Daron Shaw worked in several political campaigns as a survey research analyst. Shaw also served as a strategist in the 2000 and 2004 presidential election campaigns. He now teaches American government, campaigns and elections, political parties, public opinion and voting behavior, and applied survey research. He also serves on the national decision team for Fox News.
His interview highlighted these two points:
-
• Political science has contributed many tools and concepts to the practice of politics, but they become conventional wisdom and the “point of origin gets lost.” We do a poor job of claiming ownership.
-
• There are strong incentives not to get involved in researching certain issues, such as voter ID laws, because of the public attention and skepticism of the researcher’s political agenda.
Daron Shaw was interviewed by Steve Friess, who has written for Time, Politico, and many other outlets.
STEVE FRIESS (SF) Our broad topic is the disconnect between academic political science and the public. There is concern that the public doesn’t understand political science, that political scientists don’t communicate well with the public, and that political scientists don’t work the media very well, and this damages the prestige of the craft. In Congress, there’s been discussion about whether to fund political science research through the NSF or wherever. Do you think that political scientists have a problem with public perception?
DARON SHAW (DS) We have a problem; it’s come up multiple times over the last 10 or 15 years, and the context usually has been justifying public investment of federal dollars spent on political science research. Two things stuck me during these debates: first, by how ignorant the general public is about political science. And I say “ignorant” in the definitional sense of the word; not as an insult to the public. There’s no reason for them to have a lot of consciousness of what we do. Second, by how poor we, as political scientists, have been at communicating what our research is about, and what we contribute to improving politics. I think of those two observations simultaneously. The burden falls on political scientists; it’s not the public’s job to research what we do and come to some sort of “a-ha” moment. “Oh, these guys are doing something good! I should give them more of my tax dollars.” It’s our responsibility to make that case. I’ve watched some of my colleagues and other high-level people testify in front of Congress, and until recently, I’ve been taken aback at how poorly we’ve communicated the value of political science research. Political scientists tend to point to lawmakers, or point to the public, and say, “You don’t get it. You don’t give us enough credit. You don’t see the value of our research.” My response to that, first, is don’t insult the people who sign your checks, and second, the burden of proof is on us. It’s absolutely legitimate for representatives and senators to ask, “Tell us what it is that you’re doing. Why do we fund you? What are you contributing to democratic functioning?” That doesn’t mean that I agree that what we do is not of value—quite the opposite—but it’s absolutely appropriate for us to explain to the public what it is that we’re doing and why it matters.
SF As a professional political operative, you have conceived of and executed messaging campaigns. What would you do?
DS I’d focus on actual ideas developed by political scientists. For example, much of what we know (or what we think we know) about campaigning, about messaging, and about persuasion actually comes from the study of public opinion and from political science. But often that information, those theories, those bits of conventional wisdom get accepted, and the point of origin gets lost. I read a wonderful quote by James Carville where he claims that nothing of value in a real-life campaign has ever been learned from political science. Now, I don’t know if he said that, and I would offer that James Carville seems to be a brilliant guy. But that is a ridiculous quote. I can tick off a list of ideas political science has contributed to real-life campaigns: party identification, a core concept in understanding voting behavior, wholly derived from political science research. The spatial theory of voting—the notion that people pick candidates whose positions are closest to themselves, and that elections are battles about position—articulated and developed by political science. Ideas like issue ownership and the effects of negative advertising have all come from political science, and (to a lesser extent) marketing and sociology and psychology. And they dominate the execution of real-life campaigns. But these ideas are absorbed into campaign practice, and we’re not particularly interested in their provenance. Furthermore, political science does a poor job of claiming ownership over these concepts, and that’s our fault. As someone who’s worked in campaigns, I see these ideas in motion time and again when we’re executing campaign plans. Any plan to boost awareness of what political science does would do well to focus on those ideas—from American politics, from international relations, from comparative politics—that are so fundamental to understanding how politics works that they have become conventional wisdom.
SF Let me ask you the flip side of that idea. If you ask most people what they think about how campaigns operate these days, you would generally get a negative reaction. Is that something that any group of people wants to own?
