Yesterday at KIPAC tea, a graduate student wished everybody “a happy Mardi Gras”. This lost me, but later it made sense; when I got home to Little Britain there were pancakes for tea, for lo, it was Pancake Day, a.k.a. Shrove Tuesday. The pancakes were scrumptious but not enough for my hungry kids. Grad students are possibly even more voracious than teenagers, but luckily at KIPAC tea there is a goodly supply of donuts. This is an essential part of a PhD education of course; supply stodge and the students will turn up to the talks, absorbing some knowledge along with the sugar.
Is this investment worth while ? Over at the PeterBlog, Professor C worries that we are over producing students, because only one in ten can become an academic. Actually, without this overproduction, Peter and I would be out of a job. Why does the Government pay for astronomers ? Because we do have an economic impact. Our product is people.
We all accept this where undergraduates are concerned. Half of our job is teaching; we do our bit for the academic sausage machine, churning out the scientically literate workforce our society needs. That makes it economically feasible to pay us for the other half of our job, scientific research. We tend to think of PhD students as part of this second world; our apprentices, each producing their magnum opus after three years. But that world has long gone. The PhD is an advanced training degree. A decade ago, while I was on PPARC Council, studentships were doubled. Why ? Because we were not getting enough astronomy done ? Nope. Because the captains of industry said “we can use more of these” and yea verily the Treasury purse strings were opened.
So the donuts make sense.
Mostly this is good. The Government is happy to continue paying for pure science, because they can see that every pound spent puts three back into the economy. But they must be thinking “hmmm.. wonder if we can squeeze six out of them ?” They already did this to us on undergraduate numbers. Postgrads could be next.
The e-Astronomer 
Andy,
What you say is fine as long as the students are told in the first place that the vast majority are expected to leave research when they have their PhD. Otherwise it’s just a swindle.
And I hope this “advanced training” they get as PhD students at least teaches them how to spell “doughnut” properly.
Peter
I don’t think it is a swindle, Peter. They know that not all PhDs will get jobs in academia – but they hope they’ll be the one out of ten who do. Better odds than the lottery and it is largely up to them whether they succeed or not.
Thing that interests me though, Andy, is, if government reckons your teaching role is the key part of your job, why do they put all the emphasis on research in rating universities and largely ignore teaching? Why is someone who focuses on teaching unlikely to get promoted (see recent THE articles)?
I think it is made clear. Certainly when I was applying for Ph.D. and just afterwards at the PPARC introductory course, we had talks about careers after finishing the Ph.D., by people in other scientific areas (in industry etc), one who had done a couple of postdocs and then went into banking (derivatives) – great salary, but she burnt out after a few years. Anyway, I think I got plenty of warning then, which was 10 years ago.
Let’s get things in perspective… when I started my PhD, in ’89, the prospects were worse than they have been recently. Not only was there a good chance that I wouldn’t end up working in my favoured career, but there was a chance that I wouldn’t graduate – something that happened to at least one of my peers – and the majority of SERC students were paid peanuts (not me – I always cared too much about money to allow that to happen!)
[…] Turning donuts into PhDs « The e-Astronomer […]
It would certainly worry me if we were taking on PhD students to whom it has never occurred to do Peter’s back-of-the-envelope calculation on their career prospects: that combination of astuteness and a feel for numbers is vital in most research projects. In any case, I am pretty sure that most of us give suitable health warnings when our undergraduates come to talk to us about whether and where to do a PhD.
In practice, I am much more concerned about those who take the next step and head off down the postdoctoral route. They also face a harsh attrition rate, and, unlike those embarking on a PhD, are pursuing a path that may be actively damaging their prospects of getting a “proper job.”
Peter – I had a mental bet on whether you or John Peacock would tell me off first for writing “donuts”. But we have to give students doughnuts as we are only paying them peaghnuts.
Michael – I agree strongly about RA attrition being crueller. This links in my mind with the fact that three-year PhDs are not really fully trained up pro astronomers; people know what they are doing by the end of their first PDRA. So I like the idea (suggested to me years ago by JAP) of an expanded junior fellowship programme, explicitly tied to the PhD. One in two or one in three grads get awarded an extra two years to undertake top level research; but its still a Fellowship not a Job.
Andy: I guessed you were trying to wind Brits up by deliberately deploying Saint Anford’s local argot. At least “donut” has a certain ethnic robustness (like “sox”), and could never be mistaken for a typing error. But please don’t come home infected by less appealing transatlantic linguistic deviations (of which my current top hate is misuse of “likely” in place of “probably”: “this quasar is likely variable” etc.).
