Schöne, J., Parkinson, B., & Goldenberg, A.
(2021, January 2).
https://doi.org/10.31234/osf.io/x9e7u
Abstract
What type of emotional language spreads further in political discourses on social media? Previous research has focused on situations that primarily elicited negative emotions, showing that negative language tended to spread further. The current project addressed the gap introduced when looking only at negative situations by comparing the spread of emotional language in response to both predominantly positive and negative political situations. In Study 1, we examined the spread of emotional language among tweets related to the winning and losing parties in the 2016 US elections, finding that increased negativity (but not positivity) predicted content sharing in both situations. In Study 2, we compared the spread of emotional language in two separate situations: the celebration of the US Supreme Court approval of same-sex marriage (positive), and the Ferguson Unrest (negative), finding again that negativity spread further. These results shed light on the nature of political discourse and engagement.
General Discussion
The goal of the project was to investigate what types of emotional language spread further in response to negative and positive political situations. In Studies 1 (same situation) and 2 (separate situations),we examined the spread of emotional language in response to negative and positive situations. Results from both of our studies suggested that negative language tended to spread further both in negative and positive situations. Analysis of political affiliation in both studies indicated that the users that produced the negative language in the political celebrations were ingroup members (conservatives in Study 1 and liberals in Study 2). Analysis of negative content produced in celebrations shows that negative language was mainly used to describe hardships or past obstacles. Combined, these two studies shed light on the nature of political engagement online.