ACLAS Study Intelligence Releases Global Census of University Recognition Across Academic Databases

The Atlanta College of Liberal Arts and Sciences (ACLAS) just published something the higher-education world has never had: a global census of how academic databases recognize universities — 16,822 institutions worldwide, checked against four major academic infrastructure databases, released as open data with a DOI.

This isn't a ranking. It's a map of the invisible infrastructure that decides which institutions "exist" in the eyes of academia's data systems.

The Finding: Databases Discriminate

When a new university appears, which databases notice it — and how quickly? Our research team checked every university with an official website listed on Wikidata (16,822 institutions) against swot (JetBrains' academic domain list), ROR (Research Organization Registry), OpenAlex, and Wikidata itself. The results are striking:

41% of the world's universities are recognized by all four databases
34% are in three, 7% in two, and 18% in Wikidata alone
swot coverage is just 47% globally — but 68% in the United States vs 21% in India

There is a clear recognition ladder: new institutions are typically noticed first by swot (community-reviewed, lowest barrier), then Wikidata (notability), then ROR (curated registry), and finally OpenAlex (which ingests ROR upstream). Each rung has a higher bar than the last.

Why It Matters: The Infrastructure of Legitimacy

Academic databases are the plumbing of modern scholarship. If your institution isn't in ROR, your researchers' affiliations don't resolve. If you're not in OpenAlex, your publications are harder to attribute. These databases don't just describe academia — they shape who counts as part of it.

Yet until now, no one had measured how evenly this recognition is distributed. Our census shows it's not even at all: geography, institutional age, and database design choices all create systematic gaps.

A Curated Panel, Too

Alongside the global census, we're releasing a 35-institution curated panel spanning established online universities, new alternative institutions, controversial cases, and traditional elites as a baseline — each checked against all four databases with documented evidence. It validates the ladder: traditional elites average 2.8/4 databases, established online universities 2.6/4, new alternatives 1.7/4.

Open Data, Open Methods

Everything is published under CC-BY-4.0 with full methodology:

GitHub: github.com/aclascollege/academic-database-recognition-2026
Hugging Face: huggingface.co/datasets/ACLASCollege/academic-database-recognition-2026
Kaggle: kaggle.com/datasets/aclascollege/academic-database-recognition-2026
Zenodo (DOI): doi.org/10.5281/zenodo.23445888

The methodology documents every limitation we found — including a ~5% false-negative rate in our ROR matching and Wikidata's own coverage biases. We'd rather show our work than claim false precision.

"Academic infrastructure should be measurable. You can't fix a recognition gap you can't see — so we mapped it," said the ACLAS Study Intelligence team.

Get Involved

Are you a researcher studying higher-education systems, an institutional data manager, or just curious where your university stands? The data is yours:

Explore the 16,822-institution census on Kaggle or Hugging Face
Cite it via the Zenodo DOI: 10.5281/zenodo.23445888
Build on it — it's CC-BY-4.0, attribution only

This release joins four earlier ACLAS open datasets (AI policy audits, online degree tuition, Central Asia AI education), all published across GitHub, Hugging Face, Kaggle, and Zenodo.

About ACLAS Study Intelligence

Study Intelligence (aclas.si) is ACLAS's open research initiative for the intelligence era, dedicated to mapping the infrastructure of knowledge itself: how academic databases recognize institutions, how AI policies evolve, how education data moves. All findings are published open under CC-BY-4.0.

0 comments

More Programs

Open to Global Students