Methodological approaches and data. A Brazilian case study

Abstract

This chapter outlines the methodology employed to collect and analyze data within the theoretical framework established in Chapter 1. It argues that the chosen approach is both suitable and adequate for investigating verbal impoliteness on social media and can serve as a model for future research in linguistics, human rights, and digital media studies. We describe in detail the procedures used to compile the Corpus of Digital Impoliteness (CID – Corpus de Impolidez Digital), an open-access dataset designed to support further research in this field. To demonstrate the relevance and adequacy of this methodology, we explain our rationale for selecting Brazilian Congresswomen Sâmia Bomfim and Erika Hilton, along with Federal Supreme Court Justice Cármen Lúcia, as focal subjects. Their inclusion reflects both their institutional roles and the surge of online verbal attacks following viral videos targeting them on Brazilian social media. The chapter also details the criteria guiding the selection of specific platforms – X/Twitter and YouTube – and discusses their significance in ensuring the representativeness of the dataset. Finally, it presents the initial quantitative results that form the empirical basis for the findings elaborated in Chapter 3.