Search This Blog

Saturday, October 16, 2010

Autonomy for all - not just for IIMs and IITs

Recent news articles have indicated that the Government has promised autonomy for IIMs. They will soon be able to decide where to open new campuses, or give top up salaries to their faculty, and so on. There were also news reports about Mr. Rahul Gandhi arguing in the parliamentary committee on higher education that IITs need to be given greater autonomy so that they can compete with the best. He was participating in the discussion on the amendments to the Institutes of Technology Act, which are necessary to legalize the status of new IITs. There also have been talks of introducing the concept of "navratna" universities, that is, some universities to be given more autonomy compared to the others. We have also heard that "Innovation Universities" would have much more autonomy than other universities in the country.

All this is great news. The Ministry of HRD under the leadership of Mr. Sibal, has been consistent in its approach towards greater autonomy in education sector, and this is very admirable. However, the approach is also piecemeal. Only a few top places are to be allowed greater freedom. MHRD needs to go much beyond this.

For example, as per a news item, IITs will be allowed to recruit 10 percent faculty from abroad. Foreigners are not queuing up to work in IITs on regular government salaries. The only real need for IITs is that a few NRIs, who for whatever personal reasons want to return, but have given up Indian citizenship, should be able to work in IITs. Also, there are a few foreigners (other than NRIs) who want to spend some time in India. So, the 10 percent limit will practically not be breached in foreseeable future. So, if MHRD had said that IITs are allowed to recruit foreigners without any limit, it would not change ground reality. But MHRD would have sent a signal to the world that it is ready to give real autonomy to IITs. Also, why this 10 percent only for IITs. Why not for NITs. NITs too are having the same problem as IITs. There are some NRIs who want to return and approach NITs, but they can't be offered permanent jobs.

IIMs are to be allowed to give extra salary to its faculty, as long as that money is not coming from the government. Why can't this be done for NITs as well. Surely, NITs are not having private sources of money today, and won't be able to make use of this autonomy. But, if NITs are given this autonomy, on one hand, MHRD would have signaled its positive intent, and on the other hand, there will be NITs who will find ways to raise private moneys (like alumni gifts), and try to attract better faculty.

IITs and IIMs are too few and too small to make any significant difference to India. Giving them more autonomy will help them compete with the best in the world. But giving autonomy to the tier 2 institutes will really start making a difference in the quality of education in India. Also, these Tier 2 institutes can then feed the post-graduate programs of IITs and IIMs, creating a strong research eco-system in the country, which then would encourage the Tier 2 institutes themselves to start focusing on research and not just remain feeder institutes to IITs.

And, of course, we have to unshackle the university system as well. The argument for control is a flawed one. If we do not control a poor quality institution, then it will become even worse. And hence the need for control. But if we keep controlling, we are not allowing it to improve at all. And maintaining status quo in the age where the only constant is "change", is damaging to the university system. Just count the number of letters that regulatory authorities (like UGC, AICTE, MCI, etc.) send to a university in a year, and you will know what autonomy they enjoy.

Thursday, October 14, 2010

Grading Schemes for Evaluating Students

It is well recognized by educationists all over the world that having evaluation in terms of discrete grades is better than giving continuous marks out of 100 (or whatever be the maximum marks). IITs and several other good quality institutes have been awarding grades for decades. Slowly, other universities in India are catching up with the practice. UGC recommends this. And even school boards are now coming around to the view that giving grades is better.

There are two questions that arise. Better in what respect. What kind of grading pattern is to be selected, in particular, how many different grades be there in the grading scheme. Actually, the second question cannot be answered without understanding the first.

The grades are better than marks because they reduce stress on the students. The world over, people have realized that performance metrics are extremely important for students, and they work hard to improve their measured performance. There are very few students who learn for the sake of learning. Normally, in a grading system, the students do have a sense of where they stand in a course, and therefore, what grade they can expect. They can estimate the effort they need to put in to maintain that grade, and the effort that is needed to improve the grade.

Basically, when they are studying for marks, then every extra hour of studies before the exam, could lead to possibly a few more marks, and that pressurizes the student to study more. While in grading system, a few extra hours of efforts on the last few days may not give him an improved grade, and therefore, the student can be more relaxed. He can spend more effort on courses where the chance of improving the grade is higher.

Of course, we do want some stress on the students. We don't want them to be completely relaxed throughout the year/semester to the extent that they don't attend any classes, don't do any assignments, don't study for exams, etc. If we take the grading system to an extreme and have only two grades - Pass and Fail - there is absolutely no pressure on the student to study, and that leads to very poor performance on an average. (Many universities have tried pass/fail grades for a few courses, and have had the same result.)

On the other hand, if we have a finer grading system, and if we consider the extreme situation of a different grade for every percentage mark (basically the good old marks scheme), then there is tremendous pressure on the student and that is not good for the learning either.

So, we need to find a balance. What is the right balance - there is no unanimity on this, as can be seen by the diversity of grading schemes in universities of the world. But what is of concern to me is that the grading schemes get changed for wrong reasons.

