Help with mentoring an UG student in the lab?

[penguinshooray]

I know this forum is mostly for TAing, but I wasn't sure where the best place for this question would be. Sorry if this is not the right place.

I am a mid-year PhD student in biology. I mostly work with databases/data mining type of stuff. 

I have a summer student who has been really great, motivated, and does quality work. I am wondering, at what point are his mistakes due to my ineffectiveness as a mentor? To be more specific, my advisor placed him with me on this project that we are just starting (ie, not an ongoing project). Because it's a new project to me too, I haven't worked out all the kinks myself. So as we are going through this data, I am realizing it's more complicated than I envisioned, and I don't always have a good answer for his questions. What makes it worse is that I started my qualifying exam (written) the same week he arrived in the lab due to poor timing, so I haven't had much time to dedicate to solving these problems. It's getting to the point that we are going back and forth on how to deal with certain issues that keep popping up. I just told him to keep track of the cases that were ambiguous, but due to us having different definitions of what "ambiguous" means, he hasn't kept track of all of them and now neither of us is sure about the quality of the data. We may have to go back and re-do a week or two's worth of work because of this. Is this type of thing par for the course? I remember as an UG and early grad student having to go back and re-do things all the time after I had thought more about them, learning through trial and error, and not getting much help from advisors (on purpose, to "build independence" as they say). So I am not exactly surprised, but he is feeling visibly frustrated and I think part of it is my fault. Thoughts?

[St Andrews Lynx]

Don't beat yourself up about it. 

It sounds like you may have had to repeat your past few weeks of research anyway without the undergrad being involved - we make mistakes in our research and that's how we learn to design and execute better experiments. We aren't perfect scientists, if we were then we wouldn't need to be in grad school.

I reckon most senior people on this forum will tell you that teaching and mentorship are skills you learn, not anything innate. So yes, there is going to be mis-communications and the undergrads will make mistakes based on things you said. Or they just weren't paying attention to what you said and so its their fault.  As you go along you learn how to be unambiguous and to always think "I know that thing I said is obvious to me, but is it really obvious to an undergrad?" Or you can structure it so that the undergrad is doing simpler tasks in the beginning (make starting materials or preparing reference solutions) so that (i) they are protected from failure when they are just starting out (ii) if they make a mistake it doesn't have a major consequence to *your* key experiments. 

In this scenario, I'd give the undergrad a brief apology ("OK, I wasn't clear when I said X. That's my fault and I'm sorry about it.") then let it go and move on. It's also an opportunity to give them the talk about how research is more failures than success and that repeating experiments happens all the time. In the grand scheme of things 1-2 weeks of wasting time isn't a big deal in the average PhD. Be transparent and human: the undergrad will understand that you have competing demands and don't know everything, I think they'll be sympathetic if you admit that stuff up front. 

[Lisa_McCoy]

You are not at fault. Such things happen in research. In fact, committing errors is a natural sign of progress as you get to analyze your mistake and understand the data in hand in a better way. As for the summer student, he is bound to feel frustrated and blame his mentors for the mistakes like every other research student on earth. From what I have read, it seems like you genuinely want to help him. So, don't hold back from speaking a few encouraging words to him and make him believe that you can truly help him with his research.