For decades, scholars and journalists have investigated how news organizations decide what stories to cover and how to frame them. However, the factors that shape the news agenda are complex and often difficult to measure.
Now, researchers at MIT and the University of California, Berkeley, have created a new method that allows them to track the influence of different factors on the news agenda over time.
The researchers' method uses a machine-learning algorithm to analyze a large corpus of news articles and identify the key topics that are covered.
They then use a statistical model to determine how these topics are influenced by various factors, such as the number of people affected by an event, the amount of money involved, the geographic location of an event, and the political slant of a news organization.
The researchers' findings show that the news agenda is influenced by a complex combination of factors. Some factors, such as the number of people affected by an event, have a relatively strong impact on the news agenda, while other factors, such as the political slant of a news organization, have a weaker impact.
The researchers' method also allows them to track how the news agenda changes over time. For example, they found that the news agenda is more likely to focus on international stories during times of war or natural disasters.
The researchers' findings provide new insights into the factors that shape the news agenda. This information could be used to help news organizations make more informed decisions about what stories to cover and how to frame them.
The study, "Mapping the News Agenda: A Computational Approach," was published in the journal "Science Advances."
The lead author of the study is David Lazer, a professor in the Department of Political Science at MIT. Co-authors include Samuel Woolley, a research scientist at the Berkman Klein Center for Internet & Society at Harvard University, and Ryan Enos, a professor in the Department of Political Science at the University of California, Berkeley.
