Bharat Rakshak

While anecdotal examples are all well and good - the debate can benefit from more statistical and factual data. Career paths, post UG, is one area.

Lets look at possible post-UG career paths - Academia /Research, Corporate, Wall St / Finance & Entrepreneurship.

In each of these areas, are there determinants of 'elite' performance, as well as for entry/mid-level ?

Taking 'elite' first - what are possible statistical measures of 'elite' performance ?

- # of professors / faculty / deans from the IITs as opposed to other institutes
- # of NSF awardees or other STEM research awards across the US from IITs, as a percentage of those with foreign bachelors
- IITian percentage of Indian-American CEOs (as also next level of executive leadership) of well-regarded firms
- IITian percentage in top positions in PE /Wall St as a percentage of overseas-born immigrants
- No of IITians among Indian American billionaires, or at level below
- # of startups founded by IITis as % of immigrant founded

Coming to entry-level

- Which institutes are most preferred for recruitment by best-in-class Tech / Engineering firms?
- Which institutes are preferred for direct bachelors intake by Financial firms ?
- What is the percentage of IITian acceptance at the Ivy League for PG, as opposed to those from other institutes?
- How many startups in India by IITians ?

Most likely, the statistics would present a picture of overwhelming predominance at the elite levels - and continuing leadership at the entry levels.

If one uses the above listed metrics and say hypothetically that the numbers are of the order of 80% of IITians vs non-IITians from India in each of the category, it still would be a damning statement as the beneficiary is definitely not India. Some tangential benefits do accrue due to the success of the elite diaspora to the home country, but nothing of the scale the recipient country benefits. It only furthers the call of many who question the relevance of UGs from IITs for the Indian growth story in such a case.

The real comparison would be between the UGs from the IIT system ( 5+ campuses) and UGs from IVY league schools/MIT or even the UC system for each of these categories, properly normalized for total number of students graduated to date.

Exactly! The above listed metrics make no sense unless you compare apples with apples. So in that case the comparison needs to be on the lines of IITs vs top research universities in US (Ivy League or otherwise) and the relative impact they have had on their home countries across various fields of research. After all the question is not whether IITs are the best we have, rather the question is whether the best we have are serving their intended purpose.

Another viewpoint by looking at the Indian startup landscape and famous ones, it does not look like IITians have overwhelming contribution either when compared to others. One may even infer that IITians have marginally failed in India compared to ones who have succeeded in the west. Why is that ? It is not like that the top of class are the only ones who migrated abroad. Some who migrated out did not go to IVY league schools either, but have clawed their way up with time too. It indicates that the US system has been more helpful and beneficial for the success of IITians, more than saying the IITians have achieved spectacular results despite the US system.

In fact in academics at least there are many non-IITians who are doing equally well even in the US. This indicates the US system is able to nurture all kinds better and that IITians have no special edge as such. What they do have that is special, is a stronger alumni network when compared to other elites from less elite institutions from India living in the US.

IIESTs are back in the new again. BESU is already I think and CUSAT is on the way. Best to cut off all the tentacles to local state politics from these institutions and make them conform to central funding and governance.

http://newindianexpress.com/cities/koch ... 557080.ece

Cochin University of Science and Technology (Cusat) is all set to be upgraded into an Indian Institute of Engineering Science and Technology (IIEST). The state has also got the nod for the IIT. {This will make brand nazis fulminate.}

The proposals got the approval during the talks held between Chief Minister Oommen Chandy and Minister for Human Resource Development Kapil Sibal in New Delhi on Monday. Rural Development Minister K C Joseph was also present during the deliberations.
......
In its earlier bid, the amount was about `519 crore but now the project would require at least, `1,250 crore, Hibi Eden said. Bengal Engineering and Science University got the nod for IIEST recently.

Bade wrote:If one uses the above listed metrics and say hypothetically that the numbers are of the order of 80% of IITians vs non-IITians from India in each of the category, it still would be a damning statement as the beneficiary is definitely not India. Some tangential benefits do accrue due to the success of the elite diaspora to the home country, but nothing of the scale the recipient country benefits.

Completely untrue. India's growth spark was led by software outsourcing. Suggest you talk to folks who are from the industry to get an understanding of what role the IITs played in marketing their services to US/European clients. On the sell-side, every firm would actively market how many IITians they had within their firms. On the buy-side, the decision maker sitting in the US was in some cases an IITian or in most cases a White American CIO whose perception of Indian technology skills was formed based on direct interaction with or indirect reports on IIT engineering skills.

Irrespective of IITian or non-IITian - the diaspora plays a critical role in India's success and growth. This is true of most succesful diasporas - including that of China, whose investments in the mainland led to China's initial growth.

It only furthers the call of many who question the relevance of UGs from IITs for the Indian growth story in such a case.

So what exactly are you recommending here? What is this call that you talk about? Do detail your thoughts.

Raja Bose wrote:So in that case the comparison needs to be on the lines of IITs vs top research universities in US (Ivy League or otherwise) and the relative impact they have had on their home countries across various fields of research.

