The Supreme Court made major conservative rulings this term but did not go as far as some expected. Are Court rulings out of step with public opinion? How much do they risk public backlash for moving against the public? Joe Ura finds that the Court provokes more backlash for moving in a too liberal direction than a too conservative direction. Stephen Jessee finds that the Court has been moving rightward but that the public is slow to notice. They both doubt the Court will provoke nearly as much backlash this year as with their abortion opinion in Dobbs.
How the Supreme Court Shapes (and Is Shaped) By Its Public Support
How Court Nominations Polarize Interest Groups
Descriptive Representation in Supreme Court Nomination
How the Federalist Society Changed the Supreme Court Vetting Process
How Does the Public Move to the Right When Policy Moves Left?
Matt Grossmann: Will the Supreme Court opinions provoke public backlash? This week on the Science of Politics. For the Niskanen Center, I’m Matt Grossmann. The Supreme Court made major conservative rulings this term, but didn’t go as far as some expected. Are court rulings out of step with public opinion and how much do they risk public backlash for moving against public opinion or in a concerted ideological direction? This week I talked to Joe Ura of Clemson University about his research on public opinion in the Supreme Court. He finds that the court provokes more backlash for moving in a too liberal direction than a too conservative direction, as it provokes ideological opponents without adding support from ideological allies. His work also shows that public opinion reacts against the ideological direction of court rulings, but only for a while. I also talked to Stephen Jessee of the University of Texas about his new PNAS article with Neil Malhotra and Maya Sen. A decade long longitudinal survey shows that the Supreme Court is now much more conservative than the public.
Using surveys of public opinion on specific court rulings, he finds that the court has been moving rightward. They both doubt the court will provoke nearly as much backlash this year as their abortion opinion in Dobbs, but that doesn’t mean the court is in step with the American public. Here’s our conversation.
You’ve found that ideological judgements of the Supreme Court follow decisions, but also other factors. How do Americans judge the Supreme Court’s ideological direction?
Joe Ura: What we’ve done in some new research that we presented at the Midwest Political Science Association this last year, is look at Gallup data over the last 20 years or so that asked Americans whether the Supreme Court is too liberal, too conservative or about right. As you point out, part of the equation is that as the court gets more liberal, more people think it’s getting too liberal. As the court gets more conservative, more people think it’s getting too conservative. The other part that matters is how liberal or conservative the public is because that question asked for a relative judgment. If the court is staying consistent, but I’m getting more conservative, I might start to think at some point that the court is too liberal, even though it hasn’t changed, I have. To figure out what Americans think about the Supreme Court, we need two big pieces of information. What’s the ideological tenor of what the Supreme Court’s doing and then on the demand side, what do Americans want out of the court? If those things are in alignment, we would think that a lot of Americans would say the court is doing pretty well and that it’s about right, but if the public is more conservative than the court, then they would think the court is too liberal and vice versa.
Matt Grossmann: That assumes that they’re getting some information directly about how the Supreme Court is doing. Is there evidence that they do that or do they kind of misperceive the Supreme Court as doing the same thing that the president is doing or that Congress is doing or kind of just judge randomly?
Joe Ura: Yeah, so there is evidence that the other piece of that, whether people perceive the court as being too liberal or too conservative, is related to what the court’s doing. There is some mechanism that’s translating decisions in the aggregate to those public judgements about whether the court is getting more liberal or getting more conservative. The transmission mechanism is something we haven’t really nailed down, but there’s a couple of prime candidates that we would turn to based on what we know about how the public gets information about politics in general. One is, I don’t want to say directed, but through the mass media. The mass media is covering the Supreme Court, it’s covering its decisions. People are able to see that they’ve reversed Roe V Wade or whatever and use that information to update their beliefs about the court.
The other piece, the other mechanism is from elites, that when the Supreme Court makes decisions, newspaper columnists, presidential candidates, members of Congress, your Senator, they’re going to comment on these things to the extent that they think that they’re salient issues. People, I think, get a lot of indirect information through other political figures that they’re following. Even if they’re not attuned to the court directly, they’re paying attention to some other stream of information that produces signals about the court. Then, generally of course, are people who don’t pay attention to politics hear a lot of it secondhand. We’re all friends with somebody who pays attention to politics as their hobby, those opinion leaders and those people transmit information through into their social networks.
Oh, I lost you for a second Matt.
Matt Grossmann: You’ve also found that people do respond to the ideological direction of the court in terms of whether they approve of or like what the court is doing, but you found that they don’t get as much credit as blame or at least the people who are on their ideological side don’t seem to increase their support as much as those who are opposed to them decrease their support. Tell us about that and why you think that’s occurring.
Joe Ura Yeah, so that Gallup data I mentioned before, the too liberal, too conservative, about right. If we use that data to predict Supreme Court approval, there’s a pretty strong relationship between the share of people who say the court is about right and the share of people who say they approve of what the court’s doing, is probably what you’d expect. There’s also a negative relationship between that too liberal share and approval and the too conservative share and approval, but what’s really interesting is that there’s a much stronger relationship between the too liberal share and approval than the too conservative share and approval. What that means is that on one side of the ideological spectrum and as the court is getting more or less conservative, there are people switching back and forth between about right and too liberal and on the other side there are people switching back and forth between about right and too conservative.
The people switching back and forth between about right and too liberal, who are probably going to be people on the right side of American politics. When they change, when they see that the court is getting too liberal and they get to that point where they say, I’m not going to answer this question and say the court is about right anymore. Now I’m going to say it’s too liberal. They’re also more likely to answer the question and say, well, now I disapprove of the court because it’s too liberal. On the other side of the spectrum, people switching back and forth can see the court getting more conservative, switch from saying it’s about right to too conservative, but those people are less likely to also say they disapprove of the court and there’s this ideological asymmetry where people on the left, for some reason that we’re still digging into, have this kind of larger reservoir of goodwill towards the court. They’re willing to express approval in excess of their agreement with the court in a way that people on the other side are not willing to do.
