Bullets:
Comprehensive surveys of scientific output across over 18,000 universities and colleges around the world now show Chinese institutions leading the rankings.
Researchers at the Nature Index examined tens of thousands of top research papers in all the engineering and hard-science disciplines.
They also found that the number of top scientists in the United States peaked in 2020, and has been in decline since. Last year, China passed the United States for the first time in the number of top scientists.
The researchers concluded that while the US and China still dominate the world rankings, the trends are in opposite directions.
Report:
Good morning. This has happened suddenly, but decisively, that Chinese universities now dominate the world rankings for the hard sciences.
The Nature Index is a comprehensive ranking of over 18,000 universities and colleges from around the world, and the scores are based on quality research output. These tables are sortable, as well, by scientific discipline. For example, in Physics, the United States didn’t show up in the top 10 ranking at all.
Sichuan University in Chengdu, across all scientific and engineering disciplines, is now ahead of Stanford, MIT, Oxford, and University of Tokyo.
Here are some other takeaways. 8 out of the top 10 research institutions are Chinese. Zhejiang University is in that bunch, and where Liang went. Of the top 50 universities, 26 are from China. US has 14. Have you ever been to Xiamen? Me either. But they have a university in Xiamen that’s ahead of Cal, Columbia, Cornell, and Chicago. I know about all of those. Half the top 100 are Chinese.
The number of top scientists in China is going up, and in the US it’s going down. 2022 marked the first time that China had passed the United States, and in 2023 Chinese scientists published a third of the most influential research across the world.
In the United States, 2020 was the peak for the number of top scientists, and it’s been dropping since. Now, China has 28 percent of the world’s top scientists, while last year the US had 27%. And we’re right to wonder if we may be counting a lot of Chinese scientists here as our own, because they currently are doing work in the US for a US company or university, but who are Chinese and who can make this trend worse just by going to the airport.
Here’s the takeaway for all of it: the global landscape for high-level science has profoundly changed. It’s still China and the US that dominate, but the trends are in opposite directions.
The real-world impacts of all this is being felt deeply, tangibly. It helps explain how DeepSeek was built with an all-Chinese team for one one-thousandth of the cost that Facebook and Amazon have spent to build their AI. It helps explain why our sanctions against Huawei and SMIC were certain to fail, because every year those companies just go down the street to one of the schools on these lists and hire another thousand fresh graduates. Or ten thousand. And that’s just for semiconductors. Pick any industry, any hard-science major, and the trends are in opposite directions.
Resources and links:
Institution tables, Time frame: 1 November 2023 - 31 October 2024
https://www.nature.com/nature-index/institution-outputs/generate/all/global/all
Hua’s Substack, Whose Universities are Better – China vs. the US? Nature Magazine might upset the conventional wisdom
South China Morning Post, China surpasses US in tally of top scientists for the first time: report
SCMP, China’s Sichuan University overtakes Stanford, MIT and Oxford in high-quality research
SCMP, China leads the world in physics research as US a distant rival, Nature Index shows
With little doubt about Chinese universities' superiority and upward curve in this regard,I tend to infer it on general trends, and certain questions appear in my mind:
How are those "comprehensive surveys of scientific output" done?
Who surveys the surveyors?
About who creates the The Nature Index, its webpages are silent, but we can say with certainty that it's done by the Nature Magazine. Nature Magazine 's record is rather spotty: Nature in the past accepted and published papers with falsified data, it has retracted retractions (See for example https://retractionwatch.com/category/by-journal/nature-retractions/ ). That's a good basis for reservations about this particular output of Nature.
On my first question, the Nature Index answers with the following:
"The Nature Index tracks a small proportion of the total number of research articles published, and they cover the natural sciences and health sciences only. There are many factors that determine an institution’s output in the Nature Index. The size of the institution is a major factor, as is the composition of research specialities within the institution. The Nature Index does not attempt to normalize data for such factors. Rather it presents raw data that can contribute to an analysis. [...] Nature Index metrics alone should not be used to assess institutions or individuals. We encourage users to combine the open source Nature Index data with other information, measures and tools."
( https://www.nature.com/nature-index/brief-guide)
It was inevitable, now we shall experience it in Real Time...