say, "It's not such a big deal." And Democrats in public are going to say, "It's really, really, really awful." And they start talking past one another even when they know when they're incentivized to express the truth, that's something to worry about.

Wioletta Dziuda:

So, I think I have two things to say about it. So first of all, yes, I think we should worry if we think about this 48% who actually say they don't know. So, suppose you don't know and I know, but instead of telling you the truth, I'm going to just argue for my favorite candidate. So, I'm going to distort whether or what my beliefs are. I agree with you that might have bad consequences down the road. People might actually end up having the wrong factual beliefs, but on the positive side it seems that it took the authors very little to sway people from providing those cheerleading answers towards providing something that's a little bit more consistent with the truth. This was like $1.50, 0.17 cents. So, these are really small amounts. I mean, I think you're right to worry, but I think there's still this big question that that's not answered by this work to what extent this is going on in the privacy of our homes—

William Howell:

But it's in the privacy of our own homes as well that the expressive benefits are going to be much, much larger. The expressive benefits of talking to some anonymous person on a survey or get called. And I get to tell that person, you know, I think that the facts conform with the views of my favorite candidate are tiny, right? They've got to be tiny. So yes, you can displace them by offering somebody 0.50 cents. But when I'm talking to who I really care about and I have to ...If I'm talking to a fellow Democrat and I say, "The truth is that not everybody's going to die from the pandemic and that Trump has something and therefore we ought to think that we ought to be opening up the economy sooner rather than later." Those expressive costs of pushing back against the partisan cheerleading that may go on in our politics are pretty, pretty big. And it's going to cost a whole lot more than 0.50 cents to get me to deviate in that instance.

Wioletta Dziuda:

So here I would push back a little bit on this interpretation that the author's pursuing in their papers. So, they say the reason why people say things that they know are not factually true is because of this cheerleading, they get some excitement and happiness from just saying good things that imply good things for the politician that they like, but I think there's a host of other explanations. It might be that I'm just lazy. You asked me what was the unemployment during Obama, I have no idea, you're paying me to take the survey, I just want that click some answer. I fall back on the things that just seemed more natural to me. I like Obama, I'm going to say low unemployment. Or I might be making inferences, it may be that I truly don't know what unemployment is. Even if I think about this hard enough, but I know Obama was a good president according to me, so I'm just guessing.

You are forcing me to guess in a sense. You didn't give me, I don't know option, so I'm going to say unemployment went one down. You know, it might be that there's a different interpretation of the question. You're truly asking me what the unemployment was, but I'm interpreting the question as you asking me, do I think Obama was a good president? You're just trying to figure out whether they like Obama. So I'm going to tell you unemployment went down. So I think there's a host of explanations for the findings that they have in the paper, depending on which one you believe could make you more or less optimistic about the concern that you have.

William Howell:

Anthony, what do you think? Do the findings in this paper uniformly speak well about whether we should become more optimistic about the state of American politics or are there lingering concerns?

Anthony Fowler:

Well, I do come away more optimistic as a result of this paper in the sense that it might seem troubling when you see so many people convey factual opinions that seem to be misguided. When so many Republicans say they don't believe Barack Obama was born in the United States despite all the compelling evidence suggesting otherwise.

And so, you see these kinds of things and that seems to be concerning. And this paper suggest that actually this might be just a polling artifact, essentially. This might be an artifact of a political poll that is not an actual genuine phenomenon that is widespread in the population, it's not as if Republicans and Democrats are systematically believing the wrong things because of little bits of misinformation out there. But they say the wrong thing in the survey because they're having fun and there's no incentive to say anything else. So I come away slightly reassured, but I do take your point seriously, which is that if people know in their heart of hearts that, say, Obama was born in the United States, but nevertheless they go about their lives behaving as if they think that he wasn't at the Thanksgiving dinner table and at the rally, and they behave as if we have an imposter president who's not constitutionally, you know, that kind of thing, that that would be troubling for American politics.

And I haven't seen any compelling evidence that suggests that these kinds of survey reports correspond in really meaningful ways to behaviors in the real world that really matter. And I guess all of that seems to be kind of a damnation of survey research, which is maybe we should stop studying people survey responses as the object of interest and we should start thinking about what other kinds of important behaviors in the world we really care about and we should be studying those. So you know, we should be studying people showing up at protests and what do people say at town hall meetings and how do people vote and how do they donate and when did they sign petitions and stuff like that. And we should decide whether or not we're concerned about this kind of misinformation based on that, not based on the survey questions.

William Howell:

Yeah, it's an open question. How the views that are communicated in the context of surveys translate into observed actions. Let's just put a question mark by that. A range of views though that are in play right now. I think Greg's view is that there is learning to happen from survey research. There are reasons to do it, but what we're learning is something different from what we think we're learning. What we're observing is in large part, cheerleading and cheerleading may be consequential and it may be an endemic feature of politics, but what we're not learning is what we think we're learning, which is what people truly believe about a set of factual questions.

Anthony, your point is that maybe we're not learning anything at all, or maybe it's just an artifact and that we ought to shut down this enterprise altogether and reallocate our scarce research resources towards watching what people actually do and what people actually say in public spaces. Do I hear you right?

