Search This Blog

Sunday, June 29, 2014

Four Year Undergraduate Program

The FYUP experiment in Delhi University is over. And it makes clear that autonomy of the university system in India is only in the namesake.

UGC first agrees to the FYUP. Then the government changes, and it suddenly realizes that the under-graduate program should be of 3 years' duration. That a 4-year UG program is against the national policy. The Delhi University says that this is a 3-year UG program. Students can get a Bachelor's degree in 3 years. They need to spend 4th year only if they want an honors degree. UGC says no, you need to go back to earlier system where even an honors degree can be had in 3 years. Why can a university not offer a program which gives something extra and asks students to do extra credits. Should UGC decide the curriculum, credits, what exactly the degree say, and all that. Or should UGC be only concerned with quality and national policy. If UGC was truly concerned about the quality, our state of universities wouldn't be what it is right now.

When UGC can not force Delhi University to do anything, it starts hitting below the belt. It starts threatening affiliated colleges. They should start admission in the 3-year program. Do people in UGC not know that affiliated colleges have no authority to start programs on their own. But who cares for the rules. It was a political match and no rules are sacrosanct. The idea was that colleges should put pressure on the university and of course, UGC was playing to the gallery - the poor students who are waiting for the admission process to start.

Is it illegal in this country to have a four year undergraduate program (other than professional degrees). One of the UGC members is Prof. Sanjay Govind Dhande, who during his term as Director, IIT Kanpur, gave approval to FYUP in IIT Kanpur (that was before DU started FYUP). I wonder whether he considers DU to have violated the law or IIT Kanpur to have violated the law, or has he written his dissent notes in the commission meetings.

Those who are happy with the UGC stance claim that this blatant violation of autonomy was justified in this particular case because the VC of DU had also not followed the proper process in implementing the FYUP. And hence external pressure is justified. I am quite scared by this argument. It is essentially saying that the rule of law is not important. The process is not important. Whoever is more powerful at any point in time gets to decide everything. Why could FYUP be not challenged in a court of law if it was started without a proper process or if it was illegal. The regulator (UGC) could have initiated a probe in whether the internal processes were followed or not. They had more than one year to do all this.

I recall the saga about change of admission process to engineering colleges in 2012. I have been pretty critical of the way IITs conduct their JEE and the entire admission process, and yet I have always maintained that an external agency (even the funding agency like MHRD) should not interfere in the admission process. If they were to decide the process, it is likely to be worse than the current system. And if you read the original proposals for IIT admissions, they were as stupid as any.

So finally the DU backed down and is starting admission this week to the same original programs as they had in 2012. But what are the lessons to be learnt from this.

To me, there are two lessons. One, there has to be a proper system of checks and balances in university governance. The VC is simply too powerful in a typical university and has pretty much unlimited emergency powers. The checks are all outside the system, which should not be there. So one could force his hands by stopping funding (like UGC can do), or could force him through agitations, non-cooperation, strikes, and so on. There are alternative governance structures like that in IITs where the roles of different authorities have been defined very carefully and no single authority can take arbitrary decisions in an autocratic way. Unfortunately, UGC is encouraging this concentration of power even in Deemed universities where many had a different, more balanced governance system. (Though I am glad that recently Karnataka High Court has struck down the UGC regulations in this regard.)

The second lesson is that we should move towards dismantling this system of affiliated colleges. An affiliated college has no autonomy at all and is vulnerable to pressures from all sides. In any case, having a system where the teacher has very little role in evaluation (because we assume all teachers to be corrupt) can not result in high quality of education. The world has moved to unitary universities. Even UGC is on record saying that affiliating universities should have only a few affiliated colleges. If Delhi University was a smaller university, it would be easier for it to start new experiments.

I remember the words of Prof. P K Kelkar who started a lot of innovation in engineering education in India when he became the founding director of IIT Kanpur. One of those innovations was that the instructor will be fully responsible for all evaluations and the final grade of each student. The Directors of other institutions were horrified and warned him that a single case of corruption (in grading) would lead to his removal, and he was committing professional suicide. And he replied. If the experiments that he had launched succeeded, India will succeed, and if they failed, only one person will fail. And if educational institutions will not experiment, who else will.

We need more experiments in education, not less. But they can succeed only when they are decided in an open, transparent system with proper checks and balances and in relatively smaller systems. Till then, I will support autocracy of the VC over interference in the autonomy of the university.

