Expert Commentary

Research chat: U.T. Austin’s Talia Stroud on how to de-polarize news audiences

2013 research chat with Natalia “Talia” Jomini Stroud, assistant director of the Annette Strauss Institute for Civic Participation at the University of Texas at Austin.

Talia Stroud (

Natalie “Talia” Jomini Stroud is an associate professor of Communication Studies and assistant director of the Annette Strauss Institute for Civic Participation at the University of Texas at Austin.

The author of Niche News: The Politics of News Choice, she continues to oversee the Engaging News Project, which provides empirical research perspectives on news website features and functionality such as comment sections, social media buttons and hyperlinks.

As part of our ongoing “research chat” series, Journalist’s Resource recently talked with her to ask about issues of political polarization and the media’s role and responsibilities in serving the public. The following is an edited transcript.


Journalist’s Resource: Niche News has a lot of insights that are relevant right now, as the media struggles to operate in a highly polarized environment and as polarization has afflicted the news business itself. What are some of the key takeaways?

Talia Stroud: The book begins with the premise that people are drawn toward media that match their political beliefs. We found that across magazines, cable news, radio and websites, convention-watching behavior, politically tinged movie viewing — in so many different places you see that pattern. That is not to say that people never look at views that are not their own. It’s just that there is a tendency to look at media that you agree with more than media that articulate a different political perspective.

So the real question is: What is the democratic consequence of that behavior? Some of the things we find are: When people look at like-minded media, they develop more politically polarized attitudes. They begin to think more favorably of the candidate favoring their political party and to think less favorably of the candidate in the opposition party. The behavior really manifests itself among people who are politically knowledgeable and interested. That’s a finding that still bothers me in some ways, because you would think that citizens who are informed about politics are our great citizens. But they tend to look at media that match what they believe. Another consequence we find is that people who look at more like-minded media participate in politics more frequently. So while we might say there are some downsides of this behavior like political polarization, there are some democratic benefits. If you ask someone, “What do you envision as a great democracy?” political participation ranks pretty highly. This sort of media isn’t completely bad — it makes people participate more.

JR: If you are an editor or political journalist, what should you do with this knowledge?

Talia Stroud: There are two sides to this coin. I almost hate to say it, but we see that there is a market for media that has a partisan tinge to it. There are some strategies that editors and journalists can employ. The Engaging News Project is trying to find some real-world ways in which newsrooms could intervene to combat these tendencies. We have done four different studies over the past year looking at how newsrooms might do that. We have all of our white papers up on the website, and everything that we’re working on.

JR: If you could show a bunch of managing editors your findings and highlight a few insights for newsrooms, what would they be?

Talia Stroud: The one that has gotten the most attention is that we’ve been looking at the way buttons are used in comment sections. And right now the most common way is that there is a “Like” button or maybe a thumbs-up button. That’s great — it gets people to interact. But we wondered if whether in a news comment section, the “Like” button was really the most appropriate. So we conducted an experiment where we contrasted the “Like” button with a new button — “Respect.” We found that, at least for some topics, the “Respect” button actually garnered more clicks in comparison to the other, the recommend button. We also found that people were more likely to click on the “Respect” button for views with which they politically disagree — to respect those views, rather than like them.

engagingnewsprojectSo we think that there are small interventions that can be done in a news space that might encourage more thoughtful behavior about counter-attitudinal views. The “Like” button really engenders this partisan “agree-disagree” kind of mindset, whereas a “Respect” button opens up what you’re able to think about when you’re looking at comments.

JR: What about other aspects of the project? Any research findings that have you intrigued and you’ll explore further?

Talia Stroud: One of the other studies that we did as part of the project — which actually didn’t entirely work out, and that makes it intellectually intriguing — is an experiment where we wanted to see if changing a sentence or two on a news website could change people’s behavior. We first combed through research literature out there to find if there were theories about how to get people to think about other views. We set up a Web experiment to test this. We thought, “Wouldn’t it be great if we could find certain prompts that encouraged people to look at other views?” We had all of these wonderful aspirations for this sentence, that it would encourage people to look more at hard news than entertainment. We didn’t find any prompts that were totally positive in terms of their effects. One of them was: “Thanks for keeping up with the news. Be proud of protecting your democracy.” That led people to see the site as more credible than one without that phrase. But people clicked on fewer hyperlinks on the site with that prompt, so we can’t entirely endorse it. The thing that I find amazing, and that I want investigate, is that one or two sentences can affect people’s attitudes and behaviors. We need to know a lot more about how that works.

JR: There was an important paper by Markus Prior of Princeton published earlier this year in Annual Reviews of Political Science, “Media and Political Polarization.” It surveys the state of the research literature on how the media may or may not be driving the political polarization we’re now seeing. How do you see the line of causality here — is it that a polarized public is creating a more polarized media, or vice versa?

