Search This Blog

Monday, October 1, 2012

JEE and admission to IIT, not branch

Many of my readers criticize me for focusing on Under-graduate programs. They get more ammunition with this post, since this is again about UG admissions :-) But seriously, the UG admission and the UG programs are totally broken in this country. PG education has a problem primarily because of poor-quality UG programs. There are lesser structural problems with PG education (and I have written about them at times).

The issue that I have been thinking about is whether it is a good idea to offer admission to a branch (like IITs and all engineering colleges do right now) or just offer admission to an IIT (I will focus on IITs), and the branch selection is done sometime later (say, after a year when they have all done some basic courses, and after taking into account their performance in these courses).

This idea is not new. I have heard of this almost every year in IITK. It was actually recommended to Senate in 2012 (where unexpectedly it was referred to another committee). Many people like the idea, many others don't. What do people feel will be the negative consequences of such a move. First, if IITK takes the leadership role and does this suo moto, without other IITs joining to do the same, then the top rankers in JEE (say, in the top 500 odd ranks) will prefer to join a specific program in another IIT then joining IIT Kanpur with uncertainty of the program that will be allotted next year. But on the other hand, students with 2000-3000 ranks may prefer to try to work hard in IITK and attempt to get a branch of their choice rather than get a guaranteed branch in which they have no interest. So, the chances are that we will have students in the narrower range of 500-3000 ranks, rather than 100-5000 ranks. (I am mentioning only ranks of unreserved seats, but similar thing should happen in reserved seats as well.) Interestingly, it means that the average "quality" (if JEE measures quality) will be no different if we do branch-less admissions. However, the problem is that the popular programs may want to have those students in the 0-500 ranks. On the other hand, the less popular programs would be afraid that while they will attract better ranked student, but they will get only those students who have switched themselves off after coming to IITK (since performance in the first year will matter). And, of course, as we have said earlier, students and parents would prefer to know which program they are getting admission in. So there is really no constituency for branch-less admissions.

The other reason that people say against branch-less admissions is that it will make the first year extremely competitive , and the stress period for the student will just get extended by one more year.

Is there any advantage for branch-less admissions. Actually, I believe that if handled properly, it can reduce stress, particularly if many institutions adopt this model of admission. As I have said before, the stress in JEE comes from the fact that a small mistake can affect your future very seriously. And branch-less admission is essentially telling the student that a small mistake will not affect the future
seriously, since you could work harder later and improve your chances of a branch closer to your interest. So instead of JEE being a pressure cooker, there will be lesser stress but spread over a longer period of time. And having lower stress for a longer duration is better for the mental health than the other way round.

But even more importantly, branch-less admissions are also more fair. During the debates on changes to JEE, those opposed to changes in JEE pointed out that there is a wide error margin in the 12th class marks, and therefore its use in ranking is unfair. Well, in JEE, the error margin may not be that wide (since the error on part of grading is not there), but still it is there. And indeed, even when two students get exactly the same total, we artificially try to rank them by saying one subject is more important than the other, etc. A single exam will always have error margin (and that is why exams like GRE give scores which are bunched in multiple of 10s). How fair is it to rank students who are within the error margin of each other, and based on that rank give them a branch for the rest of his/her life.

When we have a situation where no stake holder is interested in changing the system, and yet, it seems that the current system has problems which can be addressed by those changes, the right strategy is to change very little, show that it helps a little, then make a bit more change, and so on. One way to do that will be to say that while we admit students on the basis of JEE scores (or ranks), the final branch allocation at the end of the year will be on the basis of sum of JEE scores and some measure of the first year performance. In the beginning, the measure of the first year performance can be kept very small, say making a difference to the original JEE score by no more than a couple of percent marks. This will mean that if we rank all branches by way of popularity, then the branch that a student will get after the first year will be either same as the one s/he would have got at the time of admission, or will get a branch which is one higher or one lower in that ranking of popularity. Indeed, if one performs around the average of the batch, then s/he will get exactly the same branch after one year as s/he would have got in the beginning.