I don't think that is a valid basis for comparison. The only reason the Ivy Leagues have achieved what they have is because they can afford to pay for the finest brains around the world to come over and study / research at their campus. If the IITs had similar freedom to select the best students from all over the world, and to pay top-dollar for the best global Faculty - they would produce the same or better results.

After all the question is not whether IITs are the best we have, rather the question is whether the best we have are serving their intended purpose.

If the best cannot serve their intended purpose, clearly those who are not the best would find it even more difficult.

We do need to look at which Institutes (does not have to be a single one) are the best, what challenges they face in serving their purpose and what measures would change things for the better. Like I've stated earlier - there need to be multiple institutes each with independance to define metrics for entry and attracting the best as they define it. The market is always the best judge !

Bade wrote:Another viewpoint by looking at the Indian startup landscape and famous ones, it does not look like IITians have overwhelming contribution either when compared to others. One may even infer that IITians have marginally failed in India compared to ones who have succeeded in the west. Why is that ?

Not sure if this is backed by statistics.

The IIMs lead the pack for professional management at Indian firms and financial institutions. A significant percentage of those from the IIMs are from the IITs.

As for entreprenurship, would be good to get statistics on successful & VC-backed Tech startups in India since VCs came in (say around 2000) and what percentage is led by those from the IITs. The percentage may be quite significant.

I would not expect software services to have been spawned by IITians - since it was mostly a body-shopping game to begin with. But surprisingly - there are quite a few well-known names even out there.

Arjun wrote: 2) The top star students do not consider IIT PG second rate. Its just that these folks...... have much greater choice to deal with than those from other Institutes. -----. When it comes to PG / PhD, the situation today is that the global hotspots for talent are elsewhere and therefore the UGs tend to gravitate there

Not to initiate a parallel conversation but I am intrigued by the above point. Of those B.Techs that go for Masters, given the preponderance of who do their Masters in the US i.e. at locations other than their own alma mater (I know there are exceptions) the comment is a bit surprising. You ascribe it to 'these folks have a much greater choice'. If you have a choice between A and B, and you choose B, by definition you consider choice 'B' as your second (and less desireable) choice. People have voted with their feet here, and the evidence is clear. Please have some faith in their analytical abilities forged in the crucible of JEE , they might be right.

Further, the IITs themselves have taken absolutely no pains to market their own PG programs to the UGs. The problem as I see it is actually utter lack of Marketing skills and an appreciation for supply /demand of talent among Indian Univs including the IITs. This will change slowly - but the process is too glacial.

This is also surprising that you think the problem is in marketing. Think about this.. students at IIT B.Tech would have had 4 years to see what the M.Tech program looks like, in their own backyard. The profs. are the same, the facilities are the same (mostly) and they've lived in the campus practically everyday for 4 years. If this is not sufficient for them to sense the world-class ness of the M.Tech program, and they preferred something that is literally 10,000 miles away, then either they are severely lacking in analytical skills or, (more likely) they know the standard and prefer something else. Marketing isn't going to fix this. (unrelated but GOI organizations are not exactly incentivized or known for their marketing).

SriKumar wrote: If you have a choice between A and B, and you choose B, by definition you consider choice 'B' as your second (and less desireable) choice.

'Second rate' carries a different connotation in English from 'second'. What I meant was- as far as I know, none of the UGs looked down upon the IIT PG program - they probably did not spend any time even thinking about it one way or the other. Certainly true in my case.

This is also surprising that you think the problem is in marketing.

By 'Marketing' all I meant was just the effort to at least call every 3rd year batch into one auditorium and give an inspiring speech on why they should consider the IIT PG program. If you think you can't appeal to the head, then appeal to the heart. Its like a guy at placement who receives offers from Microsoft & Google, but lets say joins up with some startup because the startup CEO comes across to explain at a placement session why startups are more exciting, more responsibility for hires etc. If that's possible - why do you think an appeal to patriotism and other techniques would not get more UGs to consider IIT PG?

I think we are wrestling with semantics here......'second' and 'second rate'. In terms of 'world-class' choices available (MIT, Stanford, Caltech, Ivy leagues etc), if students across the world (e.g. South Korea, Japan) are trying to get into Stan/MIT grad programs but not IIT M.Tech programs, then one could atleast argue that they might be second tier by comparison to the world-class institutions, prima facie. There is some documentation to support this point based on number of papers published, number of Ph.Ds graduated, and other objective measures of research. For sure, there are individual professors who are stars in their fields, but that alone is not sufficient to make a world-class program in research. When first-class talent in their own classrooms leave after 4 years, wholesale, that is a massive statement. Of what....that we can debate.

