XPAC content: "Bulletin of Miscellaneous Information (Royal Gardens Kew)"

[D.zkreet Deazoo]

Hello everyone,

I am exploring OpenAlex and I have found a weird journal entity.

The journal name is:  Bulletin of Miscellaneous Information (Royal Gardens Kew)

It has 7M documents.

Some of these documents are books from the 18th century yet listed as 2020 publication year, and have (apparently) nothing to do with that journal.

Some are highly cited documents from 2020, such as w4205483821. This work is cited a few thousand times.

I have 2 questions:

- is this because xpac needs to be further cleaned?

- Are these citations taken in consideration when calculating FWCI ?

Thank you for your  efforts.

[Gabor Schubert]

The highly cited example  (https://openalex.org/works/w4205483821) you gave now points to a record, which currently has the source "Internet Archive" in OpenAlex (not Bull Misc Info). It is actually a doublet of a New Journal England of Medicine article (https://openalex.org/works/W3113734454, DOI: https://doi.org/10.1056/NEJMoa2035389). It seems that many Bull Misc Info records have links to objects in the Internet Archive and they are erroneously indexed to this journal. Currently there are "only" 50K documents indexed to this journal without xpac (https://openalex.org/works?filter=primary_location.source.id:s4210203682), and probably even these include erroneously indexed objects. Most likely this is an indexing error of xpac (and other) objects.

Such deduplication and indexing errors can cause different problems in the automatic calculation of citation counts and derivative indices like FWCI, so one should be careful when using numbers directly from OpenAlex without further cleaning and/or error checking.

Best regards,

Gabor Schubert