Matt Grossmann: One story might be that there’s a difference between people on the left and the right and how they feel about the court and it’s sort of permanent, but another story might be that we haven’t been in an era like the one we’re in now, where a lot of people do perceive the court as moving in a very conservative direction. How much do you think this is changing or will change as the court shifts in that direction?
Joe Ura: Again, we’re now in the realm of a little bit of educated speculation and things that I haven’t nailed down in data as much as I would like, but I think part of the difference is about the kind of cues that partisans get from elites on their side. On the left, among the democratic partisans, liberal ideologues and opinion leaders, over the last, let’s say two or three generations, there’s been a strong ideological affinity between liberalism and judicial review and judicial power and it probably goes back to the civil rights era, Roe V Wade, different kinds of protections for privacy rights and contraception, expansive definitions of free speech and the right has sort of developed a politics that is more, if not antagonistic, at least suspicious of judicial review and that goes back also to the civil rights era.
We have a partisan asymmetry in how elites ideologically and for strategic policy making reasons treat judicial review and I think that’s filtered into the way that mass partisans and ordinary citizens see the court. Now that there is a kind of baked in Republican conservative majority on the Supreme Court, I think the elite signals are changing over the last couple of years. We’re handicapped a little bit in political science, in that the models we have, especially these big macro models of the kind I run, look at average effects over, for example, in that approval paper I mentioned, you’re 20 years of data.
If something has changed, as it seems to have in the last couple of years, I can look at the data and say, wow, the output of the model really doesn’t fit the last couple of years the way it has historically, but to really say that that’s a kind of permanent change is harder for us. We’ll check back in a decade, we’ll know more, but it certainly does seem to, what I can see in the data is that something really has changed in the last couple of years. The model doesn’t fit what’s been happening in terms of Supreme Court approval since 2021, as well as it has done historically and I think something about the combination of things with the last couple of Trump Supreme Court nominations and confirmations and the Dobbs decision, has shifted the data generating process and I think the next 20 years will look much different, will look much more like the last couple years than the previous 20 years.
Matt Grossmann: One trend that may predate that, but has been accelerated, is just an overall decline in approval or increase in disapproval. To what extent can we call this a real trend that’s likely to continue moving in that direction and is that a sign that the Supreme Court is less treated like a different kind of political institution than it has been?
Joe Ura: For approval, what we call specific support for the Supreme Court, which is attitudes that have to do with whether or not people like what the Supreme Court is doing in the short run. We’ve known for a while that approval of Congress and other institutions in the national government are a good predictor of approval for the Supreme Court. What that shows us is not necessarily when people think of the Supreme Court they think of Congress, but that attitudes about the court are linked to larger attitudes about the government and whether people like it and the court gets lumped in with that.
There was an idea in scholarship about the Supreme Court in public opinion, that was that the court is different and special and that Americans had a larger pool of goodwill. The court was more legitimate, there was more diffuse support for the Supreme Court than the other branches of government maybe. I think that’s probably still true, but it’s probably less true than it was. As congressional approval is absolutely at the floor, approval for the court is trending down and has for the most part over the last 15 or 20 years and there’s multiple reasons for that, but part of it is Americans are just looking more and more scant at their governing institutions generally and I think they’re learning the lesson that maybe the court isn’t as special as previous generations of Americans thought it was.
Matt Grossmann: You’ve also looked at how Supreme Court opinions change liberal and conservative attitudes in the American public overall and have found that they do respond in some ways like they do to congressional policy making, but not in all ways. Kind of explain that pattern.
Joe Ura: A thing that makes all this complicated is there is feedback from what the court and other governing institutions do in what Americans think and want from their political system. You can think of the degree of liberalism and conservatism and public opinion as a kind of measure of demand. If public opinion is very liberal at some moment, it means the American people are generally asking the government to do more liberal things. Increase social welfare spending, increase access to public healthcare services, increase social services, maybe even increase taxes to pay for things. If the public is more conservative in some moment, they’re demanding conservative changes. They want lower taxes and less spending, less generous welfare benefits and so on. When the government meets that demand, so a liberal public votes for Democrats, Joe Biden and Democratic majority come in, they pass these more progressive pieces of legislation, they’re satisfying that demand. By satisfying that demand, they’re moving public opinion back in the other direction. People don’t need as much liberalism as they needed before those laws passed and the country’s getting more conservative.
That’s the kind of basic relationship between liberalism and conservatism and public opinion, which political science has typically called mood and public policy making, in the aggregate over time. The way the Supreme Court has worked historically, that process is a little bit different. Americans have a more complicated response to ideological changes in Supreme Court decision making, in that when the court makes a decision, you get that same kind of negative effect. When the court makes more conservative decisions, you can see a response in public opinion getting more liberal, as you would if Congress passed more conservative legislation, but that effect decays over time and gets replaced by a small, but significant movement in mood towards the ideological direction of Supreme Court decision making. The way I characterize this is in a paper in the AJPS a few years back, was that there’s back a period of backlash against the Supreme Court, followed by a longer period of legitimation where the public comes to accept the ideological consequences of what the court has done. That operates at this very abstract level of ideological change, but it is very different from the patterns we see in terms of changes in response to the things that congress and presidents do.
