Field-level university and country rankings from the OpenAlex authors snapshot

[Nick Selby]

Hi all,

I'm new to the mailing list, and I wanted to share a pipeline I built on top of the OpenAlex public S3 snapshot that computes successive h-indices (Schubert 2007) up the researcher → institution → country hierarchy: h₂ per institution and h₃ per country.

The highlights:

1. The overall institution-level ranking is, in substance, a ranking of medical school size
2. China's applied-science strength is real
3. Brazil is really good at Dentistry

The full write-up with field-level tables and country rankings is on my blog: https://rupumped.github.io/h2.html. The pipeline code is on GitHub: https://github.com/rupumped/successive-h-indices.

One natural extension would be ranking university departments. The hierarchy extends cleanly to researcher → department → institution → country, with h₁ through h₄. Departments are field-specialized by construction, which would also eliminate the need to infer each author's modal field from topic weights. The missing piece is departmental affiliation data, which OpenAlex doesn't currently expose. That would be the single most valuable addition for scientometric work of this kind.

Happy to share more detail on the pipeline if it's useful. Thank you for all the great work you do!

Nick