October 31, 2014
Three publicists working at Nature revealed their study on the 100 most cited scientific articles of all time, based on data provided by Thomson Reuters from their Web of Science database. For me, the most interesting part of the article is the xlsx file, which is attached to the main content and provides data on the same subject from Google Scholar.
I agree that the Google Scholar literature base is “poorly characterized” and in contrary to Web of Science it also includes non peer-reviewed content. More importantly, however, it includes books, which changes the whole essence of the ranking. Six of the top ten positions are books, which is quite surprising. This shows that the hegemony of scientific papers is much weaker than we assume. Even more surprising is the fact that the 7th position is taken up by The structure of scientific revolutions by Thomas Khun, while in the top 100 most cited articles in the Web of Science, there are no works from the field of humanities (except for a few articles in psychology).
The ranking by Web of Science confirms the well known fact that biologists cite a lot, and that biological papers and journals are at the top of every citation ranking. However, the ranking by Google Scholar shows that researchers working in the humanities also cite a lot, although they do not cite papers as frequently as books. This fact was in my opinion neglected in discussions on citation patterns, including in those concerning the role of monographs in humanities.
Even more striking for me is that position number 12 is The Discovery of Grounded Theory by Barney Glaser and Anselm Strauss. As a sociologist I am pretty sure that grounded theory is not a dominant paradigm of conducting research in social sciences. Followers of Glaser and Strauss rather represent a well organized opposition to so called classical methodology. What we can learn from this example is that to be firmly cited it is good to create your own ‘school. It does not need to be the most popular one, it’s enough that it’s ‘yours’. What is also important, the work by Glaser and Strauss is about the method, similarly to the majority of top cited articles in the fields of life science. It seems that it is always better to develop a new method than to make new discoveries in terms of citations.