Democracy and Reporting on Elections

Tuesday, October 19, 2021
Mike Blake/ Reuters

Senior Director, Voter Study Group, Democracy Fund


Vice President for National Program and Outreach, Council on Foreign Relations


Adjunct Senior Fellow, Council on Foreign Relations

Robert Griffin, research director for the Democracy Fund Voter Study Group, discusses voter demographics at the municipal, state, and federal level and what this tells us about civic engagement and interest in local, national, and international issues. Carla Anne Robbins, adjunct senior fellow at CFR and former deputy editorial page editor at the New York Times, hosts the webinar.

This meeting is part of the Diamonstein-Spielvogel Project on the Future of Democracy.

FASKIANOS: Welcome to the Council on Foreign Relations Local Journalists Webinar Series. I’m Irina Faskianos, vice president for the National Program and Outreach at CFR.

As you know, CFR is an independent, nonpartisan organization and think tank focusing on U.S. foreign policy. This webinar is part of CFR’s local journalist initiative created to help you draw connections between the local issues you cover to national and international dynamics. And our programming puts you in touch with CFR resources and expertise on international issues and provides a forum for sharing best practices. I want to remind you all that today’s webinar is on the record and the video and transcript will be posted on our website at CFR.org/localjournalists.

Today we will talk about democracy and reporting on elections with our speaker, Robert Griffin, and host, Carla Anne Robbins. So I’m just going to give you a few highlights from their distinguished careers.

Robert Griffin is a senior research advisor at the Democracy Fund, as well as a research director and participating author for the Democracy Fund Voters Study Group. He currently serves on the editorial committee of PS: Political Science and Politics, the journal of record for the American Political Science Association. And previously he was the associate director of research at the Public Religion Research Institute, PRRI, and served as the director of quantitative analysis at the Center for American Progress.

Carla Anne Robbins is an adjunct senior fellow at CFR. She is faculty director of the Master of International Affairs Program and clinical professor of national security studies at Baruch College’s Marxe School of Public and International Affairs. And previously she was deputy editorial page editor at the New York Times, and chief diplomatic correspondent at the Wall Street Journal.

So, Robert, thank you very much for being with us today. And, Carla, I’m going to turn it now to you to start the conversation before we open up to the group for questions.

ROBBINS: Irina, thank you. It’s great to see you, as always. And one of these days we’re actually going to see each other in person at the Council. I’m looking forward to that really soon. And, Rob, it’s wonderful to meet you, if virtually, rather than just on the phone. And thank you to everybody who’s here, to all the journalists on the phone and/or on virtualandia.

So we have a lot to talk about today. And I know that for many Americans these odd-numbered years are a welcome respite from politics. But for those of us who are in the business, there is no such thing as a rest from politics. And the 2020 election may have been a year past, and the 2022 midterms may still be a year away, but there is—you know, there are a few races out there that are considered bellwethers, particularly one. The post-Census redistricting process is starting, in all of its horror in some places. The Democratic-run Congress so far has not been able to put together enough votes or agree to do away with the filibuster to override the rash of voter suppression legislation legislatures rushed into place in the wake of the 2020 defeat.

So before we get into the thing you really do for a living, the data and the poll crunching, can we start with the what’s in your queue Netflix question? You know, beyond the polls and the data, you know, what events are you watching right now to give you a sense of the country’s political mood?

GRIFFIN: So, first off, let me just say thanks for having me. And it’s also nice to see you. Appreciate the invite to speak to everyone today.

And if I understand the question, what’s in my Netflix queue to understand the mood of the country? You’re catching me at a little bit of a funny time. I just got back from vacation. But I think, like many people, I was watching Squid Games. So I don’t—I don’t know what exactly that tells us about the American political mood, but it certainly has been a hit.

ROBBINS: I mean, are you watching Virginia closely? Is that a bellwether for you?

GRIFFIN: I see. OK. You were—you were asking something slightly different. I think it’s certainly the case—

ROBBINS: You’re a literal guy, aren’t you? (Laughs.)

GRIFFIN: Listen, I take the questions as they come. And Netflix is a great show. But I think the things—there are certain things that I’m watching for, but there’s also just certain trends that I’m aware of, right? So the first trend I think you have to just be acutely aware of is that there’s a Democrat in office and in power in both the White House, and the control the Senate, and the House, right? So this is a situation in which you generally see the public turn against the party in power, to a certain degree, in off-year elections. And that has been true, you know, with very few, slight, you know, variation in that, you know, throughout most of sort of the latter half of the 20th century and into the 21st century.

So—you know, so I think there’s just a context in which we kind of expect the country to be swinging against Joe Biden somewhat naturally, and against the Democratic Party to a certain degree in these midterm elections and some of these off-year elections. Virginia is certainly, I think, a race to watch, because it’s going to, I think, start to inform us a little bit—we have to be careful not to over-read into these things—but it’s going to start to inform us about whether the Democratic turnout advantages that we saw in some of the off-year elections after President Trump became president—whether those are going to extend in any way, or whether we’re really going to see the same dynamic that we see in most places.

The reason that we might see any kind of variation there this time around is because the political demography of the coalitions has changed somewhat, right? There has very notably become this education split among White voters and, to a lesser extent, among Black, and Latino, and API voters that we’re seeing. So what that means is, as an off-year election, people with higher educations are just much likelier to turnout than people without college educations. That’s true in general elections, but that difference gets exaggerated in these sort of off-year, lower turnout elections. So that is at least a structural force that could well advantage the Democrats. But we don’t know yet, right? It’s just—it’s hypothetical. We haven’t had that many midterms like this to kind of compare to. And it’ll be a little bit of a datapoint in everybody’s head as we kind of move forward.

