They can’t even agree! Student conversations about supervisor views of the PhD Judy Backhouse School of Education, University of the Witwatersrand, South Africa e-mail: [email protected]

Abstract Supervisors in the same discipline and within the same department exhibit a range of views on what is expected of PhD students. These include different views about the nature and scope of the research proposal, the structure and style of the thesis and the role of the supervisor in the PhD process. Trying to interpret, understand and accommodate supervisor standpoints occupies the time and conversation of doctoral students. This paper draws on conversations among doctoral students in Education at the University of the Witwatersrand to explore what views supervisors hold, how students deal with these and what students learn in the process. This research forms part of a pilot study that investigates the nature of PhD learning. While differing supervisor views can create confusion, this research concludes that they provide a rich source of engagement that can enhance student learning if appropriately supported.

Introduction “Did you hear what he said? The proposal can’t be longer than thirty pages!” “Ha! My supervisor says that the literature review must be thirty pages.” “My proposal was a hundred and nine pages. How long was yours?” “About fifty - I could never have fitted it into thirty pages.” “Well my supervisor says that the proposal will become the first few chapters of the thesis, so it has to be substantial.” PhD students spend a disproportionate amount of their time and energy discussing the different views that supervisors have of the PhD. They tell stories of supervisors in the same discipline and the same department who have widely differing opinions of what PhD students must do, how they must do it and what the role of the supervisor is in the process. In this paper I explore the conversations of a group of students as they try to navigate their PhD paths. Questions that guide the investigation include: What differences in supervisor views do students report? How do students react to these conflicting views? How do they construct their own understanding of what they should do? And finally, I consider what implications this might have for efforts to improve supervision. The PhD degree in South Africa generally follows the ‘learning-by-doing’ model with a student working closely with a single supervisor to complete a substantial piece of research and produce a written thesis (Dietz et al 2006:9, Mouton 2001:xi). In this model the relationship between supervisor and student becomes a key element in the learning experience (Johnson et al 2000:137). Supervision has been established as the most important factor in the perception of the quality of the PhD experience (Dietz et al 1

2006:86, Frame & Allen 2002:100, Rugg & Petre 2004:32). However, this critical relationship has been left to individual supervisors and students to interpret and play out and unsurprisingly, supervisors often base their practice on their own experience of being supervised (Dietz et al 2006:14). Consequently there are many views of the role of a supervisor, the process of getting a PhD and what the student should produce (Dietz et al 2006:69-70, Rugg & Petre 2004:40-41, Shaw & Green 2002). This paper arose out of a pilot of a larger study of student experience. Notes were recorded from ten conversations among PhD students in Education at the University of the Witwatersrand over a period of four months from October 2006 to January 2007. The conversations commonly took place over tea in the student’s offices or at the cafeteria on the Wits Education Campus. Most of the conversations took place between a core group of four students (including the author), however some conversations included other doctoral students. In all, eight students in Education participated in the conversations. Students discussed not only their own supervisors, but also told stories of other supervisors that they had encountered either directly or had been told about by other PhD students. In the conversations a total of six different supervisors were referred to – four in Education at Wits and two in Education at other universities. As a doctoral student in Education I was privileged to have an insider view of the position of PhD students. This position made it possible to participate fully in the conversations and enjoy a high degree of trust from the other students. Students were aware of the nature of my research and willing to be part of it, however conversations were not recorded as this would have interfered with the casual nature of the interactions. Notes were written up as soon as possible after each conversation took place. Quotations in these notes are therefore not verbatim, but are as close as could be recalled to the actual words of the students. Conversations are coded chronologically as C01 to C10 and the paragraph number is indicated in references below.

