Nature Index 2018

Comments

1. bob sykes says

   February 26, 2018 at 7:21 am

   For the last 30 to 40 years, Chinese nationals have made up the largest group of foreign students in our STEM graduate programs, including the elite programs at MIT, Cal Tech et al. In many STEM graduate programs they constitute an absolute majority of all the enrolled students.

   These Chinese students are supported on externally (mostly federal) funded research projects, and they are participating in cutting edge research in many fields. In many cases they create the cutting edge. Because they talk to other students, including Americans, they even get some access to restricted military research. Almost all of these students go home, and take their advanced knowledge with them and away from the US. We lose that knowledge. Hence China now has more supercomputers than we do, they have the two fastest supercomputers, and their supercomputers are entirely homegrown, from the chips to the operating systems.

   Russia and China both have industrial sectors that are much more diverse than America’s, and China’s industrial sector is larger than ours: socks to supercomputers. In general, the quality of their manufactured products is every bit as good as ours and in many cases better. Ask Apple or any other “tech” firm. What advantages we have, mostly in military technology, are rapidly being lost. China will be at full parity with the US much sooner than our Ruling Class thinks.
2. DFH says

   February 26, 2018 at 9:27 am

   Any theories as to why the Japanese underperform relative to their population size and IQ?
3. reiner Tor says

   February 26, 2018 at 10:22 am

   So much for German science having been destroyed by Hitler kicking out the Jews…
4. reiner Tor says

   February 26, 2018 at 10:27 am

   Why is Saudi Arabia so good? Why is Slovakia so weak relative to Hungary? Even Romania seems pretty weak, relative to what I’d have expected based on prior knowledge.
5. Anatoly Karlin says

   February 26, 2018 at 10:34 am

   King Abdullah University of Science and Technology (KAUST) – most productive university in West Asia outside Israel.

   https://www.natureindex.com/annual-tables/2017/institution/all/all/regions-Western%20Asia

   I recall they hired huge numbers of very good Western professors.
6. Anatoly Karlin says

   February 26, 2018 at 10:37 am

   I suspect Germany would have probably been in the lead if such a ranking had been done in earlier decades from the time of its unification to around 1940.

   Of course kicking out the Jews didn’t help, but I think the main reasons were:

   1. Americans getting in on the university game, which Germany had dominated during the 19th century.

   2. And of course the war, consequent economic ruin, brain drain abroad, etc.
7. Anatoly Karlin says

   February 26, 2018 at 10:38 am

   Probably this: Asians: bright, but not curious?, by James Thompson
8. for-the-record says

   February 26, 2018 at 10:45 am

   Why is Saudi Arabia so good?

   That’s pretty easy to answer. Check out the faculty pages of King Abdullah University of Science and Technology, e.g.

   https://www.kaust.edu.sa/en/study/faculty?researcharea=10
   

   Not much local talent there, it would appear.

   Related question: what % of US total is “native” American as compared to “imports”? On a worldwide basis, is it possible that ethnic Chinese are already ahead of ethnic “Americans”.
9. songbird says

   February 26, 2018 at 10:49 am

   I think this is an important point to make: this is with what in the US would certainly be called bad air and drinking water. Food is somewhat similar, you may get as sick as if you visited Mexico.

   Why am I bringing this all up? Well, that is the boilerplate excuse for blacks and Hispanics in the US. I mean, if you ignore, that all those cities had dirtier water and air when they were 95-97% white. And if you don’t believe in invisible racism rays. Or school buildings that have some curse on them.
10. songbird says

    February 26, 2018 at 10:53 am

    I wonder how many of these more tropical countries are there because of research on tropical diseases.
11. Polish Perspective says

    February 26, 2018 at 11:57 am

    Slovakia & Czechia looks strange. We’re talking about two peoples who are basically as similar Norwegians and Swedes, yet even when weighted by population the massive discrepancy is striking. Some variation is normal but such a huge chasm requires some form of thorough explanation.

    Any Czech or Slovak who has any clue? (Or anyone else wishing to guess is of course welcome to do so).

    P.S. Oh and our score is pathetic. Though we do spend a pitiful amount of our GDP on R&D. I wonder how a list normalised to GDP per capita as well as percentage of GDP spent on R&D would look like. So for instance, a band of those with a per capita GDP (nominal, since research equipment often has to be imported on the international market) with, say, 10-15K, 15-20K and so on and then each group sorted by how much they spend on R&D as a proportion of GDP.

