Teaching and Learning Forum 99 [ Contents ]

An evaluation of the desirable characteristics of a supervisor

Rob Fraser and Anne Mathews
Faculty of Agriculture
The University of Western Australia
In this paper we propose a model of postgraduate supervision which broadens the traditional focus on "expertise" to include support for the student and the capacity of balance creativity with criticism in supervision. Based on this model, we report results of a survey of postgraduate students in the Faculty of Agriculture at UWA investigating the desirable characteristics of a supervisor. We find that students clearly rank non-expertise-related characteristics of supervision which provide support and which balance creativity with criticism as more important overall than expertise-related characteristics. We use these results to argue for staff development opportunities to be enhanced to enable academics to receive training in these areas of supervision competence which are ostensibly unrelated to expertise.


How do we determine the competence of an academic to supervise a postgraduate student? In most cases the answer would probably be "expertise in the area of research". And by "expertise" would be meant one or more of a range of things such as: being up-to-date in the research area; having an active research programme in the area; having a track record of research output in the area (including having done a PhD in the area); having a network of colleagues in the area; and having a track record as a research supervisor in the area.

But is "expertise" enough to determine competency for supervision? Our view is that "expertise" is an important contribution of an academic to a student-supervisor relationship, but that there are at least two other components of the supervisor's contribution which are also important and for which academics usually have no particular training in providing. Moreover, we report evidence from a survey of postgraduate students regarding desirable characteristics of a supervisor which supports our view that students desire much more from a supervisor than "expertise". We then use this evidence to argue for enhanced opportunities for academics to broaden their supervision skills beyond the confines of "expertise".

The structure of the paper is as follows. Section 1 sets out our model of supervision which identifies the main components of the contribution made by a supervisor. Based on this model, Section 2 reports the results of our survey which was designed to represent the range of characteristics of these components and to elicit student views on the desirability of these characteristics. The paper ends with a brief assessment of the implications of our research findings.

Section 1: A model of supervision

Our model has developed from our appraisal of recent literature in the area of postgraduate supervision (Cullen, Pearson, Saha and Spear, 1994; Hall, Coates, Ferroni, Pearson and Trinidad, 1997; McCormack, 1994; McMichael and Garry, 1994; Parry and Hayden, 1994). It represents the contribution of an academic to the supervisor-student relationship as having three main components:
  1. expertise in the research area
    As outlined in the Introduction, expertise can mean a range of things that could typically be characterised as one or more of the following: knowledgeable; specialist; teacher; influential; co-ordinator.

  2. support for the student
    There are times, particularly in the early part of the research, when the supervisor may need to build and support the student's self-confidence as well as to reassure and motivate the student (Delamont, Atkinson and Parry, 1997). Support for the student also covers a range of other characteristics such as: enthusiastic, available, helpful, involved, attentive, caring.
Moreover, we think there are three key phases of a research project, and that the significance of the various types of support changes between these phases:
  1. settling-in and getting going
    In this phase it is important to make the student feel confident that they have the intellectual capacity to complete the project (enthusiastic, involved).

  2. maintaining the impetus
    In this phase students often suffer a "relapse" of self-doubt. They need to be reminded of all they have achieved already, and helped to develop a clear vision both of what remains to be done and of how to achieve it (helpful, attentive).

  3. finishing-off
    Students are often distressed at the prospect of having their writing criticised and having to rework chapters at a point when they feel intellectually exhausted. It is important to try to rearrange and rebuild their work rather than knock it down, so that they can see their efforts have not been wasted and feel that they have the stamina to finish after all (available, caring).

  4. balancing creativity and criticism
    In the "getting-going" phase a supervisor should contribute ideas to the project so that the student can clearly see their commitment (stimulating, active). In the latter phases the supervisor should shift more and more towards the critic. Here the student is typically developing their own ideas about where to go and what to do and it is the supervisor's job to react constructively (critical, objective). Therefore, "both advice and criticism need to be managed in order to encourage the competent student to develop sufficient self-confidence to embark on and sustain several years of demanding independent work" (Delamont, Atkinson and Parry, 1997).
In the next section we present these components of a supervisor's contribution as a broad set of supervision characteristics and report on a survey of student views on the desirability of these characteristics.

