Government professor Brendan Nyhan’s research has gained new relevance in light of the recent presidential election. After graduating from Swarthmore College with high honors, he ran a fact-checking website for four years before pursuing a Ph.D. in political science at Duke University. His research centers on misinformation in relation to political misperception, and he is a contributor to The Upshot at The New York Times.
What brought you to Dartmouth?
BN: I applied for a job here in 2008-2009 — the year I was finishing my Ph.D. at Duke University. What was most attractive about Dartmouth was the combination of being able to teach really bright undergraduates in small classes and also have a top-tier research faculty. The teacher-scholar model was especially appealing to me because I went to Swarthmore as an undergraduate and I believe in liberal arts education, but I also have research aspirations and Dartmouth seemed like a place where I could fulfill both of those goals.
How would you define the “teacher-scholar model?”
BN: The Dartmouth ideal is someone who is committed to teaching — particularly to undergraduate teaching — while also being an active participant in their research field as well as a contributor to the collective knowledge that academia hopefully produces. The professional system in higher education tends to drive people toward one of those two roles. Most professors are focused on teaching while there’s a small set of people who are focused on research. It’s sometimes difficult to balance those roles or to find places that encourage you to do both the way Dartmouth does. I think the teacher-scholar model is something that is genuinely unique about the College. It’s one of the most appealing aspects of this institution to me.
Do you have any highlights from your time at the College thus far?
BN: I’ve taught lots of great classes with many wonderful students. I’ve worked with students in and out of the classroom, including a number of research assistants and James O. Freedman Presidential Scholars whom I’ve worked with outside of class. I’ve also taught a number of classes, including Government 10, mid-levels and a seminar. The seminar I teach, which is called “Experiments in Politics,” is particularly notable, because the students and I collaborate to design, execute and analyze a research project. When I taught the class in 2014, for instance, we published the resulting study in a real political science journal. The way the seminar works is that students each propose a study we could do that would advance our understanding of political misperceptions and conspiracy theories, the topic I study, which allows me to be a more effective advisor. In the spring of 2014, one of the students designed a project we ultimately ran that looked at the effect of redactions in government documents on belief of conspiracy theories. When people see a government document that’s had sections blacked out, does that make them more likely to believe in a conspiracy theory? There’s often an intuition that something’s being hidden by the process of redaction. We were able to test this hypothesis systematically and published our findings in the Journal of Experimental Political Science. We also wrote about the study on the Monkey Cage blog at the Washington Post. That class, which relies on the support of the office of undergraduate advising and research, is a beautiful example of the research and teaching missions of Dartmouth coming together.
Has the quarter system been of any significance to your experience?
BN: It has pros and cons. Since I’ve been here, I’ve seen how the quarter system offers students flexibility they don’t have at other places in terms of their ability to do internships and foreign study, especially for athletes or other people who have severe constraints on their schedules. However, the terms are really short, so it does mean that teaching is more intense here. There’s also a challenge to keeping students focused — the time that the term allows for them to catch up is much more limited than in a semester system. When I teach Government 10, for instance, I emphasize to students from the beginning that if you fall behind you just won’t have that much time to catch up. In a semester system, a class might be 13 or 14 weeks long. There’s a lot of time, say after a midterm exam, to do extra studying and try to catch up. But in a quarter system, you’re almost at the final by the time you get the midterm back. It’s certainly challenging. I do think the quarter system focuses students’ attention on the topic in a sustained way. The semester system is so long that it doesn’t have the same level of focus. Your instructional time is spread out over more weeks with fewer hours in the classroom together each week. I think the quarter system helps people to lock in on a topic, because you’re only taking three classes, while in a semester system you’d typically be taking four.
Is your focus on misinformation related to the election?
BN: I’ve been studying how people believe in false information for years. That interest predates this election. The campaign certainly made concerns about misinformation more salient then they’ve been before, certainly in my adult lifetime. My interest in that topic goes all the way back to when I graduated from college. I started an early fact-checking website and was one of the founders and editors of that site for four years. Then I started doing research on misperceptions and misinformation in graduate school, focusing on issues like the false belief that Iraq had weapons of mass destruction at the time of the U.S. invasion in 2003. I’ve continued to study the topic since then. During the Obama years, I studied issues like the false beliefs in death panels or that Obama wasn’t born in this country or that he’s Muslim. The 2016 campaign and Donald Trump in particular made the issue even more salient, and so I’m continuing to study and speak about it. In some ways, though, my relative level of concern about misinformation is probably somewhat lower than other concerns I have about the state of our democracy. The level of misinformation we’re seeing now is alarming, but so are some of the other political trends that we’re seeing.
And those trends are?