    At the same time, a nation’s priorities is also shown by what it spends on. IIRC, South Korea spent quite a bit on R&D as a percentage of GDP even when it was relatively poor, ditto Taiwan.
12. Polish Perspective says

    February 26, 2018 at 12:01 pm

    Related question: what % of US total is “native” American as compared to “imports”? On a worldwide basis, is it possible that ethnic Chinese are already ahead of ethnic “Americans”.

    Very good question. Some countries are being brain-drained by richer countries (yours truly) whereas the richer countries, such as the US, can reap the benefits of this. Germany has also started this process in the last 10-15 years, with the increasing ease of movement of peoples. A great deal of 30-somethings in German universities and research institutions are no longer Germans, though still a minority, a much bigger share than was the case, say, 20 years ago.

    I haven’t been to Chinese universities, but it is my impression that the research done their is almost exclusively done by locals. If you only looked at natives, I wouldn’t be surprised if the distance between them and the US shrunk even further, though at this stage it looks like a foregone conclusion at any rate. It’s just a question of time.
13. lauris71 says

    February 26, 2018 at 4:03 pm

    I think there is simpler explanation. European, and especially US/UK universities draw the brightest students/professors from much larger pool. Japan is quite isolated and much larger part of the research is done by locals.
14. anony-mouse says

    February 26, 2018 at 4:06 pm

    Age
15. anony-mouse says

    February 26, 2018 at 4:07 pm

    Age
16. A-non says

    February 26, 2018 at 4:38 pm

    Maybe this: Many Slovaks work in the Czech Republic. They are the biggest minority in this regard. As far as I know they rather work in more demanding occupations. It is not the case vice versa. And I guess, in general, more gifted Slovaks emigrate than Czechs (on per captita basis).
17. Anatoly Karlin says

    February 26, 2018 at 4:52 pm

    Even France does much better at elite research than Japan per capita, and France is sooner losing science human capital (mainly to the US) than gaining it.

    In per capita terms, Japan is basically Italy/Spain tier, with their 5-10 point lower IQs.
18. reiner Tor says

    February 26, 2018 at 7:20 pm

    Slovaks are older than Hungarians? Or younger? Or what? And what does it mean for Saudi Arabia?
19. Anonymous says

    February 26, 2018 at 8:54 pm

    The Weighted Fraction Count (WFC) of the Nature Index is is probably the single best proxy for quality-adjusted scientific output in the world today.

    “Nature Index” is garbage. Things that get into glamor mags are about an equal mix of great job and total garbage. In the long run, journal impact factor only matters in so far as ~ 95% of things worthwhile the attention are published in the top ~ 50% journals. For which the sole reason is the proliferation of total garbage damps of journals driven by oversize profits of scientific publishing industry.

    You should read this: https://www.frontiersin.org/articles/10.3389/fnhum.2018.00037/full

    That said, I just generated a list in Life Sciences (something I can judge expertly) and the result is not too bad:

    1 United States of America (USA) 8103.18 10106
    2 United Kingdom (UK) 1200.22 2222
    3 Germany 1072.95 2012
    4 Japan 844.22 1327
    5 France 632.04 1234
    6 Canada 578.44 1084
    7 China 554.16 1050
    8 Switzerland 301.43 678
    9 Australia 285.35 614
    10 Netherlands 274.79 687
    11 Spain 258.79 570
    12 Italy 221.96 572
    13 South Korea 189.9 345
    14 Israel 183.77 324
    15 Sweden 178.77 419
    16 Belgium 109.84 302
    17 Austria 99.68 242
    18 Denmark 96.91 267
    19 Singapore 81.58 189
    20 Taiwan 79.68 146

    Observation: Canada, Spain and Singapore are way higher than they should be. Otherwise, pretty reasonable – except that Germany+UK should be much closer to the total US that they are in the rankings. South Korea and Switzerland both need to be ~3 positions higher.
20. Anonymous says

    February 27, 2018 at 1:17 am

    @Anatoly Karlin
    It takes a century to build a great university. Asia’s higher education is still lagging behind, even as Japan.
    More than 95% of great scientists from 1800 – 1950 are whites, western countries are still benefiting from it, or “rest on their laurels”.
21. Yan Shen says

    February 27, 2018 at 2:15 am

    My guess is that Japan like China and other East Asian countries skew decently towards quantitative fields. A couple of years back, something like 90% of China’s total WFC for Nature was in physical sciences and chemistry alone. Japan’s output in life sciences is better than that of China and South Korea, but in general this skew is fairly clear. The overall Nature Index includes a number of non-quantitative fields, some of which I assume East Asian countries publish relatively little in.

