How to be a PhD supervisor

The relationship between PhD students and their supervisors is often said to be the most intense in the academy, with huge implications for student success. Yet most supervisors receive little if any training. Here, six academics give their take on how to approach it 

April 18, 2019
Illustration of clearing out your mind

The anomaly of academics receiving years of training in research but none in teaching is often remarked upon. But, equally notoriously, what that research training amounts to very much depends on the individual supervisor. After all, academics are not trained in research supervision either, and both supervisors and supervisees have their own set of expectations, needs and aptitudes.

How much and what variety of direction is appropriate? And to what end? Is the PhD still best conceived exclusively as an apprenticeship for would-be academics, even though many doctoral graduates end up – willingly or otherwise – in other walks of life? Indeed, is it realistic to expect all PhD students to be proficient in independent research at the end of the process? And what to do with those who are not? Is it the supervisor’s task to make sure they drop out early? Must supervisors always be optimistic and encouraging, offering whatever level of micromanagement is necessary to get the student through?

Here, six academics give their own takes on where the lines should be drawn in the doctoral quicksand.

Illustration of an argument

 ‘A PhD is about becoming an independent researcher’

If a final-year PhD researcher isn’t regularly telling me that I’m wrong, I worry; something has gone badly awry. They should argue that there’s no way my latest hare-brained experimental design could ever work, that my interpretation of the data can’t possibly be correct because it doesn’t take very basic physics into account, and that the paper published months ago by Professor W. Leader’s group at Prestige University described exactly the type of measurement I suggested in our most recent meeting. Why the heck wasn’t I aware of this? Don’t I read the literature? (They get extra credit for peppering their arguments with suitably chosen expletives.)

Demonstrating that level of project “ownership” is essential because otherwise I feel that I’ve failed in my supervision, not least because my university’s criteria for awarding a PhD – in line with those of every other university out there – stipulate that the candidate must have developed “the general ability to conceptualise, design and implement a project for the generation of new knowledge, applications or understanding at the forefront of the discipline, and to adjust the project design in the light of unforeseen problems”.

They must also demonstrate “a systematic acquisition and understanding of a substantial body of knowledge which is at the forefront of an academic discipline or area of professional practice”. I’ve never been able to square the rather perverse circle of requiring a student to be at the forefront of an academic discipline; someone whose doctoral work is, by definition, at the forefront of a discipline is most definitely not a student; they’re an expert defining the direction of their particular sub-field (or, as is more usual, their sub-sub-sub-sub-field). Still, it is clear that they should at least be in a position to swear with conviction at someone who is at the forefront of the discipline.

So I’m firmly of the opinion that we should ditch the term “PhD student” and instead use the rather more accurate “PhD researcher”. I am not alone in this. Jeff Ollerton, professor of biodiversity at the University of Northampton, has been making compelling arguments against the use of the term “PhD student” for quite some time. Retiring the “student” label would help drive a culture change that not only better recognised the core contributions made by PhD candidates to the research “ecosystem” (including those many first-authored papers submitted to the research excellence framework) but could also help improve their status in the eyes of that – hopefully dying – breed of PI who see early career researchers as the hired help, rather than the peers they should be. I was very fortunate to have a PhD supervisor with whom I genuinely collaborated; ideas were debated, discussed and dissected and I was never made to feel that I should “know my place”. I’ve since learned that this is not always the norm.  

Of course, not all PhD researchers are created equal; I am not so naive as to imagine that every doctoral candidate in every lab or library is capable of generating “a substantial body of knowledge at the forefront of an academic discipline”. But isn’t it then about time that we came clean about this? And wouldn’t phasing out the “PhD student” term help send a strong message to those who are considering taking on the formidable challenge of a doctoral programme?

A PhD is about becoming an independent researcher. The process is a universe away from the exam-and-coursework cycle that drives the vast majority of undergraduate degrees. There’s no curriculum and no textbook that map out a PhD from start to finish; no neatly annotated problems whose solutions are helpfully given in the back of the book. PhD researchers in effect write the textbook for the next generation of undergrads – what else do we mean when we say a doctoral thesis should be about expanding the body of knowledge in a particular field?    

