Would you like to help us improve this website?

Supervisory styles

Talk but also listen; lead but also cooperate; act but also interact

A balanced style can help to maintain a mutually beneficial supervisory relationship. The required balance is often between complementary activities that conflict only when one has too great an emphasis (e.g., cooperation or leadership, listening or talking, and focusing on the relationship or the task). You might need to set aside more time during supervision, or manage time differently, to attend to both sides.


Achieving a balanced style

The following table helps address a common challenge of balancing between being selfless and selfish. It compares positive aspects of relationships that can become negative if they are not balanced.

Image by Allen McInnis.

(Photo credit: Allen McInnis)


Imbalance: Selfless


Imbalance: Selfish


Co-operation vs. leadership


I disempower myself, withdraw, become more passive, rely too much on the other to take the initiative, or provide the impetus – “laissez faire.”

I believe in empowering communication, which is appropriately assertive, not disempowering but actively empowering the other person.

I disempower the other person: being overly directive, controlling or even coercive.


Listening vs. talking


I focus too much on the other person, and listen too much.

I balance attention on myself and the other person.

I am too preoccupied with my own agenda, and listen too little.


Relationship vs. task


I focus so much on feeling, needs and the relationship that I lose sight of the task.

I balance between the focus on task and the focus on the relationship.

I focus so much on the task that I lose sight of feelings, needs and the relationship.


In order to position oneself in the center, that is, to achieve a balance, it is necessary to:

  • not focus on ego-issues (i.e., taking something personally);

  • have an open attitude to embrace alternative perspectives and new information; and

  • maintain a collaborative, problem solving mindset.

Adapted from materials (the table and list) provided by Geoff Mortimore, CEDAM, Australian National University.

Why meet with a student whose research has not progressed?

The answer is not only to make progress. Research does not always yield results at regular intervals, but regular meetings create opportunities to talk about the contextual aspects of graduate education, such as careers and personal lives (issues that can relate to progress or lack thereof). Some styles of supervision can manage periods of fluctuating productivity so that even downtime is helpful (e.g., encouraging or reflective).


There are a variety of supervisory styles; they have a huge impact on supervisory relationships and thus on student success. Consider what your style might be and how that may affect your supervisee. The quote below is from a student who appreciated their supervisor’s positive and helpful style:

She's the perfect supervisor because she's the one who keeps you on track. She's the one who encourages you in what you are doing, makes sure you know what is expected of you at all times.  And…if I were to ask her a question about anything, I know she’d respond to me and she’d help me. (Jazvac-Martek, 2009, p.94)

When co-supervising you may find your style compliments or clashes with that of your co-supervisor. Below, this new supervisor talks about his experience of co-supervising a student with an experienced colleague:

We definitely have supervision styles, which are exact opposites. … He is very managerial in the sense that, “I’m the guy at the top and I’m billing you” and I’m a lot more on the one-to-one basis … it has been a bit rationed with everything else where I’m a lot more informal, so you get bursts of work—times of work where it is very productive and then you have a bit of a lull and then back again. (STEM supervisor: McAlpine & Amundsen, 2010-2015)

Another new supervisor has weekly face-to-face meetings with her students, at which she allows the students to talk about their feelings as well as their research:

I do weekly meetings. … Usually they come, they have the reporting on the work they have done in the past week.  … This is the only hour of the week where I actually sit with them—they open up their laptop or if they have something on the server … we look at the data. …Sometimes we just spend time cogitating about something.  If it is a student who is getting ready for their oral exam I sit and we put the flowchart of their theses and we see how the different theses fit.  But I also tell them and they know that even if they have nothing to say, this hour is still for them and they can still come and I don’t want them to cancel.  … Because research progress doesn’t happen on a weekly basis, right, the weekly basis is something we’ve just artificially built. And some of them… just come to talk about their frustrations and we just talk. (STEM supervisor: McAlpine & Amundsen, 2010-2015)

Contractual, directorial, laissez-faire, and pastoral styles

Supervisors whose students remain in their programs, finish their degrees on time, and report high levels of satisfaction with their graduate education tend to have a contractual style. This style involves effective time management and planning, and interpersonal skills that can increase or sustain a supervisee's motivation. This style helps supervisors to be reliable and realistically encouraging.


