Monday, 16 April 2007

Why you shouldn't do a PhD. Ever.

Have a look at this. And this. Those should both be very familiar to anyone who's done a PhD. If they make no sense, then hopefully this post will clarify. It is a long one, you have been warned. I'd get a cup of tea first.

Having read Craig's blog, I got thinking about why it's so hard to do a PhD. Suprisingly, it's not the complexity of the material, it's not the hours of work (compounded by the view of most people, who seem to think you've managed to get another 3 years of student dossing), nor is it coming up with original ideas and exploring the boundaries of your subject.

The difficult bit is when (approximately 2 years in), you realise you hate your project. You're disillusioned, tired, and want to do anything else but continue. The problem is that you've now invested two years of your life, and can't just give up.

So why does this happen? It's not just me who's been there; the experience appears to be standard, and the timing is remarkably similar (certainly amongst pure mathematicians).

One of the hardest things about research is the complete lack of punctuation. As an undergraduate, you have lectures, exams, stop. Lectures, exams, stop. Repeat a few times then graduate. There are constant targets, goals and feedback. You know what's coming next, know what you did wrong last time, and can learn from your mistakes and improve on your performance. You have mates doing the same course who are stuck with the same things. You can get help from them, and feel good about helping others.

There's virtually none of that in research.
There's no punctuation. From your first year report to the submission on your thesis (typically 2.5 years or so), the biggest comma that gingerly rears it's head is merely the gentle drift from concentrating on research to concentrating more on typing it all up.
In terms of emotional support and goals: If you're lucky, you'll have a weekly meeting with your supervisor (who, by this stage knows less about your research than you do) who will offer suggestions on directions you could take, when you should start writing up, and go through a few specifics. If you're very lucky, you'll also have some mates working in loosely related fields who you can have a moan with, and throw ideas around with (not on the hard stuff, as you'd often be there all week (literally) explaining the background to it. Apt comparisons would be that you're the only person in the room learning calculus but you can only ask when you get stuck adding up, or that you're writing an essay, but can only ask your mates about the spelling and grammar). Your peers are distanced from you due to the huge specialisation required to get a project completed in three(ish) years.
I didn't realise how much I'd missed feedback until my examiners gave their opinions on my thesis (they were extremely complementary, in case you were wondering). Nige obviously felt the material was good, and had made this clear, but having some outside feedback is a very different thing psychologically.

This all leads to a mounting sense of uncertainly. It's difficult to keep slogging away when you're always wondering:
For some of these questions, you won't get an answer until you submit, for others you can only get an approximate answer from your supervisor. Then there are some for which you just have to do the work to find out.

I didn't realise as clearly as I could have what was happening, didn't recognise how disillusioned I was, and ground to a halt roundabout Christmas '05. I attempted to slog through by putting more work in when, with the benefit of hindsight, I can see now that I'd have been better off taking a good break for a month.

So how did I do?
  • Stuff I got right: Stuck with it, kept complaining about it to anyone who'd listen (kudos to parents here, and to everyone I've played hockey, badminton, sung or drunk beer with), and most importantly, kept on doing other things (mostly hockey, badminton, singing and drinking beer) that I enjoy. Exercise is good too: I found that I got far more work done when I was cycling to and from uni than when I got the bus.
  • One I wish I'd done more of: Given talks. I hate the idea of doing talks, but actually doing them is suprisingly ok. They're too easy to avoid doing. The highlights are giving a talk on my thesis (an exercise I'll repeat in Leicester in a couple of week's time), and a short series of teaching talks on Spectral Sequences (very complication computing machines) to the then junior topologists. Why? Feedback and interaction. Talking about my thesis generated some much needed external feedback. Teaching other postgrads some complicated material has several benefits: I learnt a lot writing the talk, I felt good about helping others, and I felt good about interacting with other students - it eased off the isolation more than I realised at the time.
  • One I got right eventually: Took time off to do something different. I had a better view of what I'd achieved after a break. Taking a month out to earn some money felt like a holiday!
  • One I wish I'd thought about much earlier: Punctuation. I'm planning on cycling to Swansea when I've finished my corrections. Purely because it'll then actually feel like I've finished.
Well done for reading this far. Hope this is helpful to someone, or at least gives you some understanding of what I've been whinging about for so long. So remember kids: just say NO to postgraduate study!

That said, in a year's time when the dust has settled and I can't remember any detail of what I've been working on for all this time, I know I'll be glad I did it.


beans said...

Why does this make me want to do postgraduate studies? I'm a thicko, but I shouldn't be that thick as to not understand what NO to postgraduate studies is!

(Ok, I'm slightly fifty fifty now but I've still got plently of time to ponder on this :D)

Were you doing the trick where when you say 'No' you really mean yes (reverse physcology?).... it's late!:o

Adrian said...

Heh. Nice to see I'm stealing Craig's readers! Not really reverse psychology, I was just attempting levity.

