Reproducibility, Consensus and Reliability In Bibliometrics

Introduction Bibliometrics, and scientometrics in general, have been enjoying what seems to be an endless party. Far from stopping, the demand for bibliometric indicators from governmental bodies, administrators and researchers, is continuously growing. During this “give me the indicators” phase several solutions have been provided by the community, let say new and more sophisticated indicators, which in turn geared the transition to the present “give me the indicators, but really?” phase. The impressive penetration of bibliometric indicators in decision making processes, some of which are crucial in the development of researchers’ careers, has also brought the necessity for credibility on bibliometrics, and more specifically, on how it is practiced. Examples of improper use of bibliometric indicators have raised skepticism among users of bibliometric reports. As a scientific discipline, bibliometrics is subject to the principle of replication and corroboration of results, just like any other discipline. Precisely, the credibility of scientists goes hand in hand with the reproducibility of their results. The objective of this contribution is to bring attention to the importance of the reproducibility of the number of publications as an indicator of the quality of bibliometric reports.