In physics, we’ve become rather more open with PhD applicants in recent years about their chances of ultimately securing an academic position, should that be their goal. Let’s go one step further and be entirely upfront about the level of independence and self-direction involved in a PhD.

I’ve already followed Ollerton’s example and stopped referring to doctoral candidates as students. Why not join us? I suspect that your PhD researchers will appreciate the change.

Philip Moriarty is professor of physics at the University of Nottingham.

Illustration of man with suitcase 

‘Being a decent supervisor means being a decent human being’

My PhD supervisor taught me a valuable lesson about good supervision: it involves far more than teaching a doctoral student how to be a good writer and researcher. It is about believing in students’ academic potential, fostering their confidence and supporting them on whichever of life’s pathways they choose to take.

From the outset of my own PhD, I got a strong sense from my supervisor that he had faith in my abilities as an academic – he wanted me to succeed, and he never stopped telling me that he believed I could succeed. He was therefore committed to helping me develop the skills I would need when it came to navigating the postdoctoral job search. He advised me to do some tutoring early on, so that I could learn more about the art of teaching. He alerted me to job opportunities and gave some invaluable advice about performing well at interviews. He encouraged me to spend a semester at an overseas institution, which proved particularly fruitful, allowing me to develop my confidence on both personal and academic levels. Lastly, he instilled in me a passion for networking and collaboration, both of which have proved invaluable throughout my academic career.

Over the past few years, as I have taken up the mantle of PhD supervision, the lessons I learned from my supervisor have stayed with me. As well as guiding my students’ research and writing, I always take time to share some of the transferable skills that I acquired. I encourage them to participate in projects that will build their self-confidence as academics and in the wider world. I nudge them to present regularly at key conferences and have invited some to collaborate in writing projects with me. I seek opportunities for them to network with other academics in their field, or apply for a tutoring job, or gain experience and confidence in talking about their research. Following the example of my own supervisor, I want my students to know that I will do my best to nurture their abilities and support their endeavours as much as I can.

Nevertheless, good supervision is not solely focused on students’ future job prospects. Being a decent supervisor means being a decent human being, and showing students your respect and support. I always recall that, just before I began my PhD, my supervisor offered to assist me with my doctoral funding application; he knew I needed the funding and he wanted me to succeed. Despite being already overstretched with other academic commitments, he took time to look through my application with me, using his own rich experience to highlight its weaknesses and strengths. And throughout my studies, he supported me quietly as I went through the personal traumas of a family bereavement and a relationship break-up.

This basic sense of kindness is wrought from an acknowledgement that our PhD students are human beings, just like us. They look to us as role models – both academic and otherwise – and it is our responsibility to show them that we are good academics and good people. Hopefully, they will carry on this tradition if, in later years, they become supervisors themselves.

Kindness is something I place at the forefront of the relationships I share with all my students, and, honestly, it’s not that hard. Words of encouragement during moments of self-doubt, a listening ear when life’s stresses get in the way, a shared coffee or lunch when money is tight – small gestures that, as I recall myself, can go a long way in making the PhD journey that little bit easier.

Caroline Blyth is a senior lecturer in theological and religious studies at the University of Auckland.

 Illustration of cones

‘Providing structure is hugely beneficial to both the student and the supervisor’

Some of life’s most important roles come with no training – you have to learn everything on the job. Being a parent is one of those, and being a PhD supervisor is another. The similarity between these undertakings probably extends even further – to me, being a PhD supervisor means having the opportunity to nurture a young researcher, and to provide them with a strong and secure foundation on which they can establish a career.

This involves figuring out how best to support and encourage them, while also establishing strong expectations and boundaries. It means being present and available to provide input and feedback, while at the same time fostering confidence, independence and a sense of responsibility for their own future. It means being a role model, and inculcating skills and values (such as academic integrity) that will serve them throughout their career, whether that is in academia or beyond. It also means supporting them through the inevitable hard times, and revelling in their successes. In other words – it’s a pretty hefty responsibility, and often a difficult and stressful one.