By the way, if you're considering a PhD (I've no experience of Masters, though I think it's more like a harder & more specialised undergrad degree), you'll need to do the following:
1) get a first. You need this (in pure mathematics, anyway) to stand any chance of getting funding.
2) really love your subject. Motivation and drive are, in my experience, the hardest part, so this is vital. It also helps with getting the first!
3) be very stubborn.

I'd suggest looking for two things:
1) a PhD supervisor you think you'll get on with, and
2) somewhere with plenty of other postgrads, preferably working in similar fields.

I've been lucky enough to get/have/be all these things, and am not sure I'd have managed it without them.

Craig said...

Hi Adrian,

I think one of the hardest things is persuading yourself to have a break. When you're in the thick of things and utterly stuck, my instinct is to explore everything twice before I throw in the towel. But of course, that's where the disillusionment, frustration etc. comes in...

Hopefully in my final year I can now see it coming a little easier.

Great post by the way, I'll stick a proper link on my blog in case anyone missed the comment as I think it's something all (potential) graduate students should read.

Pete said...

Those comics are great and accurate. Anyone considering a PhD should be made to see them.

I agree with almost everything you've written.

Adrian said...

I've just added another comic (What should I be doing next?). I have to ask though: "almost everything"? I'm curious which bits you don't agree with.
I actually considered not putting the comics in, as I'm sure most PhD students will read the lot in a fit of procrastination. Anyone considering a PhD should read them before they start and get it out of the way!

Rosy At Random said...

Of course, none of that applies to *me*!


I *will* be learning so much new stuff that I'll probably find it quite hard to get bored.

Adrian said...

"I *will* be learning so much new stuff that I'll probably find it quite hard to get bored."

I was too. During my first year. I can wholeheartedly recommend the first half / first two years of a PhD to anyone. Loved it.

I appreciate that the arrangement of your PhD will be slightly different, with the additional (Masters style?) first year, but I think this stuff will become relevant to you too. Sorry.

Give us a shout when you want some tedious advice and/or a beer!

The One Who Missed The Party said...

Whoa! Dob-log overload!

Nice post, even if it's not exactly happy happy joy joy.

You'll be bloody glad you did it.

All I managed was a research-based MEng, so technically not quite an MSc*...but there were definitely those 4am** thoughts along the lines of "Is it really worth it?"

* Although we shared all our lectures with the MSc students, and were even given the same list of projects.

** Often, this coincided with (a) illegally staying in the lab overnight, (b) propping my eyelids open with matchsticks, and (c) inhaling unhealthy amounts of solvents.***

*** Actually, the thoughts that sprung from (c) were usually considerably more entertaining. Like when I started talking to my little test tubes individually.

Pete said...

I Probably agreed with all of it, but I only read it once and I don't trust my memory to remember every bit, so I put almost all as a kinda disclaimer.


I'm off to go and look at that other cartoon now that I've finished writing up a proof I've written up four times already.

beans said...

Lol, I'm everywhere me! wooohh (OK, my casper impersonation sucked!)

Well it's still early early days (only a 1st year), but thanks for your advice. Although, I'm not sure whether a first is within my grasps, I'll definitely aim towards it. :) Is funding really neccessary? I mean can you do a phd without funding (and a first)?

Well I think I love maths - but yeah I suck at the whole motivation thing! (stubborn eh- in what sense?):p

I'm going to first look at doing a master maybe- might help prepare me for the whole phd thing. Looking to do it at Manchester as well!

(This feels weird!)

beans said...

Goes with the post I reckon! :o

Pete said...


Unless you're independently wealthy, you will certainly need funding. Happily the university will often sort this out for you (This is not always the case and usually arrives along with the offer, if it is)

As for entrance requirements, a first is not the be all and end all - but it very nearly is. I know people with worse results on a PhD, but they're in the minority and very often had to fight hard to get one!

definitely do a masters if you want to do a PhD.

Adrian said...

Beans: The main thing with funding is that you've got to support yourself for three years (minimum, more than three years for mortals) without the lengthy summer holidays you get as an undergraduate where you can gain plenty of cash.

The easiest way to do this is to get an EPSRC (or equivalent) funding, but you'll (almost certainly) need the first as there is a finite amount of money to go around. This will be different for engineers, bioinformatics, etc PhDs, but for some reason they're better at attracting industrial sponsorship than pure maths...

I do have a friend who's been offered a PhD place, but no funding at present. She's looking into supporting herself with career development loans and part time work. There are other avenues of funding, but I've never had to look into these.

With regard to Masters degrees, I did one of those MMATH (four year undergrad) thingies, which I think benefitted me, but there were fewer finance issues then. You might want to have a look at Craig's blog, as he went from a BSc to a PhD with no MSc, and seems to have managed fine.

Most important thing is to work hard, and see if you're still enjoying maths more and more in a few years time.