Positive vs. Negative Emotions and Network Size: An Exploratory Study of Twitter Users

Introduction

Self-expression is one of the main motivators for sharing content in social media (Shao 2009). Self-expression is not only a way of presenting the self, but it can also be used to control the impressions of viewers and foster supportive relationships (Shao 2009). However, emotions expressed in status updates can have an impact on a user’s network size (Lin and Qiu 2012). Hutto et al. (2013), for example, found that negative emotions expressed in tweets reduce, while positive emotions facilitate, network growth. In a similar vein, Al-Saggaf & Ceric (2016), who explored the relationship between the expression of boredom and excitement on Twitter and network size, i.e. the number of followers and number of friends, found that generally speaking users who expressed boredom had smaller network sizes compared to users who expressed excitement. Similarly, Al-Saggaf, Utz, & Lin (2016), who explored the relationship between the expression of loneliness, emotion loved, sadness and happiness on Twitter and network size, found that users who expressed loneliness had smaller network sizes compared to users who expressed emotion loved and users who expressed sadness had less friends than users who expressed happiness. Al-Saggaf, Utz, & Lin’s (2016) findings are consistent with Dunder's et al. (2016) findings. The explanation given by Al-Saggaf, Utz, & Lin (2016) was that it could be because expressing negative emotions is less attractive to a user’s network (Utz 2015) or that loneliness might lead to personality attributions, i.e. the assumption that something is wrong with the lonely person. Regardless, these findings suggest not all users reap the benefits of self-expression; especially when they don’t address others in their status updates. Both Al-Saggaf & Ceric and Al-Saggaf, Utz, & Lin (2016) studies compared users who tweeted a negative emotion with users who tweeted a positive emotion using a single tweet. While these studies contributed to the limited literature on the relationship between emotions and network size, the fact that they made inferences about the relationship between expression of emotions and network size based on a single tweet weakened the strength of their inferences. This study looked at network size for each group of users, i.e. ‘bored’, ‘excited’, ‘lonely’, loved’, ‘sad’ and ‘happy,’ separately at Time 1 when they expressed these emotions and at Time 2 when they expressed these emotions again later. Comparing network sizes for each group of users at Time 1 and Time 2 allowed within group comparisons. The study also compared the network sizes of each negative emotion with its opposite emotion. The study collected 35,096 tweets posted in English to Twitter between 16 December 2016 and 24 December 2016 and performed several statistical tests to address the following research questions:

• •

  Do users who express excitement, happiness and being loved twice or more in their tweets gain more followers and friends than users who express boredom, loneliness and sadness twice or more in their tweets or less?

• •

  Do users who express positive emotions twice or more have more followers and friends compared to users who express negative emotions or less?

• •

  Do users who express boredom, excitement, loneliness, emotion loved, sadness and happiness twice or more interact more with their networks or less?