I have a rather unconventional supervision history – my first faculty position did not permit PhD student supervision – so the bulk of my experience comes from more informal supervisory relationships with people from a variety of backgrounds, from recent bachelor’s graduates, to postdoctoral fellows, to more senior researchers. But these experiences, together with my memories of my time as a student and my growing experience as an undergraduate, MSc and PhD supervisor, have taught me some important lessons.

Among the most significant is the importance of structure: setting clear and explicit expectations, including deadlines and milestones (such as submission of papers for peer review), backed up by regular meetings with a specific agenda, action items and student follow-up. Providing such structure can be a lot of work, but it is hugely beneficial to both the student and the supervisor.

For students who are making the transition from the undergraduate or taught MSc setting, one of the most challenging features of the research PhD is often the lack of deadlines and the feeling of achievement that comes with meeting them, as well as the benchmarking that feedback on assignments or exams provides. Easing that transition allows the student to gain a sense of achievement and awareness of areas of growth and weakness, but it also alerts the supervisor to any signs of difficulties that might stand in the way of completion, such as mental health difficulties that may need support and a potential revision of the PhD plan and timeline. It can also provide a clear basis for communicating with a student that things are not progressing as they should, and that completion may not be a realistic goal at all.

All that said, I have to acknowledge that I often find it difficult to follow my own advice. Academic time goes by at warp speed, and sometimes attempts at structure just fall by the wayside. I also think that the onus for managing the relationship should not fall on the supervisor alone: students must also take responsibility for their own progression.

Figuring out these roles and responsibilities can be one of the most challenging aspects of being a PhD supervisor, and seeking out formal training in good management skills (now offered by many universities) can be immensely beneficial. But, ultimately, the nature of the PhD structure is necessarily dictated on a case-by-case basis by both the supervisor’s and the student’s personalities and working styles. For some students, a “micromanagement” style may prove necessary, while for others, a lighter touch will be more effective. I naturally shy away from the former and find it difficult to implement.

No doubt many academics – who, by definition, were proficient at working semi-independently during their own doctorates – feel the same. But if we have admitted a student to a doctoral programme, we have to accept that it is incumbent on us to give them the help they need to succeed.

Sometimes, of course, a gap will remain between expectations and student progress. In such cases, structures for performance tracking and feedback can help supervisors to determine whether or not the completion of the PhD is within reach. Given the recessive job climate in academia, stepping off the PhD track can often be a sensible option, and should be thought of as goal adjustment, not defeat.

Clare Kelly is Ussher assistant professor of functional neuroimaging at Trinity College Dublin and adjunct assistant professor of child and adolescent psychiatry at New York University.

 Illustration of inner thoughts

‘We can augment research skills with practical talents that build resilience and hone interpersonal intelligence’

PhD dropout numbers are approaching 50 per cent for many scientific disciplines in the US. That is because – speaking exclusively of scientific training – the truth is that not every doctoral student is cut out for independent research. In light of that, we should reconsider the true importance of possessing a doctorate.

After all, independent research is not the only walk of life in which the deep subject expertise and critical thinking skills acquired during doctoral study have enormous value. The years spent learning complex material, performing experiments and analysing data are immensely beneficial to anyone capable of doing these things. And sharing that information is necessary for the rest of society, too.

We need credentialed experts in every field, spread broadly, to enrich the lives of others who may not be so fortunate to have such opportunities. This is especially essential today, when misinformation is more believable than fact, and people confuse opinion with scholarship. Ideally, experts of every discipline should be embedded at all levels of the community, from teachers to church leaders, from small business owners to corporate leaders, to connect people to wisdom.

Perhaps the first step towards achieving this goal would be to remove the term “non-traditional career” from the academic advising vocabulary – even while keeping it very much inclusive of the production of PIs. We are living in a time of immense innovation, so fresh thoughts about employment trajectories should be equally novel. A student who ends up working outside academia should not be regarded by their supervisor as a failed self-cloning experiment. A PhD teaches students how to think, but we can broaden the scope of our mentoring to augment research skills with practical talents that build resilience and hone the interpersonal intelligence that is valuable in a wide range of careers.