Matt Grossmann: You might expect that this applies to say, abortion, where after Roe V Wade there might be a backlash, but eventually the public kind of turns out on that side. We’ve seen the first part of that to the Dobbs decision. Is this the kind of thing where we really should expect the conservative turn in the court to have a long term benefit for conservative public opinion?
Joe Ura: This is a point where my gut and my political science say different things. My political science says and let me just argue for that here in the Political Science Podcast. Yeah, right, the model says is the court has become much more conservative than it was say between five years and a decade ago, whatever our breakpoint is. And, as the court, right? That’s, this shock to the system, we should see a backlash, right? And the public becoming more liberal than it would be otherwise, keeping in mind that there are other things going on in the political system, right? We have a democratic president for part of this period, we had a democratic majority in Congress. So they’re doing progressive things that are pushing back in the other direction. But nevertheless, right? The public would be more liberal than it would be in the absence of the more conservative Supreme Court. That effect will decay over the next couple of years. So, let’s just say the big shock was in 2022 with Dobbs, right? We’re still probably in the period of backlash.
But moving towards the period where that will begin to decay and that the public will become more accepting of the Dobbs decision than it was initially. Why does that decay happen? What I’ve always kind of speculate is that it goes back to that elite signaling, right? We’ve had… Historians and law professors have done a really good job about this, in particular, like Larry Kramer’s book called The People Themselves. He was a law professor at NYU, and became Dean of Stanford Law School. And he argued that there’s this very strong post-World War II consensus across party lines about the legitimacy and sanctity of the Supreme Court. And that politicians of all ideological stripes with some notable exceptions, acquiesced to Supreme Court decisions, even when they didn’t like them. They were still very deferential to the court. And, I’ve always taken that as an important part of this equation, that when the court makes a decision, if politicians then say, well, look, maybe we don’t like it, but this is the way things are and we just have to live with it. People get the message they need to live with it.
One of the reasons I think abortion is so important in Supreme Court politics is I think it was maybe uniquely an issue where lots of politicians refused to accept the finality of a Supreme Court decision. And not just resisting its implementation, but to really advocate for changing the outcome by changing the composition of the court, right? To change some interpretation of the Constitution by treating the court as an ordinary political institution. And I think the political system and political elites and ordinary citizens at this point are getting the message, that that’s how things work. And that’s changing the kind of underlying structure. So, to just come back to your question. I think, yes. I think over the next couple of years, the backlash will not be as strong, and that there will get to a point where, again, as I said at the beginning, there’ll be a small but significant movement in opinions towards the kind of position the court staked out.
Matt Grossmann: So you have also done a lot of work of the media as kind of the intermediary factor here in informing the public about the court. So how reliable is media coverage on what the court is doing? And is it just kind of a translation, or is there kind of an independent effect of how the media is portraying the court?
Joe Ura: What I can say is that the media generally does a pretty good job. When we have, we look at something like how liberal are the cases that the media covers in a prominent places compared with the overall liberalism of the Supreme Court, it’s a very high correlation, right? And the cases that get attention, if there’s a lot of attention to very conservative decisions, tends to be because they’re being generated by a very conservative court, and likewise when there’s lots of coverage of liberal decisions. In addition to that, the other thing that’s kind of interesting about, or another interesting finding in the literature on the Supreme Court and the media is that, Supreme Court decisions tend to generate, or at least in some circumstances, can generate attention to issues in the media that they weren’t paying attention to before. So for example, the Dobbs decision, right? And we can look at this historically. When it came to say gay rights or civil rights, these were issues that were not getting so much attention in the national media.
And when the Supreme Court began to make rulings on in those issue spaces, and particularly in ways that kind of upset the status quo, right? Changed the kind of rights claims that people could make, the media began paying more attention to those. And so when it comes to abortion, abortion is, this Dobbs is not just going to change the way people think about the abortion issue, it’s also going to change the amount of time we all spend talking about abortion, and the amount of media coverage devoted to that issue. And not just to what the Supreme Court is doing on that issue, the attention of the whole system is going to shift, right? You’ll see more bills in state legislatures, more bills in Congress, more open, I mean, it’s going to be a bigger part of the national conversation because the court has acted in this way.
Matt Grossmann: So I guess I’m struggling because I don’t see anything that is going to shift the overall direction of the court. But the one other possibility that I see is, I guess these Roberts cutesy attempts to construct coalitions that might seem to be perceived as more moderate decisions than the court could have made. And again, I’m kind of struggling with, would the public even notice that or is this just going to be, well, it’s another conservative decision. But, there might be media coverage of, well, it could have gone further. So that’s what I’m struggling with. Is that the kind of thing that the public might notice and say, well, maybe they’re not going as conservative as I thought?
Joe Ura: John Roberts is interesting as a character, right? I mean, I have written before, I wrote a couple of papers with Carla Flink from American University about the unique role of the Chief justice in preserving the Supreme Court’s institutional standing. I think John Roberts has a model of preserving Supreme Court legitimacy and preserving Supreme Court approval that guides his actions that’s different from other conservative and Republican members of the court. And I think he is taking actions to try to take the edge off some of this stuff. You could see it in his approach to the Dobbs decision, right? That he wrote a concurring opinion saying, look, we can uphold this Mississippi law without striking down Roe versus Wade. And I think if that was the headline like, hey, the Supreme Court allows stricter abortion regulations, but up Roe versus Wade survives. I think that’s a difference that registers, right? Because that’s a difference in the policymaking reality that people experience, right?