Anthony Fowler:

That's one argument. I could be persuaded that way. I'll get in trouble for saying this with some of our friends and colleagues, I could be persuaded in that direction. Or I could imagine the argument that survey research is good for certain kinds of things and not other kinds of things.

Wioletta Dziuda:

Gives different answers.

Anthony Fowler:

Yeah, I mean, certainly, I think this is one of those times for survey researchers to take a step back when they read a paper like this, they should say, "Okay, we've been asking all these factual questions and surveys, we've been making all these grand claims, why are we doing this? What do we really care about?" And one response to that is, "Let's go measure some behaviors that we really care about." Another response is, "Let's try to improve the survey instruments so that they do a better job of capturing what we really care about. Maybe we should start and maybe even as a standard course, we should start paying people to give factual responses, surveys or something like that."

So I don't want to throw away the entire enterprise of survey research, but I would say I think the field would do well to stop and say, is this new study that I'm about to write, is this more about how people answer survey questions or is this more about politics? And I think this kind of paper is very useful for pushing the field in a good direction in that regard.

William Howell:

So what does this mean for this present moment, so Wiola, you noted these factual differences vis-a-vis Covid-19 and this pandemic, differences in how Democrats or Republicans are responding to just how dangerous the disease is and I suspect it would be hard to collect a bunch of data that would document market differences between Democrats and Republicans on the toll that it's having both in terms of American's health and on the economy. What should we take away from this paper and what do you take away just independently from those kinds of differences?

Wioletta Dziuda:

Yes, of course the worry one would have seeing this kind of survey response is that if Republicans, if majority of Republicans, think that this virus is not worse than a seasonal flu, then that's going to translate in their behavior or they are not going to practice social distancing, they're going to practice lower social hygiene, wash hands less frequently, wear masks less frequently. And if that's true, that actually is huge. That has enormous consequences for the health outcomes, not only for Republicans but also for the entire society. So yes, I think it's really important to pause and think, do we believe that this translates into factual beliefs and into actual behavior. And I don't know actually how to answer this question to myself. We've seen some papers coming up recently trying to get at this question, but unfortunately and I think it just reflects the limits that we have as researchers, unfortunately those are also survey-based analysis.

So for example, there's a paper by Shanna Kushner, Sarah Wallis Goodman and Thomas Papinski that asks people, do you wash hands more frequently? Are you practicing social distancing? Are you worried that your loved ones are going to be affected? Are yourself quarantining as a result? And there seem to be differences that somehow seemed to be consistent with this first difference in answering the question of how bad the virus is. It seems that Democrats wash their hands more frequently and they are more likely to self-quarantine themselves, but again this are survey responses and perhaps people just respond to be consistent with what they responded to the first question, do you believe this is a bad virus or not? And that perhaps reflects not their true beliefs but their sort of support for the current administration and for the approach that the current administration is trying to project.

So I think Greg's paper makes me more optimistic about those are to a large extent artifacts of surveys and those reflect some chill reading and not actual behavior by people, but I still don't know actually what the truth is. And here I think I completely agree with Anthony that it would be amazing if we could get out and see to what extent these beliefs translate into behavior. And there's some anecdotal evidence you read those stories about people in Idaho who are trying to organize large groups of people to gather together at the same time just to say, "Look, we want to show to you that we don't believe this virus is so bad." So I don't know at the end.

William Howell:

It's all uncertain. It’s just question marks all the way down.

Wioletta Dziuda:

This seems to be my full answer.

Anthony Fowler:

I mean I agree with it. I guess uncertainty in the sense that it's hard to know for sure when you see a study like this. So, this study by Shanna Kushner and the coauthors, it's an interesting, provocative study, but you do worry based on evidence like the evidence in Greg's paper that in fact the differences that they're observing, and surveys are at least inflated, perhaps.

You could imagine that Democrats know they're supposed to give, that they're told to give, which is of course I'm washing my hands a lot and I'm social distancing and I'm not going out and so on. And maybe Republicans who are even doing some of these things don't want to admit it or something and maybe they, maybe they want to push in the other direction and say, Oh, it's not that big of a deal. I don't have to wash my hands. When in reality everyone's washing their hands a little bit more. No one's washing their hands non-stop.

Wioletta Dziuda:

Or not, probably, no one is washing their hands. But I want to go back to something that we have said earlier, to his worry that if people make statements that they know are factually not true in service, then they perhaps make them also at home. And that actually has negative consequences on people's knowledge down the road. So I think here is a very good example of where this concern could be real. If I actually say I'm not washing my hands and I'm not social distancing even though I do, but I say that in the survey because I want to express support for Donald Trump who says that this is not such a big threat and we shouldn't actually unnecessarily wear masks because we feel we are not sick, so why should we? So, if that's the reason why I'm saying that, that's still really bad.

Like I might be washing my hands myself, but I'm affecting people's views on what should be done, they might actually take my opinion based on the truth on some information that I have and that can have very bad negative consequences for our health. So I think, I was disagreeing with Will earlier, but here I think there's a danger in having this effect in the survey that perhaps the fact that people are willing to say things that are not factually true will have their negative consequences down the road.