In many US universities, students intent on entering the research phase of their doctorates have first to pass a two-day written and a one-day oral candidacy exam, so we already have sufficient preliminary evidence of their subject content mastery. When their suitability to be an independent researcher is in doubt, I try to direct my own students into jobs that may better suit their skills within academia. One example is student instruction; I offer teaching opportunities in my undergraduate courses in the form of 1–2 credit hours of independent study. This allows the doctoral students to prepare a topic, present it and create test questions, as well as to host and moderate the associated discussions on the electronic learning platform. The postgraduates are often surprised at how difficult and time-consuming this is.

For instance, pharmacology or biochemistry students with talents that may be more suited to the biomedical industry could be partnered with industry experts in their field who offer internships beyond the bench, to see how they fit in a non-academic environment. Many academic scientists have their own ties to industry, so it is not too much of a stretch to imagine that this could be arranged for the few students a year – perhaps as early as the second year of graduate school, before the research phase begins.

Some will say that PIs are not in a good position to prepare people for careers with which they themselves have no personal familiarity. But we were trained to develop young scientists and to get them to their destination. Whether that destination mirrors our own should be irrelevant.

Jennifer Schnellmann is associate professor of pharmacology at the University of Arizona.

 Illustration of a discussion

‘Students need to bring their interests into conversation with the discipline as a whole’

I’ve always conceived higher education to be a process of negotiation between the consumption and the production of knowledge. Students who are capable consumers of knowledge – usually those who do well in examinations – do not necessarily become equally capable producers of knowledge: that is, researchers. The reverse is also true.

Production of knowledge should start early, ideally in the form of undergraduate research papers. But, of course, it comes much more strongly into focus during the doctorate. There are debates about whether the dissertation is the last work of the research trainee or the first utterance of the professional. Perhaps it is a bit of both, but the proportions typically vary according to whether the student is working in the humanities or the natural sciences.

Research in the humanities is, for the most part, done by individuals, while scientific research is more of a group activity, led by a principal investigator. What this means is that anyone who has completed a doctorate in the humanities – my field – is, by definition, capable of independent research – unless the degree has been obtained by unfair means! Advisers share their expertise in the larger subfield, but the dissertation project itself is very much the student’s own.

I’ve known advisers who recommend that their students do not publish before completing their degree; and I’ve known advisers who encourage students to publish, present and network as much as possible even before the dissertation is completed. The conflict is between an approach that sees a doctorate as a complete, disinterested immersion in scholarship versus one that conceives it as professionalisation – which includes socialisation into the institutional lives of a given discipline, as well as more material functions such as seeking jobs, grants and fellowships. Scholarship and professionalisation are related and mutually reinforcing, of course, yet they are not the same and some people are better at one than the other.

My recommendation to my advisees is that they pursue a middle course. The origin of their scholarship, I tell them, should be something that sparks their passion, but once the fire is started, they need to bring their interests into conversation with the discipline as a whole. So, yes, they should use publication to enter into the conversation, but they should guard against letting professionalisation take over the independence and integrity of their intellectual lives. Dissertations should not be shaped in response to the jobs market. But once they have their scholarly trajectory in place, they should identify how it addresses needs and gaps in the discipline as practised today.

One of the big debates in my field, literary studies, is the book versus dissertation question. Should a doctoral student concentrate on fulfilling the traditional expectations of the very small group of people who will ever pay their dissertation any attention, adorning it with copious footnotes and a lengthy literature review that even the examiners won’t read? Or should they write with the wider book audience in mind right from the beginning?

My personal recommendation is to try to write the book. “Try” is the operative word: it is not easy to write something that coheres and reads like a book, especially for the first-time writer, yet aiming to do so will still lead to a radically more readable style. And this is important in the humanities, where the product and the process are not easily separable. It is probably OK for an article describing a new drug to be written in a dull or inaccessible style. But in literature, scholarship must also bring pleasure, even to academic readers.

