Conducted sustained multi-sited ethnographic fieldwork across Self-Help Groups in the Udaipur zone among the most intensive qualitative datasets in the project. Moved between observation, formal interviewing, and informal presence to build longitudinal relationships with respondents.
Qualitative data analysis (QDA) was conducted through a deliberate hybrid methodology: API-assisted rapid first-pass coding for thematic identification across the full corpus, meshed with long, unhurried manual re-immersion in the data reading slowly, sitting with contradictions, following threads the algorithm surfaces but cannot interpret. The two registers inform each other rather than replace one another.
The data feeds directly into the RCT running parallel to the ethnography, producing a texture the trial alone cannot capture.