A Degree of Value

Many column-inches have been devoted in the newspapers this week to the issue of University education, after provocative remarks by Phil Willis to the effect that the uncertainty over the “value” of degrees meant the system was descending into farce. Willis is the Chair of the Parliamentary Committee on Innovation, Universities, Science and Skills, which has just produced a highly critical report about the (lack of) regulation of teaching standards in UK Universities.

The Times Higher responded yesterday with an editorial accusing Universities of complacency over the issue of standards, and also ran a piece in which the Chief of the Quality Assurance Agency (QAA) tried to answer some of the criticisms of his outfit contained in the report.

There’s been a great deal of discussion over on the e-astronomer about this issue, and much of what I would say has already been said over there ,so I won’t say it all  here as well. However, there are a few points that I’d like to note.

First, most of the press coverage of this story has focussed on the fact that Universities are now awarding more first-class degrees than they used to.  Actually, the number has almost doubled within a decade. Degrees must be getting easier in order for this to the case, the argument goes. The government strenuously denies charges of dumbing down when A-level results get better every year but has a go at Universities when the same thing happens. So there’s a charge of hypocrisy for a start. However, I think the real reason for grade creep at both A-level and degree stages is that the current education system places a ridiculously high emphasis on compartmentalised learning and assessment methods that allow the students to succeed by cramming and question-spotting without any real knowledge. This has happened at Maths and Physics A-level with a particularly negative effect, and is beginning to happen in Universities too through the enforced modularisation of the curriculum that happened in the 1990s. The way to maintain and improve standards, at least in science education, is to reduce the amount of examination and make the examinations less predictable. The answer is not to entangle Universities in the clutches of a beefed up QAA.

I don’t know if the “standard” of a degree in Physics is lower now than it was ten years ago, nor even what it means to say that is the case. I certainly do think, however, that some of the papers I’m involved with now as a setter or a marker are harder in some ways than the ones I sat when I was a student about 25 years ago. I’m also conscious that I didn’t have to work to support myself most of the time when I was studying. What has changed a lot – and I hope the current generation of students believe this, because I really believe it’s true – is that Universities now put a huge amount of extra effort into teaching than they did when I was a student.

I want to make it clear that I do certainly do not think that present-day students are not as clever or as industrious as previous generations and are  just playing the system. One piece of evidence refutes that view very easily. In the questionnaires we give to students, they very often give the strongest signals of appreciation to courses they consider hard than to those they consider easy. I don’t think students don’t like dumbing down any more than staff do. They just want things to be done fairly.

I should add that I also think, within Physics, that academic standards are roughly comparable at the present time from University to University in the UK. I mean, in Physics at any rate, I honestly do believe that a First from Cardiff is worth the same as a First from Cambridge. I’ve been an external (or internal) examiner at several institutes over the last decade (including Cambridge) and, although their curricula vary a bit, I’m convinced that the academics try very hard to maintain the level of difficulty while at the same time being fair to the students by providing much more help than they used to. Many physicists, however, accept that forcing their syllabus into little modular boxes has made this circle very difficult to square.

I can’t speak for other subjects, of course. Is a first class degree in Media Studies from Nottingham Trent University worth as much (or indeed as little) as one from the University of Glamorgan? Perhaps. Perhaps not. Who knows?

However, it’s not really the issue of grades in itself that worried me most. Contained in the report is a scary section that claims that the link between “teaching quality” and research is “weak at best”. If, it says, it is essential for undergraduate teaching to be delivered within a strong research environment then research funding should be spread around. If not, then it should be concentrated.

The argument contained in the report is a masterpiece of non sequitur. Where is the evidence that research benefits from being carried out in a smaller number of departments? And if you deny a connection between teaching and research, whyshould the higher education funding agencies be involved in funding research anyway? And the evidence is always going to be “weak” when you talk about such ill-defined concepts. What does “teaching quality” mean? How do you measure it? The QAA doesn’t know and neither do I.