Added on 30th June, 2014

The success of IITs is not only because of funding (as is commonly believed) but in significant part because of its governance structure. If IIT Kanpur could withstand political interference in its admission process to a large extent, it was because the Director did not have any emergency powers, and it had to seek approval from Senate. So political interference can be reduced though not eliminated by having a proper governance structure.

The power of VC in Indian universities are unlimited because of historical reasons. The VCs of the first few universities were all British and the British gave them unlimited powers to keep Indian faculty and employees under check. And we still keep copying the Calcutta University Act whenever a new university is to be created.

Thursday, June 26, 2014

JEE Advanced 2014

So, finally, I have stopped receiving emails about JEE Counseling and I can write a blog about JEE Advanced.

I just don't understand one thing. Why are IITs as a system so opposed to any change. In 2012, we had told the government that we would bring about changes on our own and the changes should not be imposed on us. While I still subscribe to the view that changes should not be imposed on us and lack of change in the IIT system is better than a forced change from the top, it is highly disappointing that IIT system has not been able to bring forth any change in two years other than what was negotiated with the Government in 2012.

For example, for 2013 admission, it was decided that 1.5 lakh students would be shortlisted for JEE advanced. It was decided on the basis that NITs and IITs would admit students on the basis of JEE Advanced, and since there are about 30,000 seats, we should shortlist 5 times as many. But during those negotiations in 2012, it became clear that NITs would not admit students through JEE advanced, but the number remained 1.5 lakh. By the logic of the number, we should have reduced the number of shortlisted candidates to 50,000, but we did not. Remember that in 2012, we were all arguing that if there were only 50,000 odd students giving the exam, we would be able to throw in a few questions in which the students had to write a long answer, and a popular view in the IITs was that a purely MCQ (Multiple Choice Question) test is not a good admission test. But the number did not change for admission 2014 also. The test format remained objective and there was really no change between 2013 and 2014.

In 2012, it was suggested that even if NITs don't take admission through JEE Advanced, we could have a joint counseling. The idea was pursued by some people, but it could be neither implemented in 2013 nor in 2014. (And, yes, I am aware of different timelines for the result of JEE Mains and JEE Advanced, but I am also aware that several proposals were made to have coordination between the two admissions.)

Amongst the things that remained the same between 2013 and 2014 were the errors in the question paper. Why can we not have a system where we can produce an error free question paper. Why do we have to have ambiguities. Can't we have two teams - one to prepare a question paper and the other to check it. Can't the second team include a couple of teachers at the 12th class level. Is there no one in this country who could be trusted to maintain secrecy of the exam.

There was a suggestion in 2012 that there should be a committee consisting of a representative from each IIT who would not be involved with JEE operations, but would look at future JEEs and what changes can be brought in over a period of time. To expect that someone involved with JEE operation would simultaneously think of changes in that year is simply poor thinking and we know from the history of JEE that such changes are brought in mostly through court orders or Ministry orders. No such committee has yet been formed.

I am aware of instances where a bright student has filled in the choices on the ORS, later realized that the responses were shifted by one. Now, JEE insists that you must use a pen to fill bubbles. Why can't they use pencil. May be the copy will not be proper but so what. And if a student has genuinely messed up an ORS, why can't the ORS be substituted. I have myself been an IIT representative many times during JEE. I know we are carrying only a few extra ORS sheets, and therefore, we make this announcement in the beginning that ORS sheet will not be replaced under any circumstances because we don't want a large number of students asking for a duplicate ORS. But I have always given a new ORS to a student if there is a request and I have one left with me. And when I talked to my colleagues before writing this blog, I was told that they have done the same. But then there should be a clear understanding with all IIT representatives that this can be allowed. In fact, what is wrong if IIT representatives were to carry 100s of ORS extra, and actually give two to each student.

Interestingly, I had written about the need to allow students to use pen in the ORS a few years back on this blog. At that time, the local JEE Chairperson had told me that there will be a problem with pen which is what I have mentioned above, that the student may make a mistake and have no recourse to change. I had suggested to him that the right model would be to give the option to students between pen and pencil, and allow a change of ORS. I had also suggested an alternative. This was to give every student a sheet where s/he can only write an answer - A, B, C, D, or a number, as the case may be. And then after the 3 hours are over, the students are asked to put the question paper and any rough sheets away. They are given an ORS at this time, and they are given 10 minutes to transfer the answers to ORS. I believe that if IITs were interested in becoming student friendly, they can find ways to help students.

