Full Length ArticleModeling the role of message content and influencers in social media rebroadcasting

We develop a model that examines the role of content, content-user fit, and influence on social media rebroadcasting behavior. While previous research has studied the role of content or the role of influence in the spread of social media content separately, none has simultaneously examined both in an effort to assess the relative effects of each. Our modeling approach also accounts for a message's “fit” with users, based on the content of the message and the content of messages typically shared by users.As an empirical application, we examine how Twitter posts originating from top business schools are subsequently rebroadcasted (or retweeted) by other users. We employ an individual-level split hazard model that accounts for variation in rebroadcasting decisions related to (1) content, (2) the content-user fit and (3) the influence of other users. We find that the rebroadcasting a message depends not only on message content but also on the message's fit with a user. Our analysis also yields measures of influence and susceptibility to influence for each user, which can be used to identify influential social media users. We demonstrate how our approach can be used to evaluate different types of seeding strategies designed to increase the reach of social media messages.