Choosing a Research Supervisor — How to Judge a Group Before You Join
Students spend a year preparing for the exam that gets them into a research programme, and then about twenty minutes choosing the person they will work with for the next four to six years. That ratio is backwards. The supervisor sets your daily working life, the problem you solve, the equipment you can reach, how quickly you publish and whether you finish at all — far more than the institute's name on the certificate does.
This article is about how to gather real evidence on a group before you commit, what to ask, and which warning signs are worth acting on. It names no institutions and quotes no fellowship figures; those belong in the official notification and each institute's own advertisement.
What you are actually choosing
All four have to hold. Fascinating research in a group where nobody finishes is a bad choice. A wonderful supervisor with no access to the instrument your project requires is a bad choice. Students usually check only the first item — and often only from a one-line description of the group's area.
1. Research fit — check the last three years, not the webpage
Departmental profile pages go stale. What a group is genuinely doing right now is visible in its recent output. Read the titles and abstracts of its work from the last two or three years and ask three questions.
Is the group still working in the area that attracted you, or has it moved on? Research directions shift, and joining for a topic the group finished four years ago is a common and avoidable disappointment.
What techniques recur? The methods that appear repeatedly are the methods you will spend your days doing. If every project is synthesis and purification, that is your life, whatever the title of the area says. If you dislike that work, no amount of interest in the topic will carry you through year three.
Who are the first authors? If the group's students appear as first authors, students there are being given real projects and are getting credit for them.
2. Group health — the evidence that is hardest to fake
Ask about completion, not just about output. The most important question about any research group is whether its students finish, and roughly in what time. A group that produces excellent work but where students take far longer than the department's norm, or leave without completing, is telling you something.
Look also at group size relative to the supervisor's availability. A very large group around one person means less individual attention, though it often means more help from senior students — which can be excellent, if the seniors are treated well. A very small or brand-new group means much more direct access to the supervisor and usually more of your time spent building things from scratch. Neither is wrong. Know which you are joining.
3. Supervision style — and knowing your own
The single most useful piece of self-knowledge here is whether you work better with structure or with freedom. Both styles produce good scientists, and a mismatch is miserable for both people.
| Style | What it looks like | Suits you if | Costs you if |
|---|---|---|---|
| Hands-on | Frequent meetings, close direction, detailed feedback on data | You want to be trained carefully and like clear next steps | You need independence to feel ownership of the work |
| Hands-off | You set the pace; meetings when you ask for them | You are self-starting and comfortable being stuck for weeks | You need external structure to keep moving |
| Large established group | Senior students and postdocs do much of the training | You learn well from peers and want a busy environment | You want regular time with the supervisor personally |
| Small or new group | Direct daily contact; you help build the lab | You want responsibility early and enjoy setting things up | You want mature infrastructure from day one |
4. Resources — the boring question that decides your timeline
Ask plainly whether the instruments your project depends on are inside the department or have to be booked elsewhere, how long the queue typically is, whether they are currently working, and whether consumables are readily available. A project that needs a technique the group has to travel for will move at the speed of that arrangement, whatever anyone intends.
Ask, politely and directly, about the funding position of the project you would join — whether it runs on a grant with a fixed end date, and what the plan is after that. This is a normal, professional question, and a good supervisor answers it without discomfort.
How to gather the evidence
Talk to current students — this is the step that matters most, and the one students skip out of shyness. Speak to at least two, ideally away from the supervisor's presence, and ideally including one who is near the end of their degree. Ask open questions and then be quiet:
- How often do you actually meet your supervisor, and how do those meetings go?
- How long do people here take to finish, and does anyone leave before finishing?
- What happens when an experiment fails for two months?
- Who decides what you work on — and can you propose your own direction?
- How is authorship decided in this group?
- Is the equipment you need available when you need it?
- Are leave, working hours and holidays respected?
- Would you join this group again?
That last question is the most informative one you can ask, and a long pause before the answer is itself an answer.
Also try to speak to one person who has already left the group. Former students speak far more freely than current ones, and the pattern of where a group's graduates end up tells you what the training is worth.
Questions to ask the supervisor. Asking these marks you as serious, not as difficult — and a supervisor who reacts badly to them has answered a different question for you.
- What would my project be, and how does it fit the group's direction?
- What would the first six months look like in practice?
- How often does the group meet, and how do you prefer students to bring problems to you?
- Is the project funded, and for how long?
- Which techniques would I learn, and where are they located?
- How do students in your group typically move on afterwards?
- What do you expect from a student in the first year?
How to write the first email
Most enquiry emails are deleted in seconds because they are visibly circulars. A useful one is short, specific and shows you read something. Name the group's actual work in one line, say what in it interests you and why, state your background and qualification in two lines, attach your CV, and ask one clear question — whether they expect to take a student in the coming cycle. Five researched emails outperform fifty copies, every time.
Red flags worth taking seriously
- Current students avoid answering the "would you join again" question.
- Nobody has finished recently, or students routinely take far longer than the department's norm without a clear reason.
- Authorship practice sounds arbitrary or nobody can explain how it is decided.
- The supervisor cannot describe your project beyond a general area at the point of admission.
- Leave and working hours are discussed as favours rather than as normal entitlements.
- Safety is treated casually — an unlabelled, cluttered, untrained lab is a statement about how people are looked after.
- You are discouraged from talking to the group's students. In a healthy group this is encouraged.
- Every sentence is about how famous the group is and none is about how students are trained.
One flag alone is a question to ask. Three together is a decision.
Common mistakes when choosing
- Choosing by institute ranking alone. You will work with one group, not with a ranking. A well-run group in a less famous place usually beats a neglected corner of a famous one.
- Choosing the friendliest interviewer. Warmth in an interview is not evidence about supervision over four years; current students are.
- Choosing on publication counts alone without checking whether students are first authors and whether they finish.
- Not knowing your own working style and then blaming the supervisor for a mismatch you could have predicted.
- Ignoring geography, cost of living and family circumstances. These are legitimate inputs to a multi-year decision, not weaknesses.
- Accepting immediately out of relief. It is normal to ask for a few days and to speak to students first.
If it turns out wrong later
Sometimes a well-researched choice still does not work — a supervisor moves, a grant ends, a direction changes, or the working relationship simply does not function. Changing supervisor is difficult but it is not impossible, and it is a recognised situation rather than a scandal. Every institute has rules for it, usually involving the department's doctoral or academic committee, and most also have a research advisory committee whose purpose includes exactly this kind of oversight.
If you reach that point: read your institute's own regulations before saying anything, keep your own records of work and meetings, and speak first and privately to the committee member or department head responsible for research students. Acting early, calmly and through the formal route works far better than enduring silently for two more years.
The short version
Read the group's last three years of work. Ask its current students the eight questions above, and one former student anything at all. Ask the supervisor about the project, the funding and the first six months. Then decide with your eyes open — because the supervisor you choose is, in practice, the research degree you are choosing.
Selection interviews test fundamentals on the board. Panels ask you to reproduce the standard results — Nernst, Arrhenius, Gibbs energy, equilibrium constants, quantum numbers. The ABC Chemistry Calculator Suite lets you rebuild each of them with correct numbers while you revise for the interview that decides your group.
Open the ABC Chemistry Calculator Suite →Still to clear the qualifying exam? ABC Chemistry runs CSIR-NET, GATE, IIT-JAM and CUET-PG batches at the coaching centre and online for students anywhere in India — abcchemistry.in.