People in Texas or South Carolina don’t live in a state with a total abortion ban at that point. Maybe they live in a state that bans abortion after the first trimester. And that’s a real difference. And so, the way that that plays out depends on the extent to which that project is successful, and just how many times he can do it in relation to how many times he can’t do it, and on what issues he’s successful, right? So if it’s these more marginal issues and it’s infrequent, it won’t register much. If it turns into a more successful project, if somebody like Neil Gorsuch also gets nervous about the Supreme Court’s legitimacy, and he starts to worry about the plausibility of Supreme Court reforms or court packing or something like that, then it might have more legs. And right now we just don’t know how it’s likely to play out, beyond saying that, it seems that it’ll be difficult for any one justice to kind of overcome the way, just the numbers game of the kind of ideological and partisan composition of the court.
Matt Grossmann: So to what extent should the Supreme Court be paying attention to overall conservatism or liberalism of public opinion and asking itself if it’s going too far beyond that in some way, or to the direction to say, to react if there’s a liberal move in public opinion? And I guess, can you sketch us some futures that are kind of the best case of them paying attention to that in the worst case?
Joe Ura: William Riker, right? The great political scientist, has this line at the end of a paper he wrote in the American Political Science Review, which is essentially that any political system that doesn’t return the same result you get from simple majority rule is unlikely to survive in the long run, I’m paraphrasing a little bit. When that the system of supply and demand of policymaking I talked about earlier. If you have a set of institutions that reliably and in the long run fail to meet people’s demands for the policymaking they want. It’s just the disaffection builds and something’s got to give, right? Whether that is something that gives in a peaceable way within the confines of the system, big constitutional amendments as we’ve had in American history, big reforms to the way the Supreme Court operates as we’ve had in American history, right? Those are things, or something, right? Demand has built up in some way and shifted things within the system.
We’ve also had a couple of things happen where there’ve been breaks in the system that have spilled outside, right? So, Matt Hall from Notre Dame and I wrote this essay for the conversation about resistance in the south to federal courts, school desegregation decisions, right? And then spilling out violently when the system couldn’t contain them. Does the court need to follow public opinion? I think the answer is, the whole system has to follow public opinion at some point. Now there’s a question about what public opinion. I don’t think courts in particular are designed or ought to follow every short-term whim. But, a system that’s designed in a way to essentially to insulate the court from following persistent, enduring, meaningful changes in the way that Americans think about the structure of their rights, and to make decisions on important questions of liberty at odds with Americans understanding of their own freedoms and what their own constitutions mean, I don’t know that that can endure.
And so, I think the real question isn’t about whether these justices should start following public opinion, it’s a bigger question about have we designed an institution that is sufficiently responsive to public opinion over the long run in the first place, right? It’s not on these nine people’s shoulders to suddenly decide to start doing something different, it’s a bigger question of constitutional design. And I think what we’re seeing in the court as we’re also seeing in other institutions in American national politics, is that maybe the systems we have in place are really sort of at the breaking point in the stresses that our political moment is putting on them.
Matt Grossmann: So what do scholars in political science know about the courts and public opinion that practitioners and the media don’t pay enough attention to? And the reverse, is there anything where we have it wrong or at least that we should pay more attention to what close observers outside of academia are saying?
Joe Ura: There is a difference between low approval and lost legitimacy that I think that the media coverage of the court these days is not sensitive to the political scientists who study these issues, maybe have a better handle on. In terms of what the public knows that maybe scholars don’t. I think, even when I was first getting into studying this stuff as a graduate student in the early 2000s, I think there was a sensibility, a dominant sensibility among scholars that ordinary people didn’t have the Supreme Court on their radar. That it really wasn’t a prominent part of the way that people thought about politics. And maybe when people ran for president and they talked about the justices, they were point, it’s not exactly that they were wasting their time, but they were communicating kind of a narrow audience or it was like talking about foreign policy. It was a way to demonstrate mastery of an issue that presidents are expected to have mastery of, but not really about the content of it.
And, even though I think most people have it right now, I do still think there’s this kind of lingering suspicion that maybe attention to the court is lesser of a different kind than attention to other kinds of policymaking news and information. And I think whether it’s something that’s changed or whether it’s just a different sensibility, I think that seeing the court and what it’s doing as part of the same flow of information and judgments and attitudes that people have about other national policymaking institutions, is something that I think journalism and practitioners maybe are ahead of the academy and recognizing.
Matt Grossmann: Anything you’re working on now that you want to tout or anything we didn’t get to that you want to include?
Joe Ura: Yeah. We’re still digging in that team who I wrote the PRQ paper with, and I are still digging into that question of ideological and partisan asymmetries. And what we’re trying to get into is to figure out whether these differences in responsiveness on the left and the right follow from something about the structure of ideology or something about the structure of partisanship. And it may be that those things are just so closely aligned, we can’t quite pry them apart. Our initial approach was to think that this was something about the way liberalism and conservatism differ as ideologies. But it may be instead that it’s something about the positions that have been taken on the judiciary by the Republican and democratic parties. And, obviously we wouldn’t be pursuing it if we didn’t have some instinct that there’s something there. And I think our instinct is that maybe this is more about partisanship than ideology, but we’ll let the data tell us in the end.
Matt Grossmann: All right. So you have found that the US Supreme Court has become more conservative than the American public. How do you go about making that kind of determination? How can you compare the public’s opinions with the court’s views?
Stephen Jessee: Yeah, so that’s sort of the central question we’re interested in. And obviously the process by which Supreme Court justices or the court as a whole decide cases is really different from the process by which ordinary Americans think about the issues in these cases, to the extent that they think about them at all before they read our survey questions. And we think that in fact, for many, maybe most cases heard by the Supreme Court, it’s not possible for ordinary people to form meaningful opinions. So these cases might be on arcane legal issues or technicalities that maybe would be difficult for people to understand if they didn’t have a law degree, right? But the cases heard by the Supreme Court often usually have big impacts on the lives of all Americans. And for many of these cases, the issues they deal with are ones that people likely have opinions on.
