Choosing a defined set of works, texts or objects to study closely, and giving reasons for what is in and what is out.
Key thinkers: Franco Moretti, Gillian Rose, Mieke Bal
The longer story
Much humanities research does not gather data from people at all. It gathers works: forty posters, six films, one composer's sketches, every review of a single exhibition. Deciding what goes into that set, the corpus, is a method, and it needs defending like any other.
A good corpus has stated boundaries (period, place, medium, maker) and an honest account of what was left out and why. Were these works chosen because they are typical, exceptional or simply available? Are you repeating a canon or questioning it? Examiners often ask "why these?", and "because I like them" is a start, not an answer.