Retraction Watch Database
Crossref (with The Center for Scientific Integrity)
Tens of thousands of retracted papers, each tagged with why: fraud, plagiarism, fake peer review, or error.
The largest database of retractions, built by Retraction Watch and opened up after Crossref acquired it in 2023. Each row is a retraction, expression of concern, or correction, with the original paper, journal, publisher, authors, countries, and coded reasons. It is updated every working day and distributed as CSV and through the Crossref API.
Access
- Cost
- FreeFree CSV from a git repository and free Crossref API access.
- License
- License not stated
- Sign-up
- No sign-up
- How you get it
- API and Bulk downloadCSV, JSON, API
What's in it
- One row is
- One row per retraction or correction notice
- Covers
- Worldwide
- Years
- Retractions from the 1920s to present, most since 2000
- Updated
- Daily
- Size
- Tens of thousands of retraction notices
Quality
Hand-curated, with a detailed reason taxonomy, and now openly maintained daily.
- Batch retractions from paper mills can dominate counts
- Reasons depend on what the notices disclose
Connects to
- OpenAlex: Citation counts for retracted papers before and after retraction
- PubMed Baseline and Update Files: Biomedical context and MeSH topics
Questions it could answer
- Which publishers and countries account for the most retractions for fake peer review?
- How long after publication are papers typically retracted?