Rough Theory

I’ve been trying to begin to think in a more sustained way about research supervision recently – a byproduct of being involved in the research methods course, which entails an intrinsic element of short-term supervision as you work with students who are writing their thesis proposals and, in practice, also often leads to longer-term follow-ups from students seeking specific kinds of advice as they continue their research.

I’m finding it reasonably complex to think through the issue – partially because my own personal preferences for research supervision are highly idiosyncratic, and not really suitable for extrapolation (not that extrapolation from personal learning styles ever provides a very solid foundation for thinking through the teaching relationship). At the same time, because supervision is such an intense process, and tends to be restricted to a fairly small number of students at any one time, you don’t get to try out ideas with the sort of “sample size” of students that’s available from regular teaching. You don’t have as many easy opportunities to think through what works, and what doesn’t work, in practical circumstances.

Reflecting back on various bits and pieces of advice I gave to students in the Research Strategies course this term, I realise that I generally fell into the metaphor of describing a thesis as a research apprenticeship, and encouraged students to seek out supervision with someone who could mentor them through this apprenticeship process. I did this because I was reaching for a way to capture how supervision differs from most other forms of teaching: it’s more sustained, more intense, generally involves a higher degree of modelling and workshopping than most other forms of teaching, because you’re focussing on the quite individual problems that arise in research design, data collection, interpretation and writing. These problems generally take the form of an unsolved puzzle, often with “wicked problem” dimensions: supervisors might have a level of experience or wisdom that can help a student cut through complex issues more efficiently, but generally can’t rely on knowing the “answer” to a student’s question. So the supervisory relationship provides, among other things, an opportunity for students to watch how a more experienced academic muddles through the sorts of problems the student is encountering for the first time.

Looking back on the term, though, I worry that this metaphor may have focussed too much attention on supervision as a process through which specific skills are communicated – a vision of supervision that was likely to reinforce what is generally a student’s first impulse in any event: to seek supervision from the people who have the greatest knowledge in their subject area, or experience with their preferred methodology. I don’t actually believe, however, that relevant subject or methods expertise is anywhere close to being the best predictor of whether a supervisory relationship will be effective for a particular student. Students generally can, I think, seek out expert advice fairly easily via one-off or short-term interactions with academic and non-academic professionals with whom they don’t require a sustained supervisory relationship. What distinguishes particularly effective supervision, I suspect, is more likely to be a kind of mentoring “chemistry” – closely related to what one of my own supervisors describes as the “pastoral” dimension of supervisory work: can a student and a supervisor develop a sufficient level of trust that they can honestly discuss problems that arise in and around the research process, and work together to develop effective solutions?

One implication of the relational nature of supervision is that the success or failure of supervisory relationships is rarely completely one-sided: I’ve seen situations in which two students had diametrically opposed experiences with the same faculty supervisor, and situations in which a student who seemed at risk of failure under one supervisor, rapidly found their footing in a new supervisory relationship… Supervisory relationships can be quite poor without this necessarily meaning that individual supervisors or students would have experienced problems in a different relationship… This means, however, that it can be very, very difficult to generalise about what makes a “good” supervisor, since a wide range of supervisory relationships could potentially work for specific students and faculty…

What could help, though, is to foreground the concept of supervision as a mentoring relationship, when talking with students about how to think about choosing their primary research supervisor. This means, among other things, advising students to distinguish between their need for information on their subject area or technical advice on their methodology, and their need to identify an appropriate mentor who can assist them in developing into the kind of academic or professional they intend to become. I did talk a bit about this in the methods course – although I think I largely flattened the issue into one of working style (e.g., whether students prefer very structured and formal interactions, more casual interactions, very hands-on supervision, etc.). For a short project such as an Honours thesis, it probably doesn’t matter all that much. I’m conscious, though, of the number of PhD students in particular who seem to make significant supervisory changes in mid-stream, and I wonder whether some reorientation of priorities when selecting a supervisor could minimise some of this disruption, and also increase students’ ability to grow and mature as intellectuals and professionals through the supervision process.