Almost 50% of scientists give up their research within ten years, a big study shows
Almost 50% of scientists give up their research within ten years, a big study shows
A study with almost 400,000 scientists from 38 countries shows that a third of them leave science within five years after the publication of their first work, and almost half within a decade.
"We have always thought about and knew that people leave science, but the extent that happens to us was somehow unknown to us," says Marek Kwiek, co-author of the study and researcher for academic careers at the Adam-Mickiewicz University in Poznan, Poland.
The study is the greatest attempt to quantify the number of people who leave science - earlier studies were limited and mainly focused on scientists in the United States.
"If you have such large data, it will be more urgently recognizable that this is a problem," says Joya Misra, sociologist at the University of Massachusetts, at thehern, gender issues and inequality in science.
departure from the laboratory
Kwiek and his colleagues followed the publication career of two groups: 142,776 scientists (including 52,115 women) who started publishing in 2000, and 232,843 scientists (97,145 women) who started to publish in 2010.
The scientists came from countries such as the United States, Japan, South Korea and various European nations and represented 16 scientific disciplines.
The study showed that a third of all scientists in the group of 2000 had discontinued publishing within five years. This rose to about half within ten years and almost two thirds until 2019 (see "Academic Exodus"). Women had about 12% higher than men to leave science after five or ten years. Until 2019, only 29% of women published in this group, while it was the case in almost 34% of men.
The 2010 group showed a closer gender gap: Around 41% of women and 42% of men continued to publish nine years after their first publication. This improvement is promising, says Damani White-Lewis, researcher for university education and academic careers at the University of Pennsylvania in Philadelphia. "It is always good to know when we make progress because we have to be able to replicate these things."
In some scientific disciplines - especially in the life sciences - there were striking differences between men and women. For example, the likelihood of women in biology to leave science after ten years was 58%; In men it was almost 49%.
In contrast,In contrast, the likelihood of women in physics was to go after ten years (approx. 48%), almost as high as in men (47%). There were also hardly any gender differences for mathematics, engineering and computer science - all areas in which women tend to be underrepresented.
The results "Perform the necessary and important attention in the species and, as we can promote success, success and commitment in research," says White-Lewis.
reasons for leaving
Misra points out that the actual gender differences could be larger than the publication data suggests. "Often women are not recognized as collaborators in published work, and therefore we are tend to be underrepresented in the published work. There is also a certain bias. We do not know exactly who should have been listed as an author," she says.
And although the study offers some insights into where and when scientists leave the profession, she does not explain why.
There are several factors, apart from the complete abandon of research that could explain why scientists stop publication, such as switching to less research -oriented institution, entering industry or switching to an administrative role. "We don't know 100%what happened to people," says Misra. "We cannot know it without interviews and surveys," adds Kwiek.
In a study from 2023, White-Lewis and his colleagues analyzed the decision-making decisions of 773 faculty members in US universities between 2015 and 2019 and found that family reasons, the status of tenure and the main factors for the decisions were to be done.
In future studies, Kwiek plans to carry out large-scale surveys and use artificial intelligence chatbots for interviews to examine the reasons for leaving the profession.
"It would be interesting to combine the scopus data with institutional data", such as: B. Exit polls to better understand why scientists give up academic careers, ”says White-Lewis.
-
kwiek, M. & Szymula, L. High. Educ. https://doi.org/10.1007/s10734-024-01284-0 (2024).
-
White-Lewis, D.K., O’Meara, K., Mathews, K. & Havey, N. Res. High. Educ. 64, 473–494 (2023).