Inconsistency in metadata in OpenAlex

Dear all,

I hope this message finds you well.

I am noticing inconsistencies in the reference counts for some works in OpenAlex compared to the number of references in the original PDFs and the journal websites. Here are a few examples:

The work OpenAlex indicates 53 references, but both the original PDF and the journal’s website show a different count.

For the work OpenAlex, OpenAlex reports 43 references, but the PLOS ONE website and the PDF list 38.

Similarly, OpenAlex has 52 references in OpenAlex, but 49 in the original PDF and on the journal’s website.

Finally, OpenAlex has 30 references recorded in OpenAlex, while the PDF and the journal’s site show 29.

I have noticed this discrepancy occurring in thousands of cases. I would like to request help in understanding the reasons for these inconsistencies. Additionally, how might this impact the citation counts of the referenced works? For example, if there are five extra references, could these five works potentially receive citations, even though they were not cited in the original references of the citing papers?

Thank you for your attention, and I look forward to your response.

Sincerely,