 The problem underpinning this issue is that, in 1992, the (Conservative) government allowed the polytechnics to become universities. The various research assessment exercises were introduced because, prior to 1992, all Universities received research funding in proportion to their undergraduate numbers. It was assumed, you see, that a University did teaching and research. However, the new Universities (or old Polytechnics) didn’t always have research activities in the areas they were teaching, and there wasn’t enough money to fund all 120+ new Universities on the pre-1992 basis. Thus the idea was conceived to concentrate this element of research funding (called QR) in those departments that were actually doing research. That’s not unreasonable, but as bureaucracies always do, the system of research assessment has become self-serving. Sufficient  concentration was actually achieved a decade ago, but we still have to endure pointless reshuffling exercises every few years.

The big changes of 1992  left Physics in a special position. The number of Physics (or Physics & Astronomy) departments in the UK entered into the last Research Assessment Exercise was only 42. About two-thirds of UK universities do not have research activity in this area. Very few Polytechnics either taught Physics to undergraduates or did research in Physics and very few started such programmes when they became Universities.  Why? Because there is absolutely no way you can teach a modern Physics degree outside a research department. It would be impossible to keep up to date, impossible to provide appropriate projects, and impossible to retain quality  staff to do the teaching because they would clearly want to be doing physics as well as teaching it. In Physics the link between teaching and research is not “weak”. The pre-1992 situation demonstrates how crucial it really is.

I can’t speak for other subjects, but I suspect much of this applies across all disciplines. That’s why I think a University in which students are taught by people who are not doing research in the field they are teaching just shouldn’t be called a University. By definition.

The Polytechnics had much to offer this country, but their contribution was largely lost when they became second-rate Universities. But of course you’ll never find a politician who will admit that it was a mistake.

21 Responses to “A Degree of Value”

  1. Anton Garrett Says:

    One of the issues is how to determine whether standards have fallen. In my view this is best left to the opinions of experts, rather than devising metrics which merely simplify the opinions of those who devise them. And if you (Peter), as a full professor with an obvious interest in teaching, candidly admit that you don’t know then I can’t imagine who does!

    In maths I think that complaints about compartmentalisation are not wholly justified; you are either good at solving differential equations or you are not, and an exam is not so bad a way to find out.

    Incidentally I was recently in an email correspondence with a professor of “Leisure, Sport and Tourism”. I’ve heard of a gentleman of leisure but never a professor of it. The “leisure industry” has always struck me as an oxymoron.

    Anton

  2. telescoper Says:

    Anton,

    I take your point, but the problem is that anyone can be good at solving the same ODE that they’ve seen 25 times before….

    The issue of standards is a complicated one. With A-levels we can say standards have fallen because the Schools pass their “product” onto Universities and the latter have definitely noticed that certain skills are lacking nowadays. But who judges graduates? And who judges those that judge?

    Peter

  3. Anton Garrett Says:

    Lot of ODEs out there; set ’em three and the chances they’ve done those exact three (or their equivalent) before are pretty small. If they have, they are probably so fanatical about ODEs that they are expert anyway. (I was, until (1) the computer beat me – sound familiar Kasparov? – and (2) I discovered functional equations, which are much more subtle.)
    Anton

  4. telescoper Says:

    I spoke to someone yesterday who had a PhD in the field of Nursing. A Doctor of Nursing?

  5. John Peacock Says:

    Providing objective evidence on grade inflation is hard. Looking just at exam papers, you would conclude that the process has been going on forever. When I was an undergraduate, in the 1970s, I remember trying past physics exam papers from the 50s and 60s and concluding that they were impossibly difficult; the questions I was being asked to tackle seemed trivial in comparison. Given that I was a student before the numbers doing physics or the proportion of Firsts had risen, I don’t think there had been grade inflation at that stage – rather, it reflected a broadening of the syllabus, so that we were expected to do more than solve Maxwell’s equations in contrived situations.

    This broadening argument is of course what is used by the government to say that all is well with maths A-level. But I think we collectively don’t buy it in this case. This is not so much to do with comparing exam papers, but because we see the students try to deploy the knowledge that they have allegedly gained. What is apparent is not just that they have a weak grasp of many of the elements, but that they lack the confidence to deploy them in a new context. And this doesn’t improve very fast or at all with practice. So their entire intellectual foundations are rotten.