Once we do our minor tinkering with the admission process, and we do not see any major problem, we can increase the weight of the first year performance slightly.


Neo said...

sir , i doubt the fruitfulness of such a setup .

i) one year in the college will result in better decision making regarding the branch - but truly speaking , if you conduct a survey of the number of students interested in CSE(looking at the current scenario) at the time of admission and after one year ,it will only increment.
although this may not be the case for around 2-3 % of the students. but majority will definitely opt for the popular branch over their interest(most of them have none)

ii) reducing stress in JEE - As you suggested that initially less weight-age will be given to CPI, this will make 1st year really stressful for CSE-EE students with AIRs(300-400),EE-ME(500-600) and so on.

And on a final note , most of us will agree that JEE gives a better insight into 'quality' than CPI of first year since the CPI includes many irrelevant courses(HSS,TA 101 for many branches,LIF-101 for almost everyone and many more) and ambiguity in grading(10-8-6-4-0),especially in IIT Kanpur.

Ranjan Kumar said...

This business of ranks influencing branch selection is seriously flawed ab initio. The key issue is to match interests of students with the branch where they are likely to succeed best. In reality, students coming in have very limited feel of engineering. They may have better feel of sciences and social sciences. It is up to the institute to shape their thinking to see if they may get inspired into taking up the right engineering discipline. Taking a call on the basis of 'performance' in first year remains flawed as most courses are still sciences. Institutes should devise ways of judging, inspiring, counselling in helping the student pick his heart's calling in a conducive and informed setting. One way of doing this is by letting practitioners (alumni, faculty, and others) counsel/inspire, another is by making courses very flexibly available across disciplines that may lead to niche set of specializations. Degrees may state perhaps uptp two core specialization if the minimum criteria for those are met. I am not too big on this performance business being some determinant as the scale of opportunities and challenges in life after graduation matter most and should help determine right combinations. Rigid departmental selections are equally restrictive and some flexibility there would help.

Alabhya said...


The problem with going for a branchless admission format is that people come to IITs for a promise of good paying jobs. Now I am not talking about everyone but this is the story of majority of students who come to IITs. Sad as it may be hardly 5% (and I personally think that it is an over-estimation) of the students have a desire to actually become engineers.
If given a choice most would opt for CSE or EE, because those departments offer better job opportunities. Hardly anyone would select Chemical/Civil/Material Science out of interest. Also if branch allocation is done on the basis of first year's performance it would mean that poor performers would be thrown into these "low job opportunity" departments as a result their performance would further fall!
This system you propose reminds me of my school days when every new term students with >80% in the previous year were put in section A, those with >60% in section B and remaining in section C. The idea was to motivate students to get promoted to better sections and those in better sections to keep doing better or get demoted. (Un)surprisingly hardly anyone ever got promoted or demoted.

Dheeraj Sanghi said...

@Neo, @Abhaya, The suggestion is to tinker the ranking only marginally in the beginning, and see if works better, and if yes, then only move to giving more weight to the first year performance. So someone who could get CHE based on JEE rank in the current system, and does not want to work extra hard to try to get ME, can still get CHE if he remains an average student. And if there is someone who is interested in a branch which is considered unpopular at his/her rank, then even poor performance in the first year will ensure that that branch is available. So, in the first year of its implementation (if it ever does get implemented by any IIT) the proposal is to only add a maximum of 2-3% marks based on the first year performance, and then do ranking of students. What this does is to give a small extra chance to someone who did a couple of questions wrong in JEE by mistake.

As far as JEE being a better predictor of quality than first year CPI, I disagree, for one. But it does not matter, since we are only talking about students who have a very similar score in JEE to be differentiated by way of first year CPI. Since JEE has not been able to differentiate them, we have two options. The current implementation will argue that getting 51 in Physics and 49 in Maths is higher quality than 51 in Maths and 49 in Physics, and we rank them accordingly. The proposed option is that we consider both of them as equal for the first year, and rank them based on first year performance. However, someone getting 70 in Physics and 70 in Maths will be considered higher rank irrespective of first year performance.

prishu said...
This comment has been removed by the author.
Dheeraj Sanghi said...