ROBBINS: So you were sneaky. You got that demographic stuff in there already. But just one other question about bellwether elections. You know, people criticize political reporters a lot as treating everything as a horserace. And Jay Rosen, you know, has said we should really spend a lot of time asking citizens what they want covered and stop treating, you know, prediction as the essence of it. And even if you—you know, a bellwether, of course, is not just predicting whether Terry McAuliffe is going to get reelected. It’s what can we learn, whether Terry McAuliffe wins or loses, rather than the fact that someone finds Terry McAuliffe attractive or annoying. Although I’m sort of leaning in one direction; you can guess which one it is. (Laughs.) But, you know, is it useful? I mean, do bellwethers actually tell you something? And is one race going to tell us something? Because everybody’s going to get really excited by the results of that Virginia election.

GRIFFIN: Yeah. I think the reality is that an accumulation of races can start to—start to tell you things, right, multiple datapoints. But any single race can be filled with idiosyncrasies, and there are reasons it could be kind of strange to look at. I think that there—we tend to overread into these things. And I think you’re right to also just point out, right, what is the—sometimes the utility of reading everything as a reference point simply for the next major election right? What does this race tell us about this, rather than what does it tell us about the lives of Virginians, and what that’s going to look like based on who these two candidates are?

I think there’s an unfortunate supply and demand issue there as well, right? It is the case naturally, I think, also that parts of the electorate do want these horseracey type elements. Especially, by the way, if you don’t live in Virginia. What matters to you how this race goes, right? So there’s a kind of hobbyism that can come from watching this. And, you know, people can get kind of obsessed just with the horserace element of it, which I think, as you’re kind of pointing out sometimes, the utility of it can be kind of limited in some weird way. You know, it’s—we’re going to know the results of the election, you know, at some point, right? And it doesn’t—you know, the public polling that occurs around this, the commentary about who’s up, down, left, right, and sideways in these races—it ultimately doesn’t amount to much once we actually have the election results and we know the—we know—(inaudible).

ROBBINS: So now onto your day job. So we were chatting briefly last evening, and you raised a really interesting point about how different the electorate is from—you know, and how different, you know, the people who actually vote looks from the eligible voter population in the country. And how those differences, I think I got this right, get even greater as you move from the national elections to state, and all the way down to local elections. So what should we know about who actually decides to vote? And particularly if we’re local reporters, you know, if that gap is even greater, you know, can you talk about that gap and talk about what it means? You know, who’s deciding to vote and why? Maybe that should be the second set of questions. You know, who’s deciding to vote and what is that gap?

GRIFFIN: Yeah. You know, and just a little vocabulary here, right? So two groups I’ll talk about: voters, right, people who actually show up at the polls, and eligible voters, people who are over eighteen, citizens of the United States, otherwise meet the requirements to be able to vote in their state. And what we see is that pretty systematically across any election you could care to mention, you see the systematic underrepresentation of certain groups and other representation of others, right? And it falls along a lot of the same lines that we are used to seeing this kind of phenomena fall along, right?

So they are better educated, they are higher income, they are holder, they are less racially diverse. The one thing that probably cuts against it is that the electorate—voters, I should say—are actually—there’s slightly more women than men, right? So it cuts against a sort of classic feature of hierarchy in the United States. But otherwise, it follows along a lot of the dimensions that we kind of expect it to. So that’s even in a general election. It’s even in really high turnout elections, like the one we just had in 2020.

Now, once we start talking about these lower turnout elections, all of those sort of systematic biases just start to get exacerbated a bit, right? So, again, it becomes even more educated than usual, it gets even older than normal, right? It’s even less racially diverse sometimes. That varies by state in some places. But that’s just a dynamic that I think it’s important to be aware of, because ultimately these elections are, you know, who is showing up, you know, to be represented in this society really has an impact, right?

So when we’re talking about the opinions of Virginians, right, in this upcoming race, or the opinions of people sometimes in mayoral races, right, that’s only 10 to 15 percent of the eligible voters in that area sometimes, right? We’re really talking sometimes about a small slice. And I think as we’re framing up the choices that people are making, it’s always important to sort of recognize that, you know, elections are one of the mechanisms that we have to try to represent people, but the job of government is really to represent everyone, regardless of whether or not they vote. So keeping in mind some of these systematic biases I, I think, just a good leveling mechanism when we talk about these things, when we try to think about how we’re representing everyone in society.

ROBBINS: So we have notoriously low voter turnout in this country. And you’re also talking about an inequality that manifests itself. And I am actually surprised that it’s worse on a local level because, you know, it would seem to me, just in sheer civicmindedness—because we know consistently that people tend to like their local officials more than they like their national officials. We know that people like their local newspapers more than they like their national newspapers. So if people like where they live, or they like what they can see much directly, you’d think they would turn out to vote more often. So why is voter turnout so low? And why is this inequality so manifest particularly on a local level?

GRIFFIN: Yeah, you know, I think there’s a long and wide debate about this within political science. You could bring in all sorts of pieces of this puzzle. I think I tend to boil it down to some rather simple elements, which is to say most of them are structural in nature, right? We can talk about cultural differences between places and senses of civic participation and things of that nature, and I think there’s meat on them bones, but, you know, I think at the end of the day the reality is that we have systems that just sort of incentivize or de-incentivize people’s participation. And in the United States, we actually have a lot of areas where people are disincentivized to vote, right?

So there’s a long history of voter suppression in the United States. We still live with it today even. It happens—you know, there’s varying flavors of that, but the post-2020 behavior by the Republican Party often has been manifesting in a way that certainly looks like an attempt to restrict access to the ballot. I think there’s the reality that there’s actually just a lot of incentives in our system not to vote, right? So the Electoral College, right? We don’t have a national popular vote. We have these things based on states. And ultimately, if you live in Montana or in California, right, something like that, you are living in kind of a lopsided state.

The incentives that exist not just for you as a person to think through “should I vote,” but even for the campaigns and for different non, sort of, candidate organizations that might think about having get out the vote sort of drives or anything like that—the incentives for all those groups change as well, right? So we don’t have necessarily a structure that incentivizes turnout in all fifty states in the same way that some other countries do, who do have things like a national popular vote, where it does just make sense on an individual level and an organizational level and a candidate level to get people to vote as much as possible.