Diversity and standardization in doctoral studies That supervisors differ in their expectations of their role and of the student is well established (Dietz et al 2006, Frame & Allen 2002, Lessing & Schulze 2002, Malfroy 2005:169, Mouton 2001). This is to be expected across disciplines due to the different nature of the work and different ways of working (Becher 1989:108-110). Differences can also be expected across institutions where different administrative procedures and different criteria for assessing quality are in use (Shaw & Green 2002:116). However these variations are also encountered within a single discipline and within a single academic department. Particularly in the Social Sciences, the supervisor-student relationship ‘depends crucially on individuals’ (Deem & Brehony 2000:160) and these relationships ‘are full of idiosyncrasies’ (Dietz et al 2006:69). In South Africa, institutions are not selective about who supervises; there is little training or induction of new supervisors and virtually no assessment of supervision skills (Dietz et al 2006:11). Instead it is assumed that once an individual has a PhD, they are qualified to supervise the PhD. Consequently, supervision practices are often based on the individual’s own experience of being supervised - what Dietz et al (2006:14) term the ‘reproductive character of research supervision’. In addition, supervisors have different

2

strengths, skills and personalities which make them more or less able to take on certain roles. For example, Mouton (2001:16-26) identifies four different roles for a supervisor: adviser, guide, quality control and pastoral care. One supervisor may be well equipped to advise while being ill-equipped to offer pastoral care. There have been various attempts at categorizing different approaches to supervision. For example Dietz et al (2006) propose a typology of supervision based on studies in South Africa and the Netherlands. They examine three dimensions of supervision: how businesslike or personal the relationship is; the intensity of task behaviour and the orientation of task behaviour to process or product. This is used to construct six typical supervisors: delegater (low intensity, businesslike), friend (low intensity, personal), expert guide (high intensity, businesslike and process-oriented), coach (high intensity, personal and process-oriented), quality controller (high intensity, businesslike and product-oriented) and co-writer (high intensity, personal and product-oriented). Of course supervisors modify their supervision style according to the student and to the stage of the student’s work – thus a supervisor may begin as an expert guide and move towards a delegater as the student gains skills and confidence. And students are not passive bystanders. Students have preferences for one or other type of supervision relationship depending on their own personality and preferred style of working. Cases of students who share a supervisor forming informal groups ‘to evaluate and guide their relationship with the supervisor… or to try and change his or her style’ are reported by Dietz et al (2006:76). However, while there is a growing body of work on the experiences of doctoral students (Chiang 2003, Golde & Dore 2001, Lee & Williams 1999, Lessing & Schulze 2002) the issue of how supervisor differences impact on students has not been explicitly addressed. Nationally (and internationally) there are concerns to improve the quality of supervision with a view to increasing the success rates of PhDs (Department of Education 2001; Koen 2005, Scholes & Janks 2006). Efforts at improvement have suggested (among other things) documenting procedures and expectations in handbooks, agreements, ‘research contracts’ or ‘supervision plans’ (Charlesworth et al 2007:12-14, Dietz et al 2006:26, Lessing & Schulze 2002:148, Mouton 2001:18-19, Scholes & Janks 2006:3A). However, the implementation of such initiatives is not uniform. For example at Wits a wide variety of supervision practices were reported to the University Graduate Studies Committee (Scholes and Janks 2006:3D). At the same time, the need for flexibility in doctoral programs to meet the varied needs of different disciplines and different students in a changing knowledge environment has been recognised (Charlesworth et al 2007:2, Frame & Allen 2002:102, Malfroy 2005:166). Those who come from a quality assurance perspective advocate benchmarking the PhD in order to address the ‘lack of consensus and consistency’ (Shaw & Green 2002:123). However such moves to ‘standardize’ doctoral training and the supervision relationship may work counter to the need for flexibility. And, as reported below, it will be difficult to establish consensus among supervisors.

