Published March 4, 2026 | Version v1.0.0
Software Open

ncukondo-research/ai-vs-human-feedback-analysis-open-data: v1.0.0: Initial Release

Authors/Creators

Description

Open Data: AI-Generated versus Human Supervisor Feedback

Initial public release of anonymized evaluation data, analysis code, and AI feedback generation source code for:

Kondo T, Donkers J, Nishigori H, Rovers S, Heeneman S. AI-generated versus human supervisor feedback on medical students' clinical clerkship logs: convergent mixed methods study. JMIR Medical Education. (under review)

Contents

  • Anonymized data (data/): Rubric scores, thematic analysis codes, and qualitative theme mappings
  • Analysis code (analysis/): R scripts for quantitative, qualitative, and mixed methods analysis
  • AI feedback generation system (feedback-generation/): TypeScript serverless API using GPT-4o

Files

ncukondo-research/ai-vs-human-feedback-analysis-open-data-v1.0.0.zip

Files (205.2 kB)

Additional details

Related works

Is supplement to
Software: https://github.com/ncukondo-research/ai-vs-human-feedback-analysis-open-data/tree/v1.0.0 (URL)

Software

Repository URL
https://github.com/ncukondo-research/ai-vs-human-feedback-analysis-open-data