So I think, you know, add all these things up and there’s a lot, I think, less to say, necessarily, about some sense of, like, lack of civic participation. And I tend to put this at the feet mostly of the structures that people are dealing with. Again, culture and things like that are not nothing, but I think they kind of pale in comparison to a lot of the structural stuff.

ROBBINS: So that raises two questions. The first one, you used the term “voter suppression.” If I live in a state which has recently passed or is debating legislation—new voter stop the steal legislation—is there research? Because, obviously, no legislator who’s adopting this is going to describe this as voter suppression legislation. They’re going to make a persuasive argument for why twenty-four hour drop-boxes, you know, are dangerous. That, you know, nobody’s paying attention to them.

You can see why they, you know, have a ballot stuffing potential. Or, you know, I mean, there are—you know, reasonable people can make a reasonable argument for why everything, you know, in that Texas law doesn’t sound like a bad thing. Are there people who have actually done systematic research that says that if you adopt this change that it actually leads to fewer people turning out to vote, or fewer people of a certain class, race, or ethnicity turning out to vote?

GRIFFIN: So I think the reality of this stuff can actually be kind of difficult to measure in a systematic way. And let me talk about why, right? So if I study drugs, right, or something like that, I’m able to give an aspirin to everybody. And the aspirin kind of looks the same, and the context under which people take an aspirin in the same, and I can go about studying the effect of these things pretty easily. The issue is when you start to get into different types of structures around elections, it’s going to be different by every state. The context is going to be different in every state. The election itself is going to be different in every state. So it’s actually kind of difficult—I would say, even as the people who are trying to structure these things.

And I won’t make a bad faith argument and say, right, like there are people out there who are making, I think, some of these arguments in bad faith. And their intent, at the very least, is to try to get certain types of people at the margins to turnout at a lower level. And I don’t think it’s necessarily the case that they’re going to be effective in every case, right, because there’s a whole organic system that exists, right? Candidates are aware of attempts to suppress the votes of certain types of people, and therefore can raise extra money and put extra attention onto certain communities to try to raise their turnout. The same thing can happen with organizations that aren’t associated with candidates but are just sort of GOTV things.

So the—looking at the effectiveness of these things I think is slightly different than looking at the intent. The effectiveness research actually tends to home in on effects that are actually relatively close to zero sometimes. That is to say that they kind of push and pull around that zero point. Sometimes they increase turnout; sometimes they decrease turnout. But I think that’s a little—I think there’s a different conversation potentially to have from the moral implications of this. One is to—what does it mean for certain groups of people to want fewer people to vote, which appears to be intent in many of these cases? And I think there’s another reason to talk about this, which is just to say, like, the efficacy of this stuff is often blunted, I think, by the efforts of those on the ground often to counteract them. So just studying this is even kind of a difficult thing to get at.

ROBBINS: So your hair isn’t on fire about this stuff?

GRIFFIN: In the sense that I think it will be able to bet counteracted. I think it’s often the case that it can be. But I think there is a—there is a higher level at which to have this, which is not just strategic, but I think speaks to our values and our morals, and what kind of country do we want to be. Let’s, say, groups shouldn’t have to go around, you know, constantly counteracting the effects of laws that ultimately are trying to restrict the ballot. It should be easy for people to vote. We should encourage people to vote in our democracy. That is a good thing that we should ultimately—especially under the conditions we’re under, which is a country with extremely low levels of fraud, right, that it’s really bordering on nonexistent—that we should be so consumed by the need for all the security around this stuff, just strikes me as the wrong balance.

ROBBINS: So these things—so you’re not saying that these things aren’t dissuasive. You’re saying that they could be counteracted because if you had Stacey Abrams on your team you can still get people out to vote—or Taylor Swift.

GRIFFIN: That can be—yeah, that can be one story that happens, right? But then sometimes too it’s death by a thousand cuts, right? So sometimes it is the interlocking nature of various separate laws that—like, sure this law wasn’t—didn’t have an effect in Iowa, but I don’t know that it’s not going to have an effect in Kansas, right? Because maybe Kansas has an additional piece of law that it interacts with, and therefore causes a larger effect. But so, you know, it is actually quite difficult, I think, to forecast what the effect of these things are just because of that complexity that I’m trying to point at. But I think, you know, backing up from the question of can we accurately forecast, and predict, and estimate exactly how these laws are going to have an effect? I think there’s a larger conversation, I think, just to have that’s not at a strategic level but really is a question of what should we be doing, what should we be up to, what values should we be promoting within our society?

ROBBINS: So it is 1:21, and we have a whole bunch of really talented journalists on this call. So I would love for people to jump in with questions. I always have an infinite number of questions, and particularly because this isn’t my area of expertise, so I got lots of questions. So please raise your hand, you know, put questions in the Q&A, because I’ll just keep going but I would, you know, love to have more people in this conversation. So don’t wait for me to just go silent for the time. So please jump in here.

So for me, you know, when you say structural problems, I mean, lots of—lots of countries do things like they mandate people to vote. So they have very high turnout. You know, we don’t do those sort of things. So we—the word “mandate,” obviously, given—(laughs)—vaccine mandates, mask mandates—we’re, obviously, not going to go down that road. But there are other things that—I mean, New York has, like, primaries on, like, four different days. You know, there’s just—there’s lots of confusion there. People get burned out or they get confused. They’re not even sure. I mean, New York City, they’re not even sure where to vote because there’s so much confusing about changing polling places, because they have an incompetent, you know, group of people running the elections.

Is there—are there places in the country that do a better job of running elections? That if I were to write a story about how to, you know, fix the system—beyond the malign intent issue. You know, what’s a better way to run an election, to get people to vote, to find it less painful to vote, to find it easier to vote? Are there places to go and look? What’s the Camelot of elections in the United States?

