Monday, May 09, 2005

There are IITs, and then there are ...

Have you ever wondered about the widespread perception that there is a precipitous decline in the quality of engineering colleges, the moment you step outside of the IITs? There probably are many reasons, which may also be interlinked. However, let's focus on just one of them here: faculty strength. Let's also see how and why it is an important differentiator.

Take a quick look at the faculty listing in the Departments of Mechanical Engineering at IIT-M, and two NITs -- Rourkela and Trichy. Do it again with the Computer Science departments: IIT-M, NIT-R and NIT-T.

The in-your-face conclusion is this: in each subject, IIT-M has more faculty than the other two institutions combined! In Mechanical Engineering, IIT-M has more than 50 faculty members, against 19 and 12 at NIT-R and T, respectively. In Computer Science, the numbers are: 19 against 8 and 9 [1]. While my research is not exhaustive, I am reasonably certain that this huge disparity in faculty strength between the IITs and the non-IITs is also shared by other disciplines.

Well, why should this disparity matter?

Consider an academic department offering an undergraduate (UG) program. An UG student typically takes 60 courses or so -- 40 lecture courses and 20 lab courses; let us not quibble about the differences here. However, not all the courses need to be offered by this department itself; for example, engineering students do quite a few common courses: mathematics, phsyics, chemistry, etc. In addition, the students may also be asked to do one or two minors by doing enough courses in other disciplines. With these adjustments, the number of UG courses to be offered by the parent department is probably about half (i.e., 30 courses). To this, add another 20 or so courses that the department would need to offer either as electives or for its PG students.

So, the total number of courses that need to be offered is 50 in a year, or 25 in a semester. If you want the faculty to be active in research, then you can expect him/her to teach at the most one course a semester. This gives a good estimate for the number of faculty that each academic department that offers a UG degree program and one or two PG degree programs : 25 per department [2].

First consequence of the disparity in faculty strength should now be apparent: the teaching load per faculty is much higher at NITs, and inevitably, their research output is correspondingly smaller. Over the years, the smaller research output leads to an ever smaller amount of money that flows in to build research infrastructure. The vicious cycle is now complete.

A second (unrelated) problem is the pay structure, through which the governement has made these institutions second-rung (and some would say, second class) to the IITs; their faculty members earn a lower pay than their IIT counterparts with similar qualifications and achievements. Their career progression is also slower. [3].

Both these problems (larger teaching load and lower pay) make the NITs far less attractive for academics with serious interest in research. For example, one could safely assume that all the IITs receive applications from 500 to 1000 distinct (and distinguished!) individuals, who are, by definition, academically oriented. Let us say the IITs can recruit, collectively, 100 to 200 in a year. Where do the others go? Ideally, they should gravitate to NITs which are just "one step" below the IITs. On average, they don't [4]. Most of the remaining people either continue doing what they are doing (usually another post-doc), or take up R & D positions in government or industrial labs in India or elsewhere.

This lack of fresh blood into NITs leads to two related problems: NITs are forced to hire faculty with lower academic and research qualifications -- Masters degree holders. Given a pay structure that rewards higher educational qualifications, these faculty end up doing their doctoral studies at the same place where they teach. I am sure this system produced some excellent professors and researchers; but I think it is safe to say that it is extremely difficult, if not impossible, to maintain a culture of high quality over a long period of, say, 20 to 30 years in this system [5].

In my opinion, the government did a good thing in converting the RECs (which had all these problems) into NITs with somewhat more modern governance structures modeled after that of the IITs. However, I believe this change-over will be purely cosmetic, if NITs do not resolve the following interlinked issues:

  • adequate faculty strength
  • decent faculty compensation (ideally, NITs should be on par with the IITs)
  • rejuvenating their research infrastructure

While the whole country is extremely proud of what the IITs have been able to achieve, it is a sad state of affairs indeed if the next level of institutions are seen to be far lower in stature. At the same time, there is no doubt at all in my mind that NITs -- and many other institutions like them -- have extremely dedicated, hard working faculty, and attract some of the most brilliant students. They all deserve better.

Where do we go from here? What are the possible solutions to the problems (one of which is discussed in this post) faced by the NITs -- and other institutions, too. Any ideas?

[1] I have heard that the NITs also hire temporary staff as and when they need them.

[2] Computer science departments can leverage the expertise in related departments in electrical and electronic engineering, and mathematics. So, they can make do with less faculty.

[3] There are many ways in which the government goes out of its way to insult non-IIT institutions! For example, the Universities Grants Commission's website lists IITs as Institutions of National Importance, as if the others are somehow not worthy of a similar description. My rant about this issue will have to wait for another day ...

[4] "On average" is the key phrase here; don't point out logical flaws by citing exceptions. What I have offered is a statistical observation, not a "rule".

[5] The usual disclaimers apply: I have nothing against those who study for higher degrees at the same place where they work. All I am pointing out is that this is forced on them by the peculiar circumstances that they, and the NITs, face. The long-term sustainability of such a system is, as I said, quite bleak.


  1. Anonymous said...

    (1) NITs should be made autonomous like IITs, set up by a special act of the Parliament with their budgets coming from the central govt.

    (2) NITs should have student and faculty exchange programs with the IITs and IISc. Students at NITs should be able to take at least some courses from the IITs via an internet based distance education scheme. Leverage the IITs to improve quality of education at the NITs.

    (3) Professors at NITs should be free(er) to turn into enterpreneurs if they so wish.

    (4) It is useless to compete with the private industry in things like payscales, the relative disparity should be made up with greater freedom.

    (5) Turn NITs into national institutes with a pan Indian character. Admit students from all over India, instead of having a quota of 50% for locals. This takes care of the union problems in one shot.

    (6) Require NITs to accrue a varying percentage of their operating expenses from consultancies and research.

    (7) Increase fees, get at least some of the operating expenses from students :-D

  2. Anonymous said...

    Is there a 'market' in India for private universities? I know some exist, but can we create better private universities (as in America)?

  3. Anonymous said...

    I don't know if there would be a market - the NIIT system is one which has survived and grown. And many colleges are private, but affiliated to a university for accreditation. We don't have a general accreditation bodies the way the US does, so the only source for a credential or degree is the established university. Isn't IISc unaffiliated?

  4. Anonymous said...

    The industry is very active in their interaction with IITs and its professors. So they are more in tune with the world outside.