A content and frame analysis of the social media discourse on refugees and migrants, in the context of the United Nations and the European Union, 2015-2019

In this paper, we perform a cross-platform, cross-language comparison of social media content from Twitter, Reddit and YouTube related to refugees and migrants in the context of two major players in international protection: the United Nations and the European Union. The data covers a period of five years (2015-2019) and is analyzed using a combination of computational methods (word2vec) and computer-assisted content analysis (keyword-in-context) for nine languages. The results show that the most dominant categories of the global cleavage system are the globalists and the nation-statists. The most dominant topics present across the language clusters are Limiting and anti-migration, Global politics, Economy & job-market, and Humanitarian and compassionate attitudes towards migration. While these trends apply to the entire dataset, there are notable differences among language clusters and among the social media platforms, which point out the need to take into account the specific contexts when analyzing the digital discourse on migration