Data sources

Where the numbers come from.

Every source is free and openly licensed. The dot encodes how a figure is produced — counted from primary records, or compiled from curated sources. Full computation is on the methodology page.

Tier 1 — high-authority datasets (the rate data)

Retraction Watch

Weekly

Retraction events + categorized reasons (the plagiarism numerator).

Free, Crossref-hosted

OpenAlex

Monthly snapshots

Publications, ROR-resolved affiliations, disciplines (denominators + the DOI join).

Free, CC0

Crossref

Real-time

DOI metadata linking retractions to works.

Free API

ROR

Continuous

Canonical institution identifiers (the disambiguation backbone).

Free, CC0

Tier 2 — curated sources (the case data + signal)

Wikipedia

Weekly

Notable plagiarism cases (Category:Plagiarism controversies, scientific misconduct).

Free, CC-BY-SA

GDELT

Daily

Global news-volume index + recent coverage (current-events signal).

Free, DOC 2.0 API