    In general, East Asian countries skew towards physics, chemistry, and engineering and away from life sciences, social sciences, and humanities. Japan does better than other East Asian countries in life sciences and China publishes a lot more in computer science and mathematics than other East Asian countries. Japan is a world leader in patenting and high tech manufacturing and in general in East Asia a higher percentage of R&D goes towards applied industrial research. I think this can all be explained by math/verbal split. See also this Simon Marginson article below.

    http://www.researchcghe.org/perch/resources/publications/wp9.pdf

    More strikingly, in 2000 China authored just 0.6 per cent of chemistry papers ranked in the global top one per cent on citation rate in the Web of Science. Only 12 years later, in 2012, China published 16.3 per cent of the leading one per cent of papers, half as many as the US – an astonishing rate of improvement. There were similar patterns in engineering, physics and computing – where China publishes more top one per cent papers than the US – and mathematics (NSF, 2014). China, Taiwan, Korea, Japan, and to some degree Singapore, have concentrated research development in the physical sciences and related applied fields like engineering, computing and materials. In Korea and Japan this supports advanced manufacturing. China also emphasises research that supports accelerated modernisation: energy, urbanisation, construction, transport and communications. At this stage medicine and life sciences are much weaker.

    The other factor is that Japan has fairly low levels of international collaboration, low English fluency, and is a relatively closed society. While America benefits significantly from foreign Asian STEM talent, most of the STEM work in Japan is done by Japanese.

    The OECD report urges the country to better integrate itself in “global innovation networks”. In 2014 only 0.4% of Japanese research and development was financed from outside the country. It also says the proportion of scientists who have immigrated to Japan is among the lowest in the organization’s 35 member countries — in line with the mere 1.7% of Japan’s population who are foreigners. The result is very low levels of “international co-authorship of academic papers and international co-patenting”.

    As far as I can tell, based on Japan’s surge in Nobel prizes since 2000 and based on the numbers of potential future Japanese Nobel laureates, the country does fairly well in terms of elite science and world class scientists. My guess is that the factors I mentioned above can explain why that’s the case despite Japan seemingly lagging on certain other metrics. For instance, average citation rates for Japanese papers typically are fairly lowly even compared to countries like Italy and Spain, yet there’s no real sign that there’s a huge glut of elite modern day Italian or Spanish scientists!
22. Yan Shen says

    February 27, 2018 at 2:32 am

    So to summarize, once you move away from IQ and towards math/verbal split, which quite frankly as phenomenon like Spearman’s diminishing returns and history’s myriad examples of cognitively lopsided geniuses like Richard Feymann or Terence Tao suggest is much more useful for understanding elite performance.

    Japan’s output in Nature has actually declined in absolute terms over the past 5-10 years, while most other countries have gone up. I think that this can all be explained by math/verbal split, i.e. Japan does fairly well in physics, chemistry, engineering, and material sciences and invests significantly in applied industrial R&D, low levels of international collaboration and English fluency among Japanese and very few foreigners, etc.

    Nature metrics are interesting and useful but also need to be analyzed carefully. Indeed, article count numbers probably favor more internationally collaborative countries given that all it takes is one author among all of the authors for a given paper for that paper to count towards a particular nation.

    Karlin equates Japan’s elite science performance with that of Italy and Spain based on WFC. Based on average citation rates, Japan is even worse than Italy or Spain in most fields. Yet, Japan has a fairly significant presence of world class scientists and potential future Nobel laureates, while Italy and Spain have minimal such 21st century figures. Surely the factors I alluded to above explain this kind of discrepancy?
23. Yan Shen says

    February 27, 2018 at 3:00 am

    See my other comments. I believe there’s no real under-performance once you account for math/verbal split, Japanese insularity leading to much lower levels of international collaboration, and in general lower levels of English fluency among Japanese compared to other Europeans.