3

Supervisor differences In the course of the student conversations, ten areas of difference were identified and discussed by the students (see table). The differences discussed appear to reflect on three overlapping areas - different views about the role of the supervisor (points 1-4), different beliefs about the process of getting a PhD (points 4-7) and different understandings of what the student is expected to produce (the proposal and thesis) (points 7-10). 1. Supervisors differed in the extent to which they felt that they were training PhD students for research as opposed to facilitating learning (C06) 2. Supervisors acted in ways which indicated a preference for a more personal or more professional relationship with students (C07) 3. Supervisors were involved in the student’s research to greater or lesser degrees (C07, C09) 4. Supervisors gave different levels of guidance in the PhD process – such as how to do the ethics application (C09) 5. Supervisors had different ways of giving feedback (C07) 6. Supervisors displayed different attitudes towards student involvement in other activities (such as departmental research and teaching) (C05) 7. Supervisors expressed different views on the nature and purpose of the PhD proposal (C02) 8. Supervisors expressed different understandings of the literature review (C02) 9. Supervisors expressed different views of the structure of the PhD thesis (C08) 10. Supervisors exhibited different approaches to student writing and writing style (C06, C07, C09) Table 1: Supervisor differences identified by students Students discussed the styles of interaction preferred by their supervisors. One described his relationship with his supervisor as ‘very professional – not at all personal’ (C07:2) while another said that her supervisor preferred a more personal relationship (C07:1). Supervisors differ in their degree of involvement in the content of the student’s research. One will seek out relevant papers and pass them on to the student, will engage with the topic and make suggestions to the student as to how the research could proceed. Another will remain on the periphery of the student’s research, expecting the student to find their own resources and making few contributions to the thinking behind the research (C07:3; C09:2). One supervisor ‘expects you to think for yourself’ and will, for example, tell you in passing that you must apply for ethics clearance, while another will ‘hound you until it’s done’ (C09:1). Supervisors differ in the way in which they give feedback – one gives extensive written feedback in the form of annotations on the student’s work and supplements this with verbal feedback while another gives only verbal feedback and does not return written work submitted by the student (C07:3). Students tend to value written feedback as they find verbal feedback difficult to remember and even to understand at the time: ‘I just 4

write. I have no idea what he means, but I just write notes. Perhaps I will understand it later when I go back.’ (C02:14) Students discussed how one supervisor expects their students to get involved in other research projects or in teaching within the department while another kept students away from such involvement, insisting students focus only on their own research. A story was told about a student of the latter who found their work intolerably lonely (C05:2). Supervisors have different views on what PhD students are expected to produce – the research proposal, the thesis and the components of these. Supervisors express different ideas about the nature and length of the PhD proposal with one supervisor seeing the proposal as more comprehensive so that large parts of the proposal are expected to become chapters in the final thesis, another accepting a range of different proposals. One supervisor insists that the PhD proposal ‘must be 25 to 30 pages long’ while another regards 30 pages as appropriate for the literature review (C02:20-24). A student described a heated exchange between two supervisors about the appropriate structure for the thesis. One insisted that the thesis follow a defined format (specifically that findings and analysis be reported separately) while the other felt that such a format was boring and that the emerging results should dictate the structure with a wide range of structures being acceptable (C08:2-3). Such vehemently expressed views leave students confused as to how they should approach their work: ‘We just watched them. They can’t even agree. How are we supposed to know what to do?’ (C08:3) A recurring topic was the process of learning to write (‘I didn’t realize that part of doing a PhD was becoming a writer’ (C06:3)) and the level of involvement of supervisors in student writing (C06:3; C07:5; C09:3). Dietz et al (2006:74) identify the ‘co-writer’ as a type of supervisor who ‘will put a lot of time and energy into writing or designing products together’ and who places ‘a lot of emphasis on language, both on concepts and on expression.’ One of the supervisors was identified by the students as a ‘co-writer’ in this sense. One student found this involvement ‘irritating’ (C07:5) while another felt that it had improved his writing to the point that ‘it would not be possible to write a paper that a journal would not accept’ (C06:3). There was some discussion as to what extent the style that students were expected to write in was a reflection of the academic discipline or unique to the supervisor (C09:3).

Student responses Doctoral students spend considerable time and energy in trying to interpret, understand and accommodate supervisor standpoints and expectations. Of ten conversations recorded over a period of four months, eight were directly concerned with this task. Working out what your supervisor wants and how to deliver it appears to be a large part of the task of these doctoral students. By contrast four conversations concerned the expectations, norms and conventions of the academic discipline and one concerned the requirements of the institution. This focus in the conversations clearly reflects the disproportionate significance of the relationship with the supervisor in the process of obtaining a PhD.

