home icon

Control versus autonomy

What balance should I, as a PhD supervisor*, strike between guiding students' work closely and giving freedom for independent work?

Supervisors' responsibilities for developing independence in their students

'Independence' in one form of words or another is probably a requirement for a research degree in all higher education institutions throughout the world. Yet if the work were entirely independent there would be no need for supervisors.

The general consensus among experienced supervisors who have discussed the matter with me is that supervisors have a responsibility to wean their students from dependence to independence. How best to do so depends on a number of factors such as the nature of the work; the ability and expectations of the student, particularly if from a culture where it is normal to defer to authority figures; the student-supervisor relationship; disciplinary norms; and the preferences of the supervisor.

One common tactic is to set a closely supervised mini or introductory project; another is via a programme of training seminars for groups of students; and there must be many others. It is crucial that, whatever means are chosen, the supervisor and student discuss the process of moving from dependence to independence so that the student knows what to expect.

What constitutes independence in a student/supervisor relationship?

Independence is not a matter of the student working without contact with the supervisor. Many experienced supervisors seem to define independence in terms of a relationship in which they advise, warn and encourage in response to the student's own ideas and suggestions, but do not themselves direct. The North American term 'advisor' may be more descriptive of this relationship than the term 'supervisor'. The German term, which translates as 'doctor-father', is also illuminating.

When should weaning towards independence start?

Precisely when the transition from dependence to independence should start and how long it should last are determined by factors similar to those listed above. The preferences of supervisors must depend on how they perceive their role, which seems to vary somewhat with their experience.

New supervisors can tend to put in more time, effort and guidance than their more experienced counterparts, in the belief that it will prove their professionalism to themselves, their students and their academic colleagues - which can delay or even prevent students' transition to independence.

Supervisors who are more experienced often have a relatively large number of research students, to whom they cannot possibly give the same level of attention and who may have to fend for themselves somewhat from early on. Without discussion and negotiation between students and supervisors, these supervisors can be perceived as neglectful, a perception which is compounded where supervisors' work is so highly acclaimed internationally that they are away a great deal.

How much of a student's work should a supervisor do?

A related question is "How much of students' work should supervisors do themselves?"

On the face of it, the answer must be none. However there can be times when supervisor and student are working together on the same funded research project when it can be difficult to disentangle the contributions made by each of them - and perhaps also by the different members of the research team. In such situations, students need to be asked to record the boundaries of their work, in readiness for writing their theses.

Similarly tricky situations arise where students get stuck on certain parts of their work, although, in the supervisor's judgement, they would be able to continue effectively if only helped over those particular hurdles. Then there are those students who are simply not managing to progress successfully at all. Such situations provide a strong temptation, particularly if the reputation of the supervisor and department are at stake, for supervisors to step in themselves. This certainly happens, although the supervisors concerned do tend to feel extremely uncomfortable about it.

The consensus advice seems to be that the amount of work which supervisors do for their students should be a matter of what they can justify with their own consciences and not exceed what would be judged as acceptable if called into question by other members of the academic community.

edited extract from no 2 in the Guides series

Resolving Common Dilemmas in Supervision
by Pat Cryer

© Pat Cryer

* 'Supervisor' is a shorthand for 'research degree supervisor', 'advisor' or 'tutor', and applies to varying extents for all research degrees: PhD, DPhil. MPhil, Prof Doc and even undergraduate and masters' projects. In some countries, notably the USA, a 'supervisor' is known as an 'advisor'.