As far as 'looking down' is concerned, if the B.Tech entrants are so concerned about the agglomeration of world-class talent that they will go anywhere in the world to access that network(per your post); and everyone and his uncle knows the said agglomeration is not taking place at a Master's level (as you pointed out), then there are some conclusions one can come to about what they think of the post-graduate talent in their institute. Indeed, if the premise is that JEE is the best indicator of smarts and potential for future success, and we know most M.techs did not pass muster there, some simple (simplistic?) conclusions can be drawn about the type of M.Tech talent relative to B.Tech talent, whether it is explicitly vocalized or not.

If you think you can't appeal to the head, then appeal to the heart. Its like a guy at placement who receives offers from Microsoft & Google, but lets say joins up with some startup because the startup CEO comes across to explain at a placement session why startups are more exciting, more responsibility for hires etc. If that's possible - why do you think an appeal to patriotism and other techniques would not get more UGs to consider IIT PG?

I am all for getting more UGs into M.Tech program....am speaking sincerely here. But the fact is that it does not happen (all that much).....that is the point, and a point for concern atleast at one level. That an intellectual argument does not cut ice with analytical and ambitious students speaks volumes....I think it is a statement of what they think of the idea of doing PG there. If appeals have to be made to patriotism or suchlike to convince them.... I agree that _atleast_ one attempt must be made, and it is possible that might change some minds, but I say, forget patriotism....let the markets decide .

On a different note, in trying to understand your viewpoint, I see a lot of emphasis placed on the exams that mark out a student's ability (JEE/Olympiad etc), but you've not said much about the education process after the entrance. What is the value of the education itself, post-entrance? Is your position that for a smart student (i.e. who makes it through the the JEE/Olympiad type exams) the nature of education post-entrance is not that critical...they'll do well anyway, anywhere. Not putting words in your mouth...just curious. Some hold this view, some dont. My views on this have evolved.

SriKumar wrote: Indeed, if the premise is that JEE is the best indicator of smarts and potential for future success, and we know most M.techs did not pass muster there, some simple (simplistic?) conclusions can be drawn about the type of M.Tech talent relative to B.Tech talent, whether it is explicitly vocalized or not.

Don't agree with this line of thinking. Lets not forget the M.Techs at IIT have topped GATE... In fact increasingly, even on analytical aptitude the entire UG batch may not be the "best". Does it really make sense for a guy with a rank of some multiple of 1000 to join the IITs when based on some other aptitudes he ranks well enough to get a top stream at another Univ?

On a different note, in trying to understand your viewpoint, I see a lot of emphasis placed on the exams that mark out a student's ability (JEE/Olympiad etc), but you've not said much about the education process after the entrance. What is the value of the education itself, post-entrance?

The exams in themselves don't mean much to me except as markers of certain kinds of aptitude. The role of the education process after entrance, as I see it, is to equip each person to move towards the field or profession where they can reach the maximum potential, based on their personal aptitudes. For the 'self learning' types - what they learn is probably independent of the 'teaching' they receive....but that may not be true for all UGs.

Actually, what is more interesting to me is - where is Research headed? Research can be conducted at Univs, Institutes (like TiFR, CERN) and Corporates. For each STEM discipline - what kinds of research are conducted in each of these three domains? What is the demand for research talent (in manpower terms) for each discipline, by domain? How is this demand trending? The aptitudes required would differ between these three domains and also between industries. So what do we know about the specific aptitudes required for each?

Industry reports on most sectors today are widely available - but the 'Research' industry is still a little bit of a black-box - and needs to be demystified.

The big problem with JEE today is the massive amount of coaching that is starting from the 9th standard. People are paying 1 lakh per annum though the 11th standard to improve scores of their children. Many of the better centers are assuring parents their kids will score 90% provided they follow the curriculum. These same kids may not be scoring as high on their 12th board exams. This is skewing the results. There is nothing wrong with coaching in itself, but this has gone too far. I've heard of 'duffers' who have scored around 60% on 12th board exams and then with coaching score 85%+ on JEE (even that is too low for IIT acceptance).

Biggest problem with JEE is it hasn't produced a single domestic Nobel prize for Desh. What kind of system takes the brightest minds and narrowly selects the non-curious types alone. It also sets an unfortunate precedence for the rest of desh about what qualifies as 'brilliant'.

Probably less than 1% of a batch would be graduating in Pure Sciences....and even that was probably arrived at only over the last couple of decades. To expect a Nobel out of such a negligibly small pool is totally unrealistic.

Arjun wrote:Actually, what is more interesting to me is - where is Research headed? Research can be conducted at Univs, Institutes (like TiFR, CERN) and Corporates. For each STEM discipline - what kinds of research are conducted in each of these three domains? What is the demand for research talent (in manpower terms) for each discipline, by domain? How is this demand trending? The aptitudes required would differ between these three domains and also between industries. So what do we know about the specific aptitudes required for each?

Industry reports on most sectors today are widely available - but the 'Research' industry is still a little bit of a black-box - and needs to be demystified.

"Research" is actually the least mystified in reality.

