Proposal for Lateral Admission into IITs

Last couple of days, I was at Indian School of Business, where I was attending a program on "Strategies for Building Excellence in Higher Education."
Prof. Rafiq Dossani from Stanford University was the faculty for the program.

There were many interesting things that I learnt, which I may write about later. One thing that really surprised me was the information that state universities in US admit a large number of lateral or transfer students. And we are talking about the top schools only. In fact, the example that Prof. Dossani gave was about University of California, Berkeley.

Apparently, about 30 to 40 percent of UCB graduates (those who graduate with a BS degree) are those who are admitted as transfer students. A very large percentage of these transfer students are those who have done two years of college from local community colleges within the state of California. I did my own google searches, and found out that indeed this is true.

Since very many readers of this blog may not be aware of the state (government) system of higher education, let me explain that a bit. The state governments typically have three types of higher education institutions. First are the flagship state universities like UC Berkeley, UC Los Angeles, and so on. Second are state universities which are more focused on teaching,though research still takes place, and at the lowest level are the community colleges, which provide 2-year college education, directed towards either vocational training, or one could study a variety of subjects just out of interest, or one could take credits which will help in lateral entry into bigger universities after some minimum number of credits. So, there is generally, significant difference in the quality of infrastructure as well as quality of faculty between community colleges and the flagship universities. Community colleges focus on teaching alone.

As I said earlier, UCB gets 30-40% of its class through lateral admission, most of them are those who have performed exceedingly well in the community colleges. If one gets a CGPA of 3.6 (out of 4.0) in the community college, one is guaranteed admission in UC Berkeley, and all relevant credits are transferred so that the student can finish the degree program at UCB in two years.

What is interesting is that this has huge benefits for all stake holders, including UC Berkeley, Community colleges, and students.

The student wins in obvious ways. Those who could not get into UCB based on their profile at the end of high school get another chance to get into a top rated school. That too at a fraction of the cost. The tuition at Community Colleges is a fraction of the tuition at UCB. Also, University tuition for undergraduate is typically semester based, while the tuition at community colleges is typically credit based. So those who are doing part time work and part-time study, the tuition differential between the two systems will be really huge.

Community colleges win since if their students have a chance to get into UCB, they attract reasonably good students too, and therefore, they are able to maintain a decent quality of instruction, since a significant amount of learning is based on peer learning, which improves when the quality of peers is better.

Interestingly, this has significant benefits for UCB also. No admission process can really predict who will be the best performers in the 4-year of education and beyond. Invariably, some excellent students will not make it your university because your admission process could not recognize them as good enough. This is, of course, worse if your admission process is based on a single exam, and that too on a single day.

If a university is committed to excellence, it must find ways to admit those excellent students whom the initial admission process somehow missed. And UC Berkeley is doing exactly that. It has been found out that the first two years of college are better predictors of one's success in the remaining two years in college and beyond, compared to performance in high school, SAT score, and everything else one is able to consider at that time.

Of course, the quality of teaching at community colleges is inferior to the quality of teaching at UCB. But what they have noticed (and has been corroborated by large number of studies in different places) is that the toppers of even much inferior systems perform better than weaker students in much better systems. In fact, Berkeley has been doing this for a long time, and have detailed data about the performance of these lateral students. They have found out that the average performance of these students is same as the average performance of those who were admitted right after the high school. (This is true for the current mix of students. If the number of fresh admissions go up and the lateral admissions go down, then the average performance of lateral students will become better, and on the other hand, if the number of fresh admissions go down and the lateral admissions go up, then the average performance of fresh students will become better.)

The system is so well established and is so student friendly that now they have started noticing that some students who could get admission in UG program at Berkeley after high school chose to join community college (mainly to save money), and then get into Berkeley directly into junior (3rd) year.

That, of course, set me thinking, whether it would be in the interest of IITs to do the same here. We all know that our admission system, based on a single exam, on a single day, misses out on a large number of excellent students. If we could have a scheme by which we agree to admit the top 5% of the NIT students and offering them credit transfer for the relevant courses that they may have done in their respective NITs, we will have very significant advantages.

We will get students who perform better than the lower end of our current students. We will reduce stress in the society because now there is a second chance to get into the elite institutions. We will be more fair in the sense that we will be admitting students who have demonstrated excellence over a period of time, and not just on that one exam. We will improve the quality of NITs, since even though we take a few of their top students, overall they get much better students, since they all will be hoping to get into an IIT through this route. We enable government institutes (NITs) to compete with top private institutes like BITS Pilani in this fashion.

We will also reduce the number of students who give JEE twice. It might be easier for students to join an NIT, do well there to get into an IIT, than to spend one more year at Kota and play the lottery called JEE once again. And that reduces the market size of coaching too.

Also, the IIT review committees have repeatedly asked IITs to support other government institutes nearby. If news reports are to be believed, this will be included as one of the obligations of IITs when the IIT Amendment Act is introduced in parliament this month. This scheme allows us to fulfill what is likely to become our legal obligation very soon.

This will also solve the issue of having large classes for the compulsory core courses, which are taken by students of all disciplines. Normally, most institutions have a bottleneck in terms of common infrastructure, common labs, instructors for those large classes, etc. Departments generally are able to handle larger classes (since they are still a fraction of what the common core courses see). This model, therefore, makes it easier to admit more students. You don't need to offer Physics and Maths and Workshop Practice to them.

Of course, one must still come up with a detailed implementation plan. How will we allocate programs to students. How will we compare a student of one NIT to a student of another NIT. And, of course, this implementation plan must not have any subjective evaluation since I don't think that the IIT system is as yet strong enough to withstand pressures that a subjective admission process will bring. But I am convinced that if conceptually such a plan is acceptable to the stake holders of IITs, then coming up with a detailed plan would not be an impossible task.

Can anyone see a negative fall out of this?