A study of millions of online communications by members of the U.S. Congress found that negative partisan attacks received substantially more social media engagement than posts about legislative work, bipartisan cooperation or constituent services. The findings were published in the American Political Science Review.

Researchers examined about 4.7 million posts on Twitter, 2.4 million on Facebook and 184,000 official email newsletters, spanning 2009 to 2022. They used language models trained on messages reviewed by human experts to classify communications, including whether they featured partisan attacks, policy positions, legislative credit or assistance to constituents.

Across Facebook and Twitter, negative partisan messages drew more attention than other types of posts. On Twitter, attacks received 60 percent more likes, 80 percent more retweets and 79 percent more replies than the same politician’s average post. On Facebook, they received 86 percent more shares and 93 percent more comments than routine updates. Posts about bipartisan work and local services tended to attract less engagement than average.

The study also found that congressional rhetoric became more polarized between 2011 and 2022, while the share of communications devoted to negative partisan attacks rose from about five to ten percent in the early years to nearly 20 percent by the end of 2022. Messages focused on routine governance, including local funding and constituent assistance, did not keep pace. The researchers observed a sharp increase in Democratic attacks after the 2016 presidential election and a similar spike among Republicans after the 2020 election.

The authors cautioned that high online engagement should not be taken as evidence that voters generally prefer hostile political messaging. Survey research cited in the study indicates that most voters favor representatives who emphasize legislative achievements over partisan disputes. The researchers describe the risk that politicians could mistake reactions from a polarized online audience for broader public support.

The dataset ends in 2022 and does not cover the more recent growth of platforms such as Bluesky and Truth Social. The study, “Measuring Partisanship and Representation in Online Congressional Communications,” was authored by Michael Kistner, Michael Heseltine, Robert Alvarez, Maya Fitch, Lucas Lothamer and Elizabeth Simas.

Sources: PsyPost.