Evaluation in the era of team research

It is known that contemporary science no longer has anything to do with individual research, which was still the norm at the beginning of the twentieth century.^H a century. Platform-specific bibliometric data Web of Science This clearly showed: of all sciences combined, almost 90% of articles published in 1900 were signed by a single author. Today, this percentage is less than 3%, while the percentage of articles that include four or five authors exceeds 30%. The social sciences are not excluded, because only a quarter of publications in these disciplines are currently the work of one person, while 32% of them are signed by three or four researchers. The average number of authors per article is 6.2 in the “hard” sciences and technologies, and 3.6 in the humanities and social sciences. The trend is therefore general across all disciplines, and with the exception of the humanities, there is in fact no longer any individual research.

But this reality seems to conflict with the persistence of the individualistic image that is still centered around the “stars.” The Nobel Prizes in Science continue to perpetuate the myth of individual genius by praising a few people each year, often forgetting the collective nature of award-winning knowledge.

The reality of the research contradicts the persistence of an image centered around “stars.”

Even more problematic is the promotion of the magazine nature supposed to ” Rising stars “, which is mainly based on the growth in the number of publications and citations of “elected” people. But what is striking in the growth curves of the number of publications shown to these people between 2015 and 2022 is the very large number of articles published each year. In one case, we go from 5 in In 2015 to more than 60 in 2022! And in others, from approximately 30 in 2015 to more than 80 in 2022!

Of course, citations also follow an exponential curve. However, these statements leave the role of the co-authors in the shadows. We see the arbitrariness of these indicators, and to make the collective work visible, we can, for example, track the curves by dividing the number of articles by the number of co-authors and do the same for citations. We may thus discover other “stars”!

Although some (but not all) journals require that each author acknowledge their contribution, this remains vague and has no real impact on how credit is given to co-authors. What about the problem of attributing responsibility in the case of fraud? In practical terms, we then move on to the team leader, who is in a way the CEO of his ‘factory’. This issue of responsibility becomes more difficult to determine when the number of co-authors is high or when they work in different institutions, as is in fact the case in more than 70% of publications. The problem gets worse if we add the fact that currently, in more than 25% of articles, collaborators are located in different countries. Finally, we can imagine the difficulty of controlling the data of an involved researcher who has co-authored 170 articles with many different collaborators distributed in different countries – and this situation does exist.

All these aspects of research synthesis clearly show that current practices of evaluating and encouraging researchers no longer truly fit the prevailing mode of knowledge production. Only seriously questioning the individualistic—not to mention selfish—conception of research that underlies these practices will make it possible to find reliable solutions to the new demands for recognition that the research discipline faces today.