    You can see elements of this disease in our undergraduate teaching. I think the two biggest changes since I was a student are project work and format of exam papers. Project work is a much larger element of the assessment, and almost everyone gets high marks. If you graded students just on what they did in the exam hall, the proportion of Firsts would not have risen nearly as high. Now, the students like doing their projects (and not just for the high marks), but I think it would be better if they spent less time on them. If I had my way, the final year of the MPhys would have more lectures on core physics – really bedding down the basics rather than having a superficial dash through lots of options.

    As for exams, the questions now lead the students by the hand. I think many of our questions involve much the same material as they always did, but what was once a fairly terse description of a calculation will typically now be broken down into 4-5 subsections, each with hints to stop people wandering off in the wrong direction. This format is “fairer”, but you can have too much of a good thing.

    But, as with school maths, the acid test is where students try to use their knowledge at the PhD level. Do we see the same basic lack of confidence and ability to make progress unaided? I’m in two minds over this. I think PhD students are less independent these days, needing regular meetings with supervisors to keep them moving. But the subject has changed. I look at some of the (well-cited) papers I wrote in the 1980s, and they seem childishly simplistic compared to today’s research literature.

    So I think the court is out on whether there has been huge grade inflation in physics degrees. But on the other charge, I agree with Peter that Willis is completely wrong. Having been an external examiner, I do believe that a First means much the same thing, independent of where it comes from.

  6. telescoper Says:

    John,

    Some of your points raise exactly the same issues that make it difficult for me to comment on standards. Maybe it is true that analytical problem-solving demands have decreased but students now do a huge range of things that I didn’t do at all – or barely touched on – when I was a student. And if you’ve got ready access to Mathematica or Matlab why does it matter if you can’t do an integral as quickly as your supervisor used to be able?

    I agree with your comment about changes in the field too. My most cited paper (not very highly cited by your standards anyway) is embarrassingly simple-minded, but at least it was based on a reasonably original idea. Much of today’s literature seems to me to be at best incremental and at worst just stamp-collecting. This is just because the field has grown up and started to conform.

    But now I’m sounding very old again so I’ll shut up.

    Peter

  7. Adrian Burd Says:

    John, Peter,

    I agree with a lot of what you say. Comparing the present US system to what I remember of the UK system, I can see large flaws in the system over here. These seem to me to be partly a result of having no system of checks and balances (such as the external/internal examiner system), opening up the doors to (almost) everyone, removing teaching from research … in other words, a lot of the concerns you express.

    The debate between computer vs analytical problem solving is one I struggle with a lot. Many of our graduate students (in marine science) do not have the mathematical background to even start thinking about solving problems analytically. Having them use a computer may allow them to start solving problems sooner, but rapidly leads them into problems as they encounter problems that have subtle issues with them (the odes are stiff for example). So here, the problem is that students do not have the mathematical background for solving problems either analytically or by computer. However, the rapid change in techniques and technologies they use in research is requiring them to use such tools. It’s a quandary.

    Adrian (who will now shamble off back to the country village, muttering to himself).

  8. Anton Garrett Says:

    “if you’ve got ready access to Mathematica or Matlab why does it matter if you can’t do an integral as quickly as your supervisor used to be able?”

    The immediate answer is because plenty of integrals arise that require similar techniques to those covered by computer algebra, yet which aren’t covered by such routines. So one must learn those techniques to be able to crack a wider range of problems.

    More deeply, I share your view that compartmentalisation of maths and of theoretical physics is artificial – you never know what technique from one area will prove useful in another. So it is dangerously limiting to subcontract too much to machines.

