Dondena Interviews

We’re meaner than we used to be, and Shanto Iyengar has the data to prove it. Director of the Political Communication Lab at Stanford University, Iyengar says that badly behaving elected officials and flagrantly partisan news networks have contributed to a new acceptance of ordinary folks spewing venom. And with everything from social media algorithms and increasingly segregated (politically, that is) neighborhoods, our spaces for exchanging ideas with people unlike ourselves are becoming fewer and farther between.


Tell me what your talk is about today.

I’m simply going to be describing some interesting developments in the state of American politics, particularly developments concerning the level of animosity and ill-will that now seems to describe the relationship between Democrats and Republicans.


Is it much different now than it used it be?

It is different from the late 1980s the early 1990s. We see a very sharp increase in willingness to express fairly harsh views of people on the other side.


Now, you’re talking about legislators? Or me versus my conservative friends on Facebook?

No, no. I’m talking just about ordinary people. That’s why it’s interesting. Because for years people believed that the only polarization that existed in America was among legislators and elites, but we’re showing that that’s not true. We’re showing that polarization has diffused, and ordinary people now feel quite distant. So if you look at intermarriage, there is now more interracial marriage than there is interparty marriage.


Interesting. I can see that. Why do you think this is?

Well, it’s difficult to put your finger on the exact causal mechanism, but we think there are three possibilities. Number 1 is the behavior of leaders. People who hold office don’t spare any venom when they are describing their opponents. They’re extremely critical, and that kind of rhetoric is going to trickle down to their supporters. Secondly there’s the rise of new forms of media, including social media, but we’re particularly focusing in on cable news and talk because they’re very partisan. We have Fox News on the right, MSNBC on the left, and the blogosphere is completely polarized. So that’s a second possibility. And the last is that, over the last 40 to 50 years, patterns of residential segregation have developed quite starkly.


Is that right?

Oh yeah. So we have people who live in very homogeneous neighborhoods. My block in Palo Alto has one Republican. Very few people actually encounter diversity in their immediate environment. The only place you probably get political debates would be at the workplace. So, there’s this kind of sorting into micro political cultures.

Then the other overarching explanation, of course, is that it is not considered socially inappropriate to say bad things about your political opponents, whereas it is considered socially inappropriate to say bad things about racial minorities or about women or about old people. So there’s this idea that politics is a free-for-all and you can actually say what you like and not be constrained by these kinds of politically correct normative constraints.

If you look at Catholic-Protestant, black-white, north-south—the various divides that are thought to characterize American politics—and you ask people, northerners, their feelings about southerners, or African-Americans “How do you feel about whites?” the gaps are very small in relation to Democrats versus Republicans.


Tell me about the data behind the paper.

We have a ton of data. We have national surveys going all the way back to 1960. We have run a series of experiments in the last five years where we give people various tasks. So, in one experiment we asked them to read two résumés, and select the one that was more deserving of a college scholarship, and unknown to the people in the study we manipulated the résumés, so that the students’ extracurricular activities revealed their political affiliation. We found that the political activity had a stronger impact than the grade point average on the decision to award the scholarship. So the Democrats did not like the guy if he was active with the College Republicans. And so that suggested that even in this fairly innocuous, non-political task, people were using partisan cues.

And then we ran a series of studies where we gave people money. These are used by economists, basically they are behavioral games. You give people money, you say you’re playing with someone else, you’ve got $10, and you can give as much of it as you like, or you don’t have to give anything and you can keep it all. We gave them a capsule description of who they were playing with. The catch here is that if you give $2, let us say, the participant is told that the researcher will triple that amount and this player could then return an amount to you. In a series of these games we show that when people are told that they’re playing with someone from the other political party, they give much less than when they are told they are playing with someone of a different race or different gender or what have you. So trust seems to be very sensitive to these questions of politics.


How does this paper fit in with the overall picture of what you do in your research and in your career overall?

Well, my main field is media, I’ve been studying media for the last 25 years, so the connection there is with this idea that because of the diffusion of technology and the availability of thousands of news sources, people are now forced to be more selective, and so they choose to tune in to news sources that provide agreeable content. This results in a kind of echo-chamber effect by which people acquire information they are motivated to sort of seek out.


This is the problem I have with the Facebook algorithm. It shows me the things I’ve already seen, things I’d agree with. I think you should be able to tick a box that says, “I want to see things that I don’t agree with, I want to be challenged.”

You should read the latest issue of Science magazine. One of my students, who works at Facebook, has written this paper. The major claim of the paper is that the algorithm doesn’t make things worse. Friendship groups are extremely homogeneous, so when someone sends you a link, you’re presumably going to get it from someone who agrees with you on most things. But the algorithm itself, they claim, doesn’t make it worse.


Of course they would say that...Tell me about in your field, as broadly defined as you like it to be. What do you like about what you see going on, what do you wish you could see more of?

Well, in America, it’s always a rat race in terms of academic research. There’s a scarcity of funds, you spend a lot of your time writing grant proposals and the hit rate on these grant proposals is low, and so in an ideal world, you know, you wouldn’t have to do that, unfortunately that’s just the way it is everywhere. In fact, America is the country where the academic infrastructure is probably the most well-developed. Anywhere else, it’s probably even worse. And so that’s the downside.

On the other hand, I’m fortunate enough to be at a fairly well-endowed university, so you can get in-house funding to do small-scale projects and those small-scale projects, if they pan out, can be used to attract larger-scale support. So I really don’t have too many complaints.


Learn more about Shanto Iyengar here. Learn more about the Dondena Seminar Series here.
















Last updated 11 July 2015 - 07:54:30