So, a big example of this would be the court’s recent ruling in Dobbs, which overruled Roe v. Wade, and eliminated the constitutional right for a woman to choose to have an abortion in many circumstances. So in this case, the legal issues might be about whether various parts of the constitution create some right to privacy that might include the right to have an abortion or something like that. And most people may not understand these sort of legal or constitutional issues. But in terms of the impact of that decision, we’d expect that most people have an opinion. So they would have an opinion on whether there should be some constitutional right to terminate a pregnancy. And with many of the cases heard by the court each term on topics like religious freedom, protections against discrimination, say protections for the environment, we think it’s interesting and important to understand how the public’s views correspond to the court’s rulings. So to do this, we’ve conducted a series of surveys that ask a random sample of Americans for their opinions about some of the important cases heard each term, and also some questions about their general views of the court and their knowledge about it. So the first of these surveys conducted way back in 2010 with Neil Malhotra asked about important cases over a five-year period. I think it was from 2005 to 2010. And we view that as sort of a precursor to our real project that we call SCOTUSpoll.
And with SCOTUSpoll, each year since 2020 we’ve conducted an annual survey, and that’s with me, Neil Malhotra and also Maya Sen, who’s a professor at the Kennedy School at Harvard. And we choose what we think are some of the most notable cases that are heard by the court in any given term, and we sample maybe a couple thousand Americans and ask for their opinions about these cases. So this set of surveys, which is four or five surveys now, we plan to continue fielding it annually, it gives us a way of comparing the court’s rulings to the views of the public and tracking these things relative to each other over time.
So to give you an example, to make it a little more concrete, in 2020, the court heard a case called Bostock v. Clayton County, and that questioned whether employers could fire workers based on their sexual orientation or whether federal civil rights law prohibited that. And so we asked a question on this case that said some people think it should be illegal for employees to be fired based on their sexual orientation because it’s discrimination on the basis of sex. Other people think it should be legal because it’s not discrimination on the basis of sex. And we found, perhaps surprisingly, that the vast majority, 83% of respondents, including 75% of respondents who identify as Republicans, said that this discrimination based on sexual orientation should be illegal. And then we compared those responses to the court’s eventual ruling, which actually held that firing workers for being gay was indeed illegal under the Civil Rights Act.
So in each year we ask 10 or 12 questions like this about important cases. And then in terms of how we compare these things, one way to compare, simple but important, is on a case by case basis. So in Bostock, for example, we found that most Americans agreed with what was actually the court’s eventual ruling. So in that case you might say, all right, the courts went along with the majority of public opinion. Not everyone’s public opinion, but most Americans. And over the first few waves of SCOTUSpoll, we found that most of the time, somewhere around two thirds of the time, the court’s rulings were in line with the majority position. And that’s certainly a useful way to look at things. And if you’re interested in a specific case or a specific issue, you could certainly do that.
But we also want to get an idea of how the court compares to the public in terms of overall ideology. So for example, is the court more liberal than the public, more conservative than the public, pretty similar ideologically? And there are several ways of doing this. A simple way might be to just calculate the percentage of time that the court ruled in the liberal direction on these sort of major cases that we’ve chosen in a given term. So maybe the court ruled in the liberal direction on 35% of those cases. And then we could say, all right, well, on average, what percent liberal responses do the public give on these case questions? And maybe the public is more liberal in that way or not.
The way we actually do it in our research is to use a little bit of a fancier way of doing it, an approach called ideal point estimation, which in the end gives you typically estimates that are fairly similar to the simple way that you might do it, percent liberal. But it essentially conceives of this ideological scale from very liberal to very conservative, kind of a continuous scale. And the idea is to use a statistical model to estimate the ideological positions of each of the justices on the court and the court overall, as well as the positions of each survey based on the positions they take on each specific case or each case question. So we’re able to estimate the ideological position of all the justices, the court as a whole, and then each survey respondent. And we care less about individual survey respondents’ ideologies than averages, say the average ideology of the American public or the average ideology maybe of a Republican or a Democrat in the American public.
So that’s the main way that we sort of think about trying to estimate those positions and make these comparisons. And as you said, the big goal of the SCOTUSpoll project is to have a way of comparing the ideological positions of the court and its justices to the American public.
Matt Grossmann: And just state the finding, given that research, we’re looking at the ideal points and we’re comparing the courts with the public, and what do we find?
Stephen Jessee: Yeah. So in our paper, in PNAS, the finding is that although the court was ideologically pretty similar to the average position of the American public in, say, 2010, and even in 2020 after Kavanaugh replaced Kennedy, the first thing we found is that, interestingly, this was perhaps due to luck or coincidence more than anything, given that the vast majority of the justices individually held ideological positions that were quite extreme relative to the public. So there were justices that were more liberal than the vast majority of the public and there were also justices that were more conservative, but the court as a whole in 2010 and shortly before, because we had a couple of moderate justices in O’Connor and Kennedy, who tended to be the swing vote on the court, and even in 2020 when surprisingly after Roberts became the median, the court didn’t really move that much that we detected in that year.
So in both of those years, both of those time periods, it seemed like the court actually represented public opinion pretty well. And I’m guessing we’ll get to this, we’re not saying that the court is designed to do this or even should do this, but it happened to be the case that the court’s rulings in those years were on average kind of similar. The court had an ideological position that was similar to slightly more conservative than maybe but really similar to the average American’s ideological position. Then when Ruth Bader Ginsburg was replaced by Amy Coney Barrett, the court moved sharply to the right.
