grounded theory

  • a family of approaches developed in sociology as a response to hypothesis-first research models - Glaser and Strauss
  • instead of starting with a theory, you can do field work and generate ground up inductive hypothesis from the collected data
  • saturating the questions as you engage with data, develop concepts from micro to macro
  • construct a model as it emerges from the data
  • interpreting patterns and meaning
  • justifiable as by reference to the data
  • explanatory, not just descriptive
  • is it even possible to not confirm to existing models and ideas you have
  • constructivist grounded theory (Charmaz, 2006;2014) - emphasises that data and theory are co-constructed by researcher and participant - not fixed truths, interpretative, situated representations of experience
  • data collection and analysis in parallel, see emerging patterns
  • developing conceptual categories for theory building
  • theoretical sampling - decisions on who and what to study are guided by developing theories
  • use of multiple data types - elicited and non-elicited data
  • don’t look to confirm assumptions but allow the data to speak
  • line-by-line coding - each time you engage you do sense-making
    • what process is described or enacted, how is that defined by the participant
    • what actions and strategies are taking place
    • how you listen to the data may change
    • how is the participant claiming to thinking and feeling throughout - but remember inaccessibility of our own narratives, life as lived, narrative always evolving
    • shifts or turning points where the process changes
    • outcomes and effects of the process
  • active listening, active witnessing, finding the complexities without assigning meaning based on assumption
  • begin writing memos early that reflect recurring codes, early observations
  • memos as a thinking space - short analytical notes, questions that arise
  • relations - sequence, hierarchy, interaction
  • you keep collecting until the data isn’t revealing any more theories
  • show and share data to justify each theory or category, explain how they connect, create structure around what they reveal
  • final theories often communicated in everyday language, not too expert-specific
  • theories can be tested later with other methods