The aptitude of anyone above some certain low thresholds (need not be even very high as in the 99% of any classified group) can usually be measured subjectively with a one on one interview. It can never be "quantified" in a written test, tests of any kind just meet minimum objectives. As an example one could set up a entrance test that solely tests math aptitude beyond even the plus2 level, but even in such a filtered set there is no guarantee that a significant % would turn out to become string theorists in future.

Theo, as I mentioned before once a few JEE filtered one did make to Bhatnagar levels. But so have non-JEE filtered ones. So JEE as a superior classifier of talent is a weak argument. But there is no doubt that a few JEE qualified ones irrespective of their JEE ranks will make the list of achievers too whatever the criterion is for success, just like for any similar subsets from other places.

All of this can be settled by examining the achievements of JEE rankers (say first 200.. that should give a pool of approx 200 x 50 = 10,000 folks) should give a reasonable idea. Very few IITians do research. And out of that, fewer do research in pure science. Manindra Agarwal, Narendra Karmakar, Madhu Sudan and Subhash Khot are ones that have made world class contributions to their fields and I think that two of these are top rankers in the JEE. But there are many other types of achievement (in non-profits, entrepreneurship, patents, inventions, fellows of prestigious academies etc) and in all of these there are lots of people.. in the hundreds or low thousands. Some go into other fields (Raghuram Rajan for example) and there are several who have started hedge funds (Arvind Raghunathan) and others in politics (Jairam Ramesh) and social service (Arvind Kejriwal). And there are plenty more who work in obscurity like the fellow who teaches kids in HP (http://www.kogics.net/sf:kojo/). And of course there are scores of silicon valley startups with IIT founders (too many to list).

There is at least a couple of thousand in research. Still no Nobel. A system to get the best and brightest of a Billion Indians and still no Nobel. Indians from other lowly institutes have received Nobel. This business of saying not to many from IIT go to research is an excuse, thats all.

Being from IIT opens doors for you. That in itself can explain a lot of this 'success'. There are a million tuition types in India but media picks up the 'IIT' guy doing tuition.

Best test case is this one.
For 40 years there was IIT. No industrial & economic boom.
In 1990 Higher education was opened to private sector, churns out 100,000+ engineers. Voila economic/industrial boom.

We should wait for 10 more years to see what effect this "madness" called coaching is going to do to the grads. Two anecdotes -

When we were hiring some people (in India), we were quite tired of going through a lot of padded up resumes of non-IIT people, so when an IIT applicant came up, I said let us hire that person without an interview. Then my colleague - an IIT grad and single digit ranker in JEE - insisted on an interview. It turned out we did not hire that person after all.

Recently I was told with lot of pride that one once removed niece of mine got into a coaching class. Now they have entrance exams to get into coaching classes.

As for the following statement

Does it really make sense for a guy with a rank of some multiple of 1000 to join the IITs when based on some other aptitudes he ranks well enough to get a top stream at another Univ?

So, all that matters is what is currently fashionable? What the person is interested in does not matter? Sorry to say research doesn't quite work that way. it is a long haul and if somebody is not interested, then one would drop out quite early by graduating with a masters which is too common an occurrence now a days with a lot of IIT and non-IIT people who are coming to US for grad studies.

theo_fidel: I don't understand the point you are making. Why are you using Science Nobels as a benchmark of the success of top-ranked IIT undergrads? The IITs are engineering schools, duh. A better benchmark is applied research (esp patents), startups and the like. Jayant Baliga whose name came up in one of the other threads is an IIT Madras grad and his contributions have more impact than a run of the mill Nobel.

Sorry but how come a Nobel is less worthy. There are physicist/engineers who have won it too.

http://en.wikipedia.org/wiki/Jack_Kilby

Besides looking at IITs as an engineering institution is also at the core of its limitation in its current definition. This is internalized not just by UGs, but also a significant number of faculty including in the sciences within IITs. Need to broaden the horizons here to get to the league of MIT/CalTech/Stanford etc in the long run.

nsriram wrote:The IITs are engineering schools, duh.

nsriramji, Karmarkar is an EE who has become a mathematician - he is currently doing computational mathematics (IOW, experimental mathematics). Madhu Sudan arguably is a mathematician now. All the people who have become computer scientists are mathematicians. They are even listed so in the mathematics geneology project. I believe (and several great mathematicians believed in this)that the division between pure and applied mathematics is unnecessary and artificial.

One of the problems that ails our education system is to put people into slots or streams and say that they cannot cross the boundaries. It is a side effect of the desire to protect ones own academic territory. To some extent one would see this in all the education systems around the world. But now a days there is a recognition of the detrimental effects of such a POV and hence the advocacy of multi-disciplinary research.

As Bade said above, there are several EEs is who won nobel prize in physics, for example Jack Kilby for ICs. IIRC, Von Neumann was a chemical engineer at one point. Lauterbur, a chemist, won (joint) for Medicine.

matrimc wrote:

nsriram wrote:The IITs are engineering schools, duh.

nsriramji, Karmarkar is an EE who has become a mathematician - he is currently doing computational mathematics (IOW, experimental mathematics). Madhu Sudan arguably is a mathematician now. All the people who have become computer scientists are mathematicians. They are even listed so in the mathematics geneology project. I believe (and several great mathematicians believed in this)that the division between pure and applied mathematics is unnecessary and artificial.