BN: It’s not my role to speak about my own personal views as a professor or in the public sphere. I try to be very scrupulous about that, but as a political scientist, I feel compelled to speak out about threats to the norms of our democracy. Unfortunately, there’s been a long pattern of Trump violating the norms of what we expect first from a presidential candidate, and then from a president-elect, and now presumably as a president. It remains to be seen how he will govern, but those violations of the norms of our democratic system are very worrisome to me and other people who study politics professionally. It’s worrisome when a candidate is calling the results of an election into question and threatening to jail their opponent. Those are signs of potential political instability of the sort we haven’t seen here in decades. Many observers, including myself, are surprised and concerned about what we’ve seen over the last year and what that means about the American political system going forward. I’ve tried to write and speak about that issue as a political scientist to make sure that people understand how alarmed scholars and experts are. It’s not just me. There are a lot of other people with expertise in these issues who are very concerned, including some who have studied places where there has been significant damage to democracy or worse. They’re seeing signs here of the kinds of developments that occurred in places where the democratic system failed. That doesn’t mean our democracy is about to collapse, but we should take what we’re seeing very seriously.
What impact do you hope publishing your articles in the media will have?
BN: I’ve been writing non-academic articles for a long time. The goal of that writing is to bring academic knowledge and insights to the mainstream public debate. Too often academics are cloistered from the public and fail to contribute to public debates, which they then bemoan for lacking adequate sophistication. My feeling is that scholars should be a part of that debate. We have a lot of expertise to contribute. We certainly don’t have a monopoly on knowledge, though.
One of the things I’ve tried to do is to interact with political journalists and show how political science can strengthen what they do. Political science is a complement to journalism, not a substitute. It can make reporting and analysis in the media better. Just in the time that I’ve been writing professionally, we’ve seen a generation of political journalists rise up in the profession who routinely bring political science into their coverage in a way that wasn’t true in the past. That’s been driven by the rise of new publishing platforms that have helped academics share their insight — first blogs and then Twitter. Academics have become a part of the conversation in a way that wasn’t true in the past. I think that’s important and part of the service that we owe as a profession to the society that makes our work possible. I think our contributions to public debate also help demonstrate the value of the insights we can provide. We’re certainly not always right — the 2016 election showed that political science doesn’t have all the answers — but we can help our society think about the issues facing American democracy, and I think that’s a valuable contribution.
What makes misinformation believable?
BN: That’s a big question. I teach a whole class called “Political Misinformation and Conspiracy Theories” about that topic, which I’m offering in the spring. It’s a complicated question, but let me give you some quick ideas. The first point to make is that we all believe false things. There are lots of reasons why; there’s no one answer. In politics, though, we may be especially predisposed to what’s called motivated reasoning — the idea that we have a preference for which side is right which influences the information we choose to accept as valid and the information we reject. Over the history of the United States, many prominent misperceptions and conspiracy theories are not partisan or ideological. But in the contemporary period, politics is highly partisan and ideological, so the misperceptions that are often the most common are frequently related to which political tribe you want to win. The explanation that’s guided a lot of my research in this area is that people are interacting with information in a biased way. As human beings, we’re all more vulnerable to false information when it seems to confirm our prior beliefs and predisposed to resist or reject it when it seems to contradict them. People are willing to change their mind under certain conditions, but we’re all more prone to motivated reasoning than we’d like to admit. The intense partisanship of recent years has made the kind of political misinformation that exploits motivated reasoning seemingly more prevalent.
Another concern is that your preferences may not affect just what information you choose to believe, but what information you choose to consume. Not always — most people don’t follow politics closely enough or care enough to filter their information extensively. It’s sometimes overstated how many people are engaging in that sort of behavior. However, the kind of person who listens to political podcasts or reads a lot of news might make an extra effort. Those are the people who are most likely to construct a kind of information echo chamber around themselves. So it’s a subset of the population, but it’s often the most politically active ones. Those may be the folks who are most likely to be in an echo chamber and to be exposed to a lot of misperceptions and maybe not learn why they’re incorrect.
How do you feel about the rise of data journalism?
BN: On the whole, the rise of data journalism has been great. Media coverage of politics and polls is much more sophisticated about data than it used to be. Quantitative analysis is part of our toolkit for understanding the world and journalists should be using it just like every other sphere of society. With that said, I think sometimes people overstate the extent to which data journalism is a replacement for traditional journalism. They’re complementary approaches and not necessarily opposed. In a lot of cases, data journalism can inform traditional reporting and vice versa. The best journalism blends elements of both. I think the news organizations that have been the most effective in using data journalism have married it to more traditional approaches and storytelling tactics. Data by itself doesn’t necessarily provide insight or command attention. You need to bring theory to the data to help understand it. Often journalism can provide ideas about how to understand data or where to look for it. In short, data journalism is great, but traditional journalism isn’t going away.
What are you currently working on?
BN: My most recent study looked at the effect of fact checking during the 2016 campaign. The findings were encouraging, because we found some evidence of responsiveness to fact checks even among the respondents we expected to be most resistant to them. My co-authors and I conducted an experiment in which we corrected misleading rhetoric from Trump’s convention speech suggesting that crime in the U.S. had increased dramatically. It is actually way down from historical highs over the medium to long term. There was a very slight uptick in 2015, but relative to twenty years ago, it has declined dramatically. We provided corrective information to people saying that official statistics show that the crime rate is down substantially. That intervention was effective at reducing misperceptions about crime increasing even among Trump supporters whom you might expect to be most resistant. We thought that result was encouraging. Some of the findings in this field, including my own, can be pretty depressing. I take that study to mean all hope is not lost and there’s still a lot of room to learn what’s most effective in responding to misinformation.
This interview has been edited and condensed for clarity.
