Interplay of message features and source: predicting twitter users’ engagement behaviors following a terrorist attack

ABSTRACT There has been limited evidence regarding how disaster message features and source of social media communication independently and jointly affect users’ engagement behaviors. We conducted a content analysis of tweets collected from the Ariana Grande concert terrorist attack. Results from negative binomial regressions (N = 687) showed that the social support function of tweets affected reactive engagement, a higher level of content richness increased engagement, and source type moderated the effects of communication styles on engagement during the terrorist attack. For example, traditional media (vs. individuals) adopting an informal tone had a higher level of reactive and proactive engagement.