One of the problems that ails our education system is to put people into slots or streams and say that they cannot cross the boundaries. It is a side effect of the desire to protect ones own academic territory. To some extent one would see this in all the education systems around the world. But now a days there is a recognition of the detrimental effects of such a POV and hence the advocacy of multi-disciplinary research.

As Bade said above, there are several EEs is who won nobel prize in physics, for example Jack Kilby for ICs. IIRC, Von Neumann was a chemical engineer at one point. Lauterbur, a chemist, won (joint) for Medicine.

My point was that because IITs were not set up as an elite Science or Math undergrad institute, it is not meaningful to benchmark using Science Nobels (you need more realistic measures like academy memberships, patents, startups etc; that some IITians may have moved elsewhere to Economics or Math or Science or Finance or Social Work or what have you is something that cannot be prevented; ideally each individual should be free to pursue their interests and talents). For allowing individual talent to flower, the IITs could morph into broader based institutes (a la MIT or Stanford). The only way for that, IMO, is to provide more autonomy and let each institute decide their own entry criteria and have decent direction from their board of governors. I believe with the new OBC quota, the total reservations is now approaching 50 percent. According to a recent IIT director about 50 percent of present day IIT undergrads are doing time pass. There must be a strong overlap between these two categories. You cannot have excellence and inclusiveness, all at the same time. My guess is that the current IITs, with their rules and regulations, are finding it difficult to move in this direction; I agree that they should no longer be exclusively engineering schools but become regular universities. They are stuck in many ways and people like Sibal and the IIT directors are not exactly providing leadership. You can see the type of frustration that is experienced by IIT faculty here. http://dsanghi.blogspot.com/2012/07/aut ... tions.html This post has an example of the damage that is done by removing autonomy from small technical colleges in UP. The type of autonomy Mr. Sanghi desires for the IITs and other colleges seems like a pipe dream. Anyway here is his recommendation:

So, here is a simple formula to improve quality of engineering education in the country: disband technical universities, even if you have to have affiliating universities, make sure each university has no more than a handful of colleges affiliated to it. Common curriculum and common exams have ruined education in this country, since they go against the very grain of autonomy.

I completely agree that each institute should have complete autonomy as regards setting the course of education and the nitty gritty of exams and other regular activities. Very few really even in the old days wanted a central exam body to conduct UG/PG examinations, though if one went to a large university like Univ of Madras or Univ of Calcutta then it was unavoidable. But the good prof is mixing things up when making his case.

The issue here is about the number of entrance examinations (at least reducing them) an aspirant has to take. Even in the west, the Ivy leagues and other leading elite schools all use a centralized examination system and their autonomy in whom they would select is not restricted by that. I assume even IITs can have some leeway in their selection of qualified/filtered candidates for a given course, or if it does not exist now then such a path should be provided.

I dont think the PCM depts in IITs dont have the clout or size to be big players in the game.

the ones who have clout are the ones with big armies - Mech, Chem, ECE, EE.

In terms of resources (physical) Phys/Chem/Math labs in IITs are well endowed when compared to even some of the better autonomous universities with no affiliated colleges under them. This was the case two decades ago. An MSc student at any IIT would have access to very high quality labs unlike at other good but poorly funded univs. As a I recall we had open access and no rationing like I saw in JU for expensive items. Even in the 80s the crib was that students were not fully utilizing all available resources. The fact is Indian students by and large are not prepared to work with their own hands. They have to be taught after they enter college. Little different in massa environment, where I saw a 3rd EE student convert over as a Physics student and setup his muon telescope from scratch and got his bachelor's project/thesis done largely with little supervision. He is now a Asst Prof and part of the Jul4th discovery party.

Things are different now perhaps and will even out with time, but the damage done to the Univs will take time to heal.

Bade wrote:The issue here is about the number of entrance examinations (at least reducing them) an aspirant has to take. Even in the west, the Ivy leagues and other leading elite schools all use a centralized examination system and their autonomy in whom they would select is not restricted by that. I assume even IITs can have some leeway in their selection of qualified/filtered candidates for a given course, or if it does not exist now then such a path should be provided.

