Sunday, April 5, 2009


I've just spent the last week at a conference, one I went to specifically to meet certain people to get feedback I needed. I've spent the last day pulling together all the information I got, and in the process have been thinking about styles of feedback, advice, supervision.

To set the scene, a bit about my situation first. I'm in an interdisciplinary field, which in many universities (including the one where I did my Masters) is considered a sub-field of computer science. In other universities (my current one) it can be its own department, or else more closely aligned with the other major (non-technical) area, which for the purposes of this blog I am going to call the Study of Cupcakes. My own background is computer science in "a land far, far away".*

I came to Europe, to an exciting well funded department to discover that the PhD system is very very different in different countries. I have gained a lot from being in Europe, but the supervision situation, at least for our group, is... sub-optimal. The general procedure here is that only full professors can supervise PhD students. There are not many of them, and in general, they have more important things to do like run research institutes and negotiate with the EU. Hence, the convention is that someone else is designated the direct supervisor. Unfortunately, the situation with my direct supervisor became unworkable by a year and a half in, and every since I have been working basically unsupervised. I meet my official supervisor when possible, but there is no one else at the university who could take on the role of direct supervisor.

Since I am about to start (hopefully) writing up my thesis, this conference was very timely, and I managed to organise a number of one-on-one meetings. The input was invaluable and all from quite different angles (luckily complementary). The different styles made me wonder what is the ideal supervision style. Perhaps the ideal is to have it all, but on the other hand, I have become very independent due to my current situation and that is not entirely a bad thing.

My official supervisor is actually in some ways very good. His insights are legendary and with him I discuss both my research and my PhD in general, but at such high levels that it might be more accurate to say the direction of our field in general, and where my career will be in 10 years time. I tend to come out of those meetings both inspired, but also confused about what I should be doing tomorrow.

Three of the meetings I had in the last week, I would have called supervisor meetings rather than just feedback meetings and these particularly suggested different roles of a supervisor.

The first focused predominately on my research: which experiments I was missing, what I had described well, what other analyses I could do of my data. He asked why I had chosen certain methods, we discussed the story I was trying to tell, and we debated which data would best illustrate my points. We planned the next set of experiments and arranged to discuss the details by email next week.

As a side note, my reaction to his positive comments on specific details surprised me. The feedback I've had through my PhD, such as it has been, has been positive, but so general that I think I didn't believe it. I was almost embarrassed(?) by what a confidence boost I got from such small things as "nice paragraph" and "that graph is exactly what you want to show". I'm pretty sure he didn't realise the effect his comments had, but it made me hark back to the issues of my previous post. Do I crave positive feedback that much?

My next 'fill-in supervisor' concentrated on how I should structure my thesis. We discussed the idea of a one sentence summary and how expanding that makes up the introduction. She also recommended a high level to low level progression through the chapters and told me what she would expect from her students in the way of a lit review and where it would best fit in. She is planted quite firmly in the Cupcake-ology side of our field, so most of our research related talk focused on giving my analyses a certain breadth to include non-technical issues.

My third would-be supervisor wasn't talking so much about the research, or the thesis, but on how to do the research. How I should be able to articulate in one sentence the claim that I was making, and that, at every point in my work, I should be asking myself how my current experiment/reading/analysis contributed directly to proving that claim. He told me I needed a list of exactly which experiments were required and I wasn't to vary from that list unless results suggested the claim needed to be re-thought in any way. We did discuss my research in terms of what that claim might be, given the work I have done, but the major contribution from this 'supervisor' was in encouraging the sort of discipline I was going to need in order to finish.

As I said, I think ideally a supervisor might be able to take on all these roles, but what do people think is the most important role of a supervisor? Someone to discuss and plan research with? Someone to teach you how to organise and conduct research? Someone to help you learn to communicate your research? It's a question I will be keeping in mind, both when looking ahead to what I want in my postdoctoral position, but also for when I start (hopefully) supervising others.

* I am quoting Professor in Training here. I suspect we actually come from the same far away land, and since I started following her blog, that phrase, which used to evoke fairy tale lands, now just means HomeCountry.


  1. Nice to see somebody else telling tales about finishing a PhD in the EU. We have a somewhat similar structure in our department, but I think the bigger snag is that PhDs aren't allowed to supervise Master students. So nominally those students are supervised by professors, but in a lot of cases it's really the PhD students who are supervising.
    How does that work in your department?

  2. friday: I'm not actually sure what the official rules are for Masters. None of the PhD students I know are doing any supervision, though I have found myself asking supervisor type questions (why are you testing that? what are you expecting to find?) when I'm asked for technical help by Masters students. I have one friend here who finished his PhD about 18 months ago and is doing a lot of supervision for his one-time advisor, who is taking all the credit. The gap here seems to be not between student and non-student, but between full professor (habilitation + a chairs position) and everybody else.

    And thanks for dropping by. It is nice to find European academic blogs to balance out the primarily North American experiences I read about.

  3. You're welcome :)

    If you can, you might want to get experience in supervising. It gave me more perspective about what supervising is about, what style or attention is needed when to get me going.

    Are there non-full professors who have PhDs too? In my case, my daily supervisor is a researcher, but she's awesome. Also, I don't think I could deal with my official supervisor this often - his questions can be really strange (last time, we discussed curtains and coffee - why am I still not drinking any?).

  4. Actually, you are the second person in the last 2 weeks to recommend that I get supervision experience. (Though the other was thinking ahead to the postdoc I hope to start towards the end of the year.) I just have the feeling at the moment that I am supremely unqualified to supervise anyone, since I'm not sure I'm supervising myself properly.

    My direct/daily supervisor was a non-full prof PhD but meeting with her was such a negative experience that I really am better off working on my own. And other than her, there is one other researcher. He is helpful and we meet regularly but he only defended just over a year ago and so is more like a fellow student. (We started meeting when we were both students.) My official supervisor discussed Christmas traditions across cultures with me :)