GRIFFIN: Boy, I would be—you know, I am always careful to stay in my lane and know exactly when I can speak with authority. So I won’t point out a specific place. But I will say—I will generally say that places that are giving people more options, more time to be able to vote, more ways to be able to register. Maybe the state is even taking responsibility for registering people itself, right? Maybe it does it through an automatic voter registration process, like they have in Oregon. These are—these are ultimately things that are good, right? Make it easier for people to vote. Have things that just—lower requirements in general, right? We can have a reasonable level of security while also making it—the ballot accessible to folks. Again, I’d be hesitant to—especially in a moving landscape, I think, like the one we’re currently in, to sit around and sort of make too much—point to one example.

ROBBINS: So I gather we have a raised hand. So—which would be great. So Charles Robinson at Maryland Public Television, please join the conversation. Yes, Charles, are you there? Great, hi.

Q: Hi. First and foremost, thanks for this. A couple little things.

First and foremost, I’m looking at the race—the Democratic race down between—in Palm Beach County—for Representative Alcee Hastings’ old seat. There are nine Democrats that are vying for that seat. And I kind of look at what happened in the St. Louis market, where there were some older African American politicians. I want to know, are you seeing trends in that nature? And the last part of that, what are this year’s buzzwords for the elections?

GRIFFIN: So, Charles, thank you for the question. If I can ask a follow up to—when you say trends, can you just expand on that just a little bit, so I make sure I’m talking about the right thing?

Q: So in St. Louis, an incumbent was running against a newcomer—I think it’s Bailey. And she overtook him because he was not speaking to what the issue of the day was at that point. It was policing issues. Because he was not talking to that he was basically soundly defeated. In Palm Beach, just to let you know—I used to live in West Palm so I always kind of watch it and I actually knew Alcee Hastings. And Hastings was kind of this unique phenomenon. You know, he was a judge, got kicked out, then he came back and ran for the seat. And he was this kind of iconic figure. In South Florida there are nine Democrats vying for that seat. And I want to know, is that a trend? What happened in St. Louis, is that a trend or is that just, say, a blip on the screen?

GRIFFIN: Yeah. I mean, I think one of the things to talk about there is that, you know, it takes multiple data points to make a trend, if you know what I mean. Yeah, so, again, it’s a fair enough question. I tend to always back away from looking at one race as sort of indicative of anything and I try to look for multiple things. I think this is still a fight that’s being pushed within the Democratic Party, especially around policing and issues related to racial attitudes and systemic racism within the United States. I think this is a—I think this is sort of a fight that’s happening within the Democratic Party right now, where people are trying to make this, I think, more of a platform on which ty run ultimately.

I don’t think it’s clear exactly—I don’t want to say wining here in that kind of that—but I’m not sure which side is more holding sway. I think it’s certainly the case relative to even five, ten years ago that we’re having a completely different conversation, right? That this is on the map in a way that it hasn’t been before. I think it’s—I think it’s hard to classify, like, if one of, you know, those sides in that, if you want to classify it that way, is really holding more or less sway. I think that’s still something that’s being hashed out within the Democratic Party itself.

ROBBINS: Great. Charles, did that answer your question?

Q: Yes. Yes, and I just wanted to know, what are some key buzzwords we’re likely to hear this election season?

GRIFFIN: Yeah. So key buzzwords, that’s an interesting one. Whew. I have not talked to any of the Democratic Party marketing departments, but I will—I will do my best to guess at the fact that I think that the Democratic Party is going to probably try to ride out, assuming that it happens by the way, recovering economy, recovering from the pandemic. And, you know, I think the amount of money that was sent out to the American public in various forms over the course of the Biden presidency.

If I had to take a shot in the dark, you know, I think sort of touting some of those potential successes—again, we actually have to see how is the pandemic going to go? How is the economy going to go? Does inflation stay at 5 percent? Things like that. But assuming that those wins are with them, it would not surprise me to see—most of the time, you know, if you have money going out the door, good economy improving conditions, you’re going to see folks who are incumbents especially run on things like that.

ROBBINS: So this sort of segues into you have a new report coming out this week. And I know it’s embargoed until Thursday, but there’s some very interesting insights into public attitudes towards important issues, including defunding the police, and how that has changed over—you know, over the last year. But this is a much longer comparison. You found issues like greater polarization on issues of race, mainly due to Democrats shifting leftward, but also that, you know, attitudes—that the impact of the murder of George Floyd on, you know, the call to defend the police seemed to melt away over time, even among Democrats. Can you talk to us a little bit about this, with the caveat that this is embargoed? And you also—I think we can share an embargoed version of this, on pain of death, right? (Laughs.)

GRIFFIN: Yeah, no, and I’m happy to talk about it. Again, I’ll just repeat it just to let people know. It’s embargoed till the 21st. So only two days. But the—I think the short version of it, I think, is twofold. So one is, over the last decade we’ve seen this pretty big shift on a number of what you could consider sort of identity-inflected issues, right? So there are questions that come up in the American conversation around systemic racism, right, particularly in regard to African Americans. There’s questions that come up around immigration and immigrants, and their value in society. There’s questions that come up around Muslim Americans in the United States.

And these questions—on these types of questions, that we’ve been tracking for more than a decade, we’ve seen a rather significant and sustained shift leftward among Democrats, and pretty much stability among Republicans, in the United States. So that is not necessarily the most surprising thing in the world to some people. I think you all have been living through some of the same times that I have and might be able to sort of—in a qualitative way—sort of say some of the same things. We’re recording it here with numbers. But just to say that there has been this big shift, even in the ways that political scientists and sociologists and political psychologists try to measure these things. We’ve seen the shifts too.

So I think one of the things to—a couple of things to talk about there. Is how do we relate that to current events? How do we assess those long-term trends, with also talking into consideration what’s happening in the short term, particularly the murder of George Floyd as well as the protests that followed? So I think after the murder of George Floyd and during those protests, you know, just to broadly characterize how some people were thinking about the effect of these events, is some people were ultimately sort of optimistic. And they said that, OK, this is a stair-step point. We’re seeing a shift in public opinions towards police, towards systemic racism, and that, you know, we’re going to live at some new place in American public opinion and people’s attitudes will have shifted.