Sorry, but “donut” does my nut.
Anyway, the main point of my piece (to which the steady state calculation was a postscript) was that I think the transferable skills argument is very strong for undergraduate training – I’m quite happy that we have more physics & astronomy graduates than there are jobs in physics and indeed wish we had even more -but I remain very unconvinced that this argument applies as strongly to PhDs.
Looking at it purely selfishly as an academic, I have no doubt that fewer, longer PhDs would deliver better science. I think that’s in effect what Andy just said.
I’m certainly not arguing that there shouldn’t be any PhDs, but I’m not sure what calculation was done to arrive at the current number.
I don’t think most PhD application think much about their long-term career plans nor are they encouraged to do so. I certainly didn’t and wasn’t.
… and I agree with Mike M about postdocs. I’ve been worried for some while that we were appointing too many AFs and RSURFs, and that many of them were destined to fall off the end of the conveyor belt. This hasn’t really happened so far, probably because of the recent expansion in UK astro groups. But this was surely a temporary blip, driven by the RAE. Even before the general recession, the STFC crisis had called time on the idea of further expansion. So now the supply of jobs will be limited by retirements: if we start close to 10 astronomers in AFs plus RSURFs this year, there had better be 10 UK astronomers aged between 60 and 61 – but I don’t think there are this many. The age bulge of the 1960s has passed through, and largely replaced by people now in their 40s. When I was a new postdoc in the 1980s, you were lucky if there was one UK astronomy lectureship per year to apply for (thus the only reason I was able to stay in the UK was working for SRC/SERC/PPARC for 17 years). This doesn’t paint a very optimistic portrait of the tenured academic job opportunities over the next decade, and STFC/Royal Society may need to consider a scheme for post-AF/RSURF fellowships – rather like the “new blood lectureship” scheme which eventually irrigated the 1980s job desert.
John, I will never ever stop cringeing when taking a US flight and hearing that we will “take off momentarily”.
Interestingly, and entirely coincidentally, an EPSRC-area colleague mentioned to me at coffee this morning that Nottingham are piloting a scheme called EPSRC Plus, which provides one year of funding to post-submission PhD students to complete key publications from the thesis, as well as providing funding for interacting with other groups in the University to spread the word about their PhD research, outreach activities, etc, and generally start up an independent research career.
Actually it might work for a while longer — according to Wilkinson (1990, QJRAS, 31, 411), there were at that date around 100 permanent staff in UK academic institutions aged between 40 and 45 working in astronomy, of whom a little over a quarter were technical staff.
There is an interesting mismatch of expectations/requirements lurking here. The research councils think of PhDs as research training, but it’s equally clear that most universities have not changed the regulations for PhDs (still requiring significant contributions to original research, publishable results, scholarly approach and the rest). And in fact I think most academics still think this way too, really assuming that anyone doing a passable PhD will, by definition, have learnt the research skills. So are we even judging our postgrads by the appropriate metrics if the nature of PhDs is being redefined?
That 1 in 10 making it from PhD to tenure is awful scary sounding. But is it the whole story? A brief mention of ‘technical staff’ by Mike M. reminds me that there was a US study (darned if I can find it again though – anyone else know it?) that of astronomy PhDs essentially all who wanted to remain doing something astronomical were able to do so. This they got to by adding in service institutions (like the NASA centers, also the Chandra X-ray Center that paid me for ages), planetariums, non-research colleges, aerospace industry doing satellites, various outreach activities. Maybe the opportunities are wider here, but has a similar study been done in Europe or the UK?
[I’m not sure how they assessed if someone didn’t want to continue; there’s obviously the self-justifying ‘I wasn’t really interested’ effect, when you flunk out. Like me and tennis.]
And then, the 1 in 3 postdoc to tenure number is not, if we admit it, really that bad. Should the weakest 2/3 of postdocs get tenure? That is, will they produce top science, and teach the science effectively? Surely we all know (many) postdocs who doddled along, or worse, but clearly were not cut out for self-propelled research. Clearly all postdocs should be wanted up front about this. But what is the right ratio to get a tenured faculty without dead wood, based on your own experiences?
There is a danger that some of the arguments here are beginning to sound a little arrogant. Why should it be that success for a PhD student is defined as getting a permanent academic job? Surely that is just one opportunity for a bright young scientist with postgraduate experience in a rigorous, numerate discipline and with, in all probability, experience of working in international project teams. Wouldn’t we be better to say that our training of these people is so valuable that only 1 in 10 of them ends up taking an academic job?
