Dear Sachi Sri Kantha ,

Yes indeed a well-known and rather annoying matter. Your ‘counting percentages’ for number of citations in GS, RG and Scopus (100%, 77%, 55%) are pretty comparable to the ones of my own (100%, 82%, 78% and for WoS it is 75%).

Reasons are:

1. Main reason is the difference in the database used in making up the counts. Google scholar includes basically everything that can be considered as scientific literature (thesis, chapters, reports, non-indexed journals, etc.). Researchgate includes primarily articles uploaded in their own database. The same is true for Scopus, only Scopus indexed journals add to your counts.

2. Another important reason are the limitations of the search engine of the mentioned services. Google Scholar requires a specific set of information in order to be able to index items properly (a paper can be included but if the journal behind it does not provide proper info the references are not or at least improperly included in the database). The search engine behind Researchgate has some serious limitation in ‘recognizing’ certain references. So while a paper cites your publication it is not counted as citation because of this.

3. In some cases the reason is the difficulty to attribute a paper or citation to the right author (for this reason one should use ORCID id nowadays) and/or paper/journal (for this the DOI nr. system was introduced long time ago).

I am afraid that reason 1 is hard to change: One can only hope for a continuous improvement of the quality of their databases.

Reason 2 should be a matter of time: with the increasing progress in programming, speed, data handling and ‘learning ability’ of IT systems the recognition and assignment ability of those engines will become more accurate.

Reason 3 will become a problem ‘of the past’ since most journals assign DOI to the accepted papers and more and more authors have an ORCID ID.

See also:

https://www.researchgate.net/post/My_citation_rate_on_Google_Scholar_is_1300_with_415_added_since_2015_but_less_than_400_on_Research_Gate_why

https://www.researchgate.net/post/Can_anybody_explain_to_me_why_google_scholar_counts_much_higher_citations_than_Research_Gate

Article ResearchGate vs. Google Scholar: Which finds more early citations?

Method Information and tips related to search engines like Google S...

Best regards.

I appreciate the response and the three reasons mentioned by Rob Keller. As per cited reason 1, I have a follow up question, on the algorithm of Google Scholar.

I have sole-authored three books (two science reference books, and one neuroscience practical book in English). The two reference books [Prostitutes in Medical Literature (1991) and An Einstein Dictionary (1996)] were published by Greenwood Press in USA, with ISBN numbers. The neuroscience practical book (a translation from Japanese original, commissioned by the Oxford University Press) was published in 1998. Citations to these three books do not appear in my Google Scholar count. But, I have noted in the profiles of my colleagues, that their books have been included. Why such a variation in 'counting' within the same database?

If I have to add to Rob Keller response, then I just want to say that GS and RG ask users to identify papers authored by them. So if anyone verifies it wrongly by mistake or deliberately then calculations based on this anomaly will be incorrect. This becomes another reason for this variation.