    Anton

  9. Anton Garrett Says:

    There’s surely no grade inflation; you elect a Labour government and the population becomes more intelligent, truly the world is admirably arranged…

    [NB to critics: don’t think I like the Opposition merely because I dislike the Government…]

  10. John Peacock Says:

    Peter, I use Mathematica a lot. But I still think students should learn the traditional techniques, for lots of reasons. One is so that you have something to fall back on (cf. the modern hillwalkers who need rescuing because their GPS croaks and they don’t know how to read a map). Another is that Mathematica is amazing but far from perfect: there’s been plenty of times I had an integral that was clearly soluble if I could be bothered to perform a certain chain of substitutions, and yet Mathematica claimed there was no answer. It has trouble distinguishing between active variables and parameters – something like integrating Exp[-x] dx OK but falling over with Exp [-a x / (1+b^2) ] dx, which is clearly identical to the human eye (I’m sure this example is OK, but I’ve found cases of this sort where you have to lead it by the hand).

    But the real reason is that doing trig substitutions and all that is good exercise for the brain: tennis players spend time in the gym, not just hitting balls on court. Probably similar to the reasons that Latin continues to be taught.

  11. telescoper Says:

    John,

    I agree that we shouldn’t completely abandon basic mathematical skills, for the reasons that you state (and others). But what education, especially higher education, needs to do is strike a balance between giving the students skills and knowledge on the one hand, and helping them acquire a proper understanding on the other. The more important thing for a physicist – it seems to me – is not the integral at the end knowing how to set up the problem in the first place. The mathematics is important of course, but I think it’s reasonable to rein back the more esoteric stuff (e.g. confluent hypergeometric functions) in favour of more fluency with the basics.

    In any case much of the odd bits of mathematics I’ve used in research weren’t actually taught to me at University. I picked them up from books. But I was taught many ways of starting to think about problems I didn’t know how to solve and a bit of confidence to tackle difficult things without giving up at the first hurdle.

    I think a University education – in any field – has the primary role of teaching the students how to learn. Unfortunately the word “learn” is interpreted by many as meaning “memorise” rather than “acquire understanding or skill in”.

    Peter

  12. Bryn Jones Says:

    We can perhaps identify some processes that operate to level degree standards within subjects across British universities. The external examiner system brings in academics from different institutions to impose their standards in individual departments. As the years pass, external examiners are replaced, often circulating between universities, producing a healthy mix of examiners across the university system and tending to force standards to converge within subjects. (By the way, it has never been entirely clear to me whom or what the external examiners examine: the capabilities of the students, or the exam writing abilities of the lecturers. It has sometimes felt to me that it is the latter. )

    Arguably, one failure we have regarding maintaining standards in Britain has been with the quality assurance process. Departments were assessed about a decade ago by inspection teams. They did excellent work in checking teaching and assessment procedures, but I do not recall any significant concern for syllabuses (syllabi?), or academic rigour. Consequently, despite the assessment and marking of departments, we still have a House of Commons committee able to complain about differing standards across the university system. What really was the point of all the bureaucracy and effort if the teaching quality assurance system left no real assessment of academic standards?

    (And as an aside, I prefer to check results from Maple or Mathematica, either myself or using textbooks. The software can sometimes give rather odd results, failing to simplify expressions or considering wrong cases [e.g. they assume a constant a > b, whereas I’m interested in b > a].)

  13. Anton Garrett Says:

    Out of interest, the ODE which Maple cracked that I didn’t was for the variation of temperature T with time t of a light bulb filament when an alternating voltage is applied,

    dT/dt + aT^4 = b sin^2 (\omega)t

    where the terms on the LHS are due to heating of the filament, and radiation. In practice you want only periodic solutions not transients, but Maple got the general solution. It even had to define a function to do so (the inverse function of y = x \exp (-x) ), and was still implicit.

    Anton

  14. Rob Ivison Says:

    as a product of what you describe as a second-rate University, i was surprised by your generalisation. i suspect some of the folk at Hertfordshire and Liverpool might be similarly puzzled.

    there is much truth in what you have to say in the main thread of your piece – in particular, your acknowledgement that lives were often easier outside of term-time. it must astound current students to learn it was once routine to “sign on” when you arrived home for the summer! but next time you generalise, spare a thought for those that blossomed in Poly physics departments. i was taught physics and astronomy by Dunlop, Adamson, Robson, Appleton, Whittet and Bode, as was the guy looking after SCUBA2 in the office down the corridor. the environment was very good indeed, as it is in some ex-Polys to this day.

