Mapping language literacy at scale: a case study on Facebook

Abstract

AbstractLiteracy is one of the most fundamental skills for people to access and navigate today’s digital environment. This work systematically studies the language literacy skills of online populations for more than 160 countries and regions across the world, including many low-resourced countries where official literacy data are particularly sparse. Leveraging public data on Facebook, we develop a population-level literacy estimate for the online population that is based on aggregated and de-identified public posts written by adult Facebook users globally, significantly improving both the coverage and resolution of existing literacy tracking data. We found that, on Facebook, women collectively show higher language literacy than men in many countries, but substantial gaps remain in Africa and Asia. Further, our analysis reveals a considerable regional gap within a country that is associated with multiple socio-technical inequalities, suggesting an “inequality paradox” – where the online language skill disparity interacts with offline socioeconomic inequalities in complex ways. These findings have implications for global women’s empowerment and socioeconomic inequalities.