Dissemination Biases of Social Media Channels: On the Topical Coverage of Socially Shared News

Abhijnan Chakraborty, Saptarshi Ghosh, Niloy Ganguly, Krishna P. GummadiAAAI 2021 · DOI

Abstract

In a marked departure from traditional offline media, where all subscribers of a particular news media source used to get the same news stories through printed newspapers, online news media presents multiple options for the readers to consume news. For example, the subscribers of a media source can get news directly from the news website, or from what their peers share over social media sites like Facebook and Twitter. It is, however, unclear whether there are any differences in the news disseminated on these different online channels. In this work, we analyze data from a popular online news media site (nytimes.com), and show that each of these different channels tends to highlight some types of stories more than other stories. We believe that consumers of online news as well as media organizations need to be aware of such differences in various online news dissemination channels.

Introduction

While we often think of social media as a neutral pipe for information, research presented at ICWSM 2016 demonstrates that different channels—Facebook, Twitter, and Email—exhibit distinct "dissemination biases," favoring certain topics over others. In the era of print media, every subscriber to a newspaper saw the same front page. Today, news consumption is fragmented. If certain platforms systematically highlight specific topics while burying others, users may develop skewed perceptions of current events—a phenomenon the authors call "dissemination bias." This has major implications for "filter bubbles" and the design of news recommendation algorithms.

Methods and Datasets

The researchers employed a comparative analysis using the New York Times (NYT) API to track every article published over a six-month period (July – December 2015), totaling 120,231 stories. They utilized the NYT’s internal topical tagging system and tracked four distinct "popularity" lists:

  • Most Viewed: Popularity on the NYT website itself.
  • Most Emailed: Direct person-to-person sharing.
  • Most Shared on Facebook: Semi-public sharing within social circles.
  • Most Tweeted: Public sharing to a broad audience.

Key Insights: Platform Signatures

The study reveals that the "publicness" of a channel dictates its content.

  • Email (Private): Used for personal, high-utility, or sensitive topics like Health and Travel.
  • Facebook (Semi-Public): Dominated by "Opinion" pieces and "Arts," likely because these help users signal their identity to friends.
  • Twitter (Public): Functions more like a real-time news wire, favoring "Business," "Technology," and "World" news.
Topical distribution across different sharing channels shows how platform affordances influence content popularity.

Interestingly, stories on broad topics like "Sports" or "World News" are often under-represented in popularity lists relative to publication volume, while "niche" topics like "Health" and "Fashion" are shared at much higher rates.

Key Results: Topical Divergence

A significant finding is the lack of overlap between what is "Most Viewed" on the homepage and what is "Most Shared" socially. Many stories that garner millions of views never make it onto social media trending lists.

Venn diagram and bar chart illustrating the significant divergence between stories that are most viewed versus those most shared socially.

As a medium becomes more public (moving from Email to Twitter), the share of opinionated, niche content generally decreases in favor of factual, broad-interest news. However, "Opinion" remains a powerhouse for sharing across both Facebook and Email.

Limitations and Future Work

The study relies entirely on The New York Times, which may not represent the broader news ecosystem. Furthermore, while the paper identifies what is shared, it does not definitively explain why or how this bias affects the actual knowledge of consumers. Future research should explore the demographic differences of users on these platforms and the implications for news recommendation algorithms, which should be wary of using "popularity" as a universal signal.

Summary updated

Abstract

In a marked departure from traditional offline media, where all subscribers of a particular news media source (e.g., New York Times) used to get the same news stories through printed newspapers, online news media presents multiple options for the readers to consume news. For example, the subscribers of a media source can get news directly from the news website, or from what their peers share over social media sites like Facebook and Twitter. It is, however, unclear whether there are any differences in the news disseminated on these different online channels. In this work, we analyze data from a popular online news media site (nytimes.com), and show that each of these different channels tends to highlight some types of stories more than other stories. We believe that consumers of online news as well as media organizations need to be aware of such differences in various online news dissemination channels.
BibTeX
@inproceedings{chakraborty2021dissemination,
  author = {Abhijnan Chakraborty and Saptarshi Ghosh and Niloy Ganguly and Krishna P. Gummadi},
  title = {{Dissemination Biases of Social Media Channels: On the Topical Coverage of Socially Shared News}},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year = {2021},
  doi = {10.1609/icwsm.v10i1.14802},
}
Figure