ABSTRACT
A growing amount content is published worldwide every day by millions of social media users. Most of this content is public, permanent, and searchable. At the same time, the number of studies proposing different techniques and methodologies to exploit this content as data for researchers in different disciplines is also growing. This article presents an up-to-date literature review that frames available studies using Facebook, Twitter, and YouTube as data sources, in the perspective of traditional approaches for social scientists: ethnographical, statistical, and computational. The aim is to offer an overview of strengths and weaknesses of different approaches in the context of the possibilities offered by the different platforms.
