I enjoyed looking through the Nature Indexes section recently in that journal – I believe that they do this primarily as a way to make a new section in which to sell advertisements, to be honest, but the content itself is worth a look. They’re tracking publications in 82 leading scientific journals and looking for trends in publication frequency, etc.
Some of the results are not particularly surprising. Across all 82 journals, the largest number of contributions (counted by institutions of all sorts) came under the banner of the Chinese Academy of Sciences, with Harvard second. If you just count universities, then it’s Harvard, Stanford, MIT, Cambridge, Tokyo, Peking, the ETH, Oxford, Tsinghua, and Berkeley. Zooming down to just life sciences, the institution list is Harvard, NIH, Max Planck, Stanford, Chinese Academy of Sciences, MIT, UCSF, Yale, Cambridge, and Penn. And the institution list for chemistry alone reads out as Chinese Academy of Science, CNRS (France), Max Planck, Nanjing, Peking, Tsinghua, Univ. Science and Technology of China, MIT, Northwestern, and Stanford – a list with some overlaps but with some notable differences from the overall and life-sciences list, for sure. Nanjing and Northwestern are notable for having risen into those positions rather strongly as compared to last year’s figures.
There’s also an interesting section where they consider an institution’s publication output in those 82 top journals as a fraction of their total number of natural sciences papers coming out. Normalizing in this way changes things around quite a bit; you get some places that haven’t published nearly as many papers as the bigger powerhouses, but puts almost all of them into upper-tier journals. When you run those numbers, Cold Spring Harbor comes out on top, followed by IST Austria, the Weizmann Institute, the IAS in Princeton, Brandeis (whose campus, by coincidence, I am viewing from the commuter rail exactly as I write these words!), Rockefeller, the Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), the Swiss Federal Institute of Technology in Lausanne (EPFL), Princeton, and the Okinawa Institute of Science and Technology Graduate University (OIST). There are some on that list that I don’t think many people would have guessed.
And finally, if you rank things out by natural-science publications in top journals by country of origin (which is what some of you have probably been wondering about as you read through the above), the US is still in the lead. In fact, the top 7 have not changed for three years: the US, China, Germany, the UK, Japan, France, and Canada. Switzerland and South Korea have swapped places in the next two slots, and Australia moved Spain out of the top 10. So Australia clearly increased their share of the total output, but the only other country that did was China (and since this is zero-sum fractional count, these two increased at the expense of all the others in the top ten, with France and Japan dropping the most).
What the list also reveals, as those who follow the literature already know, is that there are large regions of the world that are comparatively hardly heard from. All of the Americas outside of the US and Canada, all of Africa, the Middle East and Western Asia in general outside of Israel – there are papers going into the 82 journals from those areas, but very few in comparison. That country list is very long-tailed – the US has a pretty strong position, but it and China together account for more of the total than not only the other 8 in the top ten, but (in fact) for more than the entire 159 other countries on the entire list. I don’t see that changing much, honestly. China will certainly continue to move up higher, but I would imagine that it and the US are going to dominate any such tabulation for the foreseeable future.
Something that the Nature lists don’t do is provide rankings that are normalized to population. If you do that, the standout is surely Switzerland, with only 8.4 million people total but ranked in the top ten. At the other end of the scale, India (although it produces many publications), definitely underperforms in the higher-end journals compared to its population, as do Russia and Brazil. There are, of course, many other high-population countries whose scientific infrastructures are not well developed – Indonesia is the prime example, the fourth-highest population in the world but almost invisible in the sciences, for many reasons. It takes a lot of disposable income on a national scale to fund R&D, for one thing (both public and private money), as well as a solid educational system up through the graduate-school level, and sadly throughout much of the world neither of these conditions are present. As it is now, many talented people from these areas travel to more scientifically developed nations to study and work, of course, but surely many more never get the chance to do anything of the sort. As always, one can’t help but feel that there’s a lot of human ability going to waste, and I only wish that scientific talent were as relatively straightforward to scout (not to mention as lucrative!) as talent in basketball or soccer. . .