Hello everyone,
Recently, at the University of Groningen we discovered that OpenAlex is misclassifying posters (and possibly other output types) registered in our CRIS as research articles.
During their career, the three of them have mostly carried out their research at the University of Groningen. However, the number of articles attributed to them in OpenAlex is disproportionate to the number of articles in the CRIS of the university (even when conference papers are considered as articles). Further, we found that several of the works labeled as "articles" in OpenAlex are actually outputs saved as "posters" in our CRIS.
Overview table of comparison:
The issue at hand has two important consequences already visible in the data:
a) A very large number of works in OpenAlex (due to the misclassification and possible duplication of works)
b) Large FWCI values in OpenAlex (due to large amount of non-cited works produced by the misclassification)
Has anyone come across a similar issue? This problem is only visible when examining data at individual level and not when using aggregated data.
Best regards,
Miguel Santin