Examining online social behavior changes after migration: An empirical study based on OSN big data

Abstract

With fast urbanization and decreasing transportation cost, migration becomes more common. Previous studies have shown the important role of social networks in the process of migration, but little is known about the effect of migration on social networks. To fill the research gap, this study examines the effect of migration on online social behaviors (in terms of network evolution and social interaction), as well as the moderating effect of migrants' characteristics. We collect a four-month big dataset with 2.29 million records from one of the largest online social networks in China. We apply the propensity score matching combined with the difference-in-differences method to compare online social behavior changes after migration. Our results show that, for network evolution behavior, migration positively impacts on the number of tie formation, but non-significantly impacts on the number of tie decay; for social interaction behavior, migration increases the number of contacts but decreases the number of messages. We also find some moderating effects of migrants’ characteristics, including gender, age, and degree. This study provides big data empirical evidence and some new insights to our understandings of the impact of migration on online social network behavior.