Expert Commentary

Research chat: Political scientist John Sides on election 2012

Interview with George Washington University scholar John Sides about academic research and the 2012 presidential election.

John Sides (

George Washington University professor John Sides is a leading academic voice on the Web on all things political.

His professional research has primarily focused on the dynamics of campaigns and mass political behavior, but as a contributor and co-founder of The Monkey — a blog authored by a number of social scientists — his interests take him all over the electoral and public policy map. In 2011, he even won a “blogger of the year” award for his work.

As part of our research chat series, Journalist’s Resource asked Professor Sides about the current scholarly debates around the 2012 election, and for his views more broadly about journalism and political science.


Journalist’s Resource: What is The Monkey Cage, and what is it trying to achieve in this election cycle?

John Sides: The Monkey Cage is trying to bring political science to current events. It talks about political science research and tries to make its lessons applicable to what’s going on in the world. With regard to the campaign, that means bringing previous studies and new data to bear to get a more empirically rich understanding of what’s happening.

JR: The paper you wrote with Dartmouth political scientist Brendan Nyhan for the journal The Forum discusses how there might be a better marriage between political science and political journalism. How do you see that working?

Sides: On the one hand, it’s easy to be bullish on political journalism right now because there’s just so much of it. If you think about the ecosystem that a publication like Politico has helped to create, alongside traditional mainstream outlets, there’s just a lot of reporting. A lot of reporters are working to learn things and uncover things. And I think that’s good. It makes it all the more likely that reporters will catch things that might otherwise have been ignored. So I’m positive in that sense.

I think the challenge that journalism confronts, though, is what to do with that information, how to understand it, and what context to put it in. Often, you still have a lot of the dogs chasing the same car. The same stories get written in the same ways by different outlets. These are some of the features of political journalism that have been noticed for years: pack journalism, for example, is still very much in existence.

The value of political science — although of course we can’t do this consistently, every single day — is that we can take what’s happening in the campaign and put it in context of the research that’s been done, and try to understand what’s really important and what’s not important here. So what’s the likelihood that the kinds of events we’re talking about right now are likely to be consequential a month from now, two months from now, or all the way into November? This perspective tends to sound a cautionary note, because it suggests that every twist and turn of the narrative is not ultimately that meaningful. But at the same time, that’s not meant to suggest that reporters shouldn’t report on what’s happening, or that what happens in the campaign never matters — that it’s an epiphenomenon and the only thing that matters is the state of the economy. [See this Sides post on the debate.]

So what Brendan and I have tried to argue is that political reporters could take findings from political science research and use those to provide context in their reporting about the campaign itself. So, you do the same reporting but you might put some of your learning in context.

JR: So much is changing in campaigns — the digital revolution; the increasingly bleak prospects for public financing of presidential campaigns; the campaign-finance consequences of the Citizens United decision. How relevant can much of the older, canonical political science scholarship be? Is it a brave new world? Or is the past still relevant?

Sides: What changes in campaigns and elections is the amount of money that is raised, and to some extent how that money is spent. Obviously, there have been innovations in vehicles for raising and spending campaign money. You used to have soft money from political parties; we had 527s too. Now there’s more of an emphasis on 501(c)4s and Super PACs. So those things evolve, and of course we have an evolution of campaigning strategies. Obviously, in the past this meant the advent of television advertising decades ago. It’s meant the advent of large datasets of information about voters and the targeting to locate those voters. It is true that it’s always going to take academic research a little while to be able to investigate the consequences of these innovations. They have to happen, and then we have to gather data and analyze it so we understand it.

At the same time, what we’ve learned up to this point is still relevant. To some extent, what we’ve been studying for the past 70 years is: How much does campaigning influence voters and outcomes? We’ve been studying this for a long, long time, and just because more money is spent — or because that money is being raised or spent differently — doesn’t mean that a dollar today is different than a dollar spent in 1950 or 1980 or 2000.

In particular, political science has established that there are the limitations on the amount by which campaigns can move voters, particularly in presidential general elections — although campaigns have more potential to move voters in congressional elections and presidential primaries. I don’t think there’s any reason to expect that things have changed dramatically, just because campaigns are micro-targeting and Super PACs are spending or will spend hundreds of millions of dollars. In part that’s because there’s a diminishing marginal return to all this spending. Every dollar you spend is not a new vote for the candidate. The difference between spending nothing and spending $100 million is going to be different than spending $1 billion and spending $1.1 billion. I don’t think that the system makes the scholarship obsolete so quickly. [See this related Sides post on money in politics.]

Sometimes the argument that “everything has changed” is assumed to be true even before we have evaluated whether the past is actually irrelevant. So in some sense political science gets challenged with a conjecture stated as if it’s a fact, and there’s no easy rebuttal to that conjecture because we can’t answer it yet. Should we really throw the baby out with the bath water here because a casino magnate can dump money onto Newt Gingrich’s Super PAC? Does that mean that 70 years of political science is no longer operative? I don’t think the world typically works that way. It’s very rare in science in general that something happens that necessitates throwing out everything that has been learned up until that point. There are very few paradigm-shifting events.

