Different meanings of the word ‘replicate’

In the rough-and-tumble world of science, disputes are usually settled in time, as a convergence of evidence accumulates in favor of one hypothesis over another. Until now.

On April 10 economist John R. Lott, Jr., formerly of the American Enterprise Institute, filed a defamation lawsuit against economist Steven D. Levitt of the University of Chicago and HarperCollins, the publisher of Levitt’s 2005 book, Freakonomics. At issue is what Levitt meant when he wrote that scholars could not “replicate” Lott’s results …

That’s from Michael Shermer’s Skeptic column in Scientific American. Shermer is “executive director of the Skeptics Society, bold debunkers of all things supernatural”, according to Salon which has a long interview with him [free, if you are willing to watch an ad]. I don’t know if artificial intelligence would be considered ’supernatural’, but here is an interesting article in the Skeptic magazine (flagship of the Skeptics Society) debunking the lofty claims made by AI enthusiasts.

The Oscars of Indian Science: 2006 Edition

Yes, the Shanti Swarup Bhatnagar (SSB) Prizes have been announced for the year 2006. As I said in my post last year, the SSB Prizes are the most prestigeous in India because (a) they represent peer recognition, and (b) they are rare (just one or two in each field). They do come with some money, but at Rs. 200,000 or about $5,000, it’s not much (but certainly nice!).

I’m pleased to note that Prof. S. Sampath, a colleague in the Department of Inorganic and Physical Chemistry, is among the SSB Prize winners [Congratulations, Sampath!]. He has won one of the two Prizes for Chemical Sciences; Dr. K George Thomas (RRL, Trivandrum) has won the other.

The two SSB Prizes for engineering go to Dr. Ashish Lele (Complex Fluids and Polymer Engineering, National Chemical Laboratory, Pune) and Dr. Sanjay Mittal (Aerospace Engineering, IIT-K).

I have to fault the Council of Scientific and Industrial Research (CSIR), the organization that awards these Prizes, for giving just the barest of details about the awardees. Is it so difficult to put together a news story that has details about each awardee’s important contributions? Wouldn’t it be nice — offering a higher profile for the Prize winners, and more information for the others — if full citations are available on the CSIR website? Currently, all that it offers is this press release [pdf] which deserves a prize for minimalism.

Ranking of universities across the world

Let’s face it: global rankings of universities are here to stay, despite their poor methodologies. Among them, the ranking by the Shanghai Jiao Tong University University probably deserves the award for the worst methodology. It gives a huge weight — 30 percent — to Nobel prizes won by the faculty and alumni, and a further 20 % for papers published in the journals Nature and Science.

Last year’s ranking, for example, included the University of Calcutta among those ranked between 401 and 500; IIT-Kharagpur too figured in this list, while IISc was among those with ranks between 301 and 400. The inclusion of UCalcutta was definitely because of the Nobel Prizes won by its faculty (presumably, C.V. Raman) and its alumni (Amartya Sen?). Thus, depite scoring poorly on every indicator of its current research activity — it actually scores zero for (a) publications in Nature and Science, and (b) highly cited papers — it was grouped with IIT-Kharagpur — and 98 others! — in the range 401 to 500. [1]

[Oh, just in case you are wondering, UCalcutta is out of the top 500 in this year's ranking, while IISc and IIT-Kharagpur retain their last year's rankings.]

Just because the Shanghai group has been in this business for the longest duration — four years! — its ranking is often quoted without paying any attention to its methodology. For example, Kaushik Basu used it in his recent BBC column:

A recent evaluation of universities and research institutes all over the world, conducted by a Shanghai university, has not a single Indian university in the world’s top 300 – China has six.

The Indian Institute of Science, Bangalore, comes in somewhere in the top 400 and IIT, Kharagpur, makes an appearance after that.

Now, within the US, there’s a number of rankings that (mis)lead to different conclusions about the excellence level of different institutions [2]. In India, too, all the major newsweeklies (India Today, Outlook, The Week) rank our colleges, B-schools, engineering and medical colleges — with poor (and poorly spelt out) methodologies, each pointing to a different conclusion. When the situation is so complex within a country, the ranking exercise should be hopelessly complex at the international level. After all, the higher ed systems in different countries are different; more importantly, their funding patterns are different. How can one compare, for example, a hub-and-spoke and multi-campus university like the University of Delhi and a tech-centric ’small’ university like CalTech?

Global ranking exercises, of course, don’t want to — and don’t want you to — look at the enormous complexity that arises from regional variations. By claiming to do the hard work on your behalf, they encourage your laziness!

