UKBsis Open Access Monitoring 2024 - Open Source [Datasets]

The 2024 UNL OA Monitor consolidates metadata from open sources CrossRef, DOAJ, Unpaywall and ISSN.org with institute received reports, and UKBsis datahub snapshots, applying business rules to classify publications (DOIs) into 4 OA categories

For this monitor, the project team requested the 2024 CRIS output from Dutch universities and university medical centres. This output is enriched with several open data sources and centrally aggregated into one dataset. Followed by a process of reviewing and improvement, resulting in a 2nd dataset.

The 2024 UNL OA monitor builds upon the results of the UNL OA monitor project <link to zenodo>. The monitor aims to combine a transparent and reproducible process with representative results.

Overall structure

The 2024 UNL OA monitor was created in a 2 step process. For both processes a Qlik Sense application was developed and used.

A total of 52, 893 unique DOIs were included in the analysis.

Step 1: automated processing of open metadata

The developed workflow consists of the following steps:

Importing the CRIS data provided by the universities; Cleaning this data (including removing duplicates); Enriching the CRIS data with information from open data sources; Adding the OA classification according to the business rule*;

Each DOI's ISSN is looked up within CrossRef. Each ISSN (Including the ISSNs from DOAJ) are looked up within ISSN.org. If the ISSN found per DOI appears in the DOAJ snapshot, OA Category "A" is awarded. If the ISSN of the DOI doesn't appear in the DOJA snapshot: if unpaywall states 'gold' or hybrid': OA category 'B' is awarded. If unpaywall states ‘ green’ : category 'C' is awarded. If unpaywall status 'closed' or 'bronze': category 'D' is awarded.

Step 1 results: the aggregated file is published on Zenodo; both the scripting and the used data from UPW, Crossref and ISSN are stored on github.

Result dataset: 'UNL OA monitor 2024_automated_result_open_sources_only': downloadable in the files section below

Step 2: review of correctness OA categories by UKB and institutions A review and, if applicable, correction of OA classification by both institutions and UKB.

Optionally changing the OA classification according to the business rules:

- If based on the DOI no ISSN was found, a manual check is done. If this results in an ISSN that is part of DOAJ: OA category A is added.

- If DOI has OA category status C or D but is listed as 'published open access' in the UKB datahub, the OA category is changed to B and the url to the publication is added.

- If DOI with OA category B, C or D is checked by an institution and the institution changes the OA category (B to C or D / C to B or D / D to B or C) a url to the publication or repository (when OA category is changed to C) is added.

Subsequently aggregating the OA information at the institutional levels. If requested corrections for 1 doi are inconsistent, highest ranking prevails. Reports are created at record, institutional and UNL levels.

In the final dataset column F shows the automated OA category result and column G shows the corrected OA category. To support users in checking the correctness of step 2, the transparancy of manual process step, column H shows the url to the url where the publication can be found.

Result dataset 'OA_monitor_2024_final_curated_result_20250917': link to Zenodo : https://doi.org/10.5281/zenodo.17141707

The Qlik Script & GitHub repository can be accessed here : https://github.com/UKBsis/UKBsis-2024-OA-Monitor-Project

** a.f.schalken@vu.nl