Section 2: The survey: Methodology and results

  1. Methodology

    In determining the set of characteristics of supervision perceived by students as "desirable", we made use of the list of supervisory characteristics developed by McMichael and Garry (1994) and reproduced in the Appendix to this paper. Although this list contains several pairs of characteristics which are effectively opposites (eg active/passive; involved/detached), as well as others which might be perceived as strongly over-lapping (eg friend/partner; teacher/trainer), given its apparent comprehensiveness it seemed appropriate to use this list as a base set of characteristics and to leave the judgement about which are "desirable" up to the students.

    In addition, we allocated each of the characteristics in the list to one of the three main components of the supervisor's contribution accordingly to its best fit in our judgement. On this basis we allocated seven characteristics to "Expertise", nine characteristics to "Support" and six characteristics to "Creative/Critical". Details of this allocation are also provided in the Appendix. Note that although this list contains twenty-two characteristics, the inclusion of the opposites identified above suggests that the set of desirable characteristics will be no more than twenty.

    Finally, we established a 1 to 5 scale between "Undesirable" and "Highly Desirable" for the students to evaluate each of the characteristics. Note that a characteristic would need to receive a score of 3 or more for it to be viewed as "desirable". The raw data of responses from thirty-two Faculty of Agriculture postgraduates is also included in the Appendix.

  2. Results

    Table 1 contains details of the characteristics which achieved a mean score from the student responses of 3 or more and which therefore qualified as "desirable". It can be seen that this Table contains nineteen characteristics, and excludes from the list in the Appendix the three characteristics:

    Detached (S: 2.22); Director (E: 2.56); and Passive (Cr: 1.72). Note that the first and third of these were identified previously as one of a pair of opposites, where the other of the pair is included as "desirable" (Involved (S: 4.16) and Active (Cr: 4.19) respectively). However, the second is simply not viewed as "desirable" by postgraduates.

    The desirable characteristics in Table 1 are arranged in the form of a matrix with three columns representing each of the main components of the supervisor's contribution, and two rows ranking the characteristics according to whether they achieved a mean score of between 3 and 4, or 4 and above. Based on this matrix the general pattern of results is clear: the majority of the desirable support and creative/critical characteristics achieved a score of 4 or more, while the majority of desirable expertise characteristics achieved a score of between 3 and 4. Moreover, desirable support characteristics not only dominate the set of characteristics achieving 4 or more (six of the ten), but also rank as three of the highest-scoring five characteristics.

Table 1:Desirable Characteristics of a Supervisor

Mean Score ExpertiseSupport Creative/Critical
4 and above

Knowledgeable (4.47)

Enthusiastic (4.81)
Helpful (4.53)
Attentive (4.47)
Available (4.44)

Involved (4.16)
Caring (4.00)

Stimulating (4.47)

Objective (4.38)
Active (4.19)

Between 3 and 4
Influential (3.81)
Trainer (3.66)
Teacher (3.63)

Specialist (3.16)

Coordinator (3.03)

Friend (3.34)

Partner (3.06)

Critical (3.97)

Colleague (3.59)

Section 3: Assessment of implications

On the basis of the preliminary results in Table 1, we can draw either of two possible inferences:
  1. the students clearly view non-expertise-related characteristics as more desirable in a supervisor than expertise-related characteristics, or

  2. the students took the occasion of the survey to express their relative satisfaction with the expertise-related characteristics, and their relative dissatisfaction with the non-expertise-related characteristics, of their supervisors.
If we take the position that the students responded in a detached way to the survey, then in our view the main implication of our research findings is that the traditional focus on expertise as the determinant of supervision competence is too narrow. In particular, students clearly rank non-expertise-related characteristics which provide support and which balance creativity with criticism as more important overall than expertise-related characteristics. Consequently, given that academics typically have no explicit training in providing support to students they are supervising or in balancing creativity with criticism in their supervision, it would seem appropriate for staff development opportunities to be enhanced to cater to these needs. Gaining skills to supervise effectively has benefits for both the students and the supervisors (Graham and Grant, 1997). In particular, inexperienced supervisors will benefit from training opportunities as often supervisors are not adequately prepared for being supervisors.