    To summarize, East Asian cognitive output is skewed towards quantitative STEM fields such as physics, chemistry, and engineering and away from life sciences, medicine, social sciences and humanities. In some non-quantitative scientific fields, there’s probably relatively low levels of East Asian output. When you look at the 21 major science categories listed by ISI, you’ll in fact see that the majority of these are more verbally loaded life science/medicine/social science fields.

    https://en.wikipedia.org/wiki/Institute_for_Scientific_Information#ISI_Highly_Cited

    Apart from Japan, which has pockets of historical strengths in areas of the biological sciences such as for instance immunology or STEM cell research, most of East Asian scientific output is clearly and significantly skewed towards physical sciences and engineering. When you look at the US or the UK, the opposite is clear. The Anglosphere is clearly skewed towards life sciences and medicine. I don’t expect East Asia to challenge the supremacy of the Western countries in verbally loaded scientific fields such as biology of medicine.

    https://www.nature.com/articles/515S73a

    These countries are traditionally strong in physical science and chemistry — of China’s total WFC of 5,206, for example, a WFC of 4,696 (or 90%) is from chemical and physical sciences.

    So where does the missing “excess” human capital in East Asia get allocated? Simple. Towards quantitative industrial R&D. See Karlin’s prior post about the lack of a tech sector in Europe.

    http://www.unz.com/akarlin/europe-cant-into-big-tech/

    Tech is basically California vs East Asia, with the former skewed towards software and the latter skewed towards hardware. This once again can be explained by math/verbal split.

    https://www.ft.com/content/dbb3bc26-413b-11e7-9d56-25f963e998b2

    Japan remains an innovation powerhouse, according to a geographical analysis of patenting that shows Tokyo-Yokohama is much the largest such cluster in world

    The study comes from the World Intellectual Property Organisation (Wipo), based in Geneva, which analysed the addresses of inventors named in all 950,000 international patent applications published between 2011 and 2015 under the Patent Cooperation Treaty.

    Two other Japanese clusters, Osaka-Kobe-Kyoto and Nagoya, are in the global top ten.

    The results also show strong inventive activity elsewhere in east Asia, with China’s Shenzhen-Hong Kong taking second place in Wipo’s rankings, ahead of California’s Silicon Valley in third and Seoul in South Korea..

    European clusters appear lower down the rankings, with Paris at number 10 and Frankfurt-Mannheim at 12. The UK does poorly, with London at 21, Cambridge at 55 and Oxford at 88.

    It’s a non-trivial fact that virtually every piece of modern day consumer electronics hardware that you or I use is made in East Asia. This is major major value add.

    To conclude, different groups of people will skew towards what they’re naturally good at. East Asians and Europeans today clearly seem to skew towards different areas of specialization. At a high level, East Asian countries clearly tend to prefer the quantitative physical sciences such as physics, chemistry, engineering, etc, while the Anglosphere and other Western countries skew towards life sciences and medicine. East Asian economies are primarily based on hardware and manufacturing, while Western economies tend to be based on software and services. I think all of this can be explained by math/verbal split.
24. anonymous coward says

    February 27, 2018 at 3:17 am

    Published papers and citation rank is only useful as a measure for grant distribution. It doesn’t actually measure productive scientific output. (Something that isn’t quantifiable anyways.)
25. Yan Shen says

    February 27, 2018 at 3:21 am

    This is one of the facts that usually ends up being ignored in discussions of China versus the US. A non-trivial percentage of American STEM output is done by Chinese Americans.

    Take Silicon Valley for instance. At many top tech companies like Uber or Google or the likes, anywhere between 35-50% of workers in technical roles are Asian. A safe bet would be to assume that that’s roughly half Chinese and half Indian. Ethnic Chinese have also had a big impact in hardware in America and Canada. Computer graphics today is basically Nvidia vs AMD, with AMD having purchased ATI Technologies some 12 years ago and melded it with its graphics division. Nvidia was co-founded by Jen Hsun-Huang and ATI Technologies was founded by 4 ethnic Chinese Canadians.

    Lee Ka Lau,[2] Francis Lau, Benny Lau, and Kwok Yuen Ho[3] founded ATI in 1985 as Array Technology Inc.[4]

    Also, a Chinese American, Feng Zhang, is one of the major players, along with Jennifer Doudna and Emmanuelle Charpentier in the development of CRISPR-CAS9.

    Chinese Americans are a total of 1.5-1.6% of the American population, but surely contribute significantly more than that to various areas of American STEM. What percentage of the American authors in Nature papers are actually ethnic Chinese?
26. Yan Shen says

    February 27, 2018 at 3:25 am

    Yes and no. Publications and citation rates certainly tell you something non-trivial about the state of science in a given country. But I agree careful analyses is needed. For instance, Karlin pointed out Japan’s per capita WFC is on par with that of countries like Spain and Italy. I pointed out that in fact Japan’s average citation rates in most fields are lower than those of Spain or Italy. Ergo Italy/Spain > Japan in elite science right? No.

