Quantcast

Study of headlines shows media bias is growing

A recent study conducted by researchers from the University of Rochester reveals an increasing polarization in news stories related to domestic politics and social issues.

Analyzing 1.8 million news headlines from major US news outlets between 2014 and 2022, the team, led by Jiebo Luo, a professor of computer science and the Albert Arendt Hopeman Professor of Engineering, utilized machine learning techniques to examine the headlines and present their findings on media bias at the MEDIATE workshop of the International AAAI Conference on Web and Social Media.

While it is widely acknowledged that news media outlets adopt ideological perspectives in their articles, previous studies on the differences among outlets were limited in scope and relied on small sample sizes. Leveraging machine learning, the researchers were able to study a vast sample of headlines spanning eight years, encompassing nine representative media outlets such as the New York Times, Bloomberg, CNN, NBC, the Wall Street Journal, Christian Science Monitor, the Federalist, Reason, and the Washington Times.

To measure the nuanced thematic discrepancies among headlines, the study employed a technique called multiple correspondence analysis. The researchers categorized the stories into four main groups—domestic politics, economic issues, social issues, and foreign affairs—and analyzed how media outlets on the left, right, and center differed in their language choices for headlines.

The team observed that media outlets across the US political spectrum demonstrated consistency and similarity in covering economic issues. Discrepancies in reporting on foreign affairs were attributed to the diversity of individual journalistic styles. For instance, the authors noted that the Wall Street Journal and Bloomberg primarily focused on the economic and financial implications of geopolitical tensions, resulting in divergent perspectives compared to other media outlets. However, significant differences emerged in the categories of domestic politics and social issues.

When examining the coverage of abortion issues, the researchers identified subtle differences in the language used by different media outlets. For example, Reason tends to employ the term “abortion law,” while CNN emphasizes its ideological stance by using the term “abortion rights.” Although both outlets discuss abortion issues at a higher level, the choice of words creates a discernible distinction.

Moving forward, the research team aims to delve deeper into understanding how and why media outlets select different terminology when addressing similar topics. They emphasize that comprehending these discrepancies, and identifying instances of media bias, is crucial for both media organizations and readers.

Lyu Hanjia, a computer science PhD student and lead author of the study, explains, “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.”

Other coauthors from Luo’s research group include Jinsheng Pan, Weihong Qi, and Zichen Wang. Funding for the study was provided by the Goergen Institute for Data Science at the University of Rochester.



The material in this press release comes from the originating research organization. Content may be edited for style and length. Want more? Sign up for our daily email.