QualFrame studies: documented LLM qualitative-coding evaluation methods
Repository-documented methods and configurations for LLM-assisted deductive qualitative coding of interviews and published reports. Researchers define codebooks and conditions; models propose coded passages with quotation traceability. Results are not shared in the inspected repository.
L2Research workflow (workflow level, not a quality rating) Source: Repository-documented Outcome: Not empirically evaluated here
Research applicability
Specific documented LLM-assisted qualitative coding and evaluation methods with traceable quotations; resource is methods/configurations rather than demonstrated accuracy.
Getting started
Untested editorial suggestion: read the retained primary evidence and limitations; researchers must verify source support before use.
Maintenance
Dated source snapshot only; present availability, maintenance and version parity require separate checks.
License
Source visibility is not reuse clearance; source/model/data licenses and rights require separate checking.
Workflow & human oversight
Scope: Documentation-only; no repository code executed, outcomes reproduced, private inputs fetched or production decisions changed. Human control: Researchers define conditions and interpret comparisons; evaluation scores proposals before human review. No prior approval gate is asserted. Stage connections: - Researchers select codebooks, documents, prompts and study conditions - LLMs propose codes and exact supporting passages - QualFrame locates quotations and logs provenance - QualBench compares proposals with reference annotations and reports metrics Stopping: No additional stopping policy established by this representation. Missing evidence: No autonomous planning or iterative research evidenced. Results/output data not shared in inspected repository.
Workflow level is not an evidence score; higher is not better evidence.
Approach
- Researchers select codebooks, documents, prompts and study conditions - LLMs propose codes and exact supporting passages - QualFrame locates quotations and logs provenance - QualBench compares proposals with reference annotations and reports metrics
Outcomes (as reported by the originating team)
Documented method or intended capability; no empirical research benefit established by this review.
Limitations
- Study A reference partly reviewed prior model-assisted coding, not independent ground truth. - Study B unannotated units treated as negatives are not confirmed absences; historical and model scopes differ. - Study A conditions expanded after exploration; prompt packages differ in multiple respects. - No efficacy or cost/performance improvement claimed; no private interview data accessed.
Reproducibility
AI editorial source review only. No candidate code, model, agent, experiment or reported outcome independently executed or reproduced. Documentation-only; no repository code executed, outcomes reproduced, private inputs fetched or production decisions changed.
Adoption requirements
Untested editorial suggestion: define a bounded lawful task, inspect rights/model/data requirements and verify results before use.
Comments
No comments yet.
Add a comment
Comments are held for moderation before appearing. Critical or corrective comments are welcome — especially evidence that an entry overstates what has been verified.