Friday, September 21, 2007

How do you get more students to pursue their PhDs in India?


Remember this op-ed that I linked to two days ago? In it, Jaideep Srivastava and Pankaj Jalote address the issue of increasing the number of science and engineering PhDs in India. In particular, they propose creating a new fellowship program for PhD students who could use it to spend two years -- say, their third and fourth years -- in labs of their advisers' collaborators in universities abroad. They cite China and Pakistan as the countries that have adopted this method to boost the number of PhDs. They feel that this will "pump-prime" academic R&D and PhD programs in India.

Let me come right out and say I don't like it. When we bemoan the (generally) poor state of R&D in India, we ought to be examine the bottlenecks within our system and make every effort to remove them. An option that uses an external source of help can at best be a a crutch; in my view, Srivastava and Jalote elevate this crutch and give it a privileged treatment! The US researchers are placed on a pedestal, and the opportunity to work in their labs is being cited as the 'feature' that will attract bright PhD aspirants to our universities. It demeans the expertise of Indian academics by making them, at best, second class partners in the PhD students' development. (Even if this is not what Srivastava and Jalote meant, I certainly don't see how the collaborative arrangement proposed by them can be thought of as one between equals). This is just not on.

[Aside 1: There are also other problems with their proposal: it's too small, and it's quite expensive. The numbers they cite for China (4000) and Pakistan (400) clearly are too small to make a big difference. Even their proposed numbers for India -- about 1000 every year in science and engineering -- for India represent less than 20 percent of the current PhD output! Thus, even with their program in place, India will still have to deal with the problems that plague the remaining research enterprise.]

[Aside 2: Does India really need to increase its PhD numbers? If all we want are more PhDs, we can get them -- including foreigners, and desi PhDs who are working elsewhere --- by paying the right price. If we believe this report, this price may not even be too high! Also, do we know what our current PhDs do after their graduation? For example, do we have a thriving market for PhDs in India, and if so, how big is it? Our R&D labs are notorious for selecting bachelors graduates for filling the bulk of their staffing needs. Finally, how many of our PhDs go abroad, never to return?]

[For the rest of this post, we will assume that there really is a strong need to increase India's PhD output. Read on ...]

Coming back to the proposal by Srivastava and Jalote, does India really need this external help for increasing its PhD output? In engineering, the number of PhDs is admittedly small (about 800 per year). Across all our engineering institutions, there ought to be at least 5000 faculty members who could, in principle, be graduating 2000 PhDs every year without asking for any special favours! If I may put it using industrial terminology, there is ample "spare capacity" that we can press into operation, if only an adequate supply of "raw material" were available. The raw material that is in short supply is the bright young research talent with a solid academic training at the undergraduate level.

[Aside 3: Money is certainly a very, very important factor. We know that India's support for university research has been abysmally low; we really have been running our university research on the cheap. Unless funding levels increase, asking for more PhDs is futile. It does not require great deal of smarts to realize that if you want to double the PhD output, you should be willing to double the funding for academic R&D. Given the decades-long neglect of our universities, we may actually need to more than double the funding for academic research during the initial years]. Fortunately, Prime Minister Manmohan Singh has announced huge increases in education funding over the next five years, so money for higher education may no longer be such a major constraint.]

Thus, the key question is: what are the ways in which we can ensure an ample supply of the right raw material to run India's PhD enterprise?

The supply of students with a good undergraduate training. Out of some half a million engineering graduates, less than 25 percent are deemed by NASSCOM as employable. Let us use employability as a proxy for quality of undergraduate education. Then, if we can improve our undergraduate institutions to double this employability figure to 50 percent, our pool of PhD aspirants would also double. Clearly, this requires rethinking and reforming our undergraduate programs and institutions. I have written about it before, so let me move on.

Enhanced supply need not translate into enhanced PhD enrollment. We live in an era when our bright stars have tons of options to choose from. This implies that that we ought to find ways to make doctoral studies in Indian academic institutions attractive. This, in turn, demands that we address financial and non-financial needs of our PhD students. Here are some ideas to start with:

