Tuesday, 14 March 2017

"Allocation" of students

There has always been an issue that I've not been fully able to understand ever since I've joined IIT. The problem is how should the M.Tech and PhD students in the department be paired with faculty. Currently there is no standard procedure that is followed by all IITs, even within one IIT various departments have their own policies that might change with time.

The two most common ways of pairing are:
  1. Let the students come and spend a semester or two in the department, then let them choose a person with whom they'll work on their thesis.
  2. The Head or DPGC or some admin figure in the department pairs students with faculty
Ideally the first method looks pretty good. This is also what is followed in most American universities that you first secure admission to the department and then choose the advisor once you're here. This also accounts for the students' choices and seems quite fair.
However, a general feeling among faculty is that most students aren't mature enough to take an informed decision. They tend to choose supervisors based on what their seniors tell them and thus go with the 'older' professors with established labs. It's very unlikely the student will themselves initiate and choose a young faculty member who has joined last year and still is in the process of setting up the lab.

Thus, stating this reason some departments simply allot guides to the students taking into account how many students each faculty has while trying to maintain an equitable distribution. This method, however, largely ignores students' opinion and thus can be unpopular. I've seen conflicts occurring in the department when a student makes an application for change of supervisor. This also gives un-necessary power in the hands of administrators.

In a fund-crunch scenario when the number of incoming PhD students is a quarter (or even less) of the number of faculty in the department, this question of student allocation assumes a much larger significance. I'm still undecided on which method should be used or is there a better optimized third method.
Any suggestions by the readers will be helpful.


  1. lab rotation (as practiced in the USA)

  2. Hi Anon,

    It looks like a nice idea. I've seen this happening quite often in life science departments but not so much in engineering departments. Any thoughts on the reason?

  3. In USA, student funding comes from the PI's grants. In many engineering departments, professors have to make funding commitment to the admissions committee before the student is hired. Therefore, the student arrives with an advisor in place. The disadvantages of this method are obvious, but the advantage is that students don't struggle to find a lab that can fund them.

  4. In my Masters' (in an old-IIT), I was assigned a 'guide': my experience was terrible. After this, I was too scared to do a PhD. For my PhD, I worked with an assistant prof for a year before joining - it made a lot of sense to me. We both knew what to expect of each other and I enjoyed my PhD. It also streamlines the course work one needs to take as you are specific about your research area. As a student, I would always vouch for option 1. Regarding listening to seniors, if students are choosing supervisors based on anyone else's opinion, I just find it extremely immature. It is rarely a joke when PhD is likened to a marriage!