Talia Stroud: I’ve found that both are true to some extent. That is, that people who already hold polarized attitudes are more likely to look at the partisan media out there. But we also find the opposite causal direction: That those who are watching this partisan media are displaying increasingly polarized political views. There are other scholars who have found similar patterns. For example, Matt Levendusky of Penn has a new book, How Partisan Media Polarize America, that finds similar sorts of things, as does his paper “Why Do Partisan Media Polarize Viewers?” for the American Journal of Political Science. Lauren Feldman at Rutgers University has found that there are polarizing effects of looking at media that match what you believe. So there is a good deal of evidence that it can polarize people. But this is not to say every single person is going to these outlets; it’s people that already have some level of political polarization.

JR: You have thought about how the Internet may be changing politics. Ethan Zuckerman of MIT has a good new book out called Rewire: Digital Cosmopolitans in an Age of Connection that talks about the need for “cognitive diversity” and a conscious effort to broaden one’s horizons. In your view, is the Internet exacerbating polarization further? It’s certainly been talked about.

Talia Stroud: The Internet in many ways has the potential to amplify these patterns. Two reasons were identified by Richard Davis and Bruce Bimber in their book Campaigning Online: The Internet in U.S. Elections: (a) The Internet makes it possible to find information that you make want; and (b) It actually requires you to search — people are active in the process. We see different patterns as a consequence of that ability to find information and the requirement to find information. On the one hand, you do see people looking at websites and material online that match what they believe. So they are enacting the same pattern of behavior that we see in other media formats. It also, though, gives people the opportunity to look at a different perspective, potentially.

There are some analyses that are looking at comScore data, for example, saying, “Look, people are clicking to the other side. People are looking at the other side to see what is happening, and you can see that in their browsing habits.” What we don’t yet know is why they are doing it and what is going on cognitively when they do it. If I go there for the purpose of laughing, or making fun, or bolstering my own view, then it is not helping that much that people are looking at views unlike their own. But the possibility exists. The real question for an editor or journalist is: How can you get people to look at that other view and really give it a fair shake?

JR: One pattern we’re seeing is that younger consumers are increasingly getting their news through social streams such as Twitter. The Pew Research Center has documented this trend. Of course, these are highly filtered experiences. What do you make of this? What if this trend continues to accelerate and the majority of Americans ultimately get most of their news through social streams?

Talia Stroud: It’s hard to say. But people are gathering more news and information through Twitter and other social outlets — through “push” news, rather than people going to the site and finding it on their own. There are analyses of social media that suggest people are more likely to follow or be friends with those who share their political identity. If people are looking more at news they agree with and they have a social network they tend to agree with, then we would anticipate that all of these patterns would just increase. There are some corrective forces out there; it’s not as if polarization continues unabated and that there’s nothing in the social system that might not push it the other way. For example, as we see Congress fight more and more — and this is just speculation  —  perhaps we might see people say, “We need to compromise more. We need to get more things done.” I’m talking now about the Congressional level, but that might filter down into people’s habits.

JR: So this comes down to the “filter bubble” problem. Eli Pariser’s book by the same name really coined the phrase. But others argue this is not quite as big a problem as it seems. Where do you come down?

Talia Stroud: It’s an interesting idea. I don’t know if it’s yet to the point where you’re in a complete bubble. I think that a lot of people who are designing algorithms are doing so in ways that allow for some serendipity. For example, there are some projects that are trying to organize articles in the news space by different topical categories. Designing new systems has the potential to break through any sort of filter bubble that might exist. I guess I’m concerned about people’s internal propensity, that is not machine-driven. I have kind of a mixed view on filter bubbles, then: It’s a danger and we should be aware of it, but it’s not yet to the level of where people are in perfect bubbles.

JR: Finally, you’ve studied comedic news — the sort of thing that has become very popular because of The Daily Show, The Colbert Report and other shows. Your latest paper is “Selective Exposure, Tolerance, and Satirical News” in the International Journal of Public Opinion Research. What do we know about this genre and what it’s doing to democracy? It seems you come down pretty strongly that it doesn’t help the polarization situation.

Talia Stroud: This actually relates to the Engaging News Project. I was so optimistic: I was hoping that maybe through comedy, we can become okay with other views. We can get a chuckle out of it. And perhaps we’re not so engaged in counterargument when we’re processing comedy that might be against what we believe. There was a lot of research that led me to believe that that may be the case. We did a study that contrasted a very hard-news website to one that contained articles conveyed in a comedic manner. We worked very hard to make sure they were as similar as possible in terms of the content, except one was written from a comedic viewpoint and one from a hard-news viewpoint. We also had a control group, with no website. We were monitoring where they clicked and afterward asking them about their attitudes. The evidence did not support my optimistic hopes; it was in the opposite direction. The people who had viewed the comedic site were less likely to click on counter-attitudinal articles. At the end, they were less tolerant of views not like their own. Unfortunately, I don’t think it’s the solution to getting people to think about alternative views.

Tags: communication, news, research chat

About The Author