DS That’s a good question. In political science, we observe patterns and tendencies, and then we engage in backward-looking theorizing. “Well, why would candidates do that? Why would we see negative campaigns so often? Why do incumbents always seem to win re-election?” So political science basically observes what’s going on in a campaign and then develops a theoretical conceptualization for it. It tries to offer some deeper understanding so that when you observe variation in the real world, you have some context. So we describe what “is,” and oftentimes neglect what “should” be.
We’re fairly agnostic with respect to a lot of these normative questions; for example, whether campaigns have served us well. Whether they are doing what we want them to do as people who live in a democracy. That perspective is something we ought to engage more often. For instance, in one of my research areas—political advertising—there’s not a lot research on the quality of advertising. There is a lot of research on the quantity of advertising: do lots of advertisements have an effect? But we’re reluctant to judge the content of the advertising beyond broad categorizations such as “advocacy,” “contrast,” and “attack” advertising. Judging the quality of the ads is probably where we need to go.
I’m comfortable with the idea that political scientists are uncomfortable with the normative assessment. I don’t like making those sorts of appraisals because it gets too close to advocacy and the kind of normative judgments that could be dangerous.
In contrast, I’m comfortable with the idea that political scientists are uncomfortable with normative assessments. I don’t like making those sorts of appraisals because it gets too close to advocacy and the kind of normative judgments that could be dangerous. In fact, when political scientists have tried to make those judgments, I have often been concerned about the objectivity of those researchers. Still, we need to do more if we’re to add value and understanding.
SF What do you see as the difference between a political scientist and a pundit? As a journalist, I often use political scientists as pundits.
DS Sometimes we dichotomize when dichotomies aren’t appropriate. Is George Will a pundit or is he a political scientist of some sort? Ezra Klein: pundit or political scientist? I think it’s a continuum. The extent to which people use broader, theoretical understandings of politics to inform what they are observing now is the extent to which they become more social-scientific. But if you have to distinguish between the two, an example of how to tell a pundit from a political scientist involves the use of a term like “realignment.” I would classify lots of people as pundits because in the immediate aftermath of an election they will talk about whether there was realignment. And what they almost always betray is a complete lack of familiarity with what that concept is, or what it means. They’re speaking a language that they’re really not equipped to speak. That’s not to cast blame because some of those terms have been offered into political conversation by political scientists without providing information or depth necessary for understanding. In contrast to the pundit, however, a political scientist will say, “Well, you know, we can’t really judge this. Realignment has to do with social coalition change, and that has to be observed over a period of time. It’s a very difficult thing to assess right now.” To me, this distinction has little to do with whether you have a PhD or a teaching degree. It’s about the precision of language that people use to talk about politics, and hopefully political scientists bring a more contextual and theoretical understanding to the observation of facts and events that we see in the now.
Having said all that, I go back and forth between punditry and the ivory tower all the time. If a reporter asks me, “Daron, do you think the Republicans are going to take over the Senate?” and I respond directly to that question—and I try not to, but if I do—I’m a pundit. And probably a poor pundit because I’m not as good as [MSNBC’s] Chris Hayes, or [CNN’s] Jon King, or some of the people who really pay attention to the nuts and bolts of every race on a daily basis. I’m supposed to add something to the conversation by having more historical and contextual knowledge, but I’m acting as a pundit. However, if you ask me about whether the Obama election signaled a longer-term realignment to the Democratic Party, and I talk about generational replacement and young people, over-time changes, and the rise of Hispanics, then I’m speaking as a social scientist … and I put your viewers to sleep after 10 minutes. Much of what you see on air is professional commentary that straddles that line, and some people do both very well, actually. I think Larry Sabato of the University of Virginia does a good job. I think John Sides of George Washington University does a terrific job; sometimes he’s a pundit, sometimes he’s a political scientist.
SF You’ve done both. You do TV, you’ve done campaigns, and now you’ve been teaching. Is there something in the culture of academia—and particularly in political science—that eschews publicity, that regards writing or speaking to the mainstream media, or any media, as sort of selling out or cheapening its work?