Now, I would’ve predicted, in fact, Neil and I had an opinion piece in the New York Times that actually did predict, that when Kavanaugh replaced Kennedy, the court would move well to the right of most Americans. That didn’t actually happen or didn’t happen as soon as we predicted. I think we could discuss why, I think it’s somewhat speculative, but some combination of Roberts trying to keep the court from moving too quickly and also the unique mix of cases during the time.
But at any rate, by our 2021 survey, the court’s ideological position was estimated to be well to the right of the average American’s ideological position. And it was actually estimated to be almost identical to the position of not the average American but the average Republican, so the average person who identifies with the Republican Party. So I think that’s the main finding is that we’re able to track the ideological position of the court through its rulings relative to the position of the American public in various subgroups in it, and sort of document this rightward shift in a pretty direct way.
Matt Grossmann: So let’s dig in a little bit to the methods here. It seems like you said you’re selecting the cases that the public might know the most about and trying to put them in terms that the public would be most likely to recognize. But we still get some pretty arcane issues before the court, and they seem like ones that would be most susceptible to giving us non opinions, or as you said, would be opinions developed on the fly just in response to the survey question, might be affected by wording, those kinds of things. So I guess, to what extent do you think that these are real opinions or is it something else that you’re assessing here?
Stephen Jessee: So yeah, I think it’s an important question. It’s a bit tricky and it does apply in some way to all sorts of public opinion surveys, but these types of questions might be particularly susceptible to it. I think the farther away you go from these standard meat and potatoes political questions, things that people have probably talked about, thought about, the more you might have your responses driven in part by survey artifacts like question wording or things like that.
We actually have a paper that’s in progress using the SCOTUSpoll data that looks at things like question length, question complexity, and respondent characteristics like political knowledge or education and how those relate to response order effects. So we take advantage of the fact that in each of the SCOTUSpoll questions that we fielded, there are actually two versions where we flip the response order and we rewrite the question accordingly to present the two sides in a different order. And so what we find in this paper is that, consistent with a lot of prior work on survey research, people are generally more likely to pick the first response that you give them, all else equal. It’s a small difference, less than 3% on average, but it clearly exists. But we also find that the magnitude of this difference varies with question characteristics, mainly question length, but we also look at complexity, and then respondent knowledge and things like that.
So that paper’s still in progress, but I think the findings are important not just for our work, but for survey research more broadly. And we have to make decisions on how to write questions to balance things like precision and completeness against practical concerns, like whether respondents are going to pay attention long enough to read your entire question.
So in the case of our SCOTUSpoll questions specifically, I think for most cases heard by the Supreme Court, people would be unlikely to have real opinions. And as you said, we chose notable cases that are likely to be comprehensible and that deal with important issues. A lot of these are things like abortion, environmental regulation, affirmative action, where I think people are likely to have real opinions. But there are cases where things are a little more obscure, and so I think we’re attuned to that. I think our approach to some extent would tend to make those cases less important in our estimation. If it’s a case where people don’t have real opinions or the opinions aren’t really that ideological, the likely impact of that at the end of the day would be that that case doesn’t really drive our ideology estimates that much. But I think it is something that’s important to think about and I think we would like in the future to have something to say about the level of realness of the opinions on different types of cases.
Matt Grossmann: And not to get too down into the weeds on the ideal point estimation, but talk a little bit about how that treatment might differ from others, especially because the other people we’re discussing just asks Americans whether the court is being too liberal or too conservative or about right. And you might expect that people don’t necessarily place the court on a spectrum, but they sort of just say how much is it moving in one direction or another? So how does your ideological positioning compare to that?
Stephen Jessee: Yeah, so what we really want to do is pin down the court’s position based on its actual rulings and the justices’ positions based on their actual votes, and pinned down survey respondents’ positions based on their opinions on the issues that are being impacted by these cases. So I think it is informative and useful to look at what people think in response to a general question on is the court too liberal or has the court moved to the right? But I think it’s also useful to say, well, if we can measure where a respondent stands based on their stated views, and we know what the court actually ruled, how do those correspond? So that allows us to look at what is really going on rather than what the perception is. So I think that’s an advantage.
It’s also an advantage that it’s very tricky to estimate ideological position and movement over time. People have tried to do this for a long time, asking about estimating the ideology of members of Congress with things like nominate scores. Poole and Rosenthal did a ton of great work on that, and other people have developed it further. And then people like Martin and Quinn and others have tried to do things like this with Supreme Court positions.
One really tricky thing is it’s often very difficult to compare these things over time. There are issues like, well, what if everyone in the Supreme Court just took a step to the left at the same time? You might not be able to detect that. And so one thing our approach does is it kind of uses a different anchor for our estimates. And we’re essentially estimating the court’s position and the justices’ position relative to the distribution of public opinion or ideology in the public. And so I think that gives us a unique angle on this. So I think those two aspects, really pinning things down to actual policy positions and giving that anchor, make our approach unique, even though I think these more general questions, self placements and things like that can be useful in their own right.
Matt Grossmann: And is there a clear assessment of how much these things are connected? In other words, we have this court specific measure of public opinion that you’ve identified, but it would seem to have issues that were pretty similar to other policy issues. So is it just measuring left right ideology, either self-professed or in any other kind of issue index, or is this something unique?
Stephen Jessee: Yeah, our estimates of respondents’ ideologies are pretty strongly correlated with their responses to something like the commonly used five point ideological self-identification scales. So I think that’s reassuring. I think you could also argue, well, what’s the benefit if it’s so highly correlated? But I would say that even though it’s highly correlated, we don’t actually know whether someone who says that they are somewhat conservative, say, on the five-point scale, do they have a position that is similar to the court’s position? And so that’s where our contribution comes in, is we can pin these things down relative to each other.