And I think there were people who were more pessimistic and said: This is a blip on the map, right? This is a flash in the pan. You’re going to see a lot of attention drawn towards this thing in a short period of time, and the American public is ultimately going to return to sort of a—you know, a pre-Floyd level of public opinion on a lot of issues after this happens. And I think the unfortunate story that we tell in our paper is that the pessimists were probably right, at least in the short term. That we had a weekly survey that we were running called Nationscape. And it ran from July 2019 all the way through February 2021. And we were tracking some of the same items over time, so we could on—again, on a weekly basis see how these changes happen.

And what we saw in the direct aftermath of George Floyd’s murder, as well as the rise of all the protests that were happening around the country, was this dramatic shift in attitudes—particularly towards police. But you saw this sort of trickle outward into other questions about do African Americans face discrimination. We saw it in questions that just sort of asked basic long-term questions about systemic racism. We saw shifts in those attitudes in the direct aftermath of George Floyd’s murder and, again, as the protests were going on. What we think happened is that as media coverage itself declined about these issues, as it did over time, you saw pretty much Americans belonging to every political party and every race kind of return to that status quo prior to George Floyd.

So that whatever the effect of those protests might be—and, again, there’s—public opinion is only one way to talk about these things. There’s lots of other stuff at elite levels, or within organizations, or, you know, corporate culture, all kinds of stuff like that that we could talk about. But at least when it came to public opinion, this did that have mark of sort of being a spike in change, and then a slow sort of return to status quo. So, you know, I think as we talk about some of these longer-term shifts that have occurred over the last decades, as well as some of the shifts that happened then, I think it’s just a story that we try to tell in the report that weighs some level of optimism about how we think the country is moving, as well as pessimism about how sticky some of these attitudes are that even big events like this don’t necessarily produce long-term shifts.

And that, you know, importantly, some of these shifts don’t necessarily trickle downstream to changing people’s policy attitudes, right? So we saw almost no shift, just as an example, in people’s feelings towards reparations before and after the murder of George Floyd and the protests, right? That is an attitude that did not shift at all and was relatively kind of related in the conversation that we were having. So there is a limitation in some way towards some of these higher-level attitudes that are shifting among people, and that trickling down to actually support for policies that might address some of these long-term systemic issues.

ROBBINS: Can you talk a little bit about the findings on the Republicans? I mean, you say stability. Does that mean that there was no impact of the Donald Trump era? I mean, one would have thought given the language, I mean, a presidential candidate who started his campaign, you know, talking about Mexicans as rapists and murderers, you know, certainly appealed to a certain—a certain viewpoint, would seem to feed a certain viewpoint. Good people on both sides, you know, refusing to denounce Charlottesville, seemed to—certainly seemed to bring out certain voters or confirm certain viewpoints. Are you saying that it didn’t change fundamentally attitudes among people?

GRIFFIN: So this is a—I will try to not be a survey nerd for—I will try to be for the shortest amount of time possible.

ROBBINS: Oh, go ahead. (Laughs.)

GRIFFIN: But let me put it this way: You know, every survey item that we have isn’t a perfect barometer, right? You all might have, like, a thermometer outside your house, it probably tops out at about 120, right? So there’s a point past which it cannot give you definition and does not give you greater levels of information. So the things that we have or kind of tools that we have within the political psychologist and the political scientist toolkit that have been with us for decades that we have to be able to measure these things over time, but they’re not necessarily built for all these finer-grained, sometimes qualitative changes in how rhetoric has shifted.

So, just as an example, if I had to talk about shifts that had occurred, especially into the Trump era, is that there is generally speaking an era within Republican Party politics where in their pursuit of a southern strategy but, you know, still attempting to retain some level of, you know, respectability among certain types of Whites who didn’t like overt shows of racist rhetoric, there was an era of Republican Party politics that, to a certain degree, gave more shaded versions of this, right? Slightly more veiled rhetoric towards immigrants, immigration, and towards Muslim Americans, and towards African Americans.

What we saw in the Trump era almost certainly, again, qualitatively quite, quite different, right? It has sort of pulled back the veil on a lot of this stuff. So I think this is always the point where you have to sort of do that interplay between some of these qualitative assessments and just realizing that the thermometers that you have in political science or whatever else, or pollster land, are good, right? They can be useful in a lot of ways but aren’t perfect at capturing every nuance. And I think in this case you’re pointing out something that’s worth keeping in mind as you read this. You’re going to see a flat line among Republicans, but it really doesn’t mean that everything is exactly the same as it was ten years ago, because I think even a basic survey of the landscape, just in terms of rhetoric, tells you that we’re operating in pretty different territory these days.

ROBBINS: So there was, you know, this whole question about why the pollsters got 2020 so wrong. And one of the explanations that I’ve read is that they couldn’t get Trump voters to even talk to them. You know, it used to be, you know, you couldn’t get young people because they were all on cellphones and everybody was calling on land lines. And you couldn’t get poor people because they didn’t have land lines. And now we have a new phenomenon, this is the latest explanation, is that you can’t get Trump voters to talk to pollsters because they see them as part of the elite cabal. And Trump certainly kept telling them: You can’t trust the pollsters; you can’t trust the polls. A, is that true? And, B, how does that affect the work that you do?

GRIFFIN: So it certainly could be true. This is—I mean, I’ll put it this way, in terms of saying that often the type of issue that we’re talking about or one of the potential issues in polling—because there are about 100—(laughter)—there really are. It can go sideways on you in all sorts of way. But one of the issues that we talk about is nonresponse or differential response, which is what you’re talking about. Some people are more likely to take a survey, some people are less likely to take surveys. And the ways that we try to compensate for that don’t always work.

And the reason for that is there’s this kind of amorphous, ever shifting and, this is the difficult part, kind of invisible to us response rate issue kind of going on in the background, that we don’t always know the perfect shape of. It’s a many-headed hydra. And we can run around every election beating down certain elements of it, and then the next election cycle it doesn’t even appear, for reasons that sometimes we don’t even necessarily understand perfectly ourselves as researchers. I’m being unfortunately forthright with you in this one, letting you know that pollsters actually have to deal with a lot of ambiguity. There’s a lot that we don’t know about the nature of this stuff sometimes.