DS Because I represent a public university, I have an obligation to talk to the press, to try to explain, as best I can, what I observe. I enjoy doing that. I’ve been in public school all my life. I went to UCLA for my undergraduate and graduate degrees, and have taught at the University of Texas my entire career. So, when I get a phone call, as someone who’s employed by a public university, I need to make an effort to interpret what I do for the public. Ultimately, the public pays my salary. Not everyone feels that way. Many political scientists, who love doing research and do it well, see themselves as speaking to a narrower professional audience and don’t feel obligated to translate that research to the public. They either think that other people will do that, or that they’re speaking at a level that’s never really going to have much popular resonance. So they don’t even bother. That’s an issue, but there aren’t many rewards in the discipline for talking directly to the public.
SF Is that a systemic problem? Is there a shift that needs to happen among academics, among folks who run the journals, among how tenure is determined?
DS We need to reevaluate the standards for promotion, and I think that is already happening. I’ve had conversations with many colleagues, and there is a consensus that my generation, professors currently in their 40s, may be the last generation that exists under the old “traditional” model—where the emphasis is on high-level journal publications and major university press books, and not much thought is given to communicating our ideas to a really broad audience, absent a textbook. The old-school way of communicating your ideas more broadly is you write a textbook. It’s taught to undergraduates who are interested in the subject, but no one else, really. That day is over. The rise of the Internet and all of the different forms of communication, the enormous range of outlets that are available, and the rise of politics as niche entertainment has created markets and opportunities. Consider websites and blogs like FiveThirtyEight.com and The Monkey Cage. Those outlets have narrow but extremely influential and engaged audiences. We have to compete with that, and our host universities expect us to enter that marketplace of ideas and to compete or even dominate. Again, my generation is the dinosaur generation. The old model, based on journals and technical books, is not how our research is going to be disseminated in the future. It’s now moving to different modes and different outlets.
SF There has been a transformation in the public’s interest in politics, since certainly the rise of the cable networks, the various websites, and so on. And then, certainly, during the 2008 campaign, Barack Obama and Sarah Palin became incredibly famous people who were running for office. All of a sudden, politics crossed over into our popular culture. You have Politico, and literally, you have Nate Silver, of baseball data, who becomes this key touchstone for predicting elections and assessing the public mood. In that transition, it seems like a political scientist, by the nature of the name, would be popular. People would want to know what they think, because people want to know what’s going to happen.
DS It’s very likely that some political scientists will become well-known. Take Nate Silver, for instance. I was actually glad that Nate got into politics because he realized, in a way that political scientists probably should have, that there is a market for this information. What is different is that Nate is enterprising and an entrepreneur; most political scientists aren’t. He developed an infrastructure that allowed him to appeal to the public’s appetite for election numbers. Predicting elections is something we pioneered in political science. What Nate is doing has been around for years in political science. But again, there aren’t too many of us who would dedicate our time and energy to doing what he has done because the incentive structure hasn’t rewarded that work. Publish in major journal articles, train graduate students, serve on committees, and teach our classes: that’s what we do. That’s short-sighted, because I know that the work Nate does attracts money and attention. Looking ahead, most us won’t be rock stars. We’re just temperamentally not cut out to blog every day and publicize our work. However, a few political scientists can do it and do it well, and the market is there.
SF Political scientists have a telegenic problem?
DS Yes. You put us on television, and you say, “You’ve got 30 seconds to describe polling bias.” Political scientists can barely clear their throats in 30 seconds. You can’t keep these people under 12 minutes in conference presentations. Part of our excuse is, “Well, we have complex ideas…”
SF I understand. These interviews were only supposed to be 15 to 30 minutes long. Not one has been less than 40 minutes!