5

Interestingly, students appear to accept that the supervisor’s views have to be accommodated. Students make statements like ‘you have to adapt to what your supervisor wants’ (C08:5). A student who is nearing the end of his studies says that he has ‘changed his thinking’ and ‘adapted to (his supervisor’s) approach’ (C08:5) even though it has been ‘challenging to learn’ (C06:3). Not understanding and adapting to one’s supervisor can mean failure. A story was recounted of a student who ‘spent five years trying and gave up’. Apparently this student couldn’t understand what the supervisor wanted. ‘[The supervisor] says that it was stubbornness – this student hung on to a particular view and refused to change. But… [the student] just didn’t understand what he wanted.’ (C03:7). Stories abound about the fate of other doctoral students, both successful and unsuccessful. The veracity of these stories is not at issue here – the students tell them and use them to construct their understanding of the PhD process. The differences reported in supervisor views and expectations are not always objective. Students swap their individual interpretations of how supervisors like to interact in the process of trying to make sense of the supervisor relationship, and different interpretations are evident. One who described his supervisor as having a ‘very professional’ relationship with him was countered by another who said that the same supervisor seemed ‘more personal with those who are not his students’ (C07:2). Interestingly none of the students ever made reference to the Guidelines on Supervision published by their faculty in trying to understand what supervisors wanted. Indeed the guidelines only explicitly address one of the problems that the students discussed – that ‘students should be given written as well as oral comments in response to written work’ (Graduate School for the Humanities and Social Sciences 2007:159). Similarly, according to the Wits School of Education web site, the PhD proposal should be between 3000 and 5000 words, but this was ignored by both students and supervisors and proposals submitted by the students in this study were uniformly longer than 20 000 words. It would appear that guidelines, when they do exist, are considered irrelevant or that the supervisor’s preference takes precedence. In dealing with the different expectations of their supervisors, students expressed irritation, frustration and fear of failure (C07:5; C03:7; C04:18). One student talked about being depressed and close to tears a lot of the time - ‘this thing could kill you’ (C02:14; C03:7; C04:17). Despite this, students appear largely neutral towards their supervisors. They discuss supervisors as complex mazes to be navigated. Frustration is directed at themselves for not being able to find their way rather than at the supervisor. Students say: ‘Its part of my difficulty’ (C04:19); ‘I didn’t realize’ (C06:1); ‘why didn’t I get that’ (C02:14). Students sometimes wish that they could be working with another supervisor (C05:2; C07:5; C08:5-6) but they feel that they have no choice in this (C08:6). They also recognise that they learn from being challenged to work in a way that is different from what they expected or what is comfortable (C06:3). Students swap practical advice about how to navigate supervisor expectations. A recurring theme in the discussions is that the PhD ‘is a process’ and that many things (including the convoluted contours of what a particular supervisor wants) become clear over time (C02:15-16; C03:8; C04:2-3). The group provides support and there were several comments about how nice it was to have other doctoral students to talk to – both 6

from those within the core group and from those who joined the conversations less frequently (C03:10; C04:36-37; C06:6). Students who have been on the journey longer offer empathetic comments about how they found the process challenging and what they have learned (C03:8; C06:3). In the conversations, there is a general sense of progress over time as students reflect back on how their views and strategies have shifted. Students speak frequently of what they have learned: how ideas develop; how to be patient with a long process; how to become a writer; how to approach and talk to strangers; how to plan; how to deal with frustration (go and photocopy documents); how to motivate yourself (write down and track your work); how to speak, read and write in a new language; how to use technology: ‘You learn so much, so much’ (C02:16). While the differences noted among supervisors create some confusion, it is clear that they also provide a rich source of engagement and debate and that these conversations enhance student learning in a number of ways. In fact the responses of the students point to a larger problem – that the individual supervisor looms too large in the life of the student. Increasingly the literature reflects a move away from the ‘dyadic relationship’ (Malfroy 2005:176) between a student and supervisor and calls for a vision of doctoral students ‘as self-organising agents of varying effectiveness, accessing resources, one of which is the supervisor’ (Boud & Lee 2005:502). This research confirms these views and suggests that doctoral students should be provided with an environment in which they can interact with a wide range of resources including other students, other members of the academic staff and co-supervisors.

