Google Scholar duped and deduped – the aura of “robometrics”

Abstract

PurposeThis purpose of this paper is to discuss some of the problems that exist with Google Scholar, particularly regarding content spam and citation spam.Design/methodology/approachThe paper provides an analysis of how Google Scholar has been duped by real but manipulated documents and reference lists, as well as by fake documents and references. Details of research regarding the duping of Google Scholar is presented and a possible solution is offered.FindingsResearchers showed how easy it was to dupe Google Scholar. In one case, the researchers added invisible words to the first page of one of their conference papers (using the well‐known white letter on white screen/paper technique), and modified the content and bibliography of some of their already published papers, then posted them on the web to see if Google Scholar would bite, i.e. would improve their rank position, and increase the number of citations that the targeted papers received, and the number of papers published by the authors. Google Scholar did bite. While the size of Google Scholar kept growing at an impressive rate, the intellectual growth of the Google Scholar software has been stunted.Originality/valueThe paper makes the point that the best move from Google Scholar would be to realise that the existing metadata created by competent human indexers, cataloguers, librarians and other information professionals for tens of millions of scholarly documents is far superior to the parser's results.