THE PROPERTY THAT IS FACTUALLY BEING EVALUATED WHEN THEY SAY THEY EVALUATE IMPACT

Abstract

Introduction. Analysis of the definitions and perceptions of the notion of “impact”, introduced into circulation by Yu. Garfield in 1955, does not allow to assert that in the scientometrics literature there is a strict definition of this notion at all. Since it is assumed that citedness figures are its reliable indicator, it is necessary to discover what property is actually reflected by this indicator.Materials and methods. Analytical interpretation of the scientific literature related to this problem since 1955.Results. Comparison of the notion of “impact” with the possibilities of its quantitative evaluation by citedness figures demonstrated the unreliability of this basic scientometric method for the assessment of exactly “impact” as in terms of cause-and-effect relationships, “impact” may or may not be the reason for the use of scientific documents reflected in their citedness figures. In other words, citedness is not a very reliable proxy (substitute indicator) to be used for assessing poorly defined (as it was shown in my previous article published in Scholarly Research and Information; 2019;2(1):63–73) notion of “impact”. At the same time, citedness figures reflect the value of cited scientific documents (by reflecting their use while the creation of citing documents).Discussion and Conclusions. If “impact” should not be considered as a key notion of scientometrics, then its place can naturally be taken by the notion of the value of cited documents, their totalities, creators, etc.