Conclusion Supervision is a key element in the experience and progress of PhD students. In a university environment where debate and independent thought are valued, supervisors can be expected to continue to have a range of different views about the PhD and their supervisory role. Efforts to improve supervision that focus on defining and standardizing the role of the supervisor and the processes and elements of the pedagogy are unrealistic and likely to be ignored. The variety of supervisor views of the PhD can be confusing to students, but it can also be part of a positive learning experience if the students have at their disposal a means of making sense of it. In the conversations examined, having access to a range of other students at different stages in their studies enabled PhD students to navigate different supervisor requirements and their own preferred working styles. In the process they gained support and learnt from each other. This suggests that improvements in supervision can be realised by putting in place multiple resources for students to access in the process of negotiating their own path.

Acknowledgements I would like to thank my fellow PhD students in Education for their generosity in participating in this research and freely sharing their thoughts and experiences with me. And I would like to thank our supervisors for being open to this scrutiny of their supervision practices.

7

References Becher, T. 1989. Academic tribes and territories. Milton Keynes: Society for research into higher education. Boud, D. and Lee, A. 2005. 'Peer learning' as pedagogic discourse for research education. Studies in Higher Education 30(5):501-516. Charlesworth, G., Grossman, E. et al. 2007. Strategies for Successful Supervision. Johannesburg, University of the Witwatersrand. Chiang, K. H. I. 2003. Learning Experiences of Doctoral Students in UK Universities. The International Journal of Sociology and Social Policy 23(1/2):4-32. Deem, R. and Brehony, K. J. 2000. Doctoral students' access to research cultures-are some more unequal than others? Studies in Higher Education 25(2):149-165. Department of Education. 2001. National Plan on Higher Education. Pretoria. Dietz, A. J., Jansen, J. D. et al. 2006. Effective PhD Supervision and Mentorship: A workbook based on experiences from South Africa and the Netherlands. Amsterdam: Rozenberg & UNISA Press. Frame, I. A. and Allen, L. 2002. A flexible approach to PhD research training. Quality Assurance in Education 10(2):98-103. Golde, C. M. and Dore, T. M. 2001. At Cross Purposes: What the experiences of doctoral students reveal about doctoral education. Philadelphia, PA, The Pew Charitable Trusts. Graduate School for the Humanities and Social Sciences. 2007. Graduate Studies Handbook 2007. Johannesburg, University of the Witwatersrand: 225. Johnson, L., Lee, A. et al. 2000. The PhD and the autonomous self: Gender, rationality and postgraduate pedagogy. Studies in Higher Education 25(2):135-147. Koen, C. 2005. Challenges facing the education, training and employment of South Africa’s scientific labour force, Paper two. Pretoria: Human Sciences Research Council. Lee, A. and Williams, C. 1999. Forged in Fire: Narratives of trauma in PhD supervision pedagogy. Southern Review 32(1):6-26. Lessing, A. and Schulze, S. 2002. Postgraduate supervision and academic support: student's perceptions. South African Journal of Higher Education 16:139-149. Malfroy, J. 2005. Doctoral supervision, workplace research and changing pedagogic practice. Higher Education Research and Development 24(2):175-178.

8

Mouton, J. 2001. How to succeed in your Masters and Doctoral studies: a South African guide and resource book. Pretoria: Van Schaik. Rugg, G. and Petre, M. 2004. The unwritten rules of PhD research. New York: Open University Press. Scholes, M. and Janks, H. 2006. University Graduate Studies Committee report to Senate. S2006/194. Johannesburg: University of the Witwatersrand. Shaw, M. and Green, D. H. 2002. Benchmarking the PhD - a tentative beginning. Quality Assurance in Education 10(2):116-124. Wits School of Education. 2007. Preparing your research proposal. http://www.wits.ac.za/Humanities/Education/phdproposal.html Accessed 15 July 2007.

Paper prepared for the International Conference: Postgraduate Supervision and Training Centre for Higher and Adult Education (CHAE), Stellenbosch University, South Africa 23rd-26th April 2007

9