Individual Submission Summary

Direct link:

Using Social Media to Discern Variations in the Temporal Structure of Sociability Across American Cities

Sun, May 27, 12:30 to 13:45, Hilton Prague, Floor: LL, Congress Hall II - Exhibit Hall/Posters


Though social researchers have been interested in how temporal patterns of sociability vary within and across cities, the study of city-level dynamics at short timescales has traditionally been difficult due to an absence of data or tools for analysis. Social media data provide new quantitative methods to investigate this. In this study, we use aggregated social media data collected over three months to infer the temporal structure of sociability within American cities and investigate the variation in these patterns across cities. We find that cities cluster into three distinct types and show that their patterns are strongly shared within types. This classification of American cities echoes existing understandings, but also reveals previously unidentified differences in daily social life. The results also emphasize the methodological utility of bringing to bear interpretable statistical methodologies to social media data and the power that these data hold for answering classical communication research questions.