So what I would say is that it’s certainly plausible that we were having trouble reaching certain types of Trump supporters. I think that’s a perfectly reasonable assessment. It’s a little hard to know, but we did have a presidential candidate who constantly told people: Don’t trust the polls, and why would you ever take them? And all stuff like that. So, again, seems pretty reasonable. I think, just to give you the flipside of the coin, there’s also a rather reasonable argument to be made that there’s too many, let’s call them, really high participation, high-loyalty Democrats taking, right?

So the pandemic hit. You had differential rates of people staying home, or staying in their house all day, versus going out and doing other stuff. And it might have just been the case that actually Democrats were much more likely to start taking polls, particularly once the pandemic hit. So that this is at least a possibility of what was happening under the hood. So what we started getting was all these people who were really high-participation Democrats. And being a high-participation Democrat also means that you are probably a very loyal Democrat. That the rate at which you might say, like, you know, I’m a Democrat, but I’m actually going to vote for Trump. Or, I voted for Clinton in 2016 but I’m actually going to vote for Trump in 2020.

We might have been missing some of those people too. And, again, it could actually be a mix of these two as well. There’s no reason that you can only have one problem when you can actually have tons of problems. So, you know, I think on the whole it’s an election year that I think the polling is going to continue to try to learn from and try to improve from. There’s no signs to me explicitly that things have necessarily gotten better yet, but I think we just have to kind of—you keep moving forward and you keep trying to understand the world as best you can.

ROBBINS: So I believe that people should vote, even if they don’t vote the way I like them to vote. And so Charles Robinson has asked a question and Irina has asked a question. Irina has her secret backchannel with me. And that is, well, let’s put Irina’s question first, which is: What do you expect for election participation this year, given COVID and last year’s election results? And Charles then asked: Are there issues which could flip attitudes and get people to the polls or away from the polls? Are there particular things that you think are big draws? And then I have one follow-up question, which is about the role of the press in all this. I mean, what’s our civic responsibility and our ability to—you know, to get people to participate and trust the process?

GRIFFIN: Yeah. So let’s tackle those one by one, and if I forget some in the middle just remind me.

ROBBINS: You’re young. You can’t forget things. (Laughs.)

GRIFFIN: But let’s start with—let’s start with turnout. So when it—when it comes to, like, what we can expect going forward, I think the reality in all sorts of ways is probably going to be lower, right? We just had almost a once-in-a-century level turnout. And there’s a lot of ways in which we think: This is great, right? We want people participating. But also realize that this was spurred from a high sense of crisis, that people’s participation level also came from thinking this was a very consequential, potentially crisis-level type of election that we need to be in. So there is an interplay there where we want people to participate, but we don’t necessarily want that driven by politics that feel existential in nature.

So I think on the whole it is probably the case that the temperature has lowered from the Trump era. I think that’s probably a true thing. Time will tell. But I think it’s the case that it probably has happened. So we’ll probably see lower turnout, not just as a function of the fact that, like, we’re going to have some midterm elections here and they’re going to be lower turnout than general elections, because that always happens. But even going into 2024, we had a once-in-a-century level of turnout. It was the highest it had been in a century. We’re probably going to go down from there.

So, you know, the reality is that you will that sort of fall off in pretty expected ways, right? All those groups that were slightly kind of better represented—younger voters were a little bit better represented, you saw higher levels of participation potentially among Latino and API voters this election cycle. There’s a possibility you’ll see some drop off in those groups that saw exceptionally large gains in 2020. That would be the likeliest place to look for it. Now, again, that’s not going to necessarily mean an electorate that’s less diverse, counterintuitively, just because, again, demographic trends are also changing at the same time. So you can kind of have changing turnout of the electorate at the same times as turnout rates kind of shift on you as well. So it might still be the case that these things don’t necessarily net out in a way that actually makes the electorate less diverse just than, let’s say, 2020. But turnout, almost certainly going to be lower.

ROBBINS: So existential certainly is turnout doing it. And that’s the way that Newsom ran his campaign in California. And he won, won big. And everyone knows it’s a blue state and he was going to win, but it looked a little questionable there for a while. So is that the real—is that the issue that gets people to the polls? Is that—do we run on yay Trump, oh my God Trump. Is that the guaranteed turnout?

GRIFFIN: I mean, I think there are—there’s decent evidence from the Trump era, so when he was president, that he was pushing turnout, certainly during the midterms, higher for the—midterms and off-year elections, higher for the Democratic Party and Democratic constituencies. And then in the general election, again, turnout just sort of went off the scale for everybody. And it did so during a pandemic, which is also kind of, you know, crazy. So I think there’s good evidence for it during the Trump era, that instigated higher levels of turnout, particularly among some groups.

It’s a really open question what’s going to happen now, because, you know, essentially how much is Trump on the ballot still? If you want to think about it that way. And how much is he a force within our politics is a completely open question. How much are we still going to see sort of that Democratic invigoration that we saw in the midterms in those other years translate forward? Or, again, does it just shift on us, right, in the way that it has in the past, in that parties out of power tend to see surges in turnout or surges in energy in these off-year elections? It’s hard to—it’s hard to say whether that sense of existential crisis that accompanied Trump is going to move forward, just because it’s still an open question, I think, how much he’s going to continue to be a piece of our politics.

ROBBINS: I’m feeling existential crisis.

We have a question from—and I apologize is it Lionel or Lionel Ramos?

Q: It is Lionel Ramos. And first off, thank you guys for this webinar. And I am covering race and equity for Oklahoma Watched, based out of OKC but we cover most of the state.