  1. World-class academic infrastructure: well-equipped labs, excellent internet bandwidth, great academic library that works 24x7, uninterrupted power and water supply, etc.
  2. An increased stipend: currently, it's around Rs.12,000, and it should be higher. [How much higher? Is the starting salary in a public sector company a good benchmark?]
  3. Good on-campus accommodation. They should preferably be studio apartments for everyone (it should definitely be better than hostel accommodation), and one-bedroom apartments for married students. Nobody should be made to wait in a queue for on campus accommodation (which happens routinely in many of our institutions for married couples).
  4. Academic autonomy: they should be able to work with advisors of their choice (with the advisors' consent, of course). In case they run into trouble with their current advisors, they should be able to switch to someone else without much trouble.
  5. Financial autonomy: An annual grant of, say, Rs. 20,000, placed at the disposal of each student.
  6. A comprehensive health coverage for the students and their spouses and children.
  7. A reformed administration that treats PhD students with respect. Currently, our students undergo a lot of procedural indignities, which must be removed. Payment of stipend, for example, must be automatic unless there's a good reason to withhold it.
  8. Generous travel grants, that allow a student to participate in conferences within India at least once every year and in international conferences abroad at least once during the PhD tenure. [Right now, students scrounge around for travel grants from multiple agencies.]
  9. A well-maintained non-academic infrastructure, including facilities for games, sports, yoga, dance, aerobics, a swimming pool, a well-stocked, non-technical and multilingual library, and good places for socializing (eateries, coffee houses, ...).
  10. A graduate student hall (with a refrigerator, a microwave, and a TV) in each department: PhD students do spend long hours -- even after dark -- in the Department, and they need some non-lab space to chill out.

* * *

What else do you think Indian institutions need to do if they are to become attractive destinations for a great number of bright young PhD aspirants? Feel free to pitch in with your ideas.

17 Comments:

  1. Anonymous said...

    Abi:

    I think you missed the most important criteria. I am sure the main reason you and I did our Ph.D in US is not because of any of the 10 things you mentioned; in my case one very important reason was the reputation of the university (and research group) I was joining. I guess it may be the same for you.

    Since then things have most likely improved in India (e.g. from my evaluation, in Computer Science, IIT Bombay Database group is in the world-wide top 10); nevertheless, Indian institutions need to have more such world class research groups, so that bright students do not think that they are sacrificing opportunities by not going to a top school abroad and joining an Indian program.

    =

    Somehow more students need to feel that doing a Ph.D is something important. In my batch (1983-87 Computer Sc at IIT KGP) more than half did their Ph.Ds. I am told the situation is not like that now.

    More join jobs and go for management now. May be the beauty of research needs to be drilled/conveyed to students at an earlier age.

    ==

    I have been watching the IISER web sites and they seem to be having a lot of seminars by outside speakers. See http://www.iiserkol.ac.in/ann_seminar.html
    I think that will motivate more to pursue a research career.

    I wonder if such regular seminars happens in IITs and other good institutions offering undergraduate programs. If not, having that will help.

    ==

    I also think that India should have science/math magnet schools at the high school level, especially to identify bright students from rural and semi-rural areas, and expose them the beauty of science and research. IMSA in Illinois is a bright example. See http://www.imsa.edu/ .
    There is a list of such schools world wide at
    http://jubilee.batelco.jo/Links/tabid/55/Default.aspx but, I think, none are in India.

    I proposed this to the SAC-PM http://www.orissalinks.com/?p=303 . I wish they would have considered this for the 11th plan.
    Would be happy to get feedback on this and if someone likes the idea to push it through the appropriate channels.

    cheers
    Chitta

  2. Anonymous said...

    Abi,

    I think your post itself is a reflection of the chicken and egg situation regarding the PhD issue. Without increased investment PhD production won't increase and for quality PhD education won't be possible without seeding the field with a large no. of well-trained Phds. I think China proposal is quite good to start PhD produciton at a fairly large scale which will jumpstart this process. Anways, my Rs 0.02.

    sa

  3. Sunil said...

    Abi,

    You raise many excellent points, and I agree with all of them. In addition to this long list though, one reason why people do phds in large US universities is the diversity and strength of the departments. For example, in my phd alma mater, the medical school had over 300 top notch research labs in the biomedical sciences. Most universities in India do not have a tenth of this number. Understandably, the diversity of the faculty resulted in some superb intradepartmental interactions, and the different seminar series across campus (where scientists from all over came to give talks) played a big role in the "phd experience". In addition, the diversity in faculty also resulted in a diverse range of courses and minicourses/journal clubs the phd student could attend. Again, invaluable for the student. At least in biology/chemistry and physics, the areas of research are increasingly interdisciplinary, which means the exposure to various fields are almost essential for the student. Finally, in resource and reagent intensive research (like most of the biomedical sciences), larger schools also have the ability to build core facilities with sophisticated equipment etc (for example an imaging facility for cell biology), which is essential for cutting edge research. You cannot build the "world class infrastructure" (as you suggest in your post) without having that critical mass of scientists who can fully utilize that infrastructure. This is something even well off schools (like NUS in Singapore) struggle with, since they have the infrastructure, but they don't yet have the mass of really good scientists to use it well.