@Ranjan, I agree with you, and selection of branch after the first year solves it to some extent. The mindless following of herd is largely because of parental pressure, societal pressure and because of advise by coaching places. Once a student is in IIT for a year, it is possible to organize the lectures by practitioners, and counsel them more thoroughly. I have done this experiment with one batch a few years ago, where we did counseling of the batch before the branch change, and the results were amazing. The only year in my 20 years at IIT Kanpur, when there were vacancies in Electrical Engineering department, and the first time till then when there were no vacancies in Maths. Also, the departments are more likely to be pro-active in counseling the students in such a scenario. Branch change does not solve the problem fully, since there is a default option, which most will take, that is to continue in their current branch.

prishu said...

This could well be a blog post in its own right. Nevertheless I am posting it here as a comment to convey my views.(About myself, I am currently a 3rd year undergrad at IITK.)

Why do we need a system of "competitive branch allocation" at all? The problem is to ensure that incoming students are matched to a stream that interests them. Even if the allocation is done after 1st year we are inclined on using grades/ranks as the criterion. This is not going to serve any purpose as most students would still end up not getting a branch of their choice.

Instead, the way I would like the system to evolve is by allowing students to "freely choose any branch", say after the 1st sem.I have a firm conviction that this "free choice" will lead to a healthier atmosphere.

We should try to identify why students currently tend to flock towards the popular branches. The two main reasons are:
1. Safe Choice: choosing a less popular branch may lead to fewer opportunities. Plus, changing from a more popular branch to a less popular branch
later on is easir than vice-versa. Now this problem can be solved by allowing a "free choice", meaning that a student can change his/her branch whenever s/he wants
and any number of times during the first 2 years. Consider a case of a student having to choose between CS and Civil, who is interested initially in Civil. Now
during the first year the student would not be afraid to choose Civil because s/he knows that s/he would be allowed to change to any branch later on as per his/her
choice. So there is a "sense of security". The students have greater faith in the system.

2. Placement Figures: This is more a case of societal/parental and peer pressures. This problem is difficult to solve. It requires a change of mentality on part of
the students, the society as well as the academic community. Still if students are allowed to go by interests rather than by ranks/grades, it will result in better
performance for every student.

Consider the situation where this "free choice system" is actually implemented. Maybe initially we shall have a majority of students flocking to popular branches like CS, EE etc. But the sense of security would create a better atmosphere, and save
the students would at least be saved from the heartburn of getting into a wrong branch, which is currently the case with a lot of students.
Even though the departmental constraints may have to be stretched a bit initially, there is a free choice so students would not hesitate in exploring their interests and getting into an area of choice, instead of brooding over what has been thrust upon them. And over time, after a few rounds of this experiment I expect that all department strengths will converge towards an equitable distribution. Further this would lead to a distribution of equally good students in every stream, reinforcing the placement cycle and similar figures for each stream. And this is all due to the fact that students are studying what interests them. The best thing being that nothing is forced upon anyone.

prishu said...

(This is a follow up to my previous comment. It could not fit into one because of the size limit.)

This idea of a "free choice system" may appear to be far-fetched to a lot of people. But it is certainly not so. Top universities around the world do not impose any major upon their students. They provide proper counseling mechanisms to their students, helping them to identify a suitable major during the initial undergrad years.

While there may be a lot of doubts from the feasibility angle, it is essential for a better future for Indian students. We need a process of societal mentality change for this system to be successful. Again, we should not backtrack thinking that our society is not yet ready for it.
We have to make a start somewhere. It is needed to kick start the cycle at some point of time, and eventually the cycle will begin to reinforce itself.
We need to be convinced that this experiment is worth doing. Changes need to be implemented slowly but surely.
And who better to be a leader of this change than our own IITK?, that has been leading and showing the way throughout its history.

Ranjan Kumar said...

