Working better with Research Assistants

Working in an academic environment, I’m in the fortunate position of having lots of free labor readily available. Indeed, we regularly have to turn away people who are willing to volunteer with us. It’s a great position to be in, but just because we have plenty of people willing to help doesn’t mean that we’re not without some of the same problems that anyone has when they’re looking to employ people.

Today, I took part in a panel on how to work better with RAs. I made some notes to myself beforehand based on some topics the organizers suggested, which I’ve included here.

I should preface this by saying that I’m not at all confident that everything I’m saying here is the best way or even a good way to work with research assistants. However, I do think that I’ve gotten considerably better across the years, and I think that some of this is good advice.


Advertise and Network - Our department has an annual RA recruitment night. All the labs who are looking for help send at least one representative to describe the lab’s research and what kind of RA’s they’re looking for. In advance of this event, I try to figure out what kind of work I’m going to need over the next semester or so. Based on that, I can generate a list of skills I’m hoping a potential RA would have. This allows me to have some goals in mind when it comes to finding good people.

Then I make sure I attend the recruitment event. Early in graduate school, I would just let our lab manager attend and recruit for me. However, I’ve found that maximizing the face-time with someone before they come work with me is a great way of finding people who will be great workers. This is especially true at the recruitment event, since it’s a bit like a cattle call. All the RA-hopefuls are milling around, looking at their handouts to refresh themselves on what each lab does. If someone approaches you and doesn’t really seem to know what kind of research you do (despite you having just described it for them), or they spend their time looking around the room for who they’re going to talk to next, then you odds are good that this is someone who is just looking for research experience because they know they should/they think it’s an easy way to get credit. Not that there is anything wrong with those goals, but if you’re looking for a good RA, you want someone who is there for the same reasons you are (presumably intellectual curiosity).

In short, you should kind of prepare for this thing like you’re the one looking for a job. Make sure you know what you want, tell the potential folks what you want, and see if they respond positively. If they do, then this is a good first sign.


First, make sure you do it. I used to select people based on a few emails back and forth + their cv/resume. That was a terrible idea. Make them interview. During the interview, look for a few things:

  • Do they dress up? This is an indication that they’re serious and are motivated to work for you

  • Start off with easy questions about their experiences, and what they’re looking for. Then start working up to more difficult questions. For instance, you can ask them about hypothetical situations RA’s might encounter (what would you do if you accidentally deleted a file? what would you do if a participant was acting distressed?). Increase the difficulty until you get to a point where they either confess that they have no idea, or they obviously are hopelessly fumbling, or you get to the kinds of questions for which you would expect only grad students to have genuinely good answers (i.e. how do you interpret the reproducibility crisis in science?)

  • Have them perform some tasks similar to those you might actually ask them to do if they work for you. In the last year, I’ve started giving participants two messy spreadsheets, telling them that the much larger spreadsheet A should contain information from spreadsheet B, but that I’m concerned that the information was garbled in the translation - could they find any potential problems?

    I’ve made this an extraordinarily difficult task, but it’s revealing. First, many will not know how to do this, based on my minimal instructions. Thus, they should ask questions! That’s a good thing! Or at least, it’s much better than them persisting at something and doing it incorrectly. If they do it correctly without much questioning, then that’s obviously great too. But I’m just as happy with someone who may not be experienced, but knows enough to ask the right questions in order to figure out how to do something.

  • How do they make you feel? Make sure you leave at least a little time to talk to them on a more casual basis. The biggest red flag for me is when I have a hard time talking to someone. In the past, I’ve had RAs who make me cringe every time I see them walk through the door. I’m sure this was just as much my fault as it was theirs, but regardless, it doesn’t make for a great working relationship. Make sure you like talking to them.


The biggest thing I can say here is just to meet with them regularly. Weekly meetings are key. At these meetings, you can discuss their work for the previous week and set up expectations for the next week. I’ve also found that my RAs respond well when I discuss the higher level details of the projects they’re working on. For me, this soemtimes means describing analyses I’m working on, showing them results or little bits of code I’ve written in the previous week. This makes sure they feel connected to the larger project.

I remember when I was an RA, there were more than a few projects I worked on where I just came in, ran the participants, and went home, with barely any interaction with the lead researchers. I felt like everything I did just disappeared into a void. It’s nice to see what the result of everything you work on is - even if it might be a bit over your head at first.

Somewhat relatedly, I also think it’s good for you to establish a schedule for them. We expect a minimum of 10 hours/week. If someone doesn’t show up for their shift and doesn’t get in touch to notify you, that’s a bad sign. Get in touch with them to inquire about the problem. If it happens too often, you can consider asking them to not come back in. Sometimes you can just let them continue to work sporadically - just don’t rely on them for anything important and don’t give them anything exciting.


Working with RAs and mentoring more generally is a skill that you can improve upon with effort. Sometimes this effort will go unrewarded because people leave or don’t work out. However, I think it’s much more likely that you can obtain great work from people when you put a little bit of effort into making sure that you set them up for success. Plus, you’re helping someone succeed in science and improve their skills! That should give you the warm-and-fuzzies and is, in my view, the best part.

Written on December 11, 2015
comments powered by Disqus