HomeAboutSoftwarePublicationsPostsMicroBinfie Podcast

Advice for PHD students and tips for the rest of us

Posted on July 5, 2022
an AI generated picture (Midjourney) with prompt; 'scholarship'. You can share and adapt this image following a CC BY-SA 4.0 licence.

When I was a doctoral student, I actively asked anyone I could for advice about academia and being a scientist. It has been seven years since my doctorate was awarded and thinking back, these were the five single most useful quotable pieces of advice I received. These were either advice given directly given to me by my professors, or recounted by fellow students. Many of these topics are very complex. I will touch on them only briefly here.

These are not the most important things I learnt - but those cannot be distilled into a single quote. Hopefully, these single statements will help you.

Keep firm boundaries

You're here from 9 to 5 and if you're not in the lab you should be reading papers.

This was a relayed to a friend on their first day. This was said to them as a hard reminder that a PhD candidature should be treated like a job - where there are contracted hours of work. I believe such boundaries are important to understand if you are overworking yourself and neglecting other parts of life. If you follow the nine-to-five advice above and make it clear to your team that is how you work, then no one (especially you) can say that you have not applied the effort. During the work period, you can focus on the tasks at hand. In your off-period, you can attend to other matters and relax guilt free.

The exact timings may be different for different people and will change over the course of a candidature. Have a regular dialogue about working habits and timings with your advisory team. Be deliberate with your working hours. Manage expectations of when work will be completed, given those timings. Sure, you can work after hours for the final push of a manuscript or a set of experiments, but if this workload is very demanding, then consider how to balance this later. Perhaps you work Sunday morning, so you can have a later start on Monday. Or perhaps you work hard for a few weeks, so schedule easier work (or adjust deadlines) for the week after so you can recover.

Remember why you do what you do

The P in PHD means Philosophy, so philosophize.

PhDs start with a single broad question. As we delve deeper into a PhD, we become more focused on details that serve answering the broader topic rather than directly working on the broader question itself. This is necessary to perform tractable experiments with the project's time frame. PhD candidates can spend a lot of time learning all the different techniques to collect data, which are the best methods for a situation, and how to optimise different methods in our research. We usually apply these methods to one specific facet, which is supposed to act as proxy for the question at large.

But this leads to a common trap when a candidate writes their dissertation. They frame their thesis - the singular idea present in their work - with a narrow view that fits within their day-to-day experience. Having looked so closely at the matter for so long they cannot step back and contextualise their project in the scope of the original, lofty research question.

Remember, work exists in a wider dialogue about a science and in the philosophy of a discipline. It is in this on which to ruminate in the dissertation. It is important to talk to other people outside your immediate peers and try to explain your project so that you maintain the perspective.

Simplify your message

In a 15 minutes talk, present just one interesting idea

A professor once explained to us how to structure a scientific talk. Most talks are around 10-15 minutes. He told us flatly that this is not a lot of time to explain an entire project. His approach for a presentation was to pick a single exciting and new result (presented as a figure on one slide) and then spend the rest of the slides and time explaining that figure.

It is easiest to frame your result as “if this then that” or “in this group we see this and in that group we see that because of something else”. Cause and effect. Or compare and contrast. This is much easier for an outsider to understand because, while they might not understand the nuance, they can appreciate that something is changing and there is a reason for it. If your result is “Upregulation of gene X in species Y causes phenotype Z”. You would have One (or two) slides showing this result directly. But in presenting those two slides alone, the audience has no context. You must contexualise it by explaining the following::

  • Introduction: What is species Y?
  • Introduction: What is gene X?
  • Introduction: What is the phenotype?
  • Introduction: Why is this important?
  • Introduction: How does this fit with our wider understand? (You may couch this in wider context of your PhD project or the field generally)
  • Aims: This isn't a question but you may want a slide to sum all the motivations for the work (i.e. state your aims)
  • Methods: How did you achieve this result? What was the method (briefly)?
  • The result
  • Discussion and conclusion: What are the limitations or caveats to the results and methods?
  • Discussion and conclusion: What's the implication to what we know already and what are the next steps?

Answers for each of these questions could each be a slide. And these answers, with the result itself, will fill your time.

Anything that does not directly relate understanding the figure can be cut, and usually has to be cut, to make the time limit. The worst thing you can do in a presentation is go over time.

Make time your ally

Have you tried time management methods (like Pomodoro)? I finished off three manuscripts that I have been stuck for ages, in three months.

One of my professors pointed out the value of time management. These days, there are infinite distractions with people and busy work vying for our attention. I feel time management is crucial in a PhD candidature, where you are often left to manage the day-to-day of the project. You are, in effect, a project manager. And there are techniques from project management that you can use to help. I want to mention Pomodoro as one of those techniques of being able to focus on a task and just get it done.

Pomodoro is a specific time management system, where it breaks work hours into two-hour blocks where you work solely on a single task for 25 minutes (no interruptions allowed), then break for five minutes. You can opt to continue that task or switch to something else. At the end of two hours, you take a longer break and repeat the process. This not only keeps you focused, but having the breaks embedded allow your to give yourself a rest so that you can sustain working for longer. There is a larger philosophy to the system. For instance, you can use the number of Pomodoros to complete a task to schedule how long it can take, then adjust deadlines according. Over time you can review the amount of work you have done and review where your time is being spent. And when you feel you have not done enough work in a week, you can look at your Pomodoros and know that is not true.

I find it works well for writing and administrative tasks, but not so much when you are a deep thought like programming or deep writing. Those tasks require a certain momentum (casually I would say that you need to be “in the zone”) that the Pomodoro timings disrupts.

Time management is an entire discipline with many other techniques. The field of study is well worth exploring.

Perfection is the enemy of good

Have you heard the 90/10 rule? 90% of the work gets done in 10% of the time and that last 10% of the work takes 90% of the time.

I once attended a talk from a professor who was giving career advice to our student cohort. As a self-proclaimed perfectionist, he explained that he had come to accept the “90/10 rule” with reluctance. This realisation came early in his career as he was engaged in a number of different concurrent projects. The 90/10 rule basically posits that the time to elevate something good to excellent can be an onerous task. We do not have infinite time to spend on a single piece of work. The more time we spend on one work, the less time we have for the others. You need to consider what you will accept as an acceptable first iteration of the work and how much time you can afford. If you can complete this with time to spare, then iterate again to make it better. Setting personal deadlines for tasks allows us to juggle all our different commitments and ties back with better time management. Remember, in the long run, it is better to have a few competent outputs rather than one perfect one.

Questions or comments? @ me on Twitter @happy_khan

The banner image is an AI generated picture (Midjourney) with prompt; 'scholarship'. You can share and adapt this image following a CC BY-SA 4.0 licence.