Priorities in a PhD: techniques vs topic

Question

Trying to decide between PhD offers got me thinking about the goal of a PhD. I don't mean in a general sense (see here for an example), I mean about which area of research to prioritize. Of course, ideally, in a PhD you'd have an interesting topic AND you'd be using techniques you're interested in. However, if given the choice, I would choose to focus on mastering a technique I'm interested in (which can later be applied to other topics), rather than focus on a topic I'm interested in (where different techniques can be used later).

My thinking may be naive, but I believe if you gain the necessary experience with the techniques you're interested in, it will be more useful in the future than to specialize in a niche topic because changing topics after the PhD is easier/more common than finding a job using a technique you may be interested in but have limited experience with.

So I was wondering which made more of a difference in your experience: the techniques (lab skills, programming skills) you learned during the PhD or the (sub)topic you specialized in?

Answer 1

You should be looking for the thing that glimmers. I know that may seem a bit weird. But you should be looking for the thing that sparkles and engages you. Whether it's a topic or a method or a question, it should hold you rapt when you think about it. It should be something you would be working on in your spare time even if you were not able to do a PhD in it.

You are going to be spending some years on this. Usually at least three years, often more. And usually it means your future will at least to some extent be influenced by the thesis.

So the thing you do in your thesis should hold your attention without you having to do anything about it. Don't be especially concerned about whether you are concentrating on topic or technique. Look for the glimmering, the glistening, the things that fascinate you. If you can find those then you will not have to force yourself to work. You will just want to work because the harder you work the more fun it will be.

Answer 2

However, if given the choice, I would choose to focus on mastering a technique I'm interested in (which can later be applied to other topics), rather than focus on a topic I'm interested in (where different techniques can be used later).

You haven't named a field (which is fine), but in mine, certainly both of these types of PhD students exist, those that become experts in technique X and those that become expert in sub-field y. The grad students in my department are perhaps one-third "technique experts," with the faculty somewhat less.

My thinking may be naive, but I believe if you gain the necessary experience with the techniques you're interested in, it will be more useful in the future than to specialize in a niche topic because changing topics after the PhD is easier/more common than finding a job using a technique you may be interested in but have limited experience with.

I am not so sure - although this would of course differ by fields - and at any rate, I doubt the correlation is high enough to plan your PhD around (that is, the idea that the more of a technique-expert you are, the higher your chance of getting a job).

For me, when you say "technique," I think "statistical technique," and unless you are in a stats department, you aren't going to graduate without a substantial contribution to the field you're "in".

In summary, becoming an expert in applying technique X to field Y is just as valid and needed as becoming an expert in sub-field y, and I would echo Dan's advice to follow what you like best.

Answer 3

Most people learn best (and retain knowledge best) by doing

It is certainly valuable to develop your base skills in your field during a PhD candidature, and so the more you can develop your "techniques" in a broad range of areas the better. Many people find that they are better able to develop their knowledge/skills/techniques (and retain this over the long-term) when this is done in the course of solving some applied problem of interest to them. You might be different to this, so develop your knowledge/skills/techniques in whatever way suits you best. In any case, you are certainly correct in your observation that developing your core skills is useful in the long-term, and possibly even more important than mastery of the particular topic on which you do your PhD dissertation.

In regard to what is required for a PhD program, one of the most important elements of research is that you are able to take a problem and then investigate that problem using whatever skills/methods/techniques are necessary. This may involve learning new things and it may require a great deal of tenacity in dealing with topics and methods that are unfamiliar to you at first. Consequently, there is a kind of "meta-learning" that occurs in a PhD candidature, where you learn to learn. Of course, this is something that people outside a PhD also do, but in a PhD candidature and other academic research activities it is particularly pointed, since you come into new topics where you have to go to a high level of mastery in a relatively short time.

It is worth noting that the output of your PhD candidature will ideally be some research that is valuable in its own right and not merely an exercise to strengthen your skills. The usual expectation in a PhD candidature is that you will produce some research that is sufficiently novel and valuable that it can be published in a scholarly journal, where it can contribute to the totality of scholarly knowledge of the world. So while you are correct that your skill development may be more useful to you in the longer term, you are generally still required to produce something in your topic area that is valuable in its own right. Most people learn and retain knowledge best when it is developed in the course of solving a problem of direct interest to them, so there is often a natural symbiosis between progression of a PhD dissertation in a topic area, and general development of knowledge/skills/techniques, and retention of that knowledge.