  15. telescoper Says:

    Rob,

    My comments were indeed generalisations, and I think I made it quite clear in my post that there are (and were) certainly exceptions to the rule. Indeed I could add others, such as Portsmouth, to your list of excellent departments in former Polys (although Portsmouth is Applied Maths rather than Physics). I think it’s also clear from my post that I think academic standards (in physics) are comparable across the entire sector.

    My phrase “second-rate” was aimed at the institutional, rather than departmental level. I maintain many (perhaps most?) excellent Polytechnics lost their identity in the rush to become new Universities. But that’s not to say that there weren’t excellent people working in them then or now. LIkewise, I should say, there are undoubtedly many second-rate people working in first-rate Universities!

    Peter

  16. Rob Ivison Says:

    rare for me to disagree with you, Peter, or to feeling anything but admiration for your prose. however, after another read i still struggle with your generalisations and cannot quite see why you’re so down on Willis’ report. i love your “masterpiece of non sequitur”, but i can’t see why his comments about teaching quality and research are so ridiculous. if nothing else, you’ve moved me to go and get a copy of his report!

    finally, why i should believe that “The problem underpinning this issue is that, in 1992, the (Conservative) government allowed the polytechnics to become universities”? i can see that the new system isn’t perfect, but i don’t think your argument is sufficiently compelling to justify the unusual force with which you make it.

    now i’ll return quietly to admiration mode.

  17. “I should add that I also think, within Physics, that academic standards are roughly comparable at the present time from University to University in the UK. I mean, in Physics at any rate, I honestly do believe that a First from Cardiff is worth the same as a First from Cambridge.”

    Cambridge and Oxford have a HUGELY better reputation than other universities in the UK. So presumably people from Cambridge who get a job because their degree is from Cambridge do so at the expense of those from “lesser” universities with better marks.

    What about later on, though? Would you say it does make a difference if one does a postdoc in astronomy at, say, Cambridge, as opposed to Preston? Liverpool? Nottingham? Cardiff? Manchester?

    “There’s surely no grade inflation; you elect a Labour government and the population becomes more intelligent, truly the world is admirably arranged…”

    Reminds me of how President Eisenhower was shocked when an aide mentioned that fully half the population of the U.S.A. was below average in intelligence! 🙂 Why can’t we all be in the top 1%? I used to get spam emails claiming “we have placed hundreds of websites into the top 10”.

    • telescoper Says:

      Phillip,

      Interesting point about postdocs. My feeling is that once you’re into the postdoc system people tend to judge you by what you’ve actually done, rather than by your potential (as estimated by your degree). In that sense whether it depends on the place you did your postdoc depends not on some sort of perceived reputation but on whether there is an environment there that actually allows you to produce the goods.

      I’d be interested to see statistics on the effect of one’s first PDRA location as I’m not sure whether it is important or not.

      Peter

  18. “The way to maintain and improve standards, at least in science education, is to reduce the amount of examination and make the examinations less predictable.”

    In ZEN AND THE ART OF MOTORCYCLE MAINTENANCE (worth a read, as is its successor, LILA), Pirsig describes a situation where he suggested that all grades be abolished. A student immediately objected “but that’s what we’re here for”. 😐

  19. I think in many cases, things are reversed. 100 applicants: put all the Hubble Fellows on the shortlist. When hiring someone, the concern should be with what he WILL do; what he HAS DONE is only a proxy for the former. Thus, we should expect more from someone who was a Hubble Fellow, because of the better boundary conditions compared to a typical PDRA. In practice, he is probably scrutinised less and less is expected. The argument, of course, is that he must have been good to become a Hubble Fellow. Even if that’s true, then the problem is that once one gets one good job, one goes on to the next job etc on the basis of having had a previous good job. Of course, other factors might have played a role in getting the first good PDRA.

    I think we all agree that RESEARCH quality varies enormously among institutes. If there is a link between teaching and research, then it is strange that all the external examiners think that degree standards are the same.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: