Maybe Americans are not “ideologically innocent” anymore, coming closer to matching elites’ issue positions and ideological perspectives. Martin Wattenberg finds that Americans now cite a lot more public policy issues in describing their views of politicians and parties, coherently describing their ideological perspective rather than making general statements about each party’s success or failure. Austin Kozlowski finds that citizens not only quickly match their issue positions to their party, but also increasingly line up their issue positions on economics, social and racial issues, and foreign policy on the liberal or conservative side of the spectrum. The elite messages are getting through, making Americans more polarized.

Studies: “The Changing Nature of Mass Belief Systems” and “Issue Alignment and Partisanship in the American Public.”

Interviews: Martin Wattenberg, University of California, Irvine; Austin Kozlowski, University of Chicago


Grossmann: This week on The Science of Politics, the American public is getting more politically sophisticated. For the Niskanen Center, I’m Matt Grossmann.

Americans have long been considered ideologically innocent by political scientists, failing to match elite issue positions and ideological perspectives when making their political decisions. But new research suggests that is changing fast. Canonical measures of the American public now show more ideological consistency and advanced conceptualization, and that’s all contributing to our polarization. Today I talked to Martin Wattenberg at the University of California Irvine about his American Political Science Association conference paper, The Changing Nature Of Mass Belief Systems. Coding open-ended responses on the American national election studies, he finds that Americans now cite a lot more public policy issues in describing their views rather than making general statements about each party’s success or failure. I also talked to Austin Kozlowski of the University of Chicago about his American sociological association conference paper with James Murphy, Issue Alignment and Partisanship in the American Public. He finds that citizens are not only quickly matching their issue positions to their party, but now increasingly lining up their issue positions on the liberal or conservative side of the spectrum. They both returned to measures introduced by Philip Converse to assess American political sophistication. Wattenberg saw it from the beginning.

Wattenburg: One of the most important pieces of political science research ever done on political behavior found that very few people thought in terms of ideology and policy terms when they evaluated the candidates and the parties in national elections, and that really came out of the 1956 election study done at the University of Michigan. And these questions are very general, they just ask people, what do you like and dislike about the parties and the candidates? And they continued to have been asked every four years since 1956. And more recently the exact responses have been made available for the internet on Excel files. And so I’ve been reading them and comparing them to what findings were in previous years in my own reading when I was at the University of Michigan, these comments, and things have really changed. People are a lot more ideological and a lot more policy oriented than they used to be.

Grossmann: And Kozlowski also returns to the basics to show big change.

Kozlowski:The first studies of alignment were done by Philip Converse in the 1960s. And what Converse found were very weak correlations between opinions across issues and the general public. And he also found that there was very little understanding in the general public of what the differences were between the left and the right. His conclusion from all of this was that the public was generally innocent of ideology in his terms. And what’s kind of intriguing about it is that he thought this was regrettable circumstance. His interpretation was that without ideology the public would be unable to really make sense of the political field and that lacking ideology was an indication of having an incoherent or inconsistent set of opinions. And what we find here is still working within this same basic framework of alignment or constraints, basically two words for the same thing, but in the spike of issue alignment, it seems that people are not so innocent of ideology as they once were. And an interesting point is that we now see this as being problematic in its own sense. So perhaps people know what the difference is between left and right, but they’re congealing so strongly on the left and the right that we’re losing the more creative different configurations of opinions that might diverged from the single spectrum.

Grossmann:Wattenberg was following up on reading through the voter responses where the increase in sophistication was clear.

Wattenburg: Well, I was doing a couple projects at the University of Michigan that involved actually having to go back to these interviews. I didn’t read all of them, but I read a lot of ’52, ’56, ’60, ’64, 68. And then the 1980, all of their [inaudible 00:04:26] were around in file draws and I pulled those out late at night. I wasn’t supposed to. And I just like looking at them and seeing what ordinary people said and it’s a whole different world. I mean, people are just a lot more sophisticated now than they used to be. They used to talk in real generalities, didn’t see the terms liberal and conservative very often. You did see a lot of policy. You saw a lot of, well, he’s the candidate or my party, which you don’t see that much anymore. If somebody does say that, they then go on to elaborate about it and you don’t see the prejudice that you used to.

I mean, I read a lot of 1964 and there was a lot of prejudice that came out right after the Civil Rights Act, you’d be shocked at some of the things that people said. And, in general, I’d say people, and I found that to this day, people are pretty outspoken, they say what they think. But the things that they think now are different. Mostly what I found for say 1980 was things are bad, they can’t get any worse, onto change or this guy could make things worse. And it was what Congress calls nature of the times, the economy’s good, international relations are good that you really don’t see very much of that anymore. I struggled to find any of those.

Grossmann: Kozlowski was trying to merge sociology’s interests in social transformation with political science.

Kozlowski: It seems to me that the cultural divisions that have characterized the US for a long time are increasingly overlaid by these kinds of political divisions. So as a sociologist I’m particularly interested in what political polarization reveals about the social and cultural landscape of contemporary America more so than what it reveals about the particular strategic kinds of actions that are taking place in Washington. So in the end I don’t think we can ignore either one of these aspects of polarization, the political elite side or the underlying broad social transformation side. But I think that this is exactly where a dialogue between political science and sociology can be very productive in drawing together the deep understanding of political institutions with a lot of thinking about culture and thinking about demography and these larger social indications in the general public.

Grossmann:They both say the academic view needs updating. Wattenberg says we’ve missed the rise of policy walks in the American public.

Wattenburg: From my reading in the literature it’s something that people are still hanging onto. But a lot of what they’re using, data, having data that are no more recent than the 2000 election. And then I also looked at, there’s been a lot changes since 2000 and then I re-examined the 2000 data and found that they eliminated or didn’t bother to notice this whole new category of people who don’t mention ideology, but they talk about a lot of policies, at least policies and usually more. And those people are called policy walks, and there’s a lot of them now, that’s about 10% was the electorate, a little higher percentage of voters. And overall, almost half of the people who actually vote now are either what I call concept [idealogs 00:08:00], they think about in terms of the ideology, or they think in terms of multiple policies that fit together. And that’s a very different image than has been seen in the literature until recently, which basically said, “Oh, Congress is right.” At most 15, 20% out voters, maybe 25 would fit in. So I’m pretty much doubling that.

Grossmann: Kozlowski was updating a past study finding it was mainly partisan sorting, but not issue sorting up to 2004. Now it’s both and they’re going up fast.

Kozlowski:My study is building on an article that was published about 10 years ago, so I think it would help to start with a little bit of background. Now, this previous study that was done by Delia Baldessarri and Andrew Gelman traced a couple of different forms of political polarization in the general public between the 1970s and 2004 using nationally representative survey data. Now, the first thing they looked at was called partisan alignment, which is the correlation between how an individual politically identifies themselves, Democrat or Republican, liberal or conservative. The correlation between that identity and the individual’s stances on specific issues like gay marriage or social welfare programs. And what they found is that between the ’70s and the early 2000s this correlation shot up. Basically, you would learn a lot more about a person when they tell you that they’re a Democrat or Republican in the 2000s than you would learn in the ’70s, there’s a closer correspondence between identities and issues.

The second thing that they measured is called issue alignment, which is the correlation between multiple issue opinions. And what they found there is that it was really quite low and it stayed very low between the ’70s and the 2000s. So, if you told me that you are pro life, that wouldn’t necessarily help me predict your stance on other issues like gun control or Medicare for all. So the way that they interpreted their findings was that what happened was a process of partisan sorting where we took a population that was largely not very ideological, they weren’t all hard line left or hard line right, they were leaning a little bit left, leaning a little bit right. And these people sorted themselves out better. The people leaning left increasingly took on the labels democrat and liberal, people leaning right took on the labels conservative Republican without anybody really becoming more ideological.

Now, my collaborator Jim Murphy and I looked at this article and thought maybe a lot has changed in the last 10 or so years since this article was published. Fortunately there’s been more data released since this time. Their time series only went up to 2004, we now have data for 2008, 2012, and 2016. So we basically look at these same trends in correlations over time and see what happens. What we find is that not only does the partisan alignment, the correlation between identity and opinion shoot up even faster than before, but there’s also a spike in the issue alignments, the correlation between multiple issues. So what this means is that people are in a sense becoming more ideologically consistent. Having one left leaning opinion implies having many, having one right opinion implies having many, and this is a new thing.

Grossmann: There’s still a middle, but those on each side have clearer ideas.

Kozlowski: One distinction that’s useful to make when talking about polarization is between alignment, which is what we’re studying here and bimodality, which is basically the dissolution of the center of the political spectrum. Recent studies have shown that there are still a lot of people give the neither agree nor disagree response, who don’t identify as moderates or independence. There’s still a lot of people who are right smack dab in the middle. The question is how organized are the people who are not in the middle, who are giving opinions, who are identifying as liberal or conservative? And what we find is that these people who are diverging from the center are more congealing around a single access of opposition.

Grossmann: And issue opinion correlations are rising fast.

Kozlowski: Since 2004 what we find is that the correlation between how people identify politically and their stances, what we call partisan alignment, has skyrocketed even faster than it was growing in the past. So, if somebody says that they’re a Democrat, that tells you a whole lot about what their opinions are on individual issues. But even more interestingly, since 2004 there has been this big boost in issue alignment, the correlation between opinions across issues. Now this has historically been very low and has grown only very slowly and it was last assessed up to 2004, at which point it was still low. When we carry the time series forward from 2004 to 2016, we find that this correlation between issue opinions shoots up and this in a sense is an indication of ideological consistency. If somebody has one liberal opinion, they probably have several liberal opinions. If somebody has one conservative opinion, they probably have several conservative opinions.

Grossmann: Wattenberg says open-ended responses also show the information environment has changed enabling real sophistication.

Wattenburg: What Congress is really looking for wasn’t so much ideology but a well developed sophisticated belief system. And that liberal conservative is a shortcut to talking about policy, but one can also talk about policies directly. I now say that I rarely ever saw that in the microfilm interviews that I read that people would just be talking about all sorts of different specific policies. But now you see it all the time. I think the information environment has changed and also the political environment has changed that we just get exposed to so many more policies now. And then part of that is also social issues, gun control and abortion come up all the time, gay marriage comes up a lot, various civil rights issues come up a lot. And the policy once mentioned at least three different policies, and on average they made about over six policy comments, which is, well, it’s a lot of policy. And so I isolated that. In the beginning I figured those people had well-developed belief systems, but I coded them separately to then look at their characteristics.

Wattenburg: The reason to then look at their characteristics and I’ve found that they are very politically knowledgeable, very politically interested.

Grossmann: He agrees that it’s not just sorting, but real changes in conceptualization.

Wattenburg: There’s a lot of debate and political science for quite a while now, whether the greater relationship between party and ideology is just people forwarding themselves into natural categories, and the decline of southern democrats, southern conservative democrats and northern liberal Republican, or whether it’s really people who know what they’re doing and saying, “Well, I’m a conservative, therefore I’m republican.” “I’m a liberal, therefore on the democrat,” or reversed causality. I’m not sure that it really matters. What I tried to look at by conceptualization, is that pretty much all the inquiries and certainly most of the increase in liberal democrats and conservative republicans are from the top group of sophistication, the concept audio logs or the policy walks. That means that this is a real change. On the other hand, part of the increasing relationship is the decline of McComb County conservative democrats in Michigan. That’s an easy thing I would argue to avoid. It also just might be generational over time. What I find is the mismatches are going down at all levels of [inaudible 00:16:52]. I would say I don’t take a stand saying, “It’s all [inaudible 00:16:58] or it’s all sophistication.” What’s the combination? It’s the people who are getting it right. It is sophistication and the people who were no longer getting it wrong, well that’s just throughout the electorate at all levels.

Grossmann: Increasing education explains part of the increase.

Wattenburg: Congress all along expected that as education would go up, that is higher levels of sophistication, the ideologues and I would say that [inaudible 00:17:28] policy wants with ideologues that those would go up as well. That’s another thing that I’ve definitely noticed when compared to what I read many years ago is you’re seeing a lot more times of higher education. I am some of these things that I read, I couldn’t believe the level of detail. Then I looked these people up and, “Oh yeah. Well this person has a PhD. This person was a lawyer,” and stuff like that makes sense.

You’re seeing a lot of that. I just re-estimated the data from 1980 and said, “Okay, in 1980 college educated people had this percentage of ideologues and wants. Then what we know is that over the following 36 years, the percentage with college education has gone up.” I re-estimated the 1980 data with increasing levels of education every four years. What I found was that through 2000 the increase in higher levels of sophistication was exactly what you would have projected based on education levels. Then since then the increases went even more. Any changes that I found up to 2000 were pretty predictable based on education. Since then it’s a change. There’s got to be some other factors, changes in the information environment, changes in the way candidates campaign.

Grossmann: Koslaskey also finds that increasing constraint extends beyond the highly educated.

Kozlowski: It does make a great deal of sense that we should be seeing an increase in overall ideology and issue alignments over the decades because alignment is strongly correlated with education. Education is going up overall, but an intriguing thing that we find is that there is this corresponding concurrent boost among those who are less educated. Those who do not have a college education are similarly witnessing this boost in correlation between their issue attitudes and between their partisanship and their attitudes.

While there is certainly some kind of overall gain that can be attributed to education expansion in the US, something seems to also be happening that’s pulling in the folks who are not as educated and who are less politically engaged. I think that that is going to push us to look into how do people understand the issues who don’t pay much attention to the news and who don’t have a strong education in politics because it appears that there is some way that even the politically disengaged are making sense of politics. They’re making sense of it in a way that pulls them into the same ideological patterns that we’re seeing among the more educated and more politically engaged.

Grossmann: Wattenberg finds consistency up over time as well, but he attributes it to rising conceptualization.

Wattenburg:Compared to the 1980s, there was more consistency at all levels, but where consistency is really evident is among my top groups, my concept ideologues, and policy [inaudible 00:21:00]. It wouldn’t surprise me. I would expect that consistency would be up across the board because people are more sophisticated. It’s exactly what Congress would have expected, too.

Grossmann: Koslaskey says it’s somehow expanding beyond those paying most attention to politics.

Kozlowski: One of the fascinating things about what we found is that the increasing correlations between political identity and attitudes, and correlations between multiple issue attitudes, is not contained among the most elite populations. It’s not contained among the most politically interested, the most educated, the most wealthy. It’s actually diffused among the population. This is something that’s really unanticipated by most prior research.

In general, going all the way back to the work of converse, people who are deeply engaged in politics have been ideological to some extent. That’s not entirely surprising, but the people who are not engaged in politics, there’s never really been any sign of them aligning on the left or the right. It’s just not a game that they’re playing. Yet only post 2004 and as I mentioned, especially starting 2012, we see this boost that’s across the board. The less elite groups are still climbing a little bit on these moral issues. We really see a plateau among the more elite, more educated, more affluent in their moral alignments. Those have totally petered out, whereas the less elite groups are still climbing in the moral domain and in some of these areas even matching the more highly interested, highly educated elite groups in their levels of alignment.

Grossmann: We asked different questions over time, but there are ways to account for that.

Kozlowski: The way that we get around this problem of changing questions is by looking at how alignment shifts within questions over time. Instead of test-taking one aggregate score and seeing the average correlation as it goes over time, we look within questions and see does this question increase from one year to the next? Therefore even if questions aren’t asked the entire time series, we can look from period to period and on average look at the slopes of these various questions. That in part gets around the problem of simply different questions being answered and whether these new questions are just the polarizing questions and the previous questions were non polarizing questions.

Grossmann: They found issue consistency shot up most in 2012.

Kozlowski: When you look more fine grained at year by year, it looks like the biggest spike actually happened in 2012, which is intriguing. In 2008, if anything, there was diminishing of alignments. It’s hard to know exactly what was going on in this year. There were a lot of things going on, but looking at the candidates for the presidential election, Obama and McCain, both of these were figures who were in a sense not traditional democrat or republican and were able to draw support across the aisle in various ways.

It did seem that they perhaps diminished the importance of partisanship, diminished the association between multiple liberal opinions, multiple conservative opinions. Yet only four years later in 2012 we see this big boost. It’s hard to know exactly what happened at that period. One thing that we can say is it wasn’t Donald Trump. This predates Trump. He isn’t the smoking gun in this story. It may have something to do with new kinds of media consumption, increasing online media consumption, increasing options for cable news for instance. There’s a lot of research shows that people are able to sort themselves into the kind of news that reaffirms what they already believe. This exacerbates bias in one direction or another.

Grossmann: Moral issues explain the earlier alignment, but now it’s economics and civil rights.

Kozlowski:I think the interesting thing about the issue variability is looking at the changes between the earlier time period and the more contemporary time period. Now there was a lot of talk for a long time about culture wars in politics. These conversations particularly emphasize the importance of these moral issues like abortion, and gay marriage, and drawing new people into the fold into talking about politics, and becoming the critical access on which political divisions were structured. Indeed, this is something that was important in both partisan and issue alignments from the 70s to 2004, but afterwards we find that the economic issues and civil rights issues are the key things, the key axes along which opposition is becoming structured in more recent years. This causes us to update our theses about the culture wars. This doesn’t necessarily mean that it’s not a kind of culture war, but the cultural differences that are being leveraged here in political debates aren’t just about these kind of moral/religious questions that have to do with how do we structure our economy. What are the government’s obligations to racial minorities?

Grossmann: He says that follows changes in the political debate.

Kozlowski:The directionality of this association is that the elites who are driving the mass or the mass who are driving the elites is hard to establish, but I think that we see this finding born out in contemporary political debates, that there’s a rising importance of economic and civil rights issues over the moral/religious ones.

Grossmann: Wattenberg also says changes in the debate are reflected in the interviews.

Wattenburg: Having read [inaudible 00:27:15] 1952 to as recently as 2016, messages are certainly there. My advisor, Warren Miller, that’s me, when I was starting this work many years ago, said, “What the open ended responses often reflect is the debates of last four years.” Then if you read the 1952 interviews, you’re really picking up a lot of issues and debates that came from the Truman administration construing with the president when the election is going on. You definitely see that. You definitely see people talking about current events in response to people who criticize these questions and say, “Well, people were just parroting what they heard on the news,” but it’s a really good historical insight into what people are talking about now is really different. When they’re thinking about politics, they’re really thinking about something very different than what they were doing in the 1950s, 60s, 70s, 80s [inaudible 00:13:23].

Donald Trump did change the nature of how republicans think about politics because he provided a different stimulus, somebody who wasn’t that ideological. Most republican voters in 2012 were very ideological or policy oriented. Some people were very different from the [inaudible 00:28:47] people. They just didn’t talk as much as republicans usually do about policy and especially ideology. On the other side, the democrats, they’ve been moving more ideological and more policy oriented. What we saw in 2016 was sort of a legitimation of talking in those terms, that democrats who avoided the term liberal for years, even if they were liberal. Bernie Sanders said, “Progressivism, democratic socialism, these are good things.” For the first time I’m really seeing that in interviews. Then you also see a lot more policy orientation, too. The parties in that sense have become closer together. In the long run, I expect the republicans to come back more towards normal after Trump is gone. They should be more ideological than the democrats, but I think the democrats, they’re going to continue more towards Sanders way.

Grossmann:He found that newer Trump related 2016 issues got integrated easily into others.

Wattenburg:What I found in 2016 is the policy wonks especially are very consistent in terms of the issue dimensions and they were able to relate what I call the Trump issues and take these new issues that are out there that really do cut across parties in a lot of ways like Republicans have always been free trade, but Trump takes a different stand on it, so that’s one Trump issue and then the immigration, Republicans have often been pro-immigration, of course, Trump has not been.

So, the policy wonks are very consistent in terms of they’re very conservative on what I call the Trump issues and also very conservative on social issues and economic issues and they’re fitting everything together. Concept ideologues almost as well and those two groups since they now represent about 45 percent of voters as opposed to say 25 percent or maybe even not quite that much in other years, that makes a big difference and that should increase issue consistency overall.

Grossmann: Because Kozlowski also found that Trump didn’t change too much. The increasing alignment continued.

Kozlowski: 2016 was actually not a grand departure from 2012 in what we observed. 2012 was when the biggest boost took place and this persists into 2016, but 2016 was not the year where the most action was taking place. Now, at once, you might expect Donald Trump to exacerbate alignment, but also to cut against them at the same time. Ways that Trump could heighten alignments in 2016, moving forward to 2020 is by pulling people into politics who have previously been less engaged. A lot of people who were not voters in previous elections, who were not considered likely voters came out to vote for Donald Trump in 2016, so it seems clear at this point that he’s been mobilizing a group of people who have not been engaged in politics and perhaps wouldn’t have identified with one party or another, but are now becoming politicized through the kinds of rhetoric and the kind of issue attention that Donald Trump gives.

Grossmann: Overall, Wattenberg found the 2016 responses very negative and extreme.

Wattenburg: Anything in particular about Donald Trump that would make you want to vote for him and then, a lot of people just responded with absolutely incredible negative feelings that I’d never seen before. Before you even asked them the second question, which is is there anything in particular about Donald Trump that would make you vote against him? And I have in my paper some examples of some extreme things people said and I started putting them in red in the Excel file and after awhile, it’s like I’ve seen this over and over again.

Unless somebody said something that was different, I didn’t even bother. And I sent this to a colleague who read all the Goldwater responses, I said did you ever see anything like this? And he said no, never. And then, they asked people about Hillary Clinton, same thing. No. Reasons for voting for, reasons for voting against and again, the responses were often very extreme. I heard Hillary called all sorts of things that I’d never seen before in previous years. Same thing with Trump. And then people were asked what do they like and dislike about the Republican Party, what do they like and dislike about the Democratic Party.

By the time you’ve read that, and on average it takes about 15 minutes for a respondent to get asked those eight questions and answer them, so it’s a nice little short structured interview.

Grossmann:But he also saw an explanation for the 2016 primaries where Trump voters really were distinct and there was a real divide in conceptualization between supporters of Clinton and Sanders.

Wattenburg:The Trump voters, they’re less motivated by traditional Republican goals and they are really looking to shake up the system and you really see that in some of these interviews and I must say that I saw some of these people in 2012 and 2008. I read the 2016 before I went back to reading all the 2012, 2008 in detail and I often wrote in the margin in 2008 and 2012, boy these people, this person really will like Trump when he appears, but we’re back on these old interviews, so they were there, but there was a stimulus for them.

And on the Democratic side, the Hillary people really were quite different than the Bernie Sanders people and the Hillary Clinton people thought more in traditional group terms and they’re traditional Democrats whereas the Sanders people, they like the Trump people, they really want a big change and their change is ideological and policy oriented.

Grossmann: But it all can’t just be following elites, because Laski said they found increasing constraint even among independents.

Kozlowski:The partisan trends do, in a way, mesh well with these subgroup trends where we find that the non-elite populations, the less engaged populations are showing similar increases in their political alignments. And when we do desegregate by party affiliation looking at issue alignment, we find increases in economic civil rights and moral issues among independents, maybe most strikingly. And this is surprising, because who are the independents taking their cues from if they’re not allied with either the right or the left?

And this once again draws to the fore other questions about media consumption, what kinds of sources, what kinds of information sources independents are drawing from and also brings to the fore, the question of how political issues are being framed and discussed. If there is perhaps now a way that issues are being talked about that makes them more comprehensible, intelligible, and compelling to people who previously have been disengaged with politics and still would not identify themselves as either Democrats or Republicans.

Grossmann:Wattenberg says there’s still a divide between those who do and don’t understand politics the way that elites do, but the sophisticated portion have doubled in size, a trend he finds promising.

Wattenburg:You can’t overstate it. I think Don Kinder at the University of Michigan has done a lot of research on this and he’s gone back to Converse and said well, Converse is basically right that either people have that really good, sophisticated sense or they don’t. What I think that Kinder and his co-author Kalmoe missed in their recent book was just how much the percentage who have it has gone up. I think that they definitely missed that.

So, we’re now in the case where at least when you look at people actually voted, pretty much a 50/50 wall in that about half of the electorate are really quite sophisticated and half really aren’t and they’re not that consistent, so the answer isn’t all that simple, but we’re moving in the right direction and it’s a good time for democracy. When people ask me, do you feel better or worse about the American public and American democracy after having read this? Especially in light of my recent experience, in light of the state of the research, I definitely came up more optimistic.

People are more thoughtful, I think, than political scientists have been giving them credit for recently and I expect this to continue in the right direction in the near future.

Grossmann: Elite arguments are now penetrating to the wider public.

Wattenburg: Part of what Converse was trying to say was that a lot of the debates between politicians just get lost on the American public who aren’t sophisticated enough to understand policy and ideology and that has really changed and I think it really has penetrated and people have gotten used to these arguments and they’re going to expect in the future and I think Phil Converse would say that this is a good thing. That’s something we’ll keep on following, but in this day when there’s so much criticism of American democracy, I come up more optimistic having actually read what ordinary people are saying.

Grossmann: But there is a downside. We’re mean and polarized.

Wattenburg: Downside is you think stuff like Goldwater or McGovern was extreme. Calling the other side Fascist, calling the other side criminals, saying they’re evil. I saw that over and over again. The name calling has really gotten out of hand. I don’t know whether it’s just social media or whether it’s also the candidates to blame for this, but there is certainly a lack of respect for the other side and I got to say that I see that with a lot of very sophisticated people and that in some sense, I would say, the more you know, the more you can come out as really extreme in saying why do you not just like the other side?

Grossmann: Kozlowski says the evidence should move people away from the partisan sorting explanation toward the possibility of issue-driven change.

Kozlowski:It is hard to tease out whether one is driving the other. It’s certainly very possible that some of this partisan alignment that we’re seeing now is not actually partisan parting, but is actually resulting from the issue alignment. When you openly see partisan alignment increasing, while issue alignment is low, you can pretty conclusively say that this is a consequence of sorting. People aren’t becoming more ideological, they’re not changing their attitudes on issues, they’re just picking the right party to identify with.

But if you have people who are already liberal or already conservative becoming more ideological, taking on a more platform liberal or platform conservative set of views, then you’re going to see an increase at once in partisan alignment and issue alignment through that same set of attitudinal change. So it seems that that’s really playing a key role right now and that is something that hasn’t previously been observed. Importantly, this isn’t something that can just be reduced to partisan sorting alone.

Grossmann: His next step is to ask whether cultural vision might be helping to bring the uninitiated into our political division.

Kozlowski: I would like to continue looking into the ways that political visions are mapping onto existing cultural divisions in the U.S. I think understanding how and why the less politically engaged are being pulled onto these axes of ideology requires us to look at what these groups of people care about, what they hope for from politics and how the various framings of issues resonate with them, pull them in in a way that they haven’t been able to do before.

So, I think moving forward, the big question … moving forward, we can take for granted this new political landscape that we’re dealing with with new levels of polarization and alignment, but we still need to understand how did we get here? What were the forces that brought us to this new point of political division?

Grossmann: There’s a lot more to learn. The Science of Politics is available bi-weekly from the Niskanen Center, I’m your host, Matt Grossman. Thanks to Martin Wattenberg and Austin Kozlowski for joining me. Please check out the changing nature of mass belief systems and issue alignment and partisanship in the American public and then listen in next time.