For example, they want you to ignore a key fact about excellence: it’s always the individual research groups that are the repositories of excellence. Thus, even in supposedly moribund Indian universities, there are such islands of excellence in specific subfields. A good example would be the School of Chemistry in the University of Hyderabad, or the Department of Physics in the University of Pune. The rankings also have an inherent bias against small institutions (and India has quite a few of them). For example, places like the Raman Research Institute, the S.N. Bose Natiaonal Centre for Basic Sciences, or the National Center for Biological Sciences are outside the radar screen of most rankings, in spite of their great research groups, simply because they are small and niche players.

Of course, it is certainly a great feat to put together a large university, with a solid academic and research footprint in many fields. Thus, despite their shortcomings, these rankings do do a good job of identifying the top research universities. By the same token, the top 20 (or even the top 50) universities in these lists are not very different. It’s only when you move away from these top universities that you start seeing the effects of the rankings’ individual quirks.

* * *

[1] In the mid-nineties, there was a ‘ranking’ that took India by storm. Coming from someone (whose name eludes me now) in the University of Maryland, it placed IISc at No. 17 and several IITs in the top 100. It got a lot of play in Indian press for sometime; it was, fortunately, a one-off affair.

[2] For what it’s worth, a recent issue of Newsweek put out its own list of top 100 ‘most global universities‘. Washington Monthly has a sort of anti-U.S. News listing of U.S. universities based on a measure of ‘excellence’ using a different set of critieria.

The Fame Motive

People with an overriding desire to be widely known to strangers are different from those who primarily covet wealth and influence. Their fame-seeking behavior appears rooted in a desire for social acceptance, a longing for the existential reassurance promised by wide renown.

These yearnings can become more acute in life’s later years, as the opportunities for fame dwindle, “but the motive never dies, and when we realize we’re not going to make it in this lifetime, we find some other route: posthumous fame,” said Orville Gilbert Brim, a psychologist who is completing a book called “The Fame Motive.”

From this fascinating NYTimes piece by Benedict Carey on fame as a strong motivator.

BMQ

A quick note to tell you — particularly those of you in Bangalore — about the Bangalore Materials Quiz (BMQ), an annual event organized by us for the students of Classes XI and XII. As the name suggests, BMQ covers all aspects of materials: their physics, chemistry, production, processing, properties (mechanical, thermal, electrical, magnetic, optical, …), applications and use.

I have created the BMQ blog which will be used to both disseminate information and coordinate our team’s activities.

BMQ is organized almost entirely by the wonderful graduate students of our Department. They orchestrate all aspects of the event, with some minimal guidance (and cheering from the sidelines) from me. This is the tenth year since I took over the responsibility of running this show, and I have met some of the brightest students (one of them runs this blog) through it.

BMQ is not a mega event; we usually get about 25 teams (of two each) every year. This year, we hope to attract 50 teams. On the other hand, we aren’t set up to handle a large number of teams either; so 50 is the hard limit!

The Prize we offer is admittedly small — books worth about Rs. 500 for each student! But the top two teams from BMQ get to take part in a grander event with bigger prizes at stake (see the blog for details).

* * *

Well, if you know anyone in Bangalore-based schools (higher secondary schools and pre-university colleges) who might be interested in BMQ, do please spread the word. Many thanks in advance.

What is so great about the proof of the Poincaré conjecture?

Jordan Ellenberg has a truly wonderful article in Slate.

The entities we study in science fall into two categories: those which can be classified in a way a human can understand, and those which are unclassifiably wild. Numbers are in the first class—you would agree that although you cannot list all the whole numbers, you have a good sense of what numbers are out there. Platonic solids are another good example. There are just five: the tetrahedron, the cube, the octahedron, the dodecahedron, and the icosahedron. End of story—you know them all. [...]

In the second class are things like networks (in mathematical lingo, graphs) and beetles. There doesn’t appear to be any nice, orderly structure on the set of all beetles, and we’ve got no way to predict what kinds of novel species will turn up. All we can do is observe some features that most beetles seem to share, most of the time. But there’s no periodic table of beetles, and there probably couldn’t be.

Mathematicians are much happier when a mathematical subject turns out to be of the first, more structured, type. We are much sadder when a subject turns out to be a variegated mass of beetles. [...]

[...] [Perelman's proof of the conjecture of Poincaré] means … that we can think about proving general statements about three-dimensional geometry in a way that we can’t hope to about beetles or graphs.

Manhattan Project in energy saving technologies

Wired has an article (with links) about on-going research in energy-saving technologies in MIT. Check this one out!

The research is applying new materials, new technologies and new ideas to radically improve an old concept — thermophotovoltaic (TPV) conversion of light into electricity. Rather than using the engine to turn a generator or alternator in a car, for example, the new TPV system would burn a little fuel to create super-bright light. Efficient photo diodes (which are similar to solar cells) would then harvest the energy and send the electricity off to run the various lighting, electrical and electronic systems in the car.