Richard: the issue is not really so much the attitude of PhD supervisors; it is the perception of PhD students. I am involved in Nottingham with the disbursement of our “Roberts Money,” the funding designed to give PhD students and contract research staff access to much more extensive transferable skills training. The problem we have in spending this money is not academics viewing it as a distraction for their PhD students. Indeed, many academics are actively involved in establishing a range of fairly innovative and worthwhile training programmes. No, the problem is in getting students and postdocs to engage with these activities. We have obviously tried to understand why this relative lack of engagement exists, and the fairly universal response we have got back is simple and really quite understandable, and, for astronomy students, runs along the lines of:
“I am doing a PhD in astronomy because I want to do astronomy. If I wanted to go off and work in finance/IT/administration/industry, I would have got a high-quality graduate position in a major company on the strength of my good degree in physics, and would now be earning significantly more money. I could take this training course now, but I really want to stay in astronomy, and I recognize that this training course will not enhance my prospects of achieving that goal. Indeed, since it will take time away from scientific research, writing papers, etc, which are the things that will dictate my chances of success in this field, it would actually be damaging to my prospects to do so.”
We have been banging our heads against this for quite some time, and have singularly failed to come up with an answer. You, too, must be very familiar with it as the refrain that PPARC/STFC hear at every NAM town meeting. If anyone has a solution, I would certainly be interested to hear it.
Mike
I agree with you and am indeed very familiar with the arguments. Though I remember being shocked at a NAM session to be criticised for increasing the number of PhD studentships funded by PPARC (I don’t shock so easily these days ). Anyway my point here is that the opinions voiced in this string seem to come from academics and the impression given is that failure to gain an academic position is just that; failure.
What can be done? My vote goes for further and significant increases in the number of PhDs that we train. If particle physics and astronomy are the subjects that can attract the very best students then we should use that to our advantage. We are constantly able to use the attractiveness of these subjects to both graduates and undergraduates to justify our funding. If academics are open and honest with their students, as I know you are, and are able to project a positive image of life outside of academia then I can’t see that we have a problem. After all, the students are smart enough to know that the odds are that they won’t end up in academic jobs even if they can’t, at the outset, see where their careers will lead them.
Incidentally one of the arguments at the time behind the increase in PPARC studentships was to bring the ratio of students to academics in particle physics and astronomy up to that in other subjects. I think we are still some way from achieving that goal.
To play Devil’s advocate here, it is worth considering possible downsides to an increase in PhD student numbers. What is the PhD programme for? Is it for training students, or is it for supporting academic (and academics’) research? It is both, of course, and by increasing student numbers we could certainly train more people. However, there is a risk – at some point the projects we offer would not all be cutting-edge, and may become repetitive, lacking in originality and topical interest. This completely changes, and in my view, seriously devalues the experience of a PhD for both student and supervisor. The relationship changes completely – and we may see more students with motivational problems. We should also ask the question (as the government never seems to do at undergraduate entry) of whether there are enough willing high-calibre undergraduates who are capable of pursuing a PhD successfully. If expansion continues, at some point this will no longer be true. My impression is that we are not there yet, but that it is not too far away.
One final point. If training becomes the paramount driver, one should also question whether spending c. 3.5 years on a PhD is the best way to do it. It may be more efficient to expand the MSc programmes to give students a one/two year research training.
Alan,
I agree with your comments, which give some of the reasons why I favour the introduction of a Bologna-style Masters, giving a 3+2+3 structure. The extra stage gives a chance for students to opt out before taking the full PhD but with a large part of the training needed to take skills into the marketplace. It would also provide our PhD students with a stronger background before they start and could requre them to take the transferable skills training they are reluctant to when they’ve started a PhD.
The question we need to ask is what is the cost-benefit balance for proper Masters students versus more PhD students. I don’t know how to answer this question properly, but I’m not convinced that simply creating more PhDs is the best solution either for pure science or for the economy as a whole (what’s left of it).
Peter
Alan, Peter – I think our discomfort comes from the tension between the need for advanced training degrees, which we agree with, and the continuing need for research apprentices as it were. For some years we have been trying to make these the same thing, but they should maybe be decoupled. I think this is becoming clearer as a general point, but we have not converged on the correct new model
First day at engineering school, the Dean of Academic Advising talks to the Freshmen class. “Look to your left and your right. Only one of you will graduate.”
But it was more like one in four. And most washed out by the end of their Freshman year. So half of the campus was Freshmen. No idea why anyone didn’t make it. Everyone was brilliant.
And when you got your BS, you started working.
But if you got your BS, wouldn’t you likely get your PhD? IMO, the BS is the hard part. The rest of it is fun. Or does it come down to grant proposals (seriously not fun)?