Just before I assumed the role of Director at LNMIIT Jaipur, they had decided to change the grading scheme, and many students and even faculty were not happy with that, and therefore, I had occasions to listen to lots of arguments  in favor of the older scheme. Similarly, we have gone through lots of debates on this issue at IIT Kanpur.

Basically, students want a finer grading system. Let us look at a system where there are 5 grades - 'A', 'B', 'C', 'D', and 'F'. The students in such a system would argue that if they are missing an 'A' grade and get a 'B' grade, it is too sharp a drop. If there was an intermediate grade, call it "A-" or "AB" or whatever, it would appear better on their transcript. This is a flawed argument since the assumption behind this argument is that whoever is getting an 'A' grade in the current system will continue to get an 'A' grade in the finer system. They don't imagine that if there were a finer grade, some people from higher grade will be reduced to the intermediate grade as well. In fact, if we look at a 5-grade system on a 10 point scale, where passing grades have numerical equivalence of 4, 6, 8, and 10, and if we assume that the students will be equally divided in 4 bins, then the average CPI (or CGPA) would be 7. On the other hand, if we had finer grading system, and we had grades with numerical equivalence of each number from 4 to 10, and the number of students were equally divided into 7 bins now, the average CPI would remain 7. There have been many universities who have changed from coarse to finer grading and vice versa and have seen absolutely no change in their CPI/CGPA. So, the finer grades will only mean a slightly higher grade in a few courses, and a slightly lower grade in some other courses - with the overall impact on CPI being zero. So it really does not help the students.

On the other hand, while students wouldn't agree - since comparing the impact of two systems on the stress levels is impossible for an individual student - the experts agree that the finer grading system would lead to greater stress - more competition amongst the students. So, each university has to consider whether their students are too stressed out, spending too much time on examination related learning - then make the grading system coarser. On the other hand, if the students seem to be taking the exams and other evaluation mechanisms too lightly, it is time to consider a finer grading system.

At the end, I will like to point out that the grades reduce stress primarily when the student can do a trade off between the effort and the expected grade. This trade off is possible only when there is a continuous evaluation process, whereby the student can correlated the effort and performance over a period of time before the final exam. Hence having exams/quizzes/assignments/projects and other evaluations throughout the year (or semester) is important for success of grading schemes. (Continuous evaluation is anyway a good idea, even if one is not implementing any grading scheme - so that the students learn at a smooth pace throughout the year/semester, and not having to cram everything at the last minute.)

Sunday, October 10, 2010

Does Indian Industry value merit

Being the Director of LNMIIT Jaipur for about two years, I had lots of interesting experiences, some of which I intend to narrate on this blog. One of the most eye opening experience was the attitude of industry towards quality. They just could not care less.

I was trying my best to use the IIT Kanpur alumni network to convince companies to visit us for campus placement. The technical guys I would talk to would get convinced easily that our faculty was far superior to most institutes in the country. We had the best PhD faculty to student ratio outside the IIT system. The graduating students were all those who had a 4-digit rank in AIEEE. The labs, the library, the Internet and other such infrastructure was far superior to other places. The performance in 3rd party exams like GATE was very good. With all this, one would have thought that once the companies knew about our existence, it wouldn't be hard to convince them to come to campus.

I couldn't be more wrong. The technical guys would forward my letter/brochures etc., to their campus placement or talent acquisition departments, and that will be the end of everything. To give an example, there is this company who has a large setup in Gurgaon, and is in the business of developing communication software. Our curriculum was designed for this niche market, and included all basic courses of computer science and all basic courses of communication. The degree we awarded was in Communication and Computer Engineering. We sent our brochure and an invitation to the company, but no response. A few emails, but no response. Then a few phone calls, no commitment still. Then I activated my IITK alumni network. My batchmates and my students held top positions in the company. They promised to help.

A few days later, I received a phone call from some senior manager from HR department. "We are receiving many phone calls asking us to visit your campus. Can you do something so that these phone calls stop." I told her that it was simple. She could visit the campus, and we will not need to use our contacts. She told me bluntly that that will not happen. So I made a deal. If she could explain me the reason for not selecting our campus for placement, I will not try any further. She agreed and told me that they were going to 20 institutes in her zone (which included Jaipur besides NCR), and she finds more students that meet her minimum quality level than she can recruit. So she can't go to 21st college. I suggested that she could raise her merit bar a bit higher, which possibly will help her company.


I also told her that the kind of job they do, the employees need to understand both basic CS and basic communication stuff. She agreed and told me that they recruit either CS people and make them go through a training in basic communication, or recruit ECE graduates and make them go through a training in basic CS stuff. But I was giving them students who knew both, thereby slashing the costs. I told her that only two institutes in India, to the best of my knowledge, had a curriculum which was a mix of both - LNMIIT and DAIICT, and it was surprising that they were not going to either of the two institutes.
How are you so sure that your quality is better than the places we go to, she asked me, since they would rather recruit quality people and train them rather than recruit people with the right type of courses on paper, but unable to perform in jobs.