So like I said, I think it’s reassuring that these things are correlated because I think you are right that for the most part, the way we’re considering these ideologies, at least for the public, is sort of what are your policy views? What do you want policy to be? And the fact that this is decided by a Supreme Court that’s, at least on paper, deciding these things based on what the constitution means and theories of legal interpretation, at the end of the day, there’s a policy impact of these decisions, and that’s really what we’re focusing on. So I think that’s our added contribution and what we’re able to do with our measures.
Matt Grossmann: So you also track expectations about opinions, and you use this to claim that the public has under perceived conservative change on the court. So tell us about how you do that and what the potential mechanisms for that under perception might be.
Stephen Jessee: Yeah. So in addition to asking people about their own opinions, essentially how they think the court should rule on each case, we also ask how they expect that the court will rule, not just how you want the court to rule, but what do you think the court will actually do? Right? So there are certainly people who probably said something like, “I want the court to uphold Roe v. Wade and not get rid of the constitutional right to an abortion, but I think that the court will rule the other direction.” So we wanted to capture both of those things. So in many, probably most cases in respondents, this is probably a guess, probably sometimes a pretty wild guess, maybe one educated a little bit by some understanding or impressions of the court or the case at hand. But we think these questions are useful for some things.
We just have to keep in mind that they’re pushing farther toward things that survey respondents may not have thought about that much related to some of your earlier questions. So accordingly, I think that while we can look at the absolute estimates of things like where the court public perceives the court, and that can be useful, it’s also informative and maybe even more reliable to look at changes.
So for example, if the public views the court as much more conservative on average in one year compared to the previous year, I’m more inclined to think that that movement is real and meaningful, even if there are sort of survey artifacts and people might not have real opinions. If it was totally random, why would it change systematically between the two years? So in our 2010 survey, we found that the public on average thought the court was quite liberal.
Respondents on average estimated the court’s position about half a standard deviation to the left of the average Americans ideology. So that would be more similar to the ideology of the average Democrat and the public than the average American. So they thought the court was pretty liberal relative to the American public.
And in 2020, the public updated its perceptions to think that the court was more conservative. So that was in response to Kavanaugh replacing Kennedy. There’s a lot of coverage of how that’s going to shift the court to the right. And as I mentioned it in that specific year, it didn’t estimate that the court moved that much, but the public shift actually meant that it had a much more accurate perception of the court in the end that year.
But then in 2021, when the court actually did move dramatically to the right of most Americans, the public didn’t catch on immediately. And it still perceived the court pretty similarly to how it did in 2020. As you know about the same as the average American’s position may be slightly to the right. I suspect that in the wake of the Dobbs ruling, which is obviously probably the most prominent ruling the court’s issued in a very long time, that that’s going to be something that’s going to cause people to update further. And we haven’t fully analyzed the data from the survey that occurred after the court decided that case, but it’d be really interesting to see.
Matt Grossmann: And finally, you associate these perceptions with institutional reform support. I think with the aim being that at some point if the court was seen as being very unrepresentative, that might be a reason for reform. So what do you find and why is there a relationship?
Stephen Jessee: We find that the people who perceive the court as more conservative than themselves are more supportive on average court reforms such as term limits or expanding the size of the court. So these are things obviously that there are presidential commission discussing. It seemed like there was some momentum on this. It’s unclear if that’s fizzled a bit.
But at any rate, it’s something that was sort of legitimately discussed in the media and among elites and even within the executive branch in some sense. So the fact that the people who see the court’s ideology as to the right of themselves are much more supportive of these reforms to the court, these changes, than people who see the court as similar to themselves, or especially people who see the court as to the left of themselves, that actually suggests coupled with the fact that the public misperceives on average the court as being more liberal than it is. Put those two things together and that’s pretty suggestive that if people accurately perceived how conservative the court was relative to the public or specifically to themselves, then support for these reforms would be even higher than it currently is.
So one important question will be if in the wake of the Dobbs decision or other things that the court has done recently, if people will change their perception to update and more accurately view the court as being as conservative as it is, and if that will also result in people being more supportive of term limits, packing the court or other things. So our findings are a bit suggestive on this count, but both of those two sort of associations put together I think say something interesting about what’s going on.
Matt Grossmann: So it seems like there’s a normative basis to the project to some extent that the court should at least be attending to public opinion or not be extremely far off from public opinion. But I also doubt that you all think that the court should just look at your survey and then rule accordingly as we usually think of the court as doing something different than just taking a poll and siding with the public. So what are the considerations in how much this should be driving court action and is there some kind of a threshold as to when the court is too far away from public opinion that would stimulate these kinds of institutional reform discussions?
Stephen Jessee: Yeah, this project actually, the idea for it came out of this type of question. So the court obviously was designed in large part to be insulated from public opinion that once the justices are appointed, they essentially have lifetime tenure, they don’t have to worry about public opinion or electoral concerns or things like that. And there are obviously reasons why you might want this for a court. So a classic example is so that the court can protect minority rights against the will of the majority or something like that. So people might say something like, the court isn’t supposed to just do what the public wants. And I wouldn’t disagree with you, even if you say that the court shouldn’t be affected at all by public opinion, I think it’s still an important question for political scientists and other people to ask just about how the decisions of the court compare with what the public wants.
So even if the court is supposed to be totally indifferent to public opinion, it’s interesting and important to just say, all right, well, how do its rulings compare to what the public actually wants? So for example, is the court really serving often as a big check on majority rule protecting rights by ruling against the views of most Americans? And if it does this, does it do it as often in the liberal direction as in the conservative direction or other things like that? And then as you mentioned, they are also important, but thorny questions about whether it’s a problem if the court drifts too far from public opinion in terms of its legitimacy.