I also find it interesting that a student could find his/her marks long before 18th. However, IITs did not tell you your rank before 18th. Further, they have asked you to fill in your choices by 24th (later extended to 25th), but the result or seat allocation information will be made public only on 1st July. From computing all marks to a rank should normally take one second. For this task, IITs are taking more than a week. From getting all the choices of students to deciding their seats should take a few minutes, but once again they are taking a week for that. On the other hand, the student needs a lot of time to think about the options, to learn about different IITs and their programs, may be the student want to visit a couple of IITs and talk to faculty, and for all this, they are given 6 days. So note that from marks to rank, IITs need one week. From all choices to seat allocation, IITs need one week. But all students are required to fill in choices in 6 days. How fair is this. Of course, a couple of days of margins are good. Unexpected things happen. But then they should also think of poor students and their parents. And it is in IITs' interest that students take an informed decision on the order of their choices. We should prefer happier students on our campuses.

A student sent me an email which showed that the JEE counseling software had wrongly stated the name of the science programs in IIT Kanpur as "BTech". I immediately forwarded that to Chairman, JEE at IIT Kanpur, who forwarded it to the Chairman, JEE of the organizing IIT (Kharagpur in this case). I never received any reply. Of course, that was as expected from JEE. But I was at least expecting some action. It is obvious to me that someone who receives admission to Maths and Computing program at IIT Kanpur can later claim that I was told by JEE that it was a BTech program, and now I am told that it is a BS program. If I was told that it was a BS program, I would not have filled this in at a high enough choice, and hence give me my next option. It would be very difficult for IIT system to accommodate such a request because the next in order of choices could be in another IIT. But such a student going to court will certainly win the case. Considering all this, JEE should have issued a clarification, put it in red blinking on its website, and sent email to all those who have already locked their choices. Since I did not receive any reply, I don't know if it was done, but I have been given a hint that nothing of sorts has happened.

If organizing an admission test is a very difficult task for the IIT system, they should get out of this business. Or they should do it professionally.

Monday, June 23, 2014

Importance of Placement in Admission Season

We are in the middle of admission season. Everyone wants to know which college to go to, which branch to take, and such other questions. How should one decide these things. The initial temptation of people like me is to suggest to look inside for passion and interest. And study whatever discipline interests you.

However, one finds, particularly in the IIT JEE context that students in top 100 ranks always have deep interest in CS. Their first toys were always a laptop. The next couple of hundred guys are always confused between CS and EE, the two disciplines they seem to enjoy the most (what did they do to enjoy these disciplines - did they try wiring their homes or what), the next couple of hundreds are sure that Electrical is what they want to do in life, but they would not forget to ask how easy it would be to get a branch change to CS after the first year. You get the drift.

What is behind all this. What is the most important factor to decide the discipline and the institute. It is always "scope." Do not be fooled in believing that the word means what kind of skills one would learn in the program, what kind of careers one could have and so on. It means only one thing - how much money would my son make when he graduates. You see the question is mostly asked by parents. In India, passing 12th class does not make you eligible to take your own decisions on higher education. And respecting the elders is part of our culture.

What is wrong if one wants to know if the investment one is about to make will give adequate returns. Those professors who still believe that one should study for the sake of learning are living in the previous century (and it is not that most people did not worry about jobs then). Return on investment must be an important parameter in the decision making. However,  the problem is when it is stretched a bit too far, and becomes the only parameter, and that too without much thought going into it.

So, everyone wants to know the placement information about the program. Why? Because that will help you compute the return on investment. It is believed that the placement data of 2014 will tell you what your son should expect as his first salary in July, 2018. Really! You don't need to understand your son's interests, personality, skillset. May be 2014 graduates with matching interests, personality and skillsets made more money in a different discipline. But obviously that information is too difficult to get and hence no point in even thinking about it. May be the industry will change in 4 years and different skillsets will be in demand. Once again, this information is difficult to get and hence no point in thinking about it.