In any case, students will be revising their dissertations for publication soon enough. Why waste huge chunks of time during their postdoctoral fellowship or first tenure-track job doing so? Much better to cut out the flab from the very outset.

Saikat Majumdar is professor of English and creative writing at Ashoka University. He is the author, most recently, of College: Pathways of Possibility (Bloomsbury India, 2018).

 Illustration of fishing for paper

‘The paltry credit offered in the US system disincentivises true mentorship’

Over the past half-decade I have directed well over a dozen graduate projects, and I believe more than ever that the effectiveness of the process should be replicated across the curriculum.

Working one-to-one with a senior scholar expert in the field is a unique boon for student persistence and engagement. To the extent that a thesis, dissertation or other extended research project effectively reduces the corporate scale and increasingly impersonal feel of higher education, it is a process worth celebrating. Because it is also labour-intensive, time-inefficient and messy, it offers an antidote to the brand of assembly-line education preferred by degree mills.

The thesis and dissertation serve as invaluable reminders that student engagement in the 21st century, regardless of level, has everything to do with helping scholars find authorship and agency. When students propose their topic, research it and ultimately defend it, they are more likely to experience the kind of ownership that produces the best, most accountable scholarship. Even removed from faculty guidance, extended independent research projects are valuable for the self-directed, iterative learning processes they cultivate.

However, the idea that a doctoral or master’s candidate could derive significant value from the dissertation or thesis even without faculty supervision raises the question of how meaningful, in practice, the role of director is. For example, graduate students often enrol in thesis or dissertation hours equivalent to a full-credit course, then re-enrol in those hours to work on their credit-bearing project for a second semester. However, if contact hours with supervising faculty members are minimal, as they often are, can we consider those credit hours well earned? The integrity of the thesis and dissertation process depends in large part on the activeness and availability of the director; if that director only meets in-person with their advisee once or twice a semester and again at the defence, have they not become mostly ceremonial figures, there to cement the illusion of academic legitimacy?

It isn’t that thesis advisers and dissertation directors are lazy, but that the paltry credit offered for their labours in the US system disincentivises true mentorship. While a graduate student may earn the equivalent of two full classes’ worth of credit for two semesters of thesis or dissertation work, the faculty member who signs on as director does well to earn a quarter of the credits they would otherwise earn teaching in the classroom.

Provosts may argue that because the thesis and the dissertation qualify as independent research, any credits awarded supervising faculty should reflect that supporting (read: secondary) role. Such logic works in cases where the graduate student is self-motivated and intellectually prepared, but it fails spectacularly in instances where the degree candidate needs significant retraining or academic acculturation.

The thesis or dissertation process works best with fully engaged supervisory faculty properly credited for their work. It is vital that director and degree candidate be realistic about what may be achieved in a given period of time. In my view, both grad students and their supervising faculty members are better served by shorter theses and dissertations, where greater focus and depth make for a truer simulacrum of life. PhD candidates will need to add chapters at a later date to merit publication by a scholarly or university press in any event, and the intervening years between the completion of the thesis or dissertation and its ultimate book publication serve to age the scholar and their scholarship. With candidates for advanced degrees getting younger, the postdoctoral years become more conducive to the maturative process that ripens the writer and refines the research.

For my money, the thesis or dissertation is best regarded not merely as pre-professorial training but as the beginning of collaborative intellectual enquiry designed to last a lifetime.

Zachary Michael Jack is associate professor of English at North Central College in Naperville, Illinois.

Register to continue

Why register?

  • Registration is free and only takes a moment
  • Once registered, you can read 3 articles a month
  • Sign up for our newsletter
Please Login or Register to read this article.

Related articles

Reader's comments (5)

These replies are far too long. The most important thing is to choose the right examiners.
These replies are far too long. The most important thing is to choose the right examiners.
And yet it is the university #managerialist parasites with no skin in the game who steal all credit and content of the supervisors and student when they effectively get paid bonuses for the number of completions and the supervisor gets zero. Then they push the supervisors out! At a #UniversityNearYou
Why no credit to the illustrator?
Replies are too long winded. Bullet points would suffice.