A commonly cited thing is that the court doesn’t have the power to enforce its decisions. So it largely relies on its legitimacy and prestige to hopefully force other branches of government to go along with what it decides. At the end of the day, if the Congress or the President said, “Yeah, we see a ruling supreme court, but we don’t care, we’re going to keep doing this.” It’s not exactly clear what the court could do. And the main check against that might be something like, well, the public might view that as a big violation of norms of the American politics and punish that electorally or something like that.
But given that the public’s views of the court seem to have gone dramatically downhill in a lot of ways in recent years, I think these questions are all the more important. So I think that our goal here is to characterize these things, and while we don’t directly say the court should follow public opinion or it should follow public opinion this much, I do think there are important concerns about what might happen if the court and its decisions drift further and further away from the views of the average American.
And increasingly you’ve seen democratic politicians, governors and others even begin to suggest that maybe we should ignore what the court is saying if it rules say, against abortion rights or other things. And I think it’s an important but tricky question, how that’s going to operate, how that depends on the correspondence between the court’s rulings and public opinion. And so it’s actually something we’re trying to look at a little more in future work, even if that’s pushing at least me a little farther down the normative path than I might usually be used to. I definitely think those are important questions.
Matt Grossmann: So the Dobbs decision is probably the best example of a real world mechanism that operates somewhat similarly to what you might expect from your findings that the court rules against public opinion in a major, high salient way, the public notices, there’s at least an electoral backlash associated with it. But I kind of have a hard time extending from that to 10 other issues. That is, not that you couldn’t possibly think of that happening repeatedly, but since that’s about the best example you could come up with, it kind of makes you think, how likely is this to occur in the real world that the public would notice these moves continually, react continually? So how much do you expect that kind of mechanism to operate and how unique is the abortion case here?
Stephen Jessee: Yeah, I think I agree with your characterization on that. I think Dobbs is one of these once, probably more than once in a decade, maybe once in a lifetime, decisions where virtually every American who’s paying attention at all hears about it, at least in some form, and essentially comprehends hopefully that the basic idea of what the court’s doing is essentially overruling Roe v. Wade and basically removing a constitutional right to choose to have an abortion.
So I think in the case of Dobbs, it’s likely to have a big impact on the public’s views of the court, and for that to happen pretty directly. Obviously people aren’t hearing directly from the court, they’re not going to the building and hearing the decision announced or even reading the opinion directly from that the court wrote. But they’re certainly hearing from the media and others the gist of what’s been decided.
And so I think that’s pretty directly going to operate to perhaps create backlash or at least inform perceptions of the court. We haven’t fully analyzed the wave of data that was collected since Dobbs so we can’t say exactly how that’s operating, but I have pretty strong suspicions.
On the other cases, we haven’t pinned it down and it’s something we’re interested in looking at maybe through something like content analyses of media or politician statements about the court. But I suspect that the typical mechanism is some combination of hearing indirectly about what the court’s done or even hearing characterizations in the media or by politicians of what the court is doing generally. So if you hear Biden in a speech or a press conference talking about how the court is making these radical decisions or taking these dramatically conservative positions, then that’s likely to inform your views one way or another.
So I think it’s one thing that we’d like to analyze a little more directly. It’s kind of tricky to pin down these mechanisms, but I definitely suspect that for most cases it’s not a direct impact. And it’s sort of an overall picture that’s maybe cumulative from what you hear. And as I mentioned, I think a good example of this is the sort of relentless characterization by Republicans and conservatives in past decades of the Supreme Court and really the judicial branch as sort of activist liberal courts and judges. And I think that type of thing has an impact in the long run. I suspect that that’s a big part of the reason why people sort of perceived the court as much more liberal than it was even as late as say 2010, and then again in 2021 when it shifted to the right.
But I suspect that the dominant narrative now is becoming sort of an activist or maybe to use a different term, a court that’s moving things quickly and dramatically in the conservative direction. So yeah, I suspect that mostly people are hearing about that sort of indirectly and holistically rather than about specific cases with a possible exception of things like Dobbs.
Matt Grossmann: Anything you want to tattle out what you’re doing next or anything we didn’t get to that you wanted to include?
Stephen Jessee: So the main thing is we’re just really excited, Neil and Maya and I about the SCOTUS Bull project. I think we plan on doing these annual surveys for the foreseeable future. A couple of projects in progress related to the Dobbs ruling specifically. And we also hope that the SCOTUS full data can serve as a resource for other scholars who are interested in public opinion and the Supreme Court. And now that we have sort of five waves under our belts, we’re starting to think about how we might make that happen in the future. And we’re also thinking about, we’ve started to work on a book manuscript that might hit at some of the broader issues and even some of the normative ones related to some of your questions and other things. So we’re kind of excited to just progress that forward.
Matt Grossmann: There’s a lot more to learn. The Science of Politics is available biweekly from the Niskanen Center. I’m your host, Matt Grossman. If you liked this discussion. Here are the episodes you should check out next, all linked on our website. How the Supreme Court Shapes and Is Shaped By its Public Support, How Court Nominations Polarize Interest Groups, Descriptive Representation in Supreme Court Nominations, How the Federalist Society Changed the Supreme Court Vetting Process, and How Does the Public Move to the Right When Policy Moves Left. Thanks to Joe Yura and Steven Jesse for joining me. Please check out a Decade Long Longitudinal Survey Shows the Supreme Court Is Now Much More Conservative Than the Public and Ideology and Specific Support for the Supreme Court. And then listen in next time.