More than 50,000 papers were submitted to “Science” and “Science Advances” in 2025, and most are turned down before any reviewer reads them. Researchers from the University of Vermont and the University of Colorado Boulder looked behind the scenes by analysing over 110,000 papers. The name of the institution and the size of the team turned out to count most, and the authors’ gender far less — and at the end you can test whether you understand the rules of the game in scientific publishing.
If you have ever dreamed of seeing your name in “Science”, you are not the only one with that plan. The number of submissions to leading scientific journals is at a record high today. “Science” and “Science Advances” alone, published by the American Association for the Advancement of Science, received more than 50,000 papers in 2025.
A team from the University of Vermont (UVM) and the University of Colorado Boulder (CU Boulder) decided to check which papers get through this sieve and which fall out of it. The findings, incidentally, were published in one of the journals under study, “Science Advances”.
The researchers analysed deidentified data on more than 110,000 articles submitted to the two journals between 2015 and 2020. It is a huge sample that shows what usually happens behind closed editorial doors.
“This is really a first look at this process that was previously opaque,” says Sam Zhang, assistant professor of mathematics and statistics at UVM.
The overall acceptance rate was about 6% at “Science” and about 10.5% at “Science Advances”. In other words, even in the kinder of the two journals, roughly nine papers in ten are not accepted.
What affected the chances of publication most was where the author came from. Papers by researchers from the top-ranked institutions were accepted in 11.6% of cases. Authors from the lowest-ranked institutions could count on acceptance in only 3.4% of cases.
That is a difference of more than threefold. The same kind of work, but a different address on the title page — and completely different statistics.
The second important factor was the size of the team. Papers with ten or more authors were accepted three times as often as those with no more than five people behind them.
For a would-be researcher, that is a telling lesson: in this data, collaborating in large groups clearly paid off.
The researchers also checked whether the authors’ gender affects editors’ decisions. A difference exists, but it is clearly smaller than the others. The acceptance rate was 5.3% for women and 6.4% for men. According to the authors, this gap turned out to be smaller than the differences driven by institution, team size and research topic.
The topic itself counted too. The highest success rate belonged to papers on:
At the opposite end were the political, economic, gender and societal issues involving COVID-19. These attracted the least interest.
The most surprising part, however, is what happens at the very start of the journey. Editors reject about 80% of the papers submitted to “Science” and more than 70% of those submitted to “Science Advances” before they ever reach reviewers. It is this barely visible stage of screening that decides the fate of most papers.
Zhang points out that in an age of a growing flood of AI-generated content, the role of elite journals as gatekeepers of knowledge is becoming more and more important.
Aaron Clauset, professor of computer science at CU Boulder and the study’s corresponding author, defends the very idea of peer review. “Peer review by scientific experts remains the gold standard for critically assessing whether new scientific work meets or surpasses high standards for scientific discovery,” he stresses.
He also values the data: “Our new anonymised data set … offers a first ever look at how elite peer review works.”
In his view, the current evidence suggests that these journals do quite a good job of judging research on its scientific merit. “But this is not a totally clean bill of health,” he adds. And he puts the matter plainly: “In the current environment where the integrity of science is being attacked in bad faith, we need to be thinking about whether the processes at these journals live up to the ideals the scientific community has set for them.”
For anyone taking the first steps in research, these numbers are a kind of map of the terrain. Besides the idea itself, the statistics of prestigious journals reflect the institution an author comes from, the team they work with and the topic they choose. Keep that in mind when planning your first projects and looking for people to work with.
An analysis of more than 110,000 submissions to “Science” and “Science Advances” from 2015–2020 showed that the institution’s reputation and the team’s size matter most for publication, and the gap between women and men is smaller. Most papers fall out at the editor’s desk. The authors regard expert peer review as the gold standard but, as Clauset stresses, its processes still need careful scrutiny.
5 questions · one minute
EN 
CS
DE
ES
FR
HU
IT
PL
PT
РУ
SK
TR
УК
AR
中文