I reminded her that a year earlier, they had done off-campus open recruitment of last years' graduates. 4 of LNMIIT graduates had applied, and all 4 were selected. I asked her, if there was any other college which had a 100 percent record in that recruitment exercise. She wouldn't reply to that. I then reminded her that 3 of those 4 students actually joined, and all three of them were adjudged good performers in the company, with 2 of them even getting awards for their performance. What more could she ask for. And, of course, she could come to our campus, or send someone else, and check out for herself the quality of infrastructure and anything else she wanted to know. We are only 3.5 hours from your office by car, and we would be happy to make all the arrangements.

She didn't know how to respond to this. So she said, but we have relationships with these 20 colleges for the last 5 years. We can't keep changing this list every year. I asked, you can't even add or drop one college in a year. What kind of quality process is this. But she did promise at the end that if in future they have to drop 1 of those 20 colleges, or their recruitment needs increase to a level that they need 21st college to be added, they will certainly consider us.

If this was just one example, I would have ignored it. But this happened in company after company. The technical guys visited us, talked to our students, would offer summer internships, and propose signing of MoUs, but as soon as we talked about campus placement, that was outside the domain of technical guys. In companies, where our student doing summer internships got awards for their work in the summer, and their supervisors made a strong pitch for hiring them next year, the recruitment section wouldn't listen. If you ask the company for parameters they used to evaluate campuses - these were company secrets. Absolutely no transparency anywhere in the whole recruitment process.

So, we had no problems in finding summer internships for our students - in all the top places in India and abroad. But campus placement was a much bigger challenge. And the problem was tougher because a large number of companies did not care about the quality. But at the end, it does not matter much to the students. There is enough demand that almost all of them will get placed in some company or the other. And their quality was good enough that they will rise faster than their peers. And if we looked at our alumni who graduated 3 years ago, they were doing well. But it did matter to the incoming students and their parents that many companies did not visit the campus, and many good students would not chose LNMIIT because of this reason.

I wish Indian industry will bring transparency to its campus placement process and start valuing quality. If they don't value quality, the colleges would not deliver quality. The colleges have at least learned this much from industry. You should deliver what your customers demand. For colleges, industry is the customer.

Friday, October 8, 2010

Commonwealth Games and Indian Education

We seem to do all important things at the last minute. The Commonwealth Games have been a recent example. We were cleaning up the games village, as athletes started arriving. One day before the opening ceremony, we weren't sure whether the public will be able to use metro to reach JLN Stadium. And once the ceremony is over, we all feel elated, and forget that this habit of doing things at the last minute cost us dear. We forget that sometimes tight deadlines will be missed as the Airport Link did this time. The T3 terminal of the Delhi airport opened with much fanfare in July but could not be ready for domestic traffic to shift there. And despite good sporting infrastructure in Delhi, we could not be a serious contender for Asiad 2018, because the rest of the world does not like things being done at the last minute. (And the central government saved us from embarrassment by not supporting the Delhi bid.)

The same culture is visible in education as well.Our semester starts around 1st August, and on 31st July, our admission process is still going on. In fact, no one in the country raises an eyebrow if the admission is done one month into the semester. On the contrary, if you close admission before the classes begin, people will file court cases against you. All our admission tests take place in April and May, with results being declared 4-6 weeks later. There is no possibility of completing the admission process before July.

We declare admissions open in colleges which exist only on paper. There is no faculty, no administration, no building, and sometimes we don't even know which city the college will be opened in, but we offer admissions, and we hope that like a Punjabi wedding, things will fall into place before the students actually arrive to study.

Even the post-graduate admissions, where the number of applicants is much smaller, are done at the last minute. If you look at the admission to technical programs (since I am more familiar with them), the candidates have to pass GATE. This is held in February. (Thank God, IITs do not have enough manpower and other resources to hold JEE and GATE together or very close to each other. Otherwise, it would have been held even later.) It is expected that students will give this exam in their 8th semester.

Now, there are serious repercussions of having GATE in the 8th semester. When I talk to others in academia, I get a sense that most people agree that to improve the quality of post-graduate students, we need to ensure that the student gives GATE before the campus placement season begins, and preferably, the admission offer should be made before the student gets the first job. All this would indicate that GATE should be held at the end of 6th semester, or at the beginning of 7th semester. Most of the basic courses in any discipline, in which GATE is supposed to evaluate the candidates are done in most colleges by 6th semester. So it is easier for students to give GATE early. And once the results are out, at least some enterprising institute will offer admissions immediately.

Now, if there is apparently no disadvantage of having GATE after 6th semester, and there are potentially lots of advantages of having GATE early, why don't we just do it. I have a theory for this. I think doing things at the last minute is exciting. It is a thrill. Punjabi weddings wouldn't be same if some event manager were to organize them and give a money-back guarantee against any last minute problem. There is not enough thrill in the lives of us academicians. And we are simple people. We get satisfied with simple thrills of managing admissions at the last minute.