If we make mistakes as political observers, it’s often by thinking that “this time is different” and by being insufficiently attentive to the continuities between the past and the present. It doesn’t mean that those continuities are perfect, or that there’s never change. I just mean that on average, if we’re going to make a mistake as political observers it will be to overestimate the consequence of the novel things that are happening.

JR: What are some of the salient questions that political scientists are debating in this election cycle?

Sides: One of them is certainly, “How much is the campaign going to matter?” We’re heading into the election where the broader economy doesn’t necessarily advantage either party’s nominee. This isn’t 2008. It’s a toss-up at this point. And that means there’s more room for the campaigns to affect the outcome at the margins.

In this primary, another question — and I’ve discussed this on the blog repeatedly — concerns the nature of party politics in the context of nominations. Are party leaders driving this process, or have they lost control because of alternative power bases — whether that’s the conservative media or rich benefactors of the Super PACs? In the Republican primary this time, it’s an interesting case. I don’t think it’s the best case for the theory that party leaders are really able to coordinate around a leader and promote that person. But at the same time, you clearly see party leaders trying to do that. So that’s the second debate.

Another question is: “How is race going to matter in 2012?” It mattered in 2008. People’s views about black people are absolutely important to their view of Barack Obama. But we’re interested in how much that continues to be true in campaign where you might well see advertising that has racial undertones, in part because it can come from independent groups and leave Obama’s opponent with some plausible deniability.

At the same time, we’re also interested in what’s going on with class. This campaign is taking place not only in the context of a weak economy but also economic inequality. A candidate like Mitt Romney seems to embody certain features of that inequality. We’re interested in Americans’ ambivalence about wealth, where many simultaneously celebrate it but also resent it. This is interesting particularly because Obama appears to want to make inequality a theme of the campaign.

JR: What are some of the assumptions or predictive models that have been either validated or undermined so far in the 2012 primary?

Sides: I’m not sure we have a consensus political science view about primaries. For a long time the dominant view was that a lot of success in primaries was driven by media, money and then momentum. So individual candidates leverage their ability to raise money and to get media attention; they translate that into victories in the early primaries and caucuses and then use those victories to generate momentum for their candidacy. The viewpoint that challenged that, which is in this book called The Party Decides, suggests that party insiders play a bigger role, trying to coordinate around a nominee prior to the primaries and caucuses.

2012 does not present either side with a slam dunk. On the one hand you absolutely see party leaders trying to coordinate on Romney, and trying to slap down challenges to his candidacy. They’ve done that to Gingrich not once but twice now. They did it in December, and they did it again after South Carolina. Bob Dole came out of the woodwork to criticize Gingrich, for example. But at the same time, party leaders were slow to coordinate. Lots of people didn’t endorse, and Jeb Bush is a good example. When they’re coordinating successfully, you’ll see a lot of people jump on the bandwagon earlier. Now Romney clearly has the majority of endorsements from party leaders of various kinds, but he doesn’t have as many endorsements as previous candidates have had.

The dynamics of the primary — primarily Gingrich’s rise before South Carolina — do suggest that media still plays a role. At the same time, Romney’s victory in Florida speaks to money perhaps. So I don’t think you can rule out any one of these explanations altogether. In some sense, competing ideas within political science can take some comfort in the way things have turned out but also have to grapple with some uncomfortable facts.

JR: Are their scholars’ works in particular — and findings within the scholarly literature — that have not gotten enough attention, in your view?

Sides: One of the books that probably gets under-sold is a book by Larry Bartels, published in 1988, called Presidential Primaries and the Dynamics of Public Choice. One of the things that Bartels is trying to do in that book is looking at the way momentum evolves and trying to understand how the public forms its opinions in primary elections. It’s focused on Gary Hart’s ability to challenge Walter Mondale for that race. Bartels shows how people’s opinions are formed, how the primary helps people link their preferences about candidates to their baseline political predispositions, and how momentum might shift people in particular directions. It’s a book that I think has been lost a little bit in the sands of time, and probably could be picked up and read again with interest.

The other thing that deserves more attention — and we’ve discussed this at The Monkey Cage — is research on how divisive primaries affect the nominee’s performance in the general election. In general, that research suggests that the nominees who go through a divisive primary don’t suffer as much as expected. Candidates who win after a divisive primary are often able to unite the party and go into the general election relatively strong. This is not necessarily because of the divisive primary, and not necessarily in spite of it, but ultimately it may just not be that consequential. At the moment, lots of people are asking, “Will there be lasting damage to Romney?” Well, there could be. But I’m not sure it’s likely if you go by the historical record.

Keywords: research chat, primary issue

About The Author