What Makes a Great PhD Supervisor?

By Michael Edmonds 10/11/2010 8


The PhD supervisor has an incredibly important role. Taking a student who has already demonstrated an aptitude (and hopefully a passion) for science, the PhD supervisor has three or more years to assist this student to grow into a knowledgeable, skilled and independent researcher. During the PhD there will be a focus on carrying out novel research and translating this work into publications, the ‘currency’ by which both student and supervisor will eventually be judged. However, I believe PhD supervision should be about much more than research and publications — it should be about the supervisor sharing their expertise and guiding the student in a wide range of areas.

The following is a list of behaviours I believe make a good PhD supervisor

1)   Discusses the students future with them

What does the student want to do when they finish their PhD? This is a conversation that is sometimes overlooked in the rush to get as much research done as possible. However, understanding what a students’ career goals from very early on helps the supervisor to identify opportunities for the student. Indeed, many students may not know about the many different career opportunities available to a PhD graduates other than the obvious jobs in academia and industry.

2)   Is upfront and honest about the students options and prospects

Different career goals may require different skills and abilities, not necessarily present in every PhD student. For example, to have a good chance at an academic position typically requires exceptional talent and a good publication record, and will often require a good postdoctoral position. In a student is considering such a career these facts need to be communicated clearly so the student knows what to focus on.

3)   Provides the student with opportunities for extending themselves

If a supervisor understands a student’s career goals they can identify opportunities that will allow the student to develop skills suitable for their goals. For example, allowing students interesting in teaching to run tutorials, arranging collaborations with industry, giving students responsibility for organising a group meeting. One area that I wish had been impressed on me during my PhD was the need to overcome my shyness and network at conferences. The misapprehension that one’s work will always speak for itself needs to be replaced with the idea that in science it is often who you know and not what you know that will open doors for you.

4)   Teaches the student how to communicate their research

While many students groan about having to attend and present their research at group meetings, it is vital that they learn how to present their own work clearly and confidently to a group of their peers. Group meeting also provide the opportunity to see where one’s work fits into the bigger picture, to work as a group to share and solve problems, as well as to develop interpersonal skills.

Just as important is allowing a student to develop their writing skills. I know some academics, in a rush to publish their groups’ research, may write all of the papers themselves. This deprives students of the chance to develop valuable writing skills. Also the correction of student work should not just involve correcting their writing but should be accompanied by an explanation of why the changes are necessary. Sharing the tricks of the trade around writing successful grant applications would also be desirable, although I do wonder if, in the current funding environment, whether some supervisors hold back in this area for fear of increasing their competition for funds.

5)   Teaches the student to be a truly independent researcher

Some might suggest that my earlier suggestions sound like a bit too much like hand holding, however, I believe getting alongside the student early on during their PhD allows the supervisor to steadily reduce the amount of guidance they provide to the student until they are truly independent researchers (surely one of the key aims of PhD research). This also allows the more experienced PhD students to mentor new PhD students in the supervisor’s absence.

I have derived this list from my observations of supervisor/student relationships over the past 20 years, as well as from my own personal experiences. Others may feel compelled to add their own suggestions, or perhaps to disagree with one or more of my points. I look forward to reading any comments.


8 Responses to “What Makes a Great PhD Supervisor?”

  • I’d add: Teach the student to do a thorough literature review! I encountered an example about a year ago in which a NZ researcher was excitedly announcing to the media a “new” discovery by his PhD student that is actually not new at all, but has been in standard textbooks since the 1970s at least. He also announced the follow-on work that his PhD student was about the embark on. The outcome of that research project is also already known, and has been in textbooks since the 1970s . 30 seconds on PubMed would have saved months of pointless work. I feel very sorry for that PhD student because they are having their time wasted by their supervisor. I also feel sorry for whoever is funding the research. Thanks to the Internet, it has never been easier to do a literature review, but some people apparently still don’t do them.

