Huge analysis of 320,000 careers suggests that productive researchers stay that way
-
-
Bluesky
-
-
-
-
-
X
Credit: demaerre/iStock via Getty
Researchers hoping to find late-career success are in for bad news. A study that tracked the publishing output of hundreds of thousands of scientists has found that the most important predictor of being a top performer in the late-career stage is being a high achiever early on1.
The paper, published in Quantitative Science Studies, analysed the publication record of 320,564 researchers from the 38 member nations of the Organisation for Economic Co-operation and Development (OECD). The study says that these researchers represent almost 80% of all ‘late career’ scientists — people who have at least 25 years of experience in publishing academic articles — in the world. It analysed up to 50 years’ worth of bibliographic data per researcher and considered 1.8 billion citations. The authors also investigated how career histories differed across 16 broad disciplines, including agriculture, medicine, physics and psychology, and looked for trends after normalizing for variations in publishing habits between disciplines.
Lead author Marek Kwiek, a social scientist at the Adam Mickiewicz University in Poznań, Poland, had already looked at variations in the publication output of Polish researchers in a 2018 paper2. Kwiek had found that success breeds success: the top group of performers generally stayed consistently productive. He found that the top 10% of Polish researchers produced 45% of all Polish publications, and that the people in this subset produced 12 times as many internationally co-authored publications as the average Polish academic.
A study from 1974 reported a similar phenomenon of ‘accumulative advantage’: productive scientists are more likely to stay productive, whereas scientists who have low productivity experience a decline in output3. The article hypothesized that this might be because of individual motivation or a result of extra resources being provided to high achievers.
Although the latest study by Kwiek and his colleague Łukasz Szymula is not the first to observe this relationship between productivity and success, it is by far the largest of its kind. The findings from Poland motivated the pair to investigate whether the trend would also be observed in the global research community.
How to become productive
In the study, published in September last year, Kwiek and Szymula separated researchers into ten equal groups (deciles) according to the number of papers and book chapters they had produced — which the authors used as a proxy for research productivity. They also weighted papers on the basis of a journal ‘prestige’ score, which was calculated by looking at the number of citations that a journal received between 2020 and 2024.
I’m going to halve my publication output. You should consider slow science, too