How social media predicts news viewership - The moderating role of news theme prominence

Jie Ren, Gaurav Sabnis, Hang Dong, Jeffrey V. Nickerson

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper explores the predictive power of social media with respect to news viewership in a business context. Using 0.7 million pieces of stock market news and 37.3 million stocks-related microblogs in 2013, we find that this power of social media is stronger for low prominence news themes (e.g., news about Blue Apron) than for high-prominence news themes (e.g., news about Microsoft). Specifically, the intensity of social media sentiment (either positive or negative) and social media volume are positively associated with news viewership; social media credibility is negatively associated with news viewership. More importantly, the impacts of social media sentiment, positive sentiment social media volume and social media credibility are all stronger for low-prominence news themes. Our findings quantify the power of the crowd in shaping news narratives. Especially, our findings describe how the opinions of the crowd can build up the popularity of non-elite news themes.

Original languageEnglish
Title of host publication26th Americas Conference on Information Systems, AMCIS 2020
ISBN (Electronic)9781733632546
StatePublished - 2020
Event26th Americas Conference on Information Systems, AMCIS 2020 - Salt Lake City, Virtual, United States
Duration: 10 Aug 202014 Aug 2020

Publication series

Name26th Americas Conference on Information Systems, AMCIS 2020

Conference

Conference26th Americas Conference on Information Systems, AMCIS 2020
Country/TerritoryUnited States
CitySalt Lake City, Virtual
Period10/08/2014/08/20

Keywords

  • Mass media
  • News theme prominence
  • News viewership
  • Social media

Fingerprint

Dive into the research topics of 'How social media predicts news viewership - The moderating role of news theme prominence'. Together they form a unique fingerprint.

Cite this