If, alternatively, the findings of our survey are restricted to the Faculty of Agriculture, then the above comments still apply in this context. However, the need then arises to extend our research into other discipline areas in order to determine the robustness of our results. Recent anecdotal evidence from broadly-based student support activities at the University of Western Australia suggest that this is likely to be the case.


Cullen, D.J., Pearson, M., Saha, L.J. and Spear, R.H. (1994). Establishing Effective PhD Supervision. AGPS, Canberra.

Delamont, S., Atkinson, P. and Parry, O. (1997). Supervising the PhD: A guide to success. Open University Press, Buckingham.

Graham, A. and Grant, B. (1997). Teaching more students: Managing more postgraduate research students. Oxford Centre for Staff Development, Oxford.

Hall, S., Coates, R., Ferroni, P., Pearson, M. and Trinidad, S. (1997). Tilling the field: Action research in postgraduate supervision. In Pospisil, R. and Willcoxson, L. (Eds), Learning Through Teaching, 132-43. Proceedings of the 6th Annual Teaching Learning Forum. Perth: Murdoch University. http://cleo.murdoch.edu.au/asu/pubs/tlf/tlf97/hall132.html

McCormack, C. (1994). Constructive and Supportive Postgraduate Supervision: A Guide for Supervisors. Centre for the Enhancement of Learning, Teaching and Scholarship, University of Canberra, Canberra.

McMichael, P. and Garry, A. (1994). Strategies for Supervision: A Handbook. Moray House, Heriot-Watt University, Edinburgh.

Parry, S. and Hayden, M. (1994). Supervising Higher Degree Research Students: An Investigation of Practices Across a Range of Academic Departments. DEET, Canberra.

Appendix: Postgraduate supervision characteristics: Responses of students

UndesirableHighly desirable

Active (Cr) [a] 12345
Attentive (S)[b] 41810
Available (S) 13820
Caring (S) 110912
Critical (Cr) 123179
Colleague (Cr) 114143
Coordinator (E)[c] 261572
Detached (S) 1261031
Director (E) 79952
Enthusiastic (S) 626
Friend (S) 1315103
Helpful (S) 3920
Influential (E) 110156
Involved (S) 1141214
Knowledgeable (E) 111218
Objective (Cr) 31415
Passive (Cr) 197321
Partner (S) 251672
Specialist (E) 171464
Stimulating (Cr) 21317
Teacher (E) 237137
Trainer (E) 326138

Balancing Creativity and Criticism
Support for the Student
Expertise in the Research Area
Source:McMichael, P. and Garry, A. (1994). Strategies for Supervision: A Handbook. Moray House, Heriot-Watt University, Edinburgh.

Please cite as: Fraser, R. and Mathews, A. (1999). An evaluation of the desirable characteristics of a supervisor. In K. Martin, N. Stanley and N. Davison (Eds), Teaching in the Disciplines/ Learning in Context, 129-137. Proceedings of the 8th Annual Teaching Learning Forum, The University of Western Australia, February 1999. Perth: UWA. http://lsn.curtin.edu.au/tlf/tlf1999/fraser.html

[ TL Forum 1999 Proceedings Contents ] [ TL Forums Index ]
HTML: Roger Atkinson, Teaching and Learning Centre, Murdoch University [rjatkinson@bigpond.com]
This URL: http://lsn.curtin.edu.au/tlf/tlf1999/fraser.html
Last revision: 26 Feb 2002. The University of Western Australia
Previous URL 15 Jan 1999 to 26 Feb 2002 http://cleo.murdoch.edu.au/asu/pubs/tlf/tlf99/dj/fraser.html