Published: Jan. 26, 2017

When it comes to agenda-setting, it’s out with the old, in with the new media, researchers say.

A computer monitor, smart phone and newspapers.For decades, two of America’s legacy media giants, The New York Times and The Washington Post, have had the biggest influence in setting the agenda for other news outlets. Journalists and bloggers around the country took their cues from the two outlets, creating a trickle-down effect in which content spread from these legacy media leaders toward smaller and newer publications.

However, this tradition no longer stands, according to a new paper led by CU Boulder, involving Boston University and recently published in the Journalism & Mass Communication Quarterly.

“We really wanted to test this idea that elite media still control what’s in the news,” said Chris Vargo, lead author and assistant professor of advertising, public relations and media design, specializing in big data and analytics at CU Boulder’s College of Media, Communication and Information. “We suspected that this changed and I think our results really show that.”

Vargo and co-author Lei Guo, an assistant professor of emerging media studies at Boston University, looked at media across a massive scale, analyzing 2,760 news websites and 48 million articles.

Using the Global Database of Events, Language and Tone, an open-source initiative created at Georgetown University to monitor news outlets, Vargo and Guo identified individual topics that news sites covered, as well as associations between topics. The researchers grouped the data by media type, including articles from The New York Times, The Washington Post and other traditional media (newspapers, television and radio outlets), partisan news websites, non-partisan news websites and all other archived media sources in the database.

I was shocked to see partisan media were as effective as they were,” Vargo said.

Their findings show that while no one media type controls the broader agenda, partisan media now has the strongest influence, followed by emerging non-partisan media outlets – like BuzzFeed and Gawker – that are native to online platforms.

“I was shocked to see partisan media were as effective as they were,” Vargo said.

The trend of partisan sites like Breitbart, TownHall and Daily Kos influencing the agenda of other media outlets is troubling because it could lead to greater political polarization, Vargo said.

“However, just because a story originates from partisan media, it does not mean the same partisan viewpoints will still be attached with it in other media,” said Vargo and Guo in the paper.

In addition, while partisan media may have the strongest influence overall, other types of media were influential in different areas. The New York Times, for example, has the most influence on health care coverage, while emerging media sites have more influence over issues of social justice, according to the paper.

While big data studies like this are ideal for identifying larger trends in the media landscape, questions about how and why these trends happen will need to be addressed with further research.

“We can tell you what happened, but we weren't in the newsrooms,” Vargo said. “A great follow up to this study would be to go into newsrooms and try to see if these patterns are really happening – actually interview journalists and say, ‘Why do you think a news story is important? Why did you decide to cover this story?’”

Agenda-setting in the media is an area of research Vargo continues to build on. He currently is looking at data focused on fake news and fact-checking sites, and in the past he’s analyzed social media platforms like Twitter.

While Vargo’s previous research shows signs of change on the horizon, his most recent paper shows changes that already may have taken hold.

“In 2012 we did a series of papers that looked at how people were talking about the election on Twitter,” he said. “What we found was people really did pay attention to these newer types of media, like partisan media, but those sources weren’t the leaders. Since then, the scales have been tipping and I think this paper suggests that maybe they’ve already tipped the other way.”