    So, it goes without saying that the sizes of real research departments in almost all universities need to become significantly larger for students to maximize their learning and (hopefully) their productivity. So, in addition to all the suggestions above for students, the schools in India need to make a focussed effort (say a 4 year push) to get good faculty, with the promise of all the infrastructure they need.

    Finally, I cannot say that the Chinese effort for more international collaboration is wrong. It all depends on how the exchange is set up (clearly, the "placing on a pedestal" can be problematic). It does benefit the student to see how science is done in different parts of the world. I think a better option would be to have some kind of an "externship" program, where the student has the option during the course of her PhD to spend 6 months in say a collaborator's lab abroad, or some such.

  4. Biswajit said...

    I've been to a number of universities in the US and I can't think of one in which all departments have all the facilities that you are talking about. Still most of them manage to do world class research.

    Universities in the US also have trouble attracting PhD students which is why you see so many foreigners around.

    The main reason the best students don't get into PhD programs in India is the uncertain future following the degree.

    Biswajit

  5. Anonymous said...

    PhDs must be guided by some profs. These profs need to be good enough that they are preferred over going abroad for the same degree. Where are the faculty to guide well? If a PhD student gets 12K and an assistant prof gets 20K, both figures correct as of now, how would one expect a change in the status quo?

  6. Gautam said...

    Abi: I agree with all your comments regarding how we should make it more attractive for students to pursue Ph.D's in science and research. Let me add the following:

    (a) The current status of research in most high-quality Indian institutions isn't highlighted enough. The default approximation in the minds of parents/teachers and the general public is that science done outside India is in all senses superior to science done here. The general public is mostly unaware of the fact that there are truly international level groups and individuals who work in India and set agendas for whole branches of research. One component of making it attractive for researchers to join Ph.D programs in India must also consist of highlighting the fact that the training they will get here and the problems they will work on are in no sense inferior to what they would be doing in analogous institutions outside the country. Our educational institutions have simply not done enough to present the importance and relevance of the work they do to the general public,
    in contrast to the very active effort put in by university departments in the US to present their accomplishments to the community.

    (b) As a possible solution to the problem that in many cases research programs are stalled for lack of students/post-docs - a problem that also exists in many groups in the US. - I don't think we've tried hard enough to get students from other countries in our neighbourhood and beyond, such as Iran, Thailand, Vietnam and so on. From colleagues who have interacted and taught such students, I get the impression that these countries possess a very large number of talented and highly committed students who are interested in research but prevented by the fact that research departments in universities there are not as developed as some in our country. Not only would this fulfill our need for students, but it would also contribute towards genuine south-south cooperation in the long term.

  7. Subrahmanya said...

    >Out of some half a million engineering graduates, less than 25 are deemed by NASSCOM as employable.

    It should have been 25% isn't it?

  8. Semantic overload said...

    I tried sending a trackback, but failed.
    Anywayz, I have put up a blog post as a reaction to yours. It was too long to be a comment. Hence, a blog post.
    You can find it at: http://www.semanticoverload.com/2007/09/23/phds-in-india/

    Thanks
    -semantic

  9. Abi said...

    Thank you all for your comments.

    Chitta: I am starting with the premise that we have in our institutions good, competent faculty, whose research contributions could be a lot more if only they have a greater number of first rate graduate students. This premise is largely valid for our elite institutions, as they continue to be quite selective in faculty hiring. Its validity, however, keeps decreasing as one makes one's way down to state level universities.

    Seminars happen all the time in all our elite institutions. So, this is not an issue. Similarly, while we may not have magnet schools, we do have programs -- KVPY, Young Scientist, etc -- that bring high school and college students in contact with researchers and teachers in our elite institutions.

    The job market has been scintillating. Similarly, management has also been attracting a lot of our top students. Thus, research institutions have to compete with these other places for talented students. This is all the more reason why we need to make doing a PhD in our S&T institutions more attractive.


    SA (Anon): You are mistaken. There is no chicken-and-egg problem here. We have lots of good, competent people in our institutions. These are people who would have done far better under ideal conditions (funding, facilities and students); quite a few of them are capable of being stars in their respective fields. In this post, I am just talking about one aspect of these ideal conditions: how to get good graduate students.

    The China proposal does not address the issue of ensuring that all those extra 20% PhDs will actually stay in India. But my point is that even with existing faculty strength, we can easily double or triple the number of PhDs. If only we can get those graduate students...!

    Sunil: I am completely with you on the need for large departments that would allow cross-disciplinary research encompassing many different aspects of research problems.