    Based on the surge of Japanese Nobel laureates post 2000 and on the high numbers of world class Japanese scientists and potential future Nobel laureates, Japan clearly excels relative to Spain or Italy, despite initial superficial metrics such as number of articles and average citation rates in Nature. So other factors are at play here.
27. Daniel Chieh says

    February 27, 2018 at 8:46 am

    It means that anony-mouse is a troll.
28. bb. says

    February 27, 2018 at 8:58 am

    per Slovakia-literally all the STEMs I know (about 15) do their work abroad – Germany, France, UK or US but most notably, Czechia. Also, I don’t have the source at hand, but I am pretty certain that among the V4, we spend the least on science. Actually, there are not many countries within EU, who spend less.
    So that is actually not bad – with not very significant spending increases, we can jump in output considerably.
29. enkidusvk says

    February 27, 2018 at 1:52 pm

    A-non is right. The emigration of more gifted students to Czechia is indeed one of our (Slovak) largest problems. This is even more pronounced in the case of PhD students. This and literally non-existent governmental support of R&D sector and universities gives the depressing results in the table above (similar big difference is in the rankings of universities, best Slovak ones lag by huge distance behind Charles and Masaryk Universities).
30. Anonymous says

    February 27, 2018 at 10:54 pm

    The emigration of more gifted students to Czechia is indeed one of our (Slovak) largest problems. This is even more pronounced in the case of PhD students.

    Is it even to Czechia? Weirdly, I’ve met more Slovak Ph.D students at University of Vienna than at Charles University.
31. Polish Perspective says

    February 28, 2018 at 5:41 am

    Computer graphics today is basically Nvidia vs AMD

    This is something which I thought about the other day, namely that both CEOs are Taiwanese-Americans. Taiwan is of course allied with America, but you sort of wonder if there is a Bamboo ceiling of sorts against Chinese-Americans as opposed to Taiwanese-Americans. I’m talking more about cultural self-identification rather than ethnicity(where there is much less difference).

    After all, I seem to recall a ridiculous mini-scandal a few years back where the USG basically forbade both AMD and NV to sell their hardware to China because the Chinese were gaining massively in supercomputer performance. The result was that China simply built their own supercomputers, did it better and neither NV or AMD got any business from it.

    I wouldn’t be surprised if there is a racialised panic in the upper echelons of the US establishment about specifically Chinese-Americans. We’re not just seeing panic about Russians. I’m seeing streams of articles demonising and smearing
    even something as modest as the Confucius Institutes.

    This kind of racialised panic is already leading to the mainstreaming of bigotry against Russians in the US and I wouldn’t be surprised if Chinese are next.
32. LH says

    February 28, 2018 at 10:00 am

    @Polish Perspective

    Czech and Slovak languages are mutually intelligible. Technical literature in Slovakia are often printed in Czech language, to expand the market (Czechs are not used to read in Slovak language). Slovaks are not perceived as foreigners (foreigner = bad) by the Czechs. Moving from Slovakia to the Czech Republic is not a stressful event full of uncertainty.

    As a result lot of Slovaks study at Czech universities, and young scientists tend stay there.

    In recent decade Czech government initiated construction of 48 scientific centers, from EU “for science” funds. Huge modern palaces from glass and steel were erected. However, EU funding will end in couple of years, and I predict that not every center will survive.
33. myself says

    February 28, 2018 at 2:08 pm

    There is a split, an important one, among the elite populations of the Atlantic Alliance. I use this term to refer to the United States, Canada, Britain, France, Germany, Italy and some smaller Western European nations. The same split can be seen in Australia as well.

    On the one hand, the elite business class in these lands do VERY, VERY profitable business in China. Earnings are high, business is booming, what’s not to like? The Han have rolled out a limitless, sumptuous banquet, and the Western elites have truly gorged themselves for DECADES.

    On the other hand, China, “unfortunately”, is not Western, nor White, nor even remotely interested in changing its own identity or race. As it grows ever more powerful and advanced, what will happen to the Western Hegemony of the last few centuries?

    Polish Perspective, you speak of “panic”. Exhibit A, making a big-deal about China’s investment in the 16+1 Group, saying it interferes in Europe. What? Aren’t these 16 Central-Eastern European nations sovereign states?

    I mean you can “interfere” in Prague, Warsaw or Budapest, yes, but “interference” in Brussels is NOT A THING.

    So, you have a “quasi-racialized” panic in the upper echelons all across the Western World.