Apparently there are 180 odd engg entrance exams in India. Have you heard of Amrita University? I had not. It has its own entrance exam that is taken by tens of thousands apparently (its VC is a distinguished IIT Madras alum who quit his position to found this). What this means is that the school exams are completely untrustworthy, even for new colleges. Reducing this number by a few isn't going to make much of a difference and getting all these different colleges to agree on their entrance criteria is impossible. Even in the US, a typical student might apply to half a dozen schools (some high range, some mid range, and some safety schools). They do not apply to 100 or 200 univs. The point that Sanghi is making is that institutes should be given autonomy to make good or bad choices. Lack of autonomy compromises any chance at excellence. If Institutes are trusted to do their own entrance criteria, there is a better chance of them selecting the students that are a good fit and then they can waive entrance exams etc. for those who have done other things. Unfortunately the govt sees funding and autonomy as in opposition. Perhaps the new univs that are being started now have a better chance to do things afresh, if they have the right leadership. The JEE has a role as an index for high level knowledge in PCM for high schoolers and that debate/thread was about messing with this index in dubious ways and now Sibal has backtracked (the argument that some reduction in number of exams would happen was bogus). The larger issues remain and they are independent of JEE.

The problem again is that few are willing to think out of the box of canned exams. What follows is a possible explanation for why many want to conduct their own entrance exams, like Amrita Univ that you mentioned.

What is the current merit list for iit-jee ? If I remember correctly it is 15% more than the total number of seats available around 10k now, isn't it ? I do not know if the JEE panel is willing to expand this list to have a super list that others can also select from. If they agree to it (though beyond a certain absolute cutoff it may even not make sense to expand the list), then other univs could draw their candidates from this list. Will the other institutes accept it if the first one from the expanded list that opts for Amrita Univ has a JEE rank of 10,001 ! It immediately brings down the value of the education that they are selling at a price in the public's eye. So naturally they will want to conduct their own test and create a list of AU-JEE top 500 too. It is all just a H&D exercise in the end to protect what each institute considers as its brand value.

Now, if every univ and each of the iits use a SAT like score administered by an independent agency, and on top of it, they do their own aptitude tests/interview which are all very different from each other in content as well as methods employed then it will make ample sense to have multiple ways to fill seats in different institutes that the student may apply for.

And autonomy also gets insured to some level right from the beginning in the choice of students depending on the goals of the institute as decided by the strengths of the faculty in each univ/institute.

Right, actually the AIEEE is a bigger (in terms of number of students) exam than JEE and is likely more usable in general, especially when one is talking about ranks > 10k in the JEE where the JEE may not be meaningful any more. But for the reasons you mention, places like AU will not use AIEEE or JEE ranks. Part of the reason is protecting their brand value and part of it might actually make sense (they are devising their own test that measures fit to the place), although one cannot say in what proportion.

Making a SAT like test or subject tests (called APs) is a very expensive business and hard to do in a multi-lingual country. ETS and the collegiate board are profitable ventures though and I have no doubt that they would be looking to expand into India (along with American univs).

Thinking a little more objectively, a lot depends on the actual distribution of scores in any test, JEE or any other kind. If the distribution is normal/gaussian shape then a expanded list is possible, assuming the cutoff is way ahead of the median of the distribution at present then it can definitely be taken all the way to the median of the scores as a new cutoff to create a list of ranks.

If JEE like test does not show this distribution, but has a peak or peaks (multi-modal) at two extreme ends with a shallow trough/plateau in the middle, then you do not get much of a choice to expand the list. The peak that you get closer to 100 is all that you get. This means all who fall there are a cut above due to inherent abilities or a heavily coached subset with no way to tell the difference. My suspicion is, that is what is happening at present. A few very good students will always make the cut to ensure that "all is well" when one looks back a few decades from now at the achievers and what their relative ranks were. But the system would have also pushed through a lot of undeserving ones.

Arjun wrote:
Actually, what is more interesting to me is - where is Research headed? Research can be conducted at Univs, Institutes (like TiFR, CERN) and Corporates. For each STEM discipline - what kinds of research are conducted in each of these three domains? What is the demand for research talent (in manpower terms) for each discipline, by domain? How is this demand trending? The aptitudes required would differ between these three domains and also between industries. So what do we know about the specific aptitudes required for each?

This is a broad question and the answer below is somewhat vague, as a consequence. The word 'research' is a catch-all term that can mean different things to different people. It covers a wide range of 'innovative' activity, with the Nobel prize winning, pure science-type research work at one end of the spectrum, the level of research getting more 'applied' as you wander towards the engineering side of the spectrum. In terms of institutes that do research, departments of universities are more 'pure' relative to corporate R& D labs that are geared to specific technologies that the parent companies wish to benefit from; also univs cast a much wider net for funding sources, working with private companies and govt. alike, so the research they do will reflect that. At the most applied end is the plain jane development-type (The 'D' in R&D) work in small companies e.g. saada product improvement. One does not need a Ph.d for this but simple product improvements do have a positive impact on the lives of the users, and that is a reasonable yardstick to use. Pure science wallas might not consider engineering work as not true research (too applied perhaps) but it is research in that it helps improve products and therefore has a more immediate impact on the society.

