Authors : Olivier Pourret, Dasapta Erwin Irawan, Najmeh Shaghaei, Elenora M. van Rijsingen, Lonni Besançon
The conventional assessment of scientists relies on a set of metrics which are mostly based on the production of scientific articles and their citations. These metrics are primarily established at the journal level (e.g., the Journal Impact Factor), the article-level (e.g., times cited), and the author level (e.g., h-index; Figure 1).
These metrics form the basis of criteria that have been widely used to measure institutional reputation, as well as that of authors and research groups. By relying mostly on citations (Langfeldt et al., 2021), however, they are inherently flawed in that they provide only a limited picture of scholarly production. Indeed, citations only count document use within scholarly works and thus provide a very limited view of the use and impact of an article.
Those reveal only the superficial dimensions of a research’s impact on society. Even within academia, citations are limited since the link they express does not hold any value (Tennant et al., 2019). As an example, one could be cited for the robustness of the presented work while the other could be cited for its main limitation (Aksnes et al., 2019).
As such, two articles could be cited the same number of times for very different reasons, and relying on citations to evaluate scientific work therefore displays obvious limitations (Tahamtan et al., 2016). Beyond this issue, however, the conventional assessment of scientists is clearly beneficial to some scientists more than others and does not reflect or encourage the dissemination of knowledge back to the public that is ultimately paying scientists.
This is visible in the Earth and natural sciences which has been organized to solve local community problems in dealing with the Earth system like groundwater hazards (Irawan et al., 2021; Dwivedi et al., 2022).