The Best (Cambridge, London) and Worst (Moscow, Taipei) Cities for Science
What’s the News: Many evaluations of scientific excellence singling out specific universities or departments, but two European researchers have taken a different approach: They rated the top scientific cities by looking at what proportion of published science articles are highly cited. Cambridge, Massachusetts, came out as the winner in physics and chemistry (no surprise thereâ€”MIT and Harvard) for having lots of influential papers; London was tops in psychology; Moscow was the chemistry and physics loser; and Taipei, Taiwan was the low achiever in psychology.
How the Heck:
- ResearchersÂ used a science database called Web of Science to count the number of total papers and influential papers produced in cities around the world in 2008. (In chemistry, for example, a total of 10,460 papers were published that year.)
- The expectation was that 10% of each city’s papers would appear in the top 10% of the most-cited papers. Researchers tallied up the number of actually influential papers from each city and compared that with the expected figure.
- The under-performing cities are plotted on Google Maps as red dots, while the over-performing are green. For example, on the chemistry-cities map, Moscow’s circle is the largest because it’s publishing the most chemistry papers, but it’s red because only 5 of its papers were in the top 10% of most-cited chemistry papers, far below the expected figure of 47.7 (10% of its output).
What’s the Context:
- The northernmost city with more than expected highly cited papers was Tromso, Norway, proving that science can prosper even in the icy, inhospitable stretches of the Norwegian Sea.
- While fewer in number than North America, Europe, and China, there is still some thriving science cities in countries in the Middle East, including Oman and Iran—though Iraq is noticeably blank.
- Compared to the maps of physics and chemistry, there are far more successful psychology cities.
Not So Fast: As the researchers note, the study fuzzes over any distinctions that emerge on a smaller scale than a city—for instance, the maps don’t show any difference between a city with one superstar who publishes 10 influential papers and another city with a group of 10 researchers who each publish 1. And since the scoring is based on citations, it’s subject to biases based on renown, language, and resources; the same paper published by a famous researcher at Oxford will get more notice than if it were published in Nigeria.
Reference: arxiv.org/abs/1103.3216: Lutz Bornmann and Loet Leydesdorff, Which Cities Produce Worldwide More Excellent Papers Than Can Be Expected? A New Mapping Approachâ€”Using Google Mapsâ€”Based On Statistical Significance Testing