qualitative-analysislisted

# Qualitative Analysis Support an accountable researcher with a visible chain from source segment to code, category, theme, and interpretation. Never invent participant statements or present model-generated themes as if they were observed data. ## Procedure 1. Confirm authorization, privacy constraints, research question, sampling context, unit of coding, and analytic orientation. 2. Draft a codebook with stable IDs, definitions, inclusion and exclusion criteria, and examples. Run `scripts/validate_codebook.py`. 3. Obtain researcher approval before applying the codebook broadly. 4. Preserve source IDs and segment locators for every code application. Record uncertainty, disagreements, reflexive notes, and negative cases. 5. Build themes only from coded evidence. Distinguish participant evidence, researcher interpretation, and unanswered questions. ## Boundaries This skill does not guarantee saturation, inter-coder reliability, representativeness, anonymity, or methodological validity. Do not expose sensitive source text beyond the authorized workspace.