Studying News Junkies Reveals Insights into Online Reading and Info Bubbles

It’s a weird time to be a consumer of news. There’s an overwhelming variety of websites supplying headlines and hot takes. Simultaneously, fewer newspapers and reporters are producing the local and in-depth stories that were once mainstays of journalism.

“The news industry is really volatile right now and has been for a while,” says Shoshana Vasserman, an assistant professor of economics at Stanford Graduate School of Business.

She has a unique lens on how the news environment is changing: With Gregory Martin, an associate professor of political economy at Stanford GSB, and Andrey Simonov, an associate professor of marketing at Columbia Business School and a national fellow at the Hoover Institution, she has been tracking the online habits of thousands of volunteers to get a better idea of where people get their news.

Their Attention Stream study collects anonymized data from participants who have downloaded a browser extension. It’s already recruited around 5,000 participants from around the U.S., drawing a sample of self-reported engaged news readers. You can sign up to participate.

“The broad goal of the study is to understand what drives news consumption,” Vasserman says. “How much influence do news outlets have over what people actually see?” How does someone who starts their day with sports news get national news? Do New York Times readers also check out FoxNews.com, and vice versa? And when news breaks in people’s backyards, do they turn to a local newspaper site they usually don’t spend much time on?

Although much of the online economy operates on clicks, the study seeks to understand reading itself. Its browser extension collects article text from newspaper websites (and only newspaper websites) and tracks how far down the page readers get. Martin details some of the questions the team is trying to answer: “Do they reach the bottom of the article? Do they just read the headline and stop? How long is the tab active in the browser?”

The study is also trying to understand how readers react to changes in the news landscape. “We’d like to know what happens when a newspaper lays off some of its staff or stops doing a certain type of reporting or closes altogether,” Vasserman says. “What do people who used to read that content switch to? What is the substitute — if any?”

The data Vasserman and Martin have collected so far (which includes data from an early project, Beyond the Paywall) already offers insights into how people get their news online. A few of their findings:

Read It Here First

Most participants in the study say they get their news from several sources, typically a mix of radio, podcasts, TV, and newspapers. However, they tend to have one site they rely on as their primary online news source. The median user spends more than 40% of their attentive reading time on the news site they visit most often. They spend less than 20% of their time on their second-favorite site, and the time they spend on other sites drops off from there.

Local News Matters

Another pattern Vasserman and Martin are following is how big local events affect the consumption of local news. For instance, traffic to Texas-based news outlets spiked following the school massacre in Uvalde, Texas, last May. This suggests that people still rely on local journalism, particularly when events boost its relevance. “We’re interested in understanding how much these kinds of newsworthy events change where people go because that’s another indication of the value that these newspapers provide,” Vasserman says.

Bursting the News Bubble

We hear a lot about people living in online bubbles where they only see news that reinforces their worldview. “The popular perception is that everyone is in a filter bubble and that liberals are just reading liberal sites and conservatives are just reading conservative sites,” Vasserman says. But her and Martin’s preliminary data suggests that many people seek a mix of news sources, some of which may have contradictory perspectives or biases. For example, 70% of readers who visit FoxNews.com also go to the New York Times website; half visit the NPR site. “It’s less polarized than you might expect,” she says.