As for aptitude, I guess one might generalize that the closer one is to pure research, the greater the possibility that s/he might need to work in abstract space (i.e. mathematics,...., trying to imagine/come up with a model of some arcane physical behavior for which there is no standard analogue, etc). I also think it is very tricky to make judgments in small timeframes i.e. via an aptitude exam. The student may be convinced that s/he is interested in post-graduate type research work only to lose interest a year or two down the line. In a Ph.d program (which takes years), this can be scholastic suicide. There are some (a few) who know from early on that they want to go beyond the routine type work and do a Ph.D, but for the rest, unless one spends some time in a research crucible, one cannot be sure that they can survive the long-haul. It takes a certain drive and mind-set to stay with something for 3 to 6 years, on a pittance of a stipend while their friends are earning the big bucks after their Masters, with no partial credit for dropping out half-way.

Regarding demand for each discipline, I have no idea other than to suggest that US govt. spending in this regard should offer a clue as to what might be reasonable. I am guessing the largest amounts are allocated for medical research via NIHs, etc. to fund research on cancer and other killer diseases. It is a good first priority to have.....to fund research on issues that affect the health of the people in the nation.

While singing praises for autonomy at any cost, we forget the very reason we have this discussion is due to too much autonomy at the other end of education ie, high school/plus2 level. The myriad boards and state standards with wide variance is the root cause for our takleef. So autonomy in itself has rarely solved anything. IMO, what is urgently needed is a lot more strict uniformity of primary and secondary education so that the playing field does become level before entering higher education. This in itself can reduce the bias in the filtering system without diluting real standards.

Why this has not gained attention in larger measures does confound me.

Bade wrote: If JEE like test does not show this distribution, but has a peak or peaks (multi-modal) at two extreme ends with a shallow trough/plateau in the middle, then you do not get much of a choice to expand the list. The peak that you get closer to 100 is all that you get. This means all who fall there are a cut above due to inherent abilities or a heavily coached subset with no way to tell the difference. My suspicion is, that is what is happening at present. A few very good students will always make the cut to ensure that "all is well" when one looks back a few decades from now at the achievers and what their relative ranks were. But the system would have also pushed through a lot of undeserving ones.

Agree with your analysis on the multi-peak distribution of JEE results. But I am a little curious as regards your last few points, since this theme of 'undeserving ones' seems to be a constant from your end. You mention a 'few good students' and a 'lot of undeserving ones' - Your usage of language would seem to imply something like a 20:80 ratio of good to undeserving.

By what metric would you judge that a candidate was 'undeserving'? Would the ability to get through the IIT-UG successfully in 4 years be an indication of 'deservedness'? If not, is there another metric you have in mind?

Also, why would the same argument not be applicable for GATE, AIIMS, IAS, CAT or any other high-profile Indian entrance for that matter? In any case - the consensus seems to be that Board exam success is completely 'undeserved' given the rote nature of their syllabus.

Would you also apply this logic to the South Koreans who enter the Ivy Leagues in large numbers? Presumably they are mostly undeserving candidates - who've only gotten in due to intensive coaching.

I used the term "undeserving" as in the test (JEE) is supposed to filter out gems as the claim is made. But like in any test the ratio of gems to mango men will be similar for a large sample size. But the sample size is small in this case, well at least it used to be. Small sample sizes are prone to heavy bias. Coaching provides that bias in diluting the sample and in a way defeating the purpose of the test. Nothing more.

Yes to the question on coaching in general diluting carefully calibrated tests. There is a local school which takes students from grade 8 onwards based on a written test. Recently there was an article on dilution of standards there over a period of less than 5 years. Coaching and Asian families go together in this district.

Bade wrote:The fact is Indian students by and large are not prepared to work with their own hands. They have to be taught after they enter college. Little different in massa environment, where I saw a 3rd EE student convert over as a Physics student and setup his muon telescope from scratch and got his bachelor's project/thesis done largely with little supervision. He is now a Asst Prof and part of the Jul4th discovery party.

Things are different now perhaps and will even out with time, but the damage done to the Univs will take time to heal.

This is absolutely right.

In fact people come in with love of science and drive to work and random hyena's standing in the back ground call people 'know it alls' and 'gas bags' and assorted pseudo high-funda terms and destroy peoples love for learning. Testing and scoring and 'performing' in annual reviews becomes the most important things. Very like dutiful little hamsters. Then the same folk have the temerity to claim SN Bose, I tell you. The man was fearless and beyond ridicule, which is also why he left India before getting recognized. Not for him the piffling hero worship twaddle that goes on in Indian labs and GOI type institutes. He wrote to Einstein, direct, about things Einstein and Planck had missed and despite Einstein not agreeing he proceeded with his research and published it as well. Is there anyone with such intellectual strength in India today. The answer has to be no. And a large reason is the JEE itself and its measure of what counts as brilliant. Testing and jumping through hoops some else sets up is considered the best student types. With the result that today we have a bunch of band-wagoners. Always the mridangam, never the Nagaswaram.

