Post #4406624

2024-02-09 10:54 UTC

First, praise for this #CWTS Leiden Ranking Open Edition. The Leiden Ranking was already, with U-multirank the odd one out among global rankings. Leiden is the first to try radical transparency and has put in a lot of hard and good work. It did this collaboratively, further improving data quality at OpenAlex. That is a huge. The openness extends to code, going beyond just open data and includes the algorithms for assigning topical clusters. I really like the open communication on the project!

Replies (1)

@jeroenbosman@akademienl.social 2024-02-09 11:02

Then methodological issues. There are some inherent biases stemming from selected data sources and processing of the data. These hurt HSS and institutions strong in those fields. There's already biases in OpenAlex (built on crossref, pubmed and the former MAG) and the restriction to çore journals increases bias. Book chapters were included, but not monographs and again data will likely disregard much of the long tail important in humanities.

Open ##4406626