    Please don't mistake me; I am not against collaboration with western scientists. I am all for it as long as our own faculty are fully supported and allowed to flourish enabling them to become confident, equal partners in the collaboration. Srivastava and Jalote's proposal not only does not do it, it actually elevates collaborators from rich countries to an exalted status! At our expense!

    Biswajit: You say many of the things are absent in US universities. Is it any wonder, then, that American students are not to be found in great numbers in them?

    As I see it, our institutions are competing for its graduate students against industrial and management careers, MBA and graduate studies in other countries. We ought to realize the seriousness of this competition, and do stuff to make our institutions attractive for grad studies.

    Can you elaborate on the stuff about the 'uncertain future' after a PhD? Because, as I see it, India is poised to increase its demand for PhDs for filling the faculty positions in all those new institutions that are to be created in the next five to ten years.

    Anon: Monthly stipend (and salary) is only one of the issues. Our institutions (with government funding) can do only so much on this front. However, this post is really about what they can do on all these other fronts; I think they can do a lot more!

    Gautam: (a) You may be right: our PR is probably not good enough. Any ideas on how to go about making it better?

    (b) It's interesting that you mention foreign students. But, if we are unable to convince our top students to join us in large numbers (we certainly are getting some of our top students, but I would like to see their number grow), is it reasonable to expect foreign students to come here in sufficient numbers? I am not so sure.

  10. Abi said...

    Subrahmanya: Thanks for that catch! I have corrected it.

    Semantic Overload: Thanks for the comments in your post. When I talk about PhDs, I am talking about good, solid ones, that have published in good journals. And we also have high caliber advisers, whose own research output could be far higher but for the availability of students (and, of course, funding and facilities).

    I am interested to hear more about the market for PhDs. From where I sit, I see only an excellent market in the years to come (with so many institutions that are in the pipeline). But you seem to have a different opinion (and so does Biswajit, in his comment above). Can you please elaborate on why you think the market for PhDs is not all that great?

  11. Pratik . said...

    The market for PhDs is great indeed if you believe PhD student = future faculty. But that need not always be the case, and thats not the sole aspiration that a PhD student can have.

    During my grad studies, I came across a number of bright PhD students, who, despite having an interest for research, where not interested in faculty jobs. They were more inclined towards the industry. And when I say "inclined towards the industry" I dont just mean the General Engineer Trainee jobs (for which a BTech would suffice), but the jobs oriented more towards industrial research and product development - jobs that require a rigorous training in research (and hence their need to go for a PhD).

    The market for PhD therefore is not only faculty positions, but also research openings in industry. Also helps, if these positions in industry pay well. However, the industrial R&D in India doesnt seem to be on such a high (if the rumors are to be believed). I guess this is what anon above means by "market" for PhDs.

  12. Ashutosh said...

    Two words: scientific temper

  13. Ashutosh said...

    http://ashujo.blogspot.com/2006/03/education-in-india-light-after-tunnel.html

  14. Anonymous said...

    Exposure to real life research right throughout the academic career ( from UG to PG)in terms of summer internships, tours and projects. This is severely lacking at the undergraduate levels in the universities.

  15. Anonymous said...

    Niloy,
    Speaking from an IIT BTechs point of view, what would I get out of a PhD at IIT? The only visible option seems to be a (low paying?) job as faculty at IIT. Such limited future options, along with low stipends make it a difficult choice. If you pursue your PhD abroad, you would still have the choice of returning to IIT but not vice versa.

  16. Ex-IIScian said...

    Dear Abi,

    Here's another point of view: I did my PhD from a science department in IISc. Soon after we joined, we were told by our professors that we must do postdocs abroad. They said we were already losing out in job market by doing our PhDs in India. A PhD from a middle-rank US university carried more weight than a PhD from IISc when it came to faculty appointments in my department. This trend might be changing, but has driven away some good students in the past.

  17. Subbiah said...

    Dear Abi and friends:

    It is heartening to know that so many of you - mostly teachers and researchers in some of India's better institutions but not holding administrative and managerial positions in policy making bodies - are so keen to improve higher education and research in science and engineering in India. I have a suggestion. After every topic is discussed and commented upon, one of you may write an article (an opinion piece) in Current Science or in a mass circulation daily (such as The Times of India and The Hindu) so the ideas and the discussion can reach a much wider audience. Your blogs are read by a certain class (or classes) of people. But for your ideas to have real impact we should have a much larger outreach programme. Indeed we must mount an advocacy campaign that would reach all stakeholders - from students to Secretaries to science departments and advisors to the government.

    I must thank Abi for alerting us to topics of great current relevance and always coming up with evidence-based arguments in faultless English.

    Subbiah Arunachalam