  • From experience, I would say (1), (3) and (4) are critical. After the first year I hardly saw my supervisor (he was either on sabbatical or looking for a better job in North America). For (1) I definitely needed to know what had to be done to achieve what I wished to achieve. I simply followed a path that I thought should work, but in retrospect one that could not succeed. For (3) and (4) the importance of actually selling your work cannot be understated, particularly if it is important but not fashionable. I was in the rather peculiar position of showing what is now in the text books is actually wrong. There are 63 different types of experiment that falsify it, the reviews that “settled the issue” ignored all of them, and my work. I made two critical mistakes, and paid for them. The first was that in the paper that I used to base the rest on, I left out what I considered obvious without realizing that the average chemist, having no formal training in physics beyond physics 1, would not immediately see that what I was doing would immediately follow from Maxwell’s electromagnetic theory, which is not exactly radical. I also got carried away with what I thought was a good approximation that actually permitted an analytical solution to the integrals. That doesn’t happen often in chemistry. What should have happened is that a supervisor should have made me present this in front of a physicist, not because I think it is wrong, but because it would have made the presentation more acceptable. Then a supervisor should have pointed out that if I wished to take this sort of thing anywhere, there was a specific route I should have followed, which would include conference presentations in North America. Nothing else but a solid sales program gets the unfashionable recognized, because it is only when a hundred scientists in the same room can’t find something wrong with your ideas that they start to take hold. If there is something wrong with them, you need to find out fast.

    • “because it is only when a hundred scientists in the same roomcan’t find something wrong with your ideas that they start to take hold. If there is something wrong with them, you need to find out fast.”

      Absolutely, this is why students should not shy away from giving conference talks or talking with other researchers at conferences. Giving them practice at presenting through group meetings is vital.

  • Nice post…
    In another life having taught some public speaking techniques, I make sure all my students (even summer students) get some public speaking training prior to conference presentation. There’s lots of simple tips that can give confidence (I’m happy to share them).
    I also encourage them to think about their science in lay terms… how do they explain it to their parents/peers etc. Given the public (mis)perception of science, I think that this kind of communication is a necessary part of the “training”. Perhaps they all need to write blogs! (not that I do that…. I’m amazed at the devotion some of you have to producing such thoughtful and regular posts).

  • As a student who has invested 4 years of my life in my PhD only to find out that life with the PhD is as it is supposed to be is practically impossible for my own personal circumstances, I can say that (1) and (2) are very important indeed. I was never informed about the grinding situation that you are in once you have the PhD; trying to get a postdoc and funding (which is only getting worse with the new changes in fellowship funding), plus the expectation that to get these you are willing to drag your spouse and kids around the world for these short-term projects of 1-3 years. I like science, sure I do, but I cannot expect my spouse to quit his job or my kids to change schools and maybe even countries just because its the only way I can carry on doing science. My own supervisor admits that the chances of a new PhD of getting a lecturing position are practically impossible. There are tons of people who already have a PhD *plus* many years of experience willing to kill for these positions……and now most PhDs only have contract work…..like and electrician! I wish I had thought it out better before starting. Instead, we worried about getting funding and scholarships, won some good ones and thought I had a good chance of becoming a scientist….or I wouldn’t be doing so well, would I?? (naive, naive!!)
    Now I realize I have been on a low-pay job for four years with no solid academic benefit, as I might not even get the PhD after all! I have to re-submit (after going back to the lab of course for who knows how many more months) but am feeling so bleak about it I cannot look at it even when I have nothing to do!
    Now I realize this was a BAD choice, and a BAD topic…..neither did I get the chance to meet anybody important in my field (3), and had very little of (4). I hope this may help supervisors tell the truth about becoming a scientists to new students, even though it may backfire on them…..the dilemma being that the want students to supervise, so they won’t tell them.

  • @paparalapapiricoipi

    Sounds like you have had a rough time – sorry to hear that. Don’t forget that a lecturing position is not the only option for a PhD. There is the option of working in industry for example.
    Are you in NZ or further afield? There are some funds for postdocs in NZ.
    It sounds like you haven’t received much guidance from your supervisor.

  • Proper supervision perfects students. Busy advisors often overlook this simple truth. A PhD research project is a long-distance run that requires a lot of motivation and support. The former can be instilled by the latter. All advisors: think about your time before you take on a research student.