Bad arguments about academic standards

Geoffrey Alderman (ex-University of London, now University of Buckingham) has been sounding off about academic standards. The argument is that university authorities are so anxious about their place in the league tables that they pressurize powerless lecturers to pass students who should rightly fail, to condone cheating and to lower standards across the board – all in the interests of getting more firsts and 2.1s, and so proving the success of students and staff alike.

In other words what once a 2.1 is now a first, what was a 2.2 is now a 2.1 and so on down.

Perhaps Alderman sees different sides and different areas of higher education from those that I know. But – while I share his gloom about some aspects of the league table and “outcome” culture, which is no better in the university sector than in the health service or primary schools -- overall his rant looks like a pretty feeble, knee-jerk analysis to me.

Over the last thirty years, Cambridge exam results have changed in something like the way Alderman claims. That is to say that the third class degree has virtually disappeared (except in cases of personal tragedy), and the 2.2 is now looked upon by many students as a terrible disappointment. That is despite the fact that large numbers of the middle-aged great and the good achieved no better.

But this change is much more a reflection of changing student culture and aspirations than of any collusion on our part to ‘mark up’. Fifty years ago in Cambridge there really were a still significant minority among the students, who were here for the sport and the parties, or occasionally for more honourable forms of self-improvement (art, acting, music) not wholly compatible with success in the Tripos. Many of these were happy enough with a lousy degree, if that was the price you had to pay for the other forms of experience. For better or worse (and mostly – but not entirely -- for better), that kind of student hardly exists any longer. Our students are determined to do well and so they do. Add to this the fact that (for better or worse again) we are much more careful to make it clear exactly what we expect of them – and it’s hardly a surprise that the third class has become a thing of the past. We’d be doing something frightfully wrong if it hadn’t.

If anything, Alderman’s got it the wrong way round. Our problem is that we don’t give high enough marks, not that we mark too generously.

I should stress at this point (before anyone detects a hint here) that I am not marking exams this year, and that all our papers are checked and double-checked by over-worked external examiners. The process is as fair as it could be. But our difficulty has been always to make ourselves use (as we put it) “the full mark range”. Traditionally, in my Faculty, we have marked each script out of 200, with 70% being a first class mark. For years and years when I first started teaching at Cambridge I rarely saw a mark over 145 (also expressed as “alpha minus query minus”), which  at 72.5% was generally treated as a stratospheric first.
It’s very hard to average over 70% if 72.5% is about the best you can get. And so, of course, from time immemorial we had to resort to all kinds of other strategies (like mark spread and preponderance) if we wanted to do justice to the candidates.

If we’re being any more generous in our marking now (and we do try), that’s because it’s fair, and long overdue.

I also have an anxiety about Alderman’s line on cheating. Unless there are an awful lot of un-caught criminals out there, it is in fact extremely rare. Far from it being condoned. I have always seen it treated with appropriate severity. But I’m also pleased to say that I’ve seen appropriate compassion displayed to the culprit. For in almost every case, this too is a personal tragedy:  the result of panic, desperation, fear, and in the end immature stupidity. The old principle of loving the sinner while hating the sin is probably the best guide here.

But the truth is that most students work hard, do well, treat the exams with impeccable probity and deserve their 2.1s and more.

Comments

[Paul Potts]

Since this post is still alive, let me tell you a story.

A certain academic proved that spiders have their ears in their legs. In order to prove this, he produced an average spider and tod it to walk. He then pulled its legs off and said "walk". "You see," he said, "it can't hear me."

Much of modern testing is like that, especially in language tests, and of course the IQ. I thought everyone knew this perfectly well already.

Paulo

[Tony Francis]

Dear Ken: You raise some interesting points. I know it was popularly held that black kids in the US were the victims of bad testing. Little attention was paid to the unfortunate fact that 30-70% of the children have been born to single mothers, with their fathers abandoning them. There was a movement in those days to re-vamp all tests to be "culturally sensitive". This led to parodies such as the LA Math test:
http://www.snopes.com/humor/question/mathtest.asp
While this is a joke, it is based in some reality. One must wonder whether IQ tests are really of any value in predicting success. Obviously, someone might dumb as a box of rocks, but because of personality and drive, be a cracker-jack widget salesman. Also, look at the Vietnamese who came to the US in the 1970s. I doubt many of these would have scored well on IQ tests. They took jobs as domestics, trash collectors and shrimp boat fishers. Today their kids are doctors and lawyers. The article in Wiki on IQ is of some interest. They indicate there has been "IQ Inflation" observed. It is called the Flynn Effect.
http://www.en.wikipedia.org/wiki/IQ

[Ken]

@Tony Francis

At the narrow end of the normal distribution curve, given the size of the population cohort, an increase of 1000 (0.1% of the population, an increase in the sample of 10%) would have some effect.

All the research on IQ tests show that social/cultural norms alter results. Giving an IQ test to someone from a non-Western hunter gatherer society would show that their IQs are far lower. Grammar, word based tests are the most obvious way in which even some western based groups would perform more poorly. Most people believe that the IQ gap between blacks and whites in the US is at least partially explained by social/cultural norms and by educational support. IQ tests are supposed to be about intelligence and be far less about coaching, but they still suffer from these problems. IQ does not equal intelligence, but is highly correlated with it as long as allowances are made for background.

[Tony Francis]

Dear Ken: Is an increase in Oxbridge admissions from 9000 to 11,500 statistically significant? What evidence is there that non-whites score poorly in IQ tests based on social/cultural norms?

[Ken]

@Lucy

"more students mean lower average ability". This is a statistically reliable fact unless you believe that Oxbridge by letting in more students has found equally good or better students than when they allowed in fewer students.

Say Oxbridge only let in 9000 students 25 years ago as against 11,500 now. Assume that screening technology (exam marks + interviews) is no better and no worse than before - and I would argue that it is worse. Given the spread of ability (normally distributed) the extra few thousand are much more likely to have come from a less able grouping. (I leave aside the demographics that mean that 1000 18 year olds in 1983 was a substantially smaller percentage of the total population of that age than in in 2000.)

Ah, I can see that I havent been clear enough in describing entry into Oxbridge for Lucy.

1. The overwhelming number of Oxbridge entrants were and are from independent schools and selective schools: 65%+ (Note selective schools are basically grant maintained + grammar schools, so a big chunk of state school pupils are in this category. You seem to assume the half of state students didnt fit into this category, but they do.)

2. A significant number come from non-selective state schools 25% or so.

(The rest are made up of FE/6th form colleges, which are not clearly categorisable.)

The latter group are made up of

A) Comprehensives that are not really comprehensive, but are middle class ghettoes and selective in all but name.

B) The truