How to deal with bad undergrad researchers

Question

I am a grad student in the life sciences. I have to mentor several undergrad students. One student in particular is just so bad and is always getting confused with protocols. He can't follow a protocol from start to finish without getting help from me or others, even if he has done it multiple times. I'm having to spend several hours babysitting him when he has to do simple procedures, even after nine months in the lab. I've often had to stay behind until 8 p.m. to catch up on my own work. He is utterly dependent on me, to the point where I feel suffocated.

I'm fed up. What can I do?

Answer 1

We all forget too quickly how it felt when we didn't know how to work in the lab. Others treated us like stupid, when just had not yet learned the skills. If you become a professor, you will end up spending a big part of your life just training people who don't know how to do things. And the older and more experienced you get, the harder it becomes to remember how it feels to be on that first step. So yes, you are training the other person, but you also need to train yourself in the art of training others.

With that out of the way: nine months is an awful amount of time, and what you are supposed to do is training not babysitting. Either the student develops independence, or they find someone at a lower level who can supervise them, for example, a more experienced undergraduate. By rejecting them from your supervision, you are not condemning them to lower-level status or ruining their career, but telling them that there are skills they need to learn before they can work under the supervision of a graduate student.

How did you end up with this student? Did you ask for undergrads to help you, and you hired this one? Then you can also fire them. Did your advisor assign this person to you, as part of research money the advisor is getting? Then talk to your advisor about reassigning the student, perhaps to a two-undergraduate team where this student reports to an undergraduate, an that second undergraduate then reports to you (or another grad student.)

Answer 2

Undergrads are often under-committed and overloaded with coursework. They want to experience research due to an idealized notion of what it is and are then overwhelmed by what is actually being asked of them.

It’s perhaps not too late to have a meeting where you manage expectations. How many hours/week are they devoting to the project? What are they good at? What can they expect from you in terms of mentoring? It’s useful to have a written “contract” (could be just an email summary). Some labs even have students sign those, to lend them more weight, though I personally don’t. Mentoring plans are super important in my opinion.

If you both see that you have widely disparate opinions about your work relationship, you should definitely contact your PI and let them know.

Answer 3

On top of what others have said, remember that one of the best ways people learn is by being made to do it themselves. There's no problem with telling someone to figure it out on their own after you've given them appropriate scaffolding. In fact, you might be making them codependent if you make yourself too available.

Tell the student you're leaving at 5:00 PM (or whatever is appropriate for you) and they need to have it done by then. Or if this is the kind of lab where you can leave the undergrads on their own, tell them you're going home and they need to finish themselves. Even if you have to be in the lab, tell them you're doing your own work and unavailable to help them that day. Even if they break stuff and waste lab supplies, they're also wasting money by taking up your time as a grad student or faculty (and arguably a lot more, because salary is not cheap).

Our rule of thumb is that undergrads should work on a problem for 10-20 hours before bothering the grad students about it. Grad students should work on a problem for even longer before bothering their advisor about it.

If they don't get anything done because they spend a week mastering simple procedures... well... that's undergrad research for you.

Answer 4

David mentioned setting boundaries, I want to add that setting expectations is also important.

I often have a talk with them beforehand letting them know how I will be annoyingly micro managing the first few times I teach the protocol. I expect detailed notes, then I will shadow the next few times also expect annoying nagging. Then I expect the student come to me with questions, and don't be afraid to let them fail.