Emotions, Demographics and Sociability in Twitter Interactions

Abstract

The social connections, or ties, individuals create affect their life outcomes, for example, by providing novel information that leads to new jobs or career opportunities. A host of socioeconomic and cognitive factors are believed to affect social interactions, but few of these factors have been empirically validated. In this research work, we extracted a large corpus of data from a popular social media platform that consists of geo-referenced messages, or tweets, posted from a major US metropolitan area. We linked these tweets to US Census data through their locations. This allowed us to measure emotions expressed in tweets posted from a specific area, and also use that area's socioeconomic and demographic characteristics in the analysis. We extracted the structure of social interactions from the people mentioned in tweets from that area. We find that at an aggregate level, areas where social media users engage in stronger, less diverse online social interactions are those where they express more negative emotions, like sadness and anger. With respect to demographics, these areas have larger numbers of Hispanic residents, lower mean household income, and lower education levels. Conversely, areas with weaker, more diverse online interactions are associated with happier, more positive feelings and also have better educated, younger and higher-earning residents. Our work highlights the value of linking social media data to traditional data sources, such as US Census, to drive novel analysis of online behavior.