So I’m going to—or, I’m soon to be covering a gubernatorial race here in Oklahoma, 2022. But so the incumbent, Governor Kevin Stitt, is running against the superintendent of public health instruction—or—public construction, sorry, Joy Hofmeister. But she’s switched—they’re both known Republicans in the state. Hofmeister switched parties when she announced her race. So I’m kind of wondering, you know, what are some things to look out for in a race like this, in a time like this, where everybody’s understanding of the parties is kind of very polarized? And what that might mean for turnout and the kinds of policies that come out of that kind of—that kind of governor, who is a Republican in everybody’s eyes but is running in the other party. What might that look like for the party in the future, the Democrats?

GRIFFIN: Yeah. And thank you for the—thank you for the question. I would be hesitant to prognosticate about the policies that might come out of such administration. But I guess I would say what’s going to be happening with voters in general is that they’re going to be getting this mixed signal. There’s going to be this previous knowledge relative to, frankly, all of the advertising that the campaigns and outside groups are going to try to do to stick the two candidates with their basic party labels, or to distance them from those labels in a lot of ways. So it’s a really—I think that’s a—it’s a really interesting race to actually talk about, thinking about party loyalty in an era when party loyalty is very, very high. Partisanship is very, very high. But you do have people potentially switching their labels mid-career like that.

I think that what’s unfortunate is the number of data points we have around that is actually pretty low, right? So there’s not a lot of candidates that necessarily do that all the time, particularly in sort of, you know, let’s say, the last decade or something like that. So I’m actually—I can come to you, I guess, with not a prognostication but a question, which is just to say I think this is an excellent test case for actually thinking about how people interact with individuals like that in an era of deep partisanship. Does that person’s partisan label start to override whatever previous history they’ve had, which that would then suggest to me that, you know, partisanship is exactly as bad as we kind of think it is, if not worse.

Or, do people kind of take this person as their own individual candidate and assess them as such? The reality, just to give a more historical or national context to the thing, is that people split-ticket voting—that is to say, crossover voting in some way—is almost at historic lows at present, right? Election over election, you know, the voting patterns that we’re seeing are incredibly stable. There’s something like a 99 percent correlation between 2016 results and 2020 results. And then if you also—like, that’s at the presidential level. But then also if you say between Senate races and presidential results, or House results and Senate results, or the California race and presidential results—these things have all gotten wrapped up in each other in a way that they tend to kind of move with the same trends. And it’ll be really interesting, I think, to see if there’s somebody who can buck that in some way.

ROBBINS: So—I’m sorry. Lionel, does that answer your question?

Q: Yeah. Definitely. I’m curious to know—and I’ll try not to phrase it in a way that’ll make you prognosticate, but—(laughter)—I’m curious to know how people will vote. Whether, like you said, people are going to vote—people who typically vote Republican are going to vote Democratic more so in this case, or people who—or, will Democrats have less turnout, you know what I mean, just in general? So I have some questions that I’m thinking about going forward that I guess are hard to put our finger on, you know?

ROBBINS: I mean, again, without prognosticating I’d say you have your—it feels like you’ve got your fingers on the right questions, if you know what I mean. Like, that is to say, like, what does crossover voting look like with this person who is potentially somewhat unique, or, you know, has a potential appeal or lack of appeal with these classic constituencies within this area? Again, you know, I’m always a person that tries to sort of generalize outwards from lots of data points.

But you are presenting me with sort of a unique individual that’s up against some very strong national trends and some other strong phenomena that are occurring generally within people’s voting behavior in the U.S. Again, I would at the very least say the opportunity for those things that you are highlighting, it seems like if you were looking for a situation in which people would sort of break the mold a little bit and do the types of things you’re talking about, this is about as ideal as it gets. Somebody who’s prominent enough, well known, was belonging to another party and then switching over time.

ROBBINS: Could I—could I sort of argue for thinking about it a different way? Keeping in mind that me talking about domestic politics is like driving without a license, since I’m a national security reporter. But I’m married to a political reporter. (Laughs.) Which is, when Biden ran, he made the argument that he worked with Republicans, you know, that he crossed—you know, that he was an aisle-crosser, and that this was, you know, sort of the old style of politics. I don’t know anything about this candidate, Lionel, but when you mentioned it, you know, I thought to myself it could be two things.

It could be like Richard Shelby changing parties in Alabama, and just saying: You know, the Democrats are too liberal for me. I got to go to the Republicans. You know, this is a different party and I’m walking away. Or it could be—or some Democrats could, you know, embrace her and, for completely partisan reasons, saying: Oh my God, she’s seen the light. And—or, you know, independents and some Republicans could say—and some sort of centrist Democrats—could say: Oh my God, finally. Somebody who wants to, like, bring us together here. And a lot of that might have to do how she campaigns, doesn’t it Robert? Am I missing this?

GRIFFIN: No, I think that’s right. And you’re making me think of an additional point that I think is just worthwhile. One of—there’s a really great finding within political science that essentially says moderate candidates do better in general elections relative to more extreme candidates. And what’s important to caveat with this, or how to think about it, is that the public doesn’t necessarily have, like, you know, everyone’s, you know, Medicare plans sitting in their head so they can think, ah, this person is extreme versus this person is moderate. They are looking for signals of moderation. They are looking for the appearance of moderation, if not the reality of it. Although, you know, sometimes they’re related, sometimes they’re not. In this case, though, crossing parties? That’s a pretty strong moderation trigger, right? And generally speaking, moderates do better when running in these general election races.

ROBBINS: So my question is, and I must admit not I’m thinking, like, more as an editorial writer, but I do think that this doesn’t have to be prescriptive writing. This can be explicative writing—not expletive writing—which is—you did a report on the crisis in confidence in our election system after the 2020 vote. And you had some really striking numbers in it. You said the percentage of Trump—first of all, you pointed out that there’s always this sort of sore loser phenomenon. Not your term, my term. And that, you know, people—we was robbed is a sort of common thing. But most of the time people snap back after this.