DS We’re not good at sound bites, but we can describe what we do more economically. Mark McKinnon, a famous political ad professional, came and talked with my class once, and he asked, “How many of you think it’s impossible to communicate a real, serious, political, substantive idea in a 30-second ad?” And almost everyone raised their hand. Mark said, “I want you to read what I put on the screen, and as soon as you’re done reading it, put your hand up.” Then he put the biblical passage John 3:16 up there, and after about 16, 17 seconds, everybody put their hand up. And Mark says, “That is the central concept of Christianity, and it took you under 20 seconds to read that. Don’t tell me you can’t communicate these complex ideas in under 30 seconds.” I’ve always loved that demonstration, and I apply it to political scientists. We need to take complex ideas, those core notions that we have, and explain them in a way that people find interesting, useful, and important.
We need to take complex ideas, those core notions that we have, and explain them in a way that people find interesting, useful, and important.
SF OK. I’m going to imagine a world in which it’s not Nate Silver who did what Nate Silver did, but an academic political scientist, based on publicly funded research. Grants, or whatever. It’s an election year like 2012 when half of the country doesn’t like what Nate Silver says is about to happen. Nate is right. He’s not manipulating his data but he is beaten to a pulp in advance of Election Day. If you have political scientists in the mix, doesn’t it just become another part of the political football that is our way of dealing with issues in this country today?
DS Because we rely on a common set of variables, and share a sense of how to analyze and interpret data, I think that there would be a fairly significant degree of consensus. There’s going to be some dissent, but that’s life, right? Few political scientists would have disagreed with what Nate was saying in 2012.
SF And not too many people inside the Mitt Romney campaign would’ve ever acknowledged that he was even close to being correct.
DS Yes, people’s positions can color even their judgment about how to handle data. Still, in the case of political scientists, there was a consensus that Obama was very likely to win in 2012, even among people like myself, who weren’t necessarily on that side of the aisle. But if I had built a website or blogged that, I would’ve gotten roasted.
SF By whom?
DS By people on the Right who wouldn’t have liked what I was saying. The excuse would be, “You’re making it difficult for us to fundraise.” Or, “This sort of stuff makes it difficult to get our message out.” Some version of, “Not only do we not agree with you, but you’re betraying the cause.” You’ll see that on the Left in 2014. You’ve already seen it with Nate Silver. When Nate says there is a slightly better than 50-50 chance the Republicans take the Senate, he’s gets torched by liberals. Honestly, it is our job to tell people what the data say, and then suffer the slings and arrows. But that is tough. In this polarized environment, when you teach at a public university, it becomes really tough. There is always a chance that someone who doesn’t like what you are saying will file a request for documents on your work based on the Freedom of Information Act. Then data on any research you’ve ever done—including outside work, funding sources, and more—can be turned over to them and they can do what they want with that information.
SF You’re now arguing for essentially the opposite of what you had said. If I was a political scientist who was nervous about putting my research or my name out there, and I heard you saying this, I would hide under the covers.
DS There are contradictory incentives, yes. As I said, I’m in a generation with the old, journal-oriented approach to research, who speaks to a narrow audience, and maybe gets a chance to talk to reporters from the New York Times or The Wall Street Journal from time to time. How we communicate information has changed. As you mentioned, the outlets and the venues for getting information out to the public have changed. But, your point about the overall climate and environment is an important one. If you’re talking about greater accountability and if the protections of tenure are a little less absolute, then we’ve got a fine line between becoming a player in political science but not making enemies. For example, if I’m an assistant professor, and I’m trying to get promoted, and I have some interesting way of aggregating data and predicting election outcomes, how hard do I push to get my results in the public arena? There are some potential disincentives to aggressively promoting that approach and that knowledge. You could argue, “Well, you should be fine. First and foremost, there’s a professional audience; if you’re wrong it will be discovered and corrected in the peer-review process.” But what if the work becomes public and, by some quirk, the prediction is off for the next election?
There’s a split-level to academic work: first, the old-style audience of professors and intellectuals, and second, a new audience of interested observers, political junkies, and partisans. If you survive the peer-review process, the new audience might not be interested in the appropriateness of your data or the appropriateness of your technique and methodology. They may not like what you say, and do you really have an incentive to go out and make that case aggressively? I don’t know. Sorry, we got a little off topic here, but…
SF I don’t think we did. There seems to be a couple of things going on. First, political science, and many disciplines like it, are under fire in terms of keeping their funding. Some senators, like Senator Coburn, can make a very compelling public argument that so-and-so chose political science over cancer research, and that’s an important argument when someone’s running for re-election. At the same time, to make the case that the discipline is worthwhile, you and your work would have to be in the public eye And, as you just noted, that it can be really dangerous and really scary.