A number of studies have been done of supervisory style and there seems to be general agreement that a match between student and supervisor in preferred style is supportive of a productive relationship. The studies also indicate that the notion of an “ideal” supervisor changes, depending on the context, the stage the student has reached, and the individual students involved. Two slightly different models are presented below – one from research in business and one in engineering.

Supervisor model: Business

Image adapted from Gatfield (2005).

In Gatfield’s research, 12 supervisors who had been designated as excellent by the dean of a business faculty were interviewed and asked to place themselves within one of the above quadrants. The criteria that the dean used to identify excellent supervisors were:

  • achieving high completion rates;

  • having students submit within the normally expected time frame;

  • engaging in multiple supervisions; and

  • receiving excellent supervisory reports.

Nine placed themselves in the "contractual style" quadrant, and one in each of the other three quadrants. Perhaps more important was the suggestion from the data that these excellent supervisors made a transition from one style to the other during candidature, mainly:

  • when the candidate was in crisis; and

  • when the candidate made a transition through various stages.

The ideal supervisor thus becomes one who recognises that each supervisory relationship is unique, requiring different skills and approaches.

Supervisor model: Engineering

Related research looked at the supervisory beliefs of 17 students and 17 supervisors in Engineering (Murphy, Bain, & Conrad, 2007). Four distinct orientations to supervision emerged in which beliefs about teaching, learning, research and supervision intersected; the authors argue that beliefs about teaching are central to each orientation.

Although each orientation comprised many beliefs, the orientations clearly differed in terms of two broad distinctions:

  1. Whether the supervisor should direct and take responsibility for the research (controlling beliefs) or should guide the process (guiding beliefs); and

  2. Whether the focus of supervision should be more upon the research tasks to be completed (task-focused beliefs) or upon the development of the candidates (person-focused beliefs) (p. 209).

It was found that, in the majority of cases, styles between supervisors and students were relatively congruent. This suggests the value of exploring with students their expectations of supervisory style, so that any differences do not lead to miscommunication. See the Clarifying expectations page.



2012-2013 Supervisory Surveys. Graduate and Postdoctoral Studies: McGill University.

Gatfield, T. (2005). An investigation into PhD supervisory management styles: Development of a dynamic conceptual model and its managerial implications. Journal of Higher Education and Policy Management 27(3): 311-325.

Jazvac-Martek, M. (2009). Emerging academic identities: How Education PhD students experience the doctorate. (doctoral dissertation), McGill University, Montreal, Quebec.

Murphy, N., Bain, J., & Conrad, L. (2007). Orientations to research higher degree supervision. Higher Education, 53(2), 209-234.

Further reading

Boud, D., & Costley, C. (2007). From project supervision to advising: New conceptions of the practice. Innovations in Education and Teaching International, 44(2), 119-130.

Cullen, D., Pearson, M., Saha, L.J., & Spear, R.H. (1994). Establishing effective PhD Supervision. DEST, 94/23.

Denholm C. & Evans T. (Eds.), Supervising Doctorates Downunder: Keys to effective supervision in Australia and New Zealand (pp20-27). Melbourne: ACER

Grant, B. (2000). "Pedagogical issues in research education" (pp. 31-34). In M. Kiley and G.P. Mullins. (Eds) Quality in Postgraduate Research: Making ends meet. Proceedings of the 2000 Quality in Postgraduate Research Conference, Adelaide, 13-14 April. Adelaide, Australia: Advisory Centre for University Education, University of Adelaide.

Janssen, A. (2005) Postgraduate research supervision: Otago students' perspectives on: - quality supervision; - problems encountered in supervision. Dunedin: University of Otago.

Johnson, L., Lee, A., & Green, B. (2000). The PhD and the autonomous self: gender, rationality and postgraduate pedagogy. Studies in Higher Education, 25(2), 135-147.

Lee, A., Dennis, C., & Campbell, P. (2007). Nature's guide for mentors. Nature, 447, 791-797.

Murphy, N., Bain, J., & Conrad, L. (2007). The pedagogy of 'good' PhD supervision: A national cross-disciplinary investigation of PhD supervision. Canberra: Department of Education Science and Training.

This work is licensed under a Creative Commons Attribution Non-Commercial 4.0 International License.
Graduate and Postdoctoral Studies, McGill University.

Back to top