Research case
LLM-assisted content analysis, LACA (Chew et al., RTI 2023)
Research-institute team proposed a workflow that brings an LLM into deductive coding as a co-coder, and benchmarked it on four public coding datasets.
What is reported
GPT-3.5 often reached agreement comparable to human coders, and the workflow flagged codes where the model was essentially guessing.
Before you use it
Use the “is the model guessing?” check per code; agreement varies a lot between codes in the same codebook.
Sources and review
Source checked · Not tested by this site
Checked the source record and abstract. Results describe the reported study, not an independent replication by this site.
License and reuse
No license has been recorded. Check the source’s license before reusing code, models or data.
A license note is not a license audit. Open source does not guarantee free hosting, model access or commercial reuse.