Lab / group size


Hi there,

I know that the size of a PI's lab is probably contingent on several factors - like funding, experience, what the norm is in the department, field etc. But all those things held constant, could a very small lab be indicative of anything else - e.g., not a good PI to work with? I am curious because I've come across a PI/lab whose work I am really interested in, and I'm thinking I might like to join them if an opportunity arises. But I am kind of surprised at how few students/postdocs are in the lab (two or three at most), when the PI is super successful in terms of high quality published research, chapters, etc. Should I have warning bells or am I being paranoid?

I have good reason to be a bit paranoid / cautious after my PhD experience - so just thought I'd seek opinions on here to balance me out a bit, hopefully! I've decided that working with people I regard as decent is more important than anything else (when it comes to joining a lab).

A second question - is there a way to find out what a PI is like? Would postdocs / PhDs likely be honest if I asked them what it was like to work with such and such? There may be no definitive answer to these questions but I'd love to hear what people think / have experienced.


Avatar for rewt

Small labs aren't inherently bad. As you said there a few reasons why it might be smaller, like administrative duties. I think the experience of the lab members is far more important. If the one lab member is used to publishing in high impact factor journals, they will be far more helpful than 4 team members who only publish in low-tier journals. Smaller labs let you do more varied work and do different things without encroaching on someone else's project. Though a 10 person lab group can do far more experiments or studies. So i would be paranoid if you want a high throughput post-doc. You will be fine regardless.


I can't see how the size of the group would make any difference.
I would advise you to think about what kind of career you want and then select the route which best suits that.
It sounds to me that you are looking at the wrong metrics to make a good decision in that respect.
For example, your supervisor's publication record and that of his previous students isn't helpful. It's going to be entirely down to you now. You might find you get lots of papers out or none, regardless of their previous output. There's no correlation between the two but you appear to be looking for and expecting some.
As for impact factors, you already know my feelings on that score :-D


Cheers both! I think I asked my question in a confusing way. My question isn't - is a big / small lab inherently good / bad? It is - could it possibly be any indicator on the personality / style of the PI? Or - could a small lab despite success (as I judge it - through high quality research that I enjoy reading - plus lots of grants and things) be indicative of them being a d*ck? I've come to realise that there are two types of people in academia (and maybe a third group who haven't quite yet decided), and I am very keen to avoid the latter type. But I realise, I may be being a bit paranoid / speculating too much here as I wonder, hmm, why such a small lab... are you a d*ck and no one wants to work with you and everyone ends up leaving...?!

That's my question :-) I don't care one bit about whether a lab is big or small or metrics on how others judge people's research. I want to know how I can find out what kind of person this PI is before I get myself trapped in something like I was in during my PhD (my supervisors... never again... please...). I am visiting the lab in a few weeks... but is it really possible to make a judgement through an initial meeting? Argh... help... any tips / advice on things to look out for (subtle signs of someone being a d*ck?) would be appreciated.

Hope it makes more sense now.


I think your original question made sense. Reading back, I just managed to completely misunderstand it. I was also partly answering rewt's point.

OK I'll have another go.
I can't see how the size of the group would tell you that.
The only way to know is to interview the supervisor directly as though you were hiring them for the job.
Think of all the bad scenarios you want to avoid and then ask a question to directly find out if that's what you'd face.
I would ask directly what their expectations were of me and others in the lab. I'd use that to start digging as deeply as possible. For example, is this person expecting a certain number of publications per year, to publish only in certain impact factor journals, etc. If any of those questions were answered with numbers I'd stop the interview and walk away because for me that would cross two red lines.

Other than interviewing them personally, I can't think of any other way of finding out if they are a good match for you. Most people don't do this because they are so desperate not to be rejected. That leads to the sort of trouble which could have been easily avoided. I certainly would never trust the opinion of another student or postdoc.


Thanks for your reply pm133. And I like your phrasing of whether "they are a good match" for me. That is another aspect. There are some PIs who I think would be decent people, but not necessarily a good match. I agree that asking questions and "interviewing" them would help me to discover this. But I wonder how to find out what kind of person they are - how they regard and treat others being the primary thing.

I don't want to start working with someone only to discover that they are someone who I don't think has many principles - e.g., someone who engages in bullying or treating with less respect of those who are perceived as in a position of lesser power, etc etc. I just can't stand that kind of thing. And now I've had a taste of working with folks who seem decent and respectful, and it has made me set that as a bit of a red line myself - that is what I value right now. I wonder how I could find out what kind of person they are - short of just getting to know them and interacting with them over a period of time... it's not really something you can probe by questioning...


Ooh, my reply to you has actually given me an idea...! I guess I could start a collaboration with them and see how that goes and whether I would actually want to do a postdoc with them... now there's an idea!


Quote From Tudor_Queen:
Ooh, my reply to you has actually given me an idea...! I guess I could start a collaboration with them and see how that goes and whether I would actually want to do a postdoc with them... now there's an idea!

A collaboration would be a peer to peer relationship.
You would then be asking a peer to employ you.
That sounds like an odd strategy to me although I can see why you are thinking about it.

If this was me, I would either interview them and take a chance or I would attempt to bypass the entire thing by trying for my own funding on an idea which was exciting and ambitious.


I don't agree - I've had a collaboration before which then led to a job offer (postdoc) - in my experience it's a great way to get to know whether you would want to work with / for someone in the long term or not.


A late reply but it might help. Yes, if the group is performing well and there is a few number of PhD student/postdocs in comparison with similar group, I would ring the alarming bells. Collaboration will definitely help. It might be insufficient but better than nothing. You can detect possible a**holes. Asking other lab member might help. They might be honest. The worst you can get is not getting info which is already the case now.
I heard of a group which was a "crazy" professor and only a single PhD student. I think you do not want this. Also one fast first class professor in my field was known to be an a**holes. Despite him being close to one of my favourite cities around the world, I would never consider working with him.


Eng, that is exactly the sort of thing I want to avoid!!! I am visiting the lab next week and will try get a sense of things then and possibly ask the very few lab members some questions! I am hoping I will just be able to get a sense of things - having had experience in a lab where the PI was great and the lab members happy and productive. Why don't they sell metal detectors for this sort of thing? Then again, I suppose that is what instinct is for...


This is an interesting question - and a difficult situation for you. A smallish lab may mean all sorts of things. The PI may simply value quality over quantity. I knew quite a few very good researchers who hand-picked students, collaborators and research topics in order to ensure high-quality research. They simply wanted to focus on the things they considered really worthwhile, and tried to excel at those, rather than chasing every latest research trend. On the other hand, many mediocre scientists practice empire building and try to expand their group at all cost. Of course, there are many other possible explanations as well.

The much bigger question is whether the group is actually as outstanding as it seems from the outside. Personally, I experienced huge disappointments twice after moving at great expense to seemingly leading research groups with world renowned PIs - only to find out after joining that these groups were basically run by impostors.

Unfortunately, some people are extremely gifted at charming, blackmailing and cheating their way onto authors' lists, grants, committees, etc., in some cases over decades. The best publication lists in the most renowned journals, constant presence at top conferences, etc., do not guarantee that even the most basic knowledge and competence are actually present in the research group. Unfortunately, in my experience this is almost impossible to tell from the outside because in particular such sub-par plagiarizers put a huge emphasis on presenting themselves in the best light when visited by other scientists.

In the end, the best you can do is do your research and then make a decision based on the best information you have available. Unfortunately, there is always the risk that you realize after a short while you ended up in the wrong place.

In any case, best of luck to you!


Quote From Tudor_Queen:
I don't agree - I've had a collaboration before which then led to a job offer (postdoc) - in my experience it's a great way to get to know whether you would want to work with / for someone in the long term or not.

I'm actually surprised to hear this. I genuinely thought this would be a strange tactic but obviously if you have direct experience of this working then maybe that is the right path for you to go down.