Recently I wrote a letter of recommendation for a brilliant student looking to start working on a PhD project. We discussed how to choose a good PhD project, a reasonably good PhD supervisor and what to avoid. I was reminded of my very first day as a PhD student, when Stephen Stearns was meeting my supervisor. On hearing I had just started Stephen came over to my desk and pointed me to his very own advice to PhD students. While Stephen’s advice is still very valuable, it is geared towards a US PhD experience and it is for those who already have started. But what one is looking to start a PhD, what are the things to look for and to look out for? Here are some suggestions, which, despite being primarily for the UK system are probably general enough to anyone who is thinking on embarking on a PhD.
Before we go into the details of what might be around the corner for a prospective PhD student let me spend a couple of words on what a PhD student (in STEM research) should hope to gain. Working towards a PhD should (1) teach one to think critically and, most importantly, independently. It should (2) provide a clear understanding of the rationale of how ideas and hypotheses are tested and how to actually carry out the tests in practice. Finally, it should (3) impart a deep knowledge and understanding of the topic one is working on.
Given the above, the first thing to be aware of is that in the modern British academia these standards are lapsing, because of the never ending focus on money. Supervisors have less and less time and opportunity to actually mentor their PhD students, but are incentivised to actually take on students to do work for them. This particular problem is made worse by the practice, which is becoming more and more common, of transforming parts of a research grant into a PhD scholarship. The problem is that this 'transformation' is not really meant to fund a PhD student, but to find cheap labour to achieve the work described in the grant proposal — a ‘PhD student’ would in this case be a poorly paid technician who makes up for the loss of pay and the loss of mentoring and academic freedom for the privilege of getting a PhD at the end. These kind of deals are fortunately easy to spot: every time a PhD advertisement comes out where the applicant is expected to already have a great deal of technical skills (‘must be proficient in X, Y and Z’) to analyse a very specific set of data, often already collected by a third party, with a very narrowly defined research question, people are fundamentally saying: ‘we need a skilled and experienced professional to do work we are actually not able to do, but we are trying to save and thus unwilling to pay an appropriate salary for it, and we also do not like to admit the limits of the skills present in our group, so we are advertising an 'exciting PhD opportunity’ hoping that a hard working and already very qualified student who does not already posses a PhD will fall for it’. While it is true that the work coming from a successful PhD will have a beneficial effect for the work and career of the supervisor, when people are so brazen in stating that you will be treated as (skilled) cheap labour you might want to avoid working for them in the first place.
What should a prospective PhD student look for in a good PhD position then?
a) a degree of freedom to explore what might be the most interesting aspect of the PhD topic that *you* think is important, both in terms of intellectual discussion and actual work. The whole point is your intellectual and professional growth, not meeting some externally decided targets along narrow, predetermined analysis routes.
b) a supervisor that is able and willing to carry out the supervision. In the real world this ability is not a clear cut property: your supervisor might have little time *personally*, but you might be part of a research group where you can always get proper and timely help and mentoring. As long as your supervisor, other faculty members and postdocs are actually helping, teaching and mentoring you it does not really matter who does what. On the other hand, if nobody does supervise you, you are without guidance when you need it and that will make your work much harder and stressful. In addition, even if your supervisor successfully farms out your actual supervision to a third party, your supervisor still plays a big role in *deciding with you* what to do (NOT ‘telling you what to do’), so you do not want your supervisor to hinder your work by failing to be up to date with your work and not knowing what you are doing.
c) there is (NOT 'should be’) a budget that will actually cover stuff like publication charges, travel expenses, conference fees and proper good equipment.
When I was a PhD student about 50% of PhD candidates in the UK would quit before finishing, because their supervision was so dreadful. Universities, which are big bureaucratic entities, reacted as big bureaucratic entities do: by forcing red tape on the PhD students. I am the last generation of students who started a PhD and then either finished it or not. Now you will be subject to a number of ‘upgrades’ and further requirement that are meant to force your supervisor to make sure you do work according to some sort of timetable and you also acquire a set of skills (ideally ‘transferable’) to show that you are employable within or outside academia. In practice the fulfilment of these requirements will be immediately passed on to you (‘make sure you go to enough postgrad classes to get enough gold stars’, ‘make sure you meet whatever deadline’). Do try and make the best of it: some of the postgrad courses are actually very useful, and being forced to stick to some sort of deadlines might actually help you organise your work — overall do be aware that you will have to meet these requirements and use them to your advantage to help you focus your work, stay on track and obtain timely feedback from your supervisor(s).
