LLM-in-the-loop: Leveraging Large Language Model for Thematic Analysis

Evidence & verification

Source level

Peer-reviewed source
Grounded in a peer-reviewed publication. This describes the source; it does not mean the reported results have been independently reproduced.

Outcome verification

Not empirically evaluated here
A documented resource or intended workflow, not a demonstrated academic outcome. No empirical benefit is established by its inclusion.

Sources checked

2026-10-16 (UTC)

Primary-source abstract/metadata review only.[18] Performance, uptake and promotional claims are not treated as validated research gains.

Source-checked does not mean independently reproduced. See the Method page for how these labels are assigned.

Sources (1)

Research applicability

Peer-reviewed framework for thematic analysis in which a human and an LLM collaborate via in-context learning to generate a codebook.[18]

Getting started

Read the paper (arXiv:2310.15100); the collaboration procedure is described as an in-context-learning prompting workflow for generating a thematic-analysis codebook; consult the paper for any released materials.

Maintenance

Maintenance status not applicable; static publication, no released software audited.

License

Not audited; consult the arXiv record and the venue version for terms.

Workflow & human oversight

Resource-level classification is intentionally unspecified; instantiated workflows may differ. Humans define objectives, control data access, and check outputs.

Workflow level is not an evidence score; higher is not better evidence.

Approach

Prompts frame discussions with an LLM (e.g., GPT-3.5) to produce a final thematic-analysis codebook, demonstrated on survey datasets about music-listening experience and password-manager use.[18]

Limitations

Two case studies only (survey datasets); coding quality reported similar to human coders but not independently reproduced. Abstract-level review only - full-text methods and any released materials were not inspected, and no independent performance validation.

Reproducibility

Live arXiv abstract/metadata check, accessed 2026-10-16, for this review. No software, model or claimed outcome reproduced.

Adoption requirements

LLM access for in-context prompting (paper uses GPT-3.5 as an example) and qualitative research expertise to assess generated codebooks.

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.