Date published:

Science dropped out

Prof Marek Kwiek and his doctoral student Łukasz Szymula from Adam Mickiewicz University, Poznań examined how members of the scientific community abandon academic science and how this phenomenon of drop-out (defined as the discontinuation of publishing) varies by gender, academic disciplines and over time. The paper has just been published in the highly prestigious journal Higher Education.

"Worldwide ( also in Poland), we tend to believe that women continue to drop out of science noticeably faster and in higher percentages than men. However, it is no longer the case today," says Professor Marek Kwiek. - "We tend to overlook the fact that all scientists leave science (both men and women) and have not realised until now on a global scale in most STEMM disciplines (science, technology, engineering, mathematics and medicine), with every new research cohort entering science, the difference between men and women in the probability of quitting science is smaller. The issue lies in the attractiveness of the academic profession to all and the abandonment of science by all, not just women scientists. The underlying problem is widespread and growing. We begin an academic job, and we leave it. And the analyses of structured Big Data show it," adds Prof Marek Kwiek.

The researchers tracked the fate of individual scientists from their first publication (a longitudinal or longitudinal study). They measured the phenomenon traditionally referred to as "quitting science". Using publication metadata from the raw Scopus database (provided by Elsevier) - a global bibliometric database of publications and citations - they pursued the publication career details of scientists from 38 OECD countries who started publishing in 2000 and 2010, totalling around 400,000 individuals. The study is limited to 16 STEMM disciplines. The researchers analysed the individual scientific output of both groups up to 2022. - "With an increasing number of women in science and an increasing number of female scientists in individual academic cohorts between 2000 and 2010, the phenomenon of quitting science is becoming less gender-differentiated," emphasises Prof. Marek Kwiek and M.Sc. Lukasz Szymula. "Furthermore, there are substantially different variations at the discipline level and over time. As the researchers say, quitting science means something different for men and something different for women depending on the discipline (particularly in heavily male-dominated disciplines such as mathematics, physics, computer science or engineering); moreover, it implies something different for scientists from different age cohorts entering the scientific workforce.

2.1 million scientists were surveyed in detail year by year with the help of Big Data analytics infrastructure. It emerged that a third of the scientists in the first group (cohort 2000) quit science after 5 years, and half left after 10 years. In contrast, two-thirds of scientists had stopped publishing - quit science - within 20 years, with the percentage leaving for all disciplines being consistently lower for men. - "Behind the aggregate changes for all disciplines combined, there are nuanced changes at the level of individual disciplines," the study's authors emphasise. Furthermore, they add: - "In the mathematised disciplines, there are no differences between men and women in drop-outs. The same is the case for all STEMM disciplines combined. If one already publishes in these disciplines, chances of staying in science (i.e. continuing to publish) do not differentiate scientists into men and women.

Remarkably, scientists in Poland "disappear from science" less frequently than their colleagues in the richest economies of the world. - "Poland has meagre drop-out rates regardless of the selected cohort and in all analysed disciplines; surprisingly, in the global context, the differences between men and women even, for the oldest cohorts, are not high," says Prof. Marek Kwiek. As the researchers add, the high rate of remaining in science may indicate both the high attractiveness of an academic career and the lack of good employment opportunities for scientists outside the educational sector.

More in the article: Kwiek, M., Szymula, L. Quantifying attrition in science: a cohort-based, longitudinal study of scientists in 38 OECD countries. Higher Education (2024). https://doi.org/10.1007/s10734-024-01284-0