Career advice: how to supervise a PhD student for the first time

Starting out
Supervision will give you a chance to share the accumulated wisdom of your own PhD journey and anything else that has followed. However, you need to start at ground zero with each new student to help build a shared sense of what good practice looks like. 

A good first step is for both of you to take a small batch of seminal papers and agree to read them before swapping notes. This simple exercise will allow you the chance to demonstrate how to scrutinise the key ideas, assumptions, limitations and contributions that each author or authoring team makes in its paper. Doing so in the style of a collaborative, worked example will help to set a particular tone that will pay rich rewards in the months and years ahead. 

Being clear about the level of depth and the practicalities of note taking is as important as showing how you approach the basic task of getting to grips with the literature.

Give the feedback you wish you’d received
Bemoaning the failings of your supervisor represents one of the most common ways of establishing rapport among a group of doctoral students. “They’re never there”, “they don’t give detailed comments”, “they’re always in a rush” and so forth. Each new supervisory relationship, however, represents your opportunity to break the cycle.

Recall your own anxieties and needs as a PhD student and try to offer your new student the kind of supervision that you wish you had received. Draw on your own supervision experiences, whether these were of being micromanaged or of Zen-like levels of uninterest. These formative experiences probably mean that you know what you should offer to your new student. Be bold and strive to provide the right balance between nurturing and challenging. You’ll also need to balance the other demands that arise in modern academic life – maybe you’ll find yourself reflecting on the reasons that your supervisor was always in a rush.

Beware unrealistic expectations
As a new supervisor, one of the worst mistakes you could make would be to overlook quite how inexperienced you were as a new PhD student. Unfortunately we tend to airbrush out our early, bumbling incompetence and concentrate on the latter-day, polished professionalism that we now exhibit. 

Do not set supervisory expectations around the version of you that completed your own PhD some time ago. Rather, set them at the more modest level of the version of you that started your PhD journey even longer ago. Visiting unrealistic expectations on your new student is a recipe for unhappiness. You’ll be disappointed; they’ll be confused.

Be patient, supportive yet demanding
Newly qualified supervisors can be among the most demanding because they remember the intensity of writing up and preparing for a viva. Having recently watched their own work being subjected to unforgiving scrutiny in the context of a viva, new doctoral graduates can, in turn, impose demands when they come to supervise and/or examine. 

However, a PhD is more expedition than sprint. Try to remember this, particularly in the early months, because your new student will no doubt experience plenty of false dawns and blind alleys as they grapple with the literature, realise that accessing data might be tricky and worry about their methodological preferences.

Simply being there and empathising is not enough either. You face the particular challenge of finding the right times and the right issues over which to demand higher quality work than your student feels that they can produce. Done well, later this will be recounted as providing inspiration. Handled badly, you will be seen as the uncaring taskmaster who made the whole thing unnecessarily tense. 

Notice your own foibles
It is natural for us to develop particular quirks and preferences in our reading, reviewing and supervising. As you offer feedback on written work, draft presentations, posters and the like, see if you can spot common themes. Once you know the common themes, it is incumbent on you to offer some exemplars when students ask the not unreasonable question, “so what would good look like”?

Cultivate a little stockpile of excellent literature reviews, contribution statements and analyses of data. Have these to hand and offer them as a complement to the red ink in your feedback. 

These examples don’t have to be in exactly the same subject area, methodological tradition or empirical context. Indeed, it may be helpful if they aren’t. They don’t even have to be particularly contemporary. But you should be able to play the Graham Norton role while fronting the imaginary TV show called The Top 10 Research Papers Ever. 

Robert MacIntosh is head of the School of Social Sciences at Heriot-Watt University. He writes regularly about academic life on the Heriot-Watt blog thePhDblog.com.