We even had some here call Nobel prizes 'run of the mill', like IIT's and IITians are above it! What manner of mill is this, more like mill stone that, around the entire desh. The whole purpose of sacrifice by desh was to bring scientific honor. And then we had matrimc demonstrate that 'single digit' JEE topper (apparently a badge of honor that has been revealed to all and sundry) is now random management type working on job interviews. Nothing against matrimc, but really. Is this is end product of the JEE selection process. How asinine is this, like Feynman telling everyone his grades in High school as a badge of honor. What can one say....

WRT Arjun's claim that without IITians there would be no outsourcing to India, and apparently no economic boom either, it simply took my breath away. Wow!!

While we are at it, Chennai does have two world renowned institutes. MCC and Loyola College, note IIT is not include and to my shame neither is CEG. Students do come from around the world to study here. From Europe, Korea, even Japan & China. Their name far out shines others.

Bade wrote:I used the term "undeserving" as in the test (JEE) is supposed to filter out gems as the claim is made. But like in any test the ratio of gems to mango men will be similar for a large sample size. But the sample size is small in this case, well at least it used to be. Small sample sizes are prone to heavy bias. Coaching provides that bias in diluting the sample and in a way defeating the purpose of the test. Nothing more.

From my vantage, JEE is probably the least susceptible to coaching in comparison to GATE, IAS or other such tests. So I would be far more concerned about the quality of our M.Techs and bureaucrats, if I were to go by this metric.

imo some posters who work in sciences are falling prey to the assumption that the best n brightest always need to work in sciences. in No other country or system is this the case - people go into sc, engg, finance, business, teaching ... so in that sense we can expect IIT UGs to be distributed across the spectrum too. no one vocation is "better" than the other and just because a presidents gold medalist chose to join google doesnt mean his work or contribution there is "inferior" to that of a gold medalist working in physics somewhere. there are challenges in all fields. a physicist might do one big thing in a career spanning decades or none at all. a engineer can atleast take some satisfaction in producing something that reasonably works and is useful to humans periodically

given the relatively smallish number of pure science r&d and teaching jobs at tier-1 places worldwide vs that in engg/finance we should anyways expect and will get far more IIT UGs in these vocations.

even soviet union with its command driven system channelled its best into sciences, aerospace and some other areas considered crucial, not just sciences.

Theo_Fidel wrote:WRT Arjun's claim that without IITians there would be no outsourcing to India, and apparently no economic boom either, it simply took my breath away. Wow!!

If that's the 'claim' that you've inferred from my post - who am I to stop you? You are free to 'infer' what you want from my post, as I am to infer what I can from your logic.

Thanks, Srikumarji. Some random thoughts-

SriKumar wrote:At the most applied end is the plain jane development-type (The 'D' in R&D) work in small companies e.g. saada product improvement. One does not need a Ph.d for this but simple product improvements do have a positive impact on the lives of the users, and that is a reasonable yardstick to use. Pure science wallas might not consider engineering work as not true research (too applied perhaps) but it is research in that it helps improve products and therefore has a more immediate impact on the society.

Product development at corporates is a two-staged process that answers- 1) what features are required in order to improve the product and 2) how does one incorporate these features that are determined to be important. The former is more the domain of Market Research, requiring understanding consumers and competition. More of an MBA-ish role, whereas the latter is more engineering-oriented.

As for aptitude, I guess one might generalize that the closer one is to pure research, the greater the possibility that s/he might need to work in abstract space (i.e. mathematics,...., trying to imagine/come up with a model of some arcane physical behavior for which there is no standard analogue, etc).

Agree. I would also think Univ/ Institute research in the STEM areas requires greater aptitude for abstract thinking and mathematical manipulation, whereas the Corporate type would be more engineering oriented. Requires different kind of aptitudes.

I also think it is very tricky to make judgments in small timeframes i.e. via an aptitude exam.

I understand - but it should be possible to define the traits that one is looking for even in a subjective process. So, in addition to a certain level of analytical skills - one would look for intellectual curiosity / ability to persist in an undertaking / not be as focused on making money / big picture thinking perhaps? Would be good to arrive at a list of the traits that one would typically look for.

Regarding demand for each discipline, I have no idea other than to suggest that US govt. spending in this regard should offer a clue as to what might be reasonable.

Some areas are probably heavily dependent on government grants - whereas others may have moved to a situation where they are largely funded by corporates. IBM may be doing far more high-end research in the area of computing than say many Ivy League Univs, excluding areas like Theoretical Computer Science.

We are moving to a world where even high-end research will increasingly be conducted by corporates...budgets for research, broken up by industry, may offer a clue on where the demand lies.

breakup of industry r&d spending in 2007 (usa)
http://www.plosone.org/article/fetchObj ... tion=PNG_M
chemicals + pharma get 40% of industry spend

breakup of federal govt r&d spending in 2007 (usa)
http://www.plosone.org/article/fetchObj ... tion=PNG_M

military r&d consumes 50% of federal r&d spending and NIH gets 25%
NSF gets 3% , NASA 7%

so it would appear fields like biology, chemistry, chem engg, automobiles, medicine and the military cos is where most of action is and physics/maths attracts little to no funding from industry and govt.