Upgrade StorageRetracted articles: Solely identified by the Retraction Watch database or also by Crossref API?
Retracted scientific articles can be identified via Retraction Watch’s CSV-based database and via the Crossref API.
According to the YouTube video Exploring Retraction Watch Data in Crossref that was posted by Crossref an article may be listed as retracted in
a) Retraction Watch only,
b) Crossref only, or
c) both Retraction Watch and Crossref.
Retraction Watch was acquired by Crossref in 2023; since then, articles marked as retracted by Retraction Watch have also been marked as retracted on Crossref.
Additionally, a publisher can update metadata on Crossref for an article published by that publisher and mark it as retracted. It is unclear to me whether Crossref updates from publishers are also fed back into the Retraction Watch database.
At the same time, there has been a really great integration for Retraction Watch in Zotero since 2019, which means that articles marked as retracted on Retraction Watch are flagged in Zotero with a corresponding warning message.
I suspect that this integration means that in Zotero, only those articles are marked as retracted that are also marked as retracted in the Retraction Watch database. In other words, articles that are marked as retracted only on Crossref are likely not marked as retracted in Zotero.
For example, the following article is marked as retracted both on the publisher’s official website and on Crossref:
Publisher: https://doi.org/10.1016/j.iswa.2025.200599
Crossref: https://api.crossref.org/works/10.1016/j.iswa.2025.200599
When I add this article to Zotero, it is not displayed there as retracted.
According to Crossref, this article was marked as retracted by the publisher as early as December 1, 2025; the CSV-based Retraction Watch database was last updated on March 30, 2026, at the time of this post. This means that an automatic mechanism ensuring that articles marked as retracted by a publisher on Crossref are also reported as retracted on Retraction Watch should presumably have been in place long ago.
Dear Zotero Team, can you confirm that, when it comes to retracted articles, Zotero currently only checks the Retraction Watch database and not Crossref? If so, wouldn’t it make sense to expand the check to include Crossref as well?
According to the YouTube video Exploring Retraction Watch Data in Crossref that was posted by Crossref an article may be listed as retracted in
a) Retraction Watch only,
b) Crossref only, or
c) both Retraction Watch and Crossref.
Retraction Watch was acquired by Crossref in 2023; since then, articles marked as retracted by Retraction Watch have also been marked as retracted on Crossref.
Additionally, a publisher can update metadata on Crossref for an article published by that publisher and mark it as retracted. It is unclear to me whether Crossref updates from publishers are also fed back into the Retraction Watch database.
At the same time, there has been a really great integration for Retraction Watch in Zotero since 2019, which means that articles marked as retracted on Retraction Watch are flagged in Zotero with a corresponding warning message.
I suspect that this integration means that in Zotero, only those articles are marked as retracted that are also marked as retracted in the Retraction Watch database. In other words, articles that are marked as retracted only on Crossref are likely not marked as retracted in Zotero.
For example, the following article is marked as retracted both on the publisher’s official website and on Crossref:
Publisher: https://doi.org/10.1016/j.iswa.2025.200599
Crossref: https://api.crossref.org/works/10.1016/j.iswa.2025.200599
When I add this article to Zotero, it is not displayed there as retracted.
According to Crossref, this article was marked as retracted by the publisher as early as December 1, 2025; the CSV-based Retraction Watch database was last updated on March 30, 2026, at the time of this post. This means that an automatic mechanism ensuring that articles marked as retracted by a publisher on Crossref are also reported as retracted on Retraction Watch should presumably have been in place long ago.
Dear Zotero Team, can you confirm that, when it comes to retracted articles, Zotero currently only checks the Retraction Watch database and not Crossref? If so, wouldn’t it make sense to expand the check to include Crossref as well?