Do not forget that supervisors do not cease to be people because they have students to look after, thus:
d) looking after you comes after looking after children and spouses, and after their health — in fact anybody neglecting their family for you is almost guaranteed not to be a good supervisor: students need to become independent scientists and taking a protective and paternalistic approach to supervision will not help anybody.
e) even the best supervisor will have time and resource allocation conflicts that might not be resolved in your favour all of the time.
f) if you are doing cutting edge stuff it is likely that *you* are the world leading expert on what you are doing, and when you need help or advice, the best you might get is a well informed and experienced ’sounding board’. There will be times when even the best and most helpful supervisor will not be able to answer your questions or help solve your problems.
Let me finish with two more points that also need spelling out. You have every right to have expectations from your supervisor, but this right is based on the fact you work hard, and are proactive in both your work and your learning. Do not ever take the approach that if something that you need to do is difficult or boring you should find a way of dumping it on somebody else, or getting your supervisor to do the difficult stuff for you. You are meant to become a professional and independent researcher, and as such you should be able to handle the endless stream of work that is difficult or boring *by yourself*. This is particularly important because one of the difficult stuff in science is *thinking and coming up with good ideas and solutions to problems*. If you go back with your tail between your legs to your supervisor every time things get hard, your supervisor will start doing the thinking for you, and you will learn very little. Do not do that.
Finally, choose your supervisor wisely, if for no other reason than that your supervisor will provide references for you at the start of your career, and you need those references to be good. In fact, in an ideal world, first choose a supervisor, then discuss a possible project you’d like to do, let the would-be supervisor find the money and finally get started. Given that this is not the ideal world, chances are you will not be in a position to do what I suggest, but do not forget that one of the most important parts of your PhD is getting proper supervision, so do look into that very very carefully. Your PhD experience will almost surely be harder than you imagined, even under ideal conditions, but by keeping the above in mind you should get the most out of it. And remember that a PhD is a test of perseverance.
Before we go into the details of what might be around the corner for a prospective PhD student let me spend a couple of words on what a PhD student (in STEM research) should hope to gain. Working towards a PhD should (1) teach one to think critically and, most importantly, independently. It should (2) provide a clear understanding of the rationale of how ideas and hypotheses are tested and how to actually carry out the tests in practice. Finally, it should (3) impart a deep knowledge and understanding of the topic one is working on.
Given the above, the first thing to be aware of is that in the modern British academia these standards are lapsing, because of the never ending focus on money. Supervisors have less and less time and opportunity to actually mentor their PhD students, but are incentivised to actually take on students to do work for them. This particular problem is made worse by the practice, which is becoming more and more common, of transforming parts of a research grant into a PhD scholarship. The problem is that this 'transformation' is not really meant to fund a PhD student, but to find cheap labour to achieve the work described in the grant proposal — a ‘PhD student’ would in this case be a poorly paid technician who makes up for the loss of pay and the loss of mentoring and academic freedom for the privilege of getting a PhD at the end. These kind of deals are fortunately easy to spot: every time a PhD advertisement comes out where the applicant is expected to already have a great deal of technical skills (‘must be proficient in X, Y and Z’) to analyse a very specific set of data, often already collected by a third party, with a very narrowly defined research question, people are fundamentally saying: ‘we need a skilled and experienced professional to do work we are actually not able to do, but we are trying to save and thus unwilling to pay an appropriate salary for it, and we also do not like to admit the limits of the skills present in our group, so we are advertising an 'exciting PhD opportunity’ hoping that a hard working and already very qualified student who does not already posses a PhD will fall for it’. While it is true that the work coming from a successful PhD will have a beneficial effect for the work and career of the supervisor, when people are so brazen in stating that you will be treated as (skilled) cheap labour you might want to avoid working for them in the first place.