Such a light-based system would not replace the car’s engine. Instead it would supply enough electricity to run subsystems, consuming far less fuel than is needed to keep a heavy, multi-cylinder engine running, even at low speed. Also, the TPV system would have no moving parts; no cams, no bearings, no spinning shafts, so no energy would be spent just to keep an engine turning over, even at idle.

What they don’t teach you in graduate school

Here are the links to the complete series by Paul Gray and David E. Drew:

Part 1
Part 2
Part 3
Part 4

Ben Barres’ Commentary: The Aftermath

Following Ben Barres’ explosive commentary in Nature, reactions are pouring in. First, there was media coverage (AP, SFChronicle, WaPo, WSJ, Science Daily) including a NYTimes interview.

There are at least two blog reactions that are a must read. The first is by Sean Carroll and the second is by JoAnne Hewett (whose post also has a lively comment thread with more than 150 comments!), both at Cosmic Variance.

In its latest issue, Nature itself carries reactions from academics. Two of them (Steven Pinker and Peter Lawrence) defend themselves against some accusations from Barres. Two others (Margaret McCarthy and Donna Dierker ) present additional perspectives on gender differences. Nature’s News Blog also has tons of comments.

Revenge and retribution: Beware the tricks our minds play

In his NYTimes op-ed, Harvard psychologist and author of the recently published Stumbling on Happiness Daniel Gilbert says:

… In virtually every human society, “He hit me first” provides an acceptable rationale for doing that which is otherwise forbidden. Both civil and religious law provide long lists of behaviors that are illegal or immoral — unless they are responses in kind, in which case they are perfectly fine.

After all, it is wrong to punch anyone except a puncher, and our language even has special words — like “retaliation” and “retribution” and “revenge” — whose common prefix is meant to remind us that a punch thrown second is legally and morally different than a punch thrown first.

That’s why participants in every one of the globe’s intractable conflicts — from Ireland to the Middle East — offer the even-numberedness of their punches as grounds for exculpation.

The problem with the principle of even-numberedness is that people count differently. [...]

[Bold emphasis added]

People count differently! Gilbert goes on to describe some interesting experiments that show clearly the asymmetry with which we perceive — and justify — our own ’second blow’ and that of the other guy. It may not help resolve any of the conflicts (or wars) we (or our societies) get into, but it certainly helps us understand one of the important ways in which our minds trick us into not-quite-rational ways of perceiving ourselves and the world around us.

* * *

Here’s the link to Gilbert’s blog post where he provides some additional perspectives behind this op-ed.

Depression, migraines and conflicts of interest

July 13:

The latest incident, disclosed in letters to the editor and a correction in Wednesday’s journal, involves a study showing that pregnant women who stop taking antidepressants risk slipping back into depression.

Most of the 13 authors have financial ties to drug companies including antidepressant makers, but only two of them revealed their ties when the study was published in February.

July 18:

the editor of The Journal of the American Medical Association said she was misled again, this time by the authors of a study linking severe migraines to heart attacks in women.

All six authors of the study have done consulting work or received research financing from makers of treatments for migraines or heart-related problems.

Here’s a profile of Dr. Catherine D. DeAngelis, the tough-talking chief editor of the Journal of the American Medical Association (JAMA).

She is the first woman to edit the journal, commonly known as JAMA, in its 123-year history. Her résumé runs 25 pages, excluding minor details like these: On dull days she blasts the “William Tell Overture” across the cubicles, and the bird in her office ficus tree has a motion detector that makes visitors jump.

[...]

… Dr. DeAngelis said she had dug up conflicts she suspected, that she had questioned the authors, and that when she believed an author intended to deceive her, she asked that the author be disciplined by the dean of his or her medical school.

She also publishes corrections in which she names doctors who evade her rules. She said one such doctor from Harvard, whom she declined to name, called her in tears.

“This follows them the rest of their life,” Dr. DeAngelis said.

Further, she has tightened the journal’s guidelines, making authors sign statements that describe any possible conflicts.

“It’s very rare that it’s deliberate,” she said. “In most cases, they just don’t get it. But the rule is: You reveal, then let me or the other editors decide whether it’s relevant.”

Dr. Joseph B. Martin, dean of the Harvard Medical School, said he completely concurred with Dr. DeAngelis’s argument that doctors should disclose “more rather than less, and ideally everything.”

Dr. DeAngelis says she gets 6,000 submissions a year with an average of six authors each, and she cannot check them all. “I’m not the F.B.I.,” she said.

Asked if she could give the rules even more teeth, perhaps even arrange with editors of other top journals to blacklist cheaters, she said, “Have you ever heard of the Sherman Antitrust Act? I’ve talked to three lawyers about that. They all said: ‘You want to end up in jail? Don’t go there.’ ”

* * *

Benedict Carey’s commentary on willful non-disclosures by scientists who should know better is here. The NYTimes editorial on these recent revelations is here.