DS Well, here is another example of how different incentives come together. I have been involved in expert testimony in a voter ID case. The evidence I looked at suggested voter ID laws aren’t having much impact on turnout. But it’s a difficult question to examine, and honestly, there are strong incentives for me to not get involved in researching this issue for this or any other court case: you’re deposed, and the opposing side talks about how unqualified you are, what an idiot you are, and how your research is horrible. Then, you get done with the case, and lots of people find out about you, and they dislike what you’re arguing, and they question your motivation and objectivity. In the end, I had no interest in being “Mr. Voter ID.” I have thoughts on it. I’m confident on where I think the data lead. But I don’t want to publicly promote those data or do additional research in the area. I don’t want people to say, “Oh, look, this guy’s at the University of Texas, and he’s worked in Republican campaigns. He’s obviously trying to prove a point. Why are we paying him a public salary?” This scenario has to enter into your mind.
SF Let’s talk more about this example. The notion that voter ID laws won’t make a difference, won’t affect turnout that much: can’t that be argued both ways?
Can’t a Democrat ask, “Why are we bothering with them? It’s clearly mean-spirited and inconvenient.” And the Republicans can ask, “Why do you care? It’s not going to change anything.” It seems like you can take that, pull it apart in any direction you want, and it doesn’t matter what your bias might be.
DS That’s absolutely right. Plus, most of us academics make lousy partisans because all of our arguments are laden with qualifiers and caveats. In fact, when I characterized what I know about voter ID for the court case, I was ambivalent. I said that on the one hand, most election fraud occurs at the level of registration, not on Election Day. But on the other hand, the data shows that the sorts of people who don’t have ID aren’t likely to be voters anyway.
In light of all this, I’m optimistic and bullish on electoral politics. The cross-fertilization between academia and the real world has been good. The campaigns being waged now are better than there were 15 or 20 years ago.
SF Any other thoughts, especially being in academia but having done so much work for Republican groups that tend to often use academia as one of its villains?
DS Academic work is much more visible today in real-life campaigns. Back in the 1990s, the number of academics who were called on, or who had any connections with the practical political field was almost zero. They were uninterested in us; we were largely uninterested in them. We operated in two different realms. That has completely changed since the work that Karl Rove and the Republicans did in the early 2000s. It accelerated substantially with the development of analytics and big data for the Obama campaign. Many academics and trained graduate students moved into the practical realm and are major players in politics, both in campaigns and in public policy. They are well-versed in political science theories and methods. I don’t want to overstate the extent to which it has impacted the election results. Obama won in 2008 and 2012 because of the fundamentals and because he was a tremendous candidate running in good years for Democrats. But at the edges, in a battleground state, this applied knowledge could be worth a point or two. Of course, there’s a larger question about how this has affected the public. Earlier we talked about negative advertising and how the public is turned off. But I think the content and targeting of political communication is improving, partly thanks to superior research. Candidates do a much better job in articulating messages, in targeting voters, and in getting information to voters than they did 15 or 20 years ago. Voter turnout increased in 2000, 2004, and 2008, and that’s positive. In light of all this, I’m optimistic and bullish on electoral politics. The cross-fertilization between academia and the real world has been good. The campaigns being waged now are better than there were 15 or 20 years ago when they just put out these big, broad television advertisements. Similarly, the quality of political science research, the back-end data that we have access to, and what we’ve been able to investigate has dramatically improved from where it was 20 years ago. So, I’ve really enjoyed where the discipline’s gone during the last 15 or 20 years. We’re more practical; we know a lot more about what really goes on, we’re able to test a lot more, and what we say has more credibility. Consequently, the feedback loop to professionals has been improved.
SF I appreciate your time.