Dheeraj, my suggestion is to get some alumni to counsel as well. In this day and age it is not difficult and it would enrich perspectives from beyond the boundaries of IIT/K. Successful alumni can serve as great magnets. Enlisting services of alumni is a common practice in good universities. I know you think it is difficult to mobilize alumni seriously but I think it is critical for IIT/K to systematically move in this direction. IIT/B has an active program where graduating students are pt in touch with seniors in their target companies/universities from the institute for mentoring.

Dheeraj Sanghi said...

@Ranjan, What makes you think that I find mobilizing alumni to be difficult. In fact, it is a very easy thing to do. I made use of my connections in alumni network very effectively when I was Director of LNMIIT. When said that it is possible to organize lectures by practitioners, I meant mostly alumni.

Ungrateful Alive said...

Is there any need to reinvent the wheel? The business of declaring a major is a well-tuned machine in colleges worldwide. IITs have finite capacity. Each department has a finite capacity. JEE is an imperfect but usable input to IIT admission. First k semesters' performance is an imperfect but usable input to a department accepting a student for major. There's nothing technically interesting about the setup, so we can just move on.

Giri@iisc said...

In IISc, we allot a branch to the BS students only after the third semester. Initially, this was not opposed but people have now come to realize its usefulness.

When the initial batch joined, more than 60% wanted Physics but after two semesters, the interest seems to have changed and it is become equitable with students expressing interest in other branches.

Of course, IISc also allows two majors, one major with two minors etc allowing students to choose what they want.

Best regards,


zizou said...

i) I think the problem can be solved by having a five year undergraduate degree AND having a second counseling at the end of the fourth semester. this way students will get an in depth exposure to the various branches.Right now i think the problem is that students do not get enough exposure by the end of the first year so they are unprepared by the end of the first year to decide their branch.

ii) Also for the first two years all grades should be pass/fail.this way students are not under any pressure and they will try to EXPLORE the subjects instead of just thinking of the exams as something routine and tedious.

iii)The first two years can also function as an that i mean after the end of second year most of the educational disparities between students will be removed. also two years will be enough time to improve ones English skills.

iv) By the second year all students will have finished engineering mathematics and will have deeper exposure to all branches.This will also be helpful because it is very difficult to teach concurrent courses In indian universities as the classes are never coordinated properly.In my university in the third sem we were supposed to take three different subjects that were interdependent and because of a lack of coordination the weaker students really struggled.

iv)I also think indian students are not as mature and independent as the students in united states so they should be given an additional year to select the appropriate branch.

Ungrateful Alive said...

First two years pass/fail is not desirable. Among other problems, it will demoralize the high fliers. Unlike JEE, evaluation in IIXs is fairly continuous and insensitive to toothaches on final exam day. As an employer, I am certainly interested in knowing quite closely how an IIX student conducted his/her life during the first two years.

Prashant said...

One method could be this -

1) Have a liberal branch change regime. Be prepared to expand branches by 20-30% or so.

2) Yes, some branches are terribly unpopular and this will not change.
Might as well warn a student that he is going to be stuck with Civil/Metallurgy/Textile well before he steps in, so that he knows what he is stepping into, and exercises whatever options he has judiciously. Otherwise, if some 18 year old kid gets stuck with these branches after the first year, and can't even leave for some other institute, I won't be surprised if the suicide rate shoots up.

3)IIT needs to have a serious discussion about which branches should have how many seats. Resources should be allocated to departments to expand, in proportion to their demand ( well, somewhat ). Given that the demand is for branches like CS and EE where a lot can be accomplished via software (except for power electrical engineering) you might want to explore a Coursera like system, with credits, to cater to the hundreds of students who desire to study those subjects. Think of using good undergrads as TAs. IIIT-Hyd seems to handle a batch size of 150 students per year in Computer Science. Surely some creative solutions could be used over here.

3) Have unrestricted entry to a general Engineering Sciences program where students can mix and match courses across couple of branches, pure science, mathematics and design. Even if they don't always get exactly the courses they like, there is a much higher probability that they might actually discover something they like. Much better than being forced deep down into metallurgy or textile for four years against one's wishes.

And seriously, some departments like Textile and Agriculture and Mining could be shut down at least for the undergraduates, and those resources could be transferred to other departments.

4) I am grateful to the professors at KGP in this regard. I was AIR ~2000, did not have a spectacular GPA; but the professors in Comp Science were always very encouraging and I did a whole lot of CS courses with labs(with credit, not audit ). At the end of my stay in IIT, I felt I had done a fair bit of coursework which I was interested in. I have heard that things are very different at IIT Kanpur and there is so much rigidity that even a Mathematics department student has to battle to enroll for electives in the Comp Science department.But I don't know how much these anecdotal tales can be relied on.

Ungrateful Alive said...

Parents of JEE aspirants in an undeveloped, floundering society, who are mostly illiterate about the beauty in all engineering, should be counseled right at the point they send their child to Kota or Pace or whatever, to sensitize the family to the fact that, even as, say, Civil Engineering remains unglamorous, our buildings crumble within a decade while buildings in civilized societies are built to last a century; our roads are a national shame; our trains move at one-quarter of the typical speed in Europe, etc. This country is on the verge of ruin, and only massive indoctrination can save it. Once Indians do what they (learn to) like and not everyone has an "angle" to a job (stepping stone, money, prestige, ...), there will be little need to doctor IIT major policies etc. Hand in hand, India also needs to protect its currency, and make sure the pricing is right: why are Indians happy to spend 35kINR over a silly smart phone upgrade with negligible marginal benefits, while letting the road in front of their house rot, and curse daily as their cars hit the potholes?

Dheeraj Sanghi said...

@Prashant, Great ideas. There are certainly multiple options to achieve the same goal. But all options face resistance to change. And hence multiple options are necessary, who knows what might get accepted by the Institute.

You are absolutely right that we need to find solutions to enable and support large classes in popular programs. But for doing this, certain incentives are necessary in the system. A system which treats everyone as equal will have very little reason for change. In fact, we had asked for opinions of various departments on the admission numbers. There was no surprise. Departments with better faculty-student ratio oppose any restructuring. Departments with poor faculty-student ratio cite their excessive teaching load to demand reduction of student intake. And you have to take every decision by consensus in an academic institution. In such a situation, one has to come up with an incentive mechanism, which encourages departments to agree to higher class sizes. May be some part of the budget can be linked to enrollment.

IIT Kanpur has just approved two Engineering Science programs with a maximum of 10 seats each (though they are not general ESc programs which would have been great, but inter-disciplinary programs at the UG level).

In terms of a non-CSE student being able to do CSE courses, we used to be the only department offering some of the basic courses as part of minor program for outside the department students, but that stopped a little over 10 years ago. We are again likely to start a minor program very soon, probably this year itself, but yes, for about 10 years, the number of non-CSE students in CSE courses have been very low at IITK. But as I said above, it is difficult to convince someone who is already having amongst the largest classes on campus to take a few more without any incentives what so ever.

chitta said...

The IISERs at Kolkata, Pune, Bhopal, Mohali, and Trivandrum and NISER Bhubaneswar currently take in 5 yr integrated M.Sc students without pre-assigning branches/majors, which are allocated later. They currently have Physics, Chemistry, Maths and Biology. [Kolkata also has Earth Science. Pune does not seem to allocate branches at all.] All of them plan to have Computer Science and Engineering Science in the near future. It will be interesting to see how things go when these two branches are introduced.

Umesh said...

I support Not having any branches at all in IITs. Students should be free to mix and match courses from multiple branches and they will get major/minor as per the number of requisite courses. I am from MME in IIT Kanpur and NO CSE professor will ever allow me to even audit the Data Structure course. They want a very high CPI for you to register in the course as open elective and even then you will have a hard time registering for it.
Only, one of our senior was able to register for it and he was turned down thrice. In pre-registration, during registration and ultimately allowed after multiple rounds to professor. No one can dispute that Data structure is the most basic course and anyone demonstrating a good grasp of it stands a good chance to get in technology companies. But to our CSE professor the pride mattered more.
Would Dr. Sanghi have these opinion if you were a professor of Civil or MSE. The country needs Civil and MSE engineers as much as the CSE guys. IITs are primarily funded by Indian Government and I do not see a reason why CSE should become the most dominant of branches. We have IIITs for that who output only CSE students and we have private colleges mushrooming all over India to cater for those people who want to study only CSE/EE. I will rather ask IITK CSE/EE department to be more open in accommodating students from other departments so that they can be prepared for all walks of life. Then may be we will see the power of technology applied to all sectors of India rather than producing software for just US MNCs.
My two cents.

Umesh said...

Continuing my previous comment. IITK CSE department contributes only one course to whole UG curriculum(Except for EE). The CSE student's get to do courses from all departments and that prepares them to be a better engineers in real life, able to build software for all sectors. But IITK CSE department is an island of exclusivity when it comes to contributing to other departments, calling them non glamorous departments. They are very happy to grab the best students of other department based on CPI but will quickly disown a lower CPI student even if he has huge interest in learning CS subjects. IITK CSE's best offering is a Summer Certificate Course in Data structure taught by final year CS students and they charged Rs 2000 (back in 2002). What you is a certificate to show and very rarely actual learning. I did that course, learnt nothing and yet got a satisfactory grade in that Certificate. Easy money for final year CS guys without really contributing much.

Dheeraj Sanghi said...

@Umesh, I am sure you have no idea that every year, the number of non-department students handled by any engineering department in the Institute, CSE department has the maximum numbers. And this is true whether you consider all courses (including core, that is), or just consider electives. Of course, if you include non-engineering departments, then HSS handles the maximum out-of-department students, followed by Maths, then Physics, and Chemistry.

You will not also know that most of the electives that CSE students do are in non-engineering departments.

I have never been associated with that summer course, but I wonder why you did that course. Didn't you ask your seniors whether they learnt anything. Every time, a batch of graduating students have decided to call it off, there is tremendous pressure on them to continue, and their remuneration from the course is enough to pay for their food bill, and actually the course is subsidized.

But of course, we live in a free country, and everyone has a right to criticize without knowing the statistics.

Umesh said...

@Dheeraj Sanghi :
My comments are based on personal experience and specific to MME. I scored a A in Java in ESC101 in very 1st semester. In fact I had one of 13 As, but CPI wasn't good enough for registering in an elective course. It was futile to ask also.
The only senior from MME who was allowed to do Data Structures mentioned in last post is Mayank Goel. After that he did 2 more CSE courses as open electives and got a minor degree in CSE. Now he is founder of and has been role model for many of MME Students.
I did pretty good in Software once out of B.Tech. I was in a startup for 3.5 yrs and was one of key people. Of course, I couldn't clear interviews in Microsoft/Amazon even with 4 yrs of experience because they focused on DS & Algo. I then learned DS & Algo from Online courses CS 61B (Berkley), Stanford's CS107, Berkley's OS course, MIT's 6.006: Introduction to Algorithms. I found their Board work specially helpful in Video courses, which is something that was missing in IIT videos. Within 6 months, I cleared interviews for Amazon & Microsoft and now I am with Flipkart, another technology company.
I sometimes ask, if IITK CSE had offered the DS & Algo course to whole of UG students, wouldn't that have made me a better Engineer and I agree. As a MME student, I was doing a project in Modelling of Steelmaking and was unable to complete it. Now looking back, I realize my DS & Algo skills were poor and that was primary reason for not being able to code. It was same in startup that I worked for 3 plus years. I did code, but DS & Algo would have really helped me in writing far better code.
So, while some professors may allow some student's to register for open elective. Is IITK CSE as a department doing enough for all UG students get the benefits. I feel nopes. I strongly feel that a course in Data Structure and Algorithms should be at least offered as Science Elective and open to all, if it can't be made compulsory in curriculum.

PS: I really like your blogs and the blogs have some real depth. However, we all look at the world from our LENS of own experiences. I have seen and experienced a different side of coin and hence the comments. My opinion is not because it is a free democracy and I should write just because I can write.

Umesh said...

@Dr. Sanghi :
I did that course, because I was grabbing at every opportunity that presented me to grab CSE knowledge. I was passionate about learning programming and DS & Algo was the stepping stone to CS world. I had got a really bad CPI in 2nd semester and so there was no way, I would have been allowed to register for a CS course. Of course, the final year guys weren't able to explain the concepts so well, so I ended up not learning anything. And that was a big set back. I thought, may be DS & Algo is just too tough. Of course, I take full responsibility of making the conclusion and may be I could have worked harder.
PS: My experiences by definition is colored. I see from my LENS and mine is just one data point. Others experiences might have been different.

Dheeraj Sanghi said...

@Umesh, You are right that we look at from different lenses. The difference is that while in my position I get so many complaints, suggestions, etc., that I can understand the other viewpoint, but your comments made me feel that you perhaps don't understand the other viewpoint.

This semester, for example, the average student registration per faculty in the Institute is 60, while for CSE department, it is 75. 25% extra load is not a joke, and as I said in my previous comment, it has been this way for as long as I have been in the Institute.

So, one way to look at the situation is that CSE faculty allows maximum number of non-department students of all engineering departments. The other way to look at the situation is that CSE faculty rejects the maximum number of non-department students of all engineering departments.

The second statement is correct (and hence you are technically correct), but without the first statement, it gives a completely wrong picture (and you did use those words - island of exclusivity, the pride matter more, etc.) And I get this feeling that even with these statistics that I am giving, you are not going to be convinced :-)

And my point about proper incentive mechanisms in an earlier comment was that if there was some incentive for teaching more, while CSE may not be able to offer more "seats" than it currently does, but it is definitely conceivable that some other faculty member in the Institute will come forward and volunteer to teach the first year programming course, and that reduction in load can be used to offer Data Structures and other courses to a much larger number of non-CSE students than what happens currently.

Umesh said...

@Prof Sanghi
Thanks for bringing the per faculty registration statistics. That is a figure I wasn't aware of and that does affect my perception. I have read your blog on challenges of teaching 500 students in a class and can relate to it. My own experience in job has taught me that it is really hard to learn programming unless you implement it, endorsed by by attempts to teach my younger brother and mentor some juniors at company. So I do appreciate your concern.
I am not sure about the solution. Your suggestion of bringing in an external faculty from Industry or external universities is really a great idea and implementable also (if IITK agrees of course). The industry does add a lot of value. Guest lectures and courses are something that is already implemented in institutes in Metro. R&D Head of my first company, an IITM alumunus, had maintained close connections with IITM and even offered a full elective course in SRM University. I am positive that IITK with its brand value can find one or two visiting faculty and good ones from Industry.
I feel a really good way to offer such courses will be through Science electives, but someone at institute will be able to better know how to best offer it.
PS: Maths department does offer some course (like Combinatorial Optimization) but stress is not on Implementation/program and that makes a lot of difference.

Prashant said...

Now somewhat off-topic, but speaking of Admissions, have you taken a look at the 20 percentile cut-offs across boards ? Your prediction about students of better boards being disadvantaged is absolutely true.

The cut-off for state boards in mostly in the sixties, (even when not all of them are all that difficult) whereas it is near / above 80 percent for CBSE/ICSE (82% for ICSE if I'm not mistaken). If people are so concerned about school education, why have they started off with a strong disincentive for students to study in schools of better boards ! I hope this matter gets some more thought next year.

If cut-offs spiral north every year, class 12 pressure might really end up sending even more burnt out students who'd be all the more likely to switch off early.

Umesh said...

