For me, easily the most jarring change from being a postdoc to becoming a lecturer was the supervision of undergraduate dissertation students. I’d certainly help supervise dissertation students before as a PhD student, and taken a leading supervisor role in Canada as a postdoc. But these were single hand-picked students. As a lecturer you will receive somewhere between 6 and 12 undergraduate dissertation students to supervise all at the same time, each with varying degrees of enthusiasm for the task at hand, all working on the same timelines. Now, I consider myself equal parts a scientist and an educator, and I find that dissertation supervision is the part of the job which best blends these two (often discordant) roles most enjoyably. As such, I enjoy this particular aspect of his job immensely and invest much more time than many academics do perhaps largely because I have few doctoral students to pontificate towards. But it requires a fair bit of planning to get through the year, making sure that everyone limps over the finish line.
Most places will have some cryptic method of assigning students to supervisors. Sometimes this is merit-based (best students get their first choice), whereas sometimes its entirely choice-based (everyone gets their first choice, within reason). In your first year though, when none of the undergraduate students know you, you are unlikely to have many people knocking at your proverbial door. The task of wrangling this many students all at once becomes all-the-more daunting when you realize they were all hoping to get Dr Popular next door but couldn’t get their paperwork sorted in time. But this matters not – in my experience the most successful student/supervisor relationships have little to do with how enthusiastic they appeared to be at first – I certainly was an awful undergraduate until I started getting acquainted with my new lab setting during my dissertation.
So, two paragraphs in and I’ve provided no tips about managing these students. Ok. Bear in mind that these tips, more so than my previous posts, will be pretty subjective.
- Write all the students names on the whiteboard along with their project titles and various markers of progress (submitted ethics form, gained ethical approval, started data collection etc). This will help you keep track of them all (not trivial) and will highlight progress discrepancies quite visibly to the ones who are behind the pace.
- Nag your students regularly about their progress. The students who fail are never the ones who got ethical approval for their projects before Christmas. This also covers you in case of an appeal – showing you have emailed them more than they have emailed you makes it pretty unambiguous where the fault lies.
- Try to avoid email conversations with project students – have them meet with you when they want to ask questions. This, I find, encourages self-learning a lot more because they often cannot be bothered to make an appointment for something trivial. Keep your meetings with the students brief and focused. I only allow mine to book 15 minute slots with me (with exceptions for early project design meetings) which keeps everyone on task.
- Use a booking system to arrange all meetings. Some like Doodle polls. I use Calendly, which is free and integrates with Outlook or Google calendar in a smooth way. As a general rule, email is just such a time sink for an academic anything you do to streamline will save you tons of energy in the long run.
- Keep all your students on track, but try to stagger their data collection if you can. You’ll find there’s a lot of drama that can occur when 7 individuals are all trying to collect data in one room at the same time. I use Calendly for student-facing lab booking, along with some rules (block booking for only a half day etc).
- Avoid the temptation to have all your students each doing cool individual projects. You’ll sink. Some supervisors like to group their students to maximize data collection. I’m not such a fan of this – it makes the writing of an independent dissertation difficult for the student, and starkly highlights differences in how much time you spend with each student if they all end up talking to one another (but many make this strategy work very well). I prefer to try to have approx half my students doing ‘cool Gavin stuff‘ and half doing data collection which doesn’t require the lab.
- Create a lab manual or wiki which outlines standard operating procedures (SOPs) for all experiments (or common protocols) you run. If you find yourself explaining how to do something year on year, you need to outsource this to a document!
- Teach good habits. Make them back up their own data, regardless of your own backup strategies. Demand all your students keep a physical lab book, which they arrive to meetings with you prepared to take notes in and properly write up after they leave, and use to document stuff during data collection. Few of them will engage properly with this, but it’s good practice for a scientist and occasionally helpful for writing up projects for publication. When they are finished, ask them to scan all the pages of the lab book and email it as a single .pdf along with a .zip file of their data and a copy of the submitted thesis itself. Using OSF seems like a more efficient way to do this, but many students are surprisingly tech naive…
- Keep your emotions and frustrations in check. You will enjoy supervising some more than others. You then might well be disappointed with the final product of these. On the flip side, you may be pleasantly surprised with the final product of others who completely failed to engage in the process. Try to detach yourself from this at marking time. Mark the document dispassionately, and keep this distinct from whatever proportion of the marks you allocate for ‘engagement with the process’. If you aren’t sure, ask for it to be 2nd marked (most dissertations usually will be anyway).
- Be aware that many students are going through a difficult time in their final (and most stressful) year while doing their dissertation. Their lack of engagement in the process is rarely anything to do with the supervisor/project.
- Try to create a lab culture. Bring all your students together for a group meeting to present their ongoing work to one another a couple of times. This can be used to motivate everyone to be on a similar timeline, and can generate a bit of peer tutoring. If your students are helping one another out with their stats, then this is one less job for you and they are learning good life skills.
- Students freak out about the stats. Their number 1 nightmare is forgetting how to enter data into SPSS, forgetting which menu to click, and forgetting which table to read and which to ignore (and thus have no results section). I tell mine up front that they’ll impress me if they come to my office with SPSS printouts showing the analysis that they’ve tried to do without my help. I also tell them that if it’s wrong, fine, but I’ll be impressed they’ve tried and I will eventually help them do their stats properly if necessary.
- If your institution has a policy of reading drafts (the ever-popular ‘formative feedback’), set your students a deadline for this if there isn’t a fixed internal one. You cannot imagine the drama that will unfold if you tell students they need to get a draft to you tomorrow because you’re going on holiday and wont be back until after the hand-in deadline. I provide feedback in Word using the comment bubbles (lots of them!). Don’t use track changes to edit their text.
- If you see their results and start to get excited, keep it inside. Students already worry that they will fail if they find an ‘insignificant’ result. Tell them that they have asked a question worth asking in a sensible way and communicated it well, that’s all you’re looking for. I’ve made this mistake before and I can assure you that in no uncertain terms, any talk of publication should wait until after they have received their degree classification.
Well I didn’t really expect to write that much, but there you go. Story of this blog so far.