What should a prospective PhD student look for in a good PhD position then?
a) a degree of freedom to explore what might be the most interesting aspect of the PhD topic that *you* think is important, both in terms of intellectual discussion and actual work. The whole point is your intellectual and professional growth, not meeting some externally decided targets along narrow, predetermined analysis routes.
b) a supervisor that is able and willing to carry out the supervision. In the real world this ability is not a clear cut property: your supervisor might have little time *personally*, but you might be part of a research group where you can always get proper and timely help and mentoring. As long as your supervisor, other faculty members and postdocs are actually helping, teaching and mentoring you it does not really matter who does what. On the other hand, if nobody does supervise you, you are without guidance when you need it and that will make your work much harder and stressful. In addition, even if your supervisor successfully farms out your actual supervision to a third party, your supervisor still plays a big role in *deciding with you* what to do (NOT ‘telling you what to do’), so you do not want your supervisor to hinder your work by failing to be up to date with your work and not knowing what you are doing.
c) there is (NOT 'should be’) a budget that will actually cover stuff like publication charges, travel expenses, conference fees and proper good equipment.
When I was a PhD student about 50% of PhD candidates in the UK would quit before finishing, because their supervision was so dreadful. Universities, which are big bureaucratic entities, reacted as big bureaucratic entities do: by forcing red tape on the PhD students. I am the last generation of students who started a PhD and then either finished it or not. Now you will be subject to a number of ‘upgrades’ and further requirement that are meant to force your supervisor to make sure you do work according to some sort of timetable and you also acquire a set of skills (ideally ‘transferable’) to show that you are employable within or outside academia. In practice the fulfilment of these requirements will be immediately passed on to you (‘make sure you go to enough postgrad classes to get enough gold stars’, ‘make sure you meet whatever deadline’). Do try and make the best of it: some of the postgrad courses are actually very useful, and being forced to stick to some sort of deadlines might actually help you organise your work — overall do be aware that you will have to meet these requirements and use them to your advantage to help you focus your work, stay on track and obtain timely feedback from your supervisor(s).
Do not forget that supervisors do not cease to be people because they have students to look after, thus:
d) looking after you comes after looking after children and spouses, and after their health — in fact anybody neglecting their family for you is almost guaranteed not to be a good supervisor: students need to become independent scientists and taking a protective and paternalistic approach to supervision will not help anybody.
e) even the best supervisor will have time and resource allocation conflicts that might not be resolved in your favour all of the time.
f) if you are doing cutting edge stuff it is likely that *you* are the world leading expert on what you are doing, and when you need help or advice, the best you might get is a well informed and experienced ’sounding board’. There will be times when even the best and most helpful supervisor will not be able to answer your questions or help solve your problems.
Let me finish with two more points that also need spelling out. You have every right to have expectations from your supervisor, but this right is based on the fact you work hard, and are proactive in both your work and your learning. Do not ever take the approach that if something that you need to do is difficult or boring you should find a way of dumping it on somebody else, or getting your supervisor to do the difficult stuff for you. You are meant to become a professional and independent researcher, and as such you should be able to handle the endless stream of work that is difficult or boring *by yourself*. This is particularly important because one of the difficult stuff in science is *thinking and coming up with good ideas and solutions to problems*. If you go back with your tail between your legs to your supervisor every time things get hard, your supervisor will start doing the thinking for you, and you will learn very little. Do not do that.
Finally, choose your supervisor wisely, if for no other reason than that your supervisor will provide references for you at the start of your career, and you need those references to be good. In fact, in an ideal world, first choose a supervisor, then discuss a possible project you’d like to do, let the would-be supervisor find the money and finally get started. Given that this is not the ideal world, chances are you will not be in a position to do what I suggest, but do not forget that one of the most important parts of your PhD is getting proper supervision, so do look into that very very carefully. Your PhD experience will almost surely be harder than you imagined, even under ideal conditions, but by keeping the above in mind you should get the most out of it. And remember that a PhD is a test of perseverance.