CBSE and ICSE boards are not affordable to most of people who go to State Boards. The school fees and Board Fees are too high, the medium of education also doesn't make it easier. Also, the number of CBSE/ICSE affiliated schools and their reach is limited in cities or small towns only. They do not cater to rural background and definitely not cater to student's from socially and economically underprivileged background. Also, it will be incorrect to judge the merit or intelligence of a student based solely on Marks in School/Board Conducted exams. Many of the student's in State Board help in family business or they make their own arrangements for finances .. The teacher quality is not good and most of government schools have 1/3rd of the sactioned staff strength and even less. In one word, State Boards student's in most cases are already lagging behind CBSE/ICSE in terms of infrastructure and faculty availability.
I agree that 80 percentile does make CBSE/ICSE board a lot more competitive and I will not dispute that. I can see a side effect of this that some students from better background will start going to State Boards and general quality of State Board Education will increase. I see 80 percentile as an incentive to send your kids to state board but only if you want to send them to IITs .. Now a days, India has enough opportunities in terms of Education that you will still send your kids to CBSE/ICSE.
Also, somewhere I read, Empirically it has been observed that most of student who clear IITJEE also fall in 80 percentile bucket .. Anyways, it will improve the school education also, because it will reduce the importance of coaching.

PULKIT said...

If IITs opt for admission to IIT and not branch then I see a potential problem in this system. In recent JEE, out of top 1000:
IITB: 285
IITD: 224
IITK: 166
IITM: 138
Others: 51

Clearly, in recent years JEE aspirants have given utmost preference to IITB. This year 86 out of top 100 had IITB as first choice and 12 went for IITD as top priority. (98%). IITB and IITD total have 847 seats. This can be clearly assumed that no seat in these IITs would be available after a rank of, say, 1200 and while IITK, KGP, M have a fair amount of students in top 1000, the number would decrease significantly with each having about only 50 to 75 top in top 1000. So branches like Chemistry IITB would get students near 1200 and even Mechanical KGP would be surely after 1500 in this scenario.

Similarly old 5 IITs in total have 2391 seats. Anyone would opt for these 5 IITs before rest 11 IITs (R+G+BHU+8 new). So these 2391 seats would not be available after a rank, say, 3000. Now IITR, IITG or IITBHU which this year had a significant number of students in top 1700 would now hardly have any population above 2500.

So this would in a way lead to sudden supremacy of few IITs and sudden deterioration of few others. It will create tiers like B,D / K,KGP,M / R,G,BHU / New ones. In fact the 16 IITs have 4251 GE seats in total. ISM would not get a single student till all IITs fill up in this case.

Further if only IITs have this system and not NITs, better brains would flock to NITs/IIITs etc where they atleast get a confirmed branch. This would completely devastate the IIT system depriving us of the creme of the nation.

Further the sort of studies we have in first year somehow encourage mugging practices. And if JEE is not the right way to assign branches then first year CG is out of question.

If it is proposed that students are allowed to choose any IIT and any branch after first year, then again this is not feasible because just like Boards can't be compared, we can't compare marking of different IITs and just like JEE+Board combination cant work, similarly JEE+CGPA also cant work.

In fact what we must do is to have an aptitude test much before JEE say in November. The aptitude test may be compulsory and would suggest the potential interest of student. In the end we cant force a student to take a branch. Even if the branch is allotted after first year the top pointers would rush for CSE,EEE only.

The IITs also need to expand more with more courses of interest and more inter branch flexibility. For example I have interest in Railways but I cant see a course where I can pursue my interest at large.

Vijayant Singh said...

Good point, only thing to add to this is... Either make the movement of students between intra-IIT branches a norm rather than exception, or, introduce multiple degrees all across so that the guy who had to do with M&M and couldn't get Particle Physics can still earn those two BS degrees by spending an extra year.

Scully said...

Great Blog Post! Interesting Comments.
I got an A* in ESc101. I know it is not a big achievement, I know that many students have no idea about programming before they come here but the number of those who do is not very low,its considerable..
I have been reading a lot about Computer Science for 2-3 years now, studied some from MIT OCW ,etc.
I think that given a chance I would be at least in the top 15% in CSE department.
But No.I just don't deserve a chance.Just because I fell short of getting an A in a few irrelevant courses in 2 semesters,I am stuck with a department I have no interest in.
I am interested in HSS courses very much. Is there a way to take a course even if one does not have it in her template for that semester?