Anthony Fowler:

So, there's the question of do these survey responses spill over into other areas of our lives?

Wioletta Dziuda:

And even if they don't, they are published or we read them. And then, we might say, "Oh, you know, half of the people actually say that we shouldn't be wearing masks. So, perhaps they know something that I don't know." And I'm more likely to, to now not to wear a mask.

Anthony Fowler:

It could be. Let's hope people aren't getting their public health advice from polls, but I guess to the extent that's true that’s problematic.

Wioletta Dziuda:

Where do we go that would probably count for advice?

William Howell:

No, but I don't think it's like getting public health advice from polls. It's that when we see the images of all kinds of people on beaches in the aftermath and in interacting with one another and then flippantly saying, "If I get Covid, so be it, I'll be fine." Like, we kind of respond to that viscerally, which then can exacerbate societal divides between different populations, just as it may have implications for our willingness to do things like go to the grocery store and our willingness to go to restaurants and it may have behavioral consequences.

If I believe that there are all these kinds of people who aren't taking this seriously, who aren't doing the things that they're supposed to do, that may have meaningful implications for how I understand the world and how I engage it. No? I think that's not because they're foolish, that they think that the survey responses are more meaningful than formal CDC reports about the state of the pandemic, but it is about us trying to understand each other and our understandings of one another have implications for how we engage politics and in this case, how we conduct ourselves on our private lives.

So, look, we've touched on a bunch of different issues and some of which have to do with understanding the findings on offer money of which have to do with trying to assign meaning to them. Having read this paper, where at the end of the day do you come out?

Anthony Fowler:

I would say that I'm largely convinced by the study that differences in factual beliefs between partisans are overstated. That surveys overstate the extent to which partisans really disagree with each other on factual questions.

And for the most part that reassures me that things are not as bad as we previously thought. People do learn things from the news and people have a rough sense of when the economy's doing well and when things are going badly in a war and when a pandemic is going badly and so forth. And when the responses are good or bad and those things probably do affect people's votes. But when it comes time to answer a survey, they give the partisan answer, which is not actually their genuine belief. And so that reassures me, but as Will suggested, I think there are lots of other interesting normative questions that we should continue to ask ourselves going forward. But I'm going to leave with some optimism about the state of public opinion and Democrats and Republicans in the US.

Wioletta Dziuda:

So, I agree with Anthony. I think the paper convincingly shows that the differences between Republicans and Democrats in factual knowledge are not as large as just some surveys have us believe and that's good. No matter what the consequences of that are for personal behavior, voting behavior, and so on, which still remains to be studied and has been studied in various settings. I think this finding itself is extremely reassuring that despite of all this polarized media, we seem to know the same things.

William Howell:

And I guess I would say I'm not reassured. I think that this paper is a really important corrective and it's a really smart corrective to what we think we may have been learning from surveys about what people truly believe about a set of factual matters that have clear political relevance, but I'm not reassured that our politics are healthier than they might otherwise be. In part because I don't know what the relationship is between the kind of expressive value that is being documented in surveys, the extent to which that reflects our larger politics.

I worry that it reflects a lot of features of our politics wherein there's a lot of social sanctioning. There is a lot of expressive benefits to be had from interacting with people whose views about yourself you worry a whole lot about. And then, I also worry a bit that if everybody has a shared set of factual understandings about the state of the world and yet they are so willing to divide politically that then what does that say about this capacity of facts to discipline our thinking as a factual matter because I don't think we're as a result of this survey updating our views about the extent of polarization between Democrats and Republicans.

Those disagreements persist and they're profound. And so I don't know. I guess I'm not reassured at the end of the day that our politics is healthier than I thought, but I do buy this really important, I believe this really important point, which is that we haven't been learning the things that we think we've been learning from the surveys.

Anthony Fowler:

Will, would you recommend that we go out and advertise the results of this study? Do you think that would actually help some of our public discourse? So, when people say, can you believe all the crazy things people are writing on 4Chan or something like that, would it be helpful to the American public to know that some of those things might not actually be people's beliefs, but they're just having fun with us? I mean in some sense, maybe we had a sense of that already, but would it be helpful? Would that actually in and of itself be a valuable thing to inject into public discourse?

William Howell:

I think it's a great idea. Here’s what they should do. I think it'd be really interesting to say, "Look, dear Democrats, you think that Republicans are all crazy, but they actually know more than you think that they know." And likewise, "Dear Republicans, all the Democrats aren't just unhinged, they too have a set of true factual beliefs that converge with yours." And to see whether or not... Yeah, we can see what sort of second order effects that intervention has.

Wioletta Dziuda:

That might actually backfire because if you tell me that my Republican friends are arguing certain positions, not because they truly believe that this position is are right, but for some other reasons I might actually become even more partisan. So, my first reaction was, yes, let's recommend this, but my second reaction is, okay, let's study that first, yes.

William Howell:

Thanks for listening to Not Another Politics podcast.

Wioletta Dziuda:

Our show is a podcast from the Harris School of Public Policy and is produced by Matt Hodapp. Thanks for listening.