But people didn’t snap back after 2020. And you notice that the percentage of Trump voters in 2020 who said they were not at all confident that their vote was tallied accurately was more than four times as high as the percentage of Clinton votes who said the same in 2016—35 percent versus 8 percent. And that after this last election just 34 percent of Trump voters said they would accept Biden as the legitimate president. You know, and even in 2000, after the mess of 2000, you know, 68 percent of people said they would accept—you know, 68 percent of the people who were Democrats said they would accept Bush, who voted for Gore.

That’s pretty striking, this difference. Now, it’s not surprising, given all the things that Trump has said. We, as reporters, really struggle with this question of how do you deal with the disinformation that’s coming out? The, you know, it’s all—it’s all rigged, it’s all the dishonesty that comes out, the big lie stuff that’s coming out? How do you do this—how do we cover these things if we want people to trust the system, which I think is a civic good, and not necessarily taking sides? How do you get people to just trust the basic mechanics of elections, given the fact that we basically have a year between now and the election? You know, what stories do you cover and how do you cover them? Because I think we do have a civil responsibility as journalists to do that.

GRIFFIN: Yeah, you know, I think—I think the thing to keep in mind here is that the biggest moving piece is really going to be the parties, at some level, right? The American—like, every election cycle you’re going to get people who are just naturally not going to feel good about an election because they lost. That’s a natural thing that happens to people. And they’re looking for reasons why that didn’t happen, and therefore they’re coming up with all this stuff. But it’s ultimately the rituals of our democracy, right, that the parties that they ran against, they know it was a legitimate election. This person actually did win. You know, they’re the president now. We all have to get behind them. You know, they’re president for everybody.

This is a really common feature of our democracy, and it just didn’t happen this time. And in fact, it didn’t just happen in the sense that you had a person like Trump just say, well, you know, I’m not sure. Or, you know, like not be conciliatory. But really going on the offensive and trying to spread misinformation and disinformation about this election. So what’s really happening here is this is going to, in some ways, be an elite story, right? So I think there’s a way in which holding elected officials accountable for this, leading with information, right, with the truth and not necessarily what some of these folks are saying often, is going to be some of the more powerful tools that we have at our disposal because it’s really going to be these candidates and the party apparatuses that are going to have a big effect on what people end up thinking.

Again, it’s not unusual to see people be sour grapes. But there are rituals that help close the wound after an election. And this really takes place on the part of elected officials. And this time there was just a total failure to do this at the highest levels in significant portions of the Republican Party and significant portions of the conservative media. And that effort is, frankly, ongoing. It’s a thing that we are still talking about because there are folks who have still not taken—have still not done the responsible thing here, which is to help close the wounds after an election, so you know this was actually legitimate.

So, again, I think—again, I am not a journalist. But if I were to think about the intervention points there, there are going to people who are trying to spread mis- and disinformation. I think lead with the truth early and often. I think calling lies lies I think is, just from my own perspective, an important thing. And I think there’s really trying to hold people accountable, if they are individuals who at some point have taken part in trying to spread mis- and disinformation, that they don’t kind of get a free pass in the future, right? This is ultimately an attack on the lifeblood of our democracy, which is elections. And that’s sort of, you know, a very offensive thing within civil society, that they would sort of take this move and make that action.

ROBBINS: Well, we got a million more questions but we’re running out of time. So what—rather than your answering them, what we’re going to advertise that we’re going to share—because you’re going to make you share it with us—is information on where journalists can—you mentioned to me that there were tools that they could use to take a look at their redistricting process to make sure that it’s fair, which we’re going to share. We’re going to push out to people. And also where journalists can find relative data about voting demographics in their communities and on the national level. Any other sort of cool data or information, you know, tools that they can be using as they—as they follow elections in our communities?

GRIFFIN: Sure. And I’ll try to hit up a bunch of those. Anyone here, if you want to feel free to provide my email address, I do a lot of work on political demography. Anyone should feel free to reach out to me if you’re looking for information on that. Sometimes I’ll have it; sometimes I won’t. But I’ll always try to point you the right direction if I can. We didn’t get to talk about it today, but redistricting is a really big story. For anybody who’s like more information about that from experts, I would suggest PlanScore.org. Again, it’s PlanScore.org. It’s an excellent organization that is trying to, in real time, take the maps that are being proposed in all these states and try to walk through some of the expert metrics that we have for trying to say whether these things are fair or not, whether they really represent the people in that state or not. I think the site itself is rather excellent at explaining this stuff. It also can connect you with experts. And just in general, even if there’s any other requests, we at the voter study group always try to be a resource to journalists. So feel free to reach out at any time.

ROBBINS: And we will—we’ll go to Robert with some other questions. And if you have questions, send it to us or to Robert, and we’ll pass them on, about how to find data. And, you know, FiveThirtyEight has a—they’re updating regularly what’s going on state-by-state on redistricting. So what the deadlines are, all of that. Redistricting is really a hot topic, which we didn’t have time to get to. But any other questions that you have, push them to us. We can push them to Robert. And we can squeeze him for data, which he’s really, really good at.

So, Robert, Rob, thank you so much for doing this. It was great. And, Irina, back to you.

FASKIANOS: Thank you both for doing this. We really appreciate it. And to echo what Carla said, we will share the resources that Rob mentioned, as well as his email address. You can follow Carla on Twitter at @robbinscarla, and Robert at @rp_griffin. And please visit CFR.org, ForeignAffairs.com for the latest developments and analysis on international trends and how they are affecting the United States. And as always, we encourage you to email us for suggestions of future webinars. You can email [email protected]. And we would be happy to stand those up and help you with any other resources that you may be needing.

So thank you, again, for being with us. And stay safe out there.


Top Stories on CFR

NATO (North Atlantic Treaty Organization)

The war in Ukraine marks a new era of instability in Europe. Countering Russia’s efforts will require a stronger, more coordinated NATO.


After the rise of Chinese power during the 2010s and failed U.S. policies in the Indo-Pacific, the United States should renew the Pivot to Asia and place the region at the center of its grand strategy.*


Far-right advances in the European Parliament elections have destabilized politics in France, a longstanding pillar of the European Union, and highlighted fault lines in the bloc.