ROCHESTER, N.Y. — Does it feel like news headlines lately are intentionally designed to sow discord? It’s not all in your head. News stories covering both domestic politics and social issues are becoming more and more polarized along ideological lines, according to a new study by researchers at the University of Rochester. The team analyzed 1.8 million news headlines from major U.S. news outlets spanning 2014 to 2022 to reach their findings.
Led by Jiebo Luo, a professor of computer science and the Albert Arendt Hopeman Professor of Engineering, the study authors made use of machine learning to analyze news headlines across the country. Study authors explain that while there is a broad societal consensus that news media outlets adopt ideological perspectives in their articles, earlier research investigating the differences among outlets were limited in scope and usually used small sample sizes.
The introduction of machine learning — a branch of artificial intelligence — in this latest study allowed the research team to assess a vast sample of headlines over an eight-year period across nine representative media outlets: The New York Times, Bloomberg, CNN, NBC, the Wall Street Journal, Christian Science Monitor, the Federalist, Reason, and the Washington Times.
This project utilized a technique known as multiple correspondence analysis to gauge and measure the fine-grained thematic discrepancies between news headlines. The study grouped news stories into four categories: domestic politics, economic issues, social issues, and foreign affairs. From there, study authors analyzed how left, right, and central media outlets differed in the language they used across their headlines.
Researchers report that U.S. media outlets across the political spectrum were generally consistent and similar when covering economic issues. Discrepancies, however, appeared regarding foreign affairs. Study authors attributed this trend to diversity in individual journalistic styles. For instance, the Wall Street Journal and Bloomberg primarily focus on the economic and financial implications of geopolitical tensions, thus resulting in differing perspectives in comparison to other media outlets.
However, headlines in the domestic politics and social issues categories displayed important differences.
“We observed a lot of subtle differences in the words they choose when they cover the same high-level topics,” says Hanjia Lyu, a computer science PhD student who was the lead author of the study, in a university release. “For example, when covering abortion issues, Reason tends to use the term ‘abortion law,’ while CNN underscores its ideological position by using the term ‘abortion rights.’ On a higher level they are both talking about abortion issues, but you can feel the subtle difference in the words that they choose.”
Moving forward, researchers want to dig deeper on these topics and continue researching news bias in order to better understand how and why media outlets use different words to cover the same kind of topics. They believe that understanding these discrepancies, and when they may indicate media bias, is key for both media outlets and the general public.
“For consumers, it’s useful to know this information because the echo chamber effect is very strong and people are used to only listening to things they like to hear. Showing the divergence and the increased partisanship may make them aware that they need to be more conscious consumers of news,” Prof. Luo concludes.
This research was presented at the MEDIATE workshop of the International AAAI Conference on Web and Social Media.