Network Agenda Setting, or Networked Framing? (Non)correspondence Between User and Right-Wing Media Semantic Networks on YouTube

How does media shape and reflect right-wing rhetoric in the U.S.? Theories of media effects have moved towards networked approaches to agenda setting and framing, but it remains uncertain how issue attributes or frames emerge in the U.S. media ecosystem in which users themselves can shape political rhetoric through discussion on social media. We provide the largest test to date of the different predictions of networked agenda setting (NAS) theory and networked framing, through a semantic network analysis of all 19,112 video transcripts and 661,958,464 user comments posted on the YouTube channels of four major U.S. conservative media outlets between January 2019 and March 2021. Both overall, and within key topics like COVID-19 or Black Lives Matter, we find that user comments diverge strongly from video transcripts, with users repeatedly introducing associations, emotionally charged rhetoric, and conspiracy theories not originally present. Our results challenge claims by network agenda setting scholars that “objects and attributes can be transferred simultaneously in bundles” from the media agenda to the public agenda, but are more consistent with scholarship on networked framing. We argue that future work should strive to synthesize both approaches.