Do you even know and understand 2014 data. Do institutes and universities give you complete data that you can look at. Sorry, they don't give you adequate data, only that part which indicates a rosy picture. One institute says they had 400 companies come to campus, while the other had 200. Is the first one necessarily better. What if these 400 companies gave only 1000 job offers, while the other 200 companies gave 1200 job offers. I saw a recent news item when one institute claimed highest number of job offers. Needless to say that it hid the fact that the number of students were also highest in that category of institutes, and the percentage of students getting jobs was less than other similar institutes. Do you know when most institutes say that 90% students got placed, this is a fraction of students who "registered" for placement and had more than 60% marks. What happens to those who have 59% marks, and may be they did not register some students who would not have got jobs anyway. There is very little verifiable information available in public domain, which is so small that comparing such information and guessing the future rate of return is totally irrational.

But more importantly, should you really be interested in salary of July, 2018, or salary of July, 2058. Note that the salary in July, 2058 is likely to be two orders of magnitude higher than the salary in July, 2018, even in constant rupees. Looking back at that time, the first month salary will not even look like peanuts. So the return on investment (unless one has taken loan, and there is a serious cash flow problem at home) should take into account the salary of July, 2058 and not July, 2018. And is there any study that looks at correlation of income across 40 years. Is there any industry which will continue to be at the top in terms of paying high salary for 40 years. Could anyone 40 years ago predict what types of jobs would be highest paying jobs in 2014. Sorry to disappoint people, but the answer to all these questions is in negative.

So figuring out which discipline would result in the highest rate of return on your investment is an impossible task.

Should we completely give up on this criteria? Of course, not. Four years of time and a substantial expense is too much of an investment to be made for the love of learning. One does need to worry about whether adequate return will be there for this investment.

But the key here is "adequate" return and not "maximum" return of all possible options. The utter confusion that you see amongst parents is because they are trying to maximize return and not be satisfied with adequate returns, and as we have said above, it is next to impossible to ascertain what will give maximum return over the next 50 years.

Let us look at it another way. Would the return be considered adequate if you were to become richer than 99% Indians within 5 years of your graduation. I would guess that is pretty good return on your investment. But if you were to argue that a CS degree is likely to make you richer than 99.2% Indians, while a Chemistry degree will only make you richer than 99.1% Indians, and hence I will choose CS over Chemistry, then that is the beginning of confusion. Then someone will suggest that may be a CS degree from IIT-X will make you richer than 99.21% Indians, while a CS degree from IIT-Y will only make you richer than 99.19% Indians. And now the confusion is really big time. Inside your heart you know that the basic data for 2014 is not available, that projections over the next 50 years (even 5 years) is impossible, that the exact career will depend on the individual capabilities and skillsets, and hence it is very difficult to guess which program will make you richer than 99.21% Indians and which program will make you richer than 99.20% Indians. And yet, you want that differentiation, you want the best for your son, and the best only means one thing for you - richer than most other people.

Anytime you consider richer than others as a criteria, you and your son will never be happy. If you are not happy being richer than 99.19% Indians, you would not suddenly become happy if you are richer than 99.20% Indians. On the other hand, if you had chosen "adequate" return as the criteria, you would immediately realize that a large number of programs in most government funded colleges would meet that criteria, and hence one could talk about other dimensions like how much fun would one have learning.

Also, when one looks at "adequate" return as the criteria, one is admitting that the Institute/Program will do only so much, and the career depends a lot on the individual. So one would actually work hard and do better in life. On the other hand, when one looks at "maximum" return as the criteria, then one is assuming that it somehow depends primarily on the Institute/program, and the individual has very little to do. It makes the individual less enterprising and indeed the returns will be lower in such a scenario. Thus the proponents of "maximum" return actually don't get maximum returns. So in their own interest, they should switch to the model of "adequate" returns.


(Note that I have used "he" above since parents do seem to consider other criteria for their daughters like "closeness to homes and return on investment is a lesser criteria.)

I had written a very similar blog in 2011. That is linked here.
I had also written a blog on how to choose your IIT/Branch in 2011. That is linked here. It is pretty outdated in terms of specific data though the algorithm perhaps remains the same today.

I used to write the counseling article/blog every year, but haven't done so since 2011, since I have assumed the role of Dean of Academic Affairs at IITK in December 2011. It is politically difficult for me to say that any other IIT is doing things better than IITK in any aspect, and if I were to argue in favor of IITK, it will be discounted by my readers because I will be expected to say that. Even my postings on facebook saying that IIT Gandhinagar (which I have visited more than 25-30 times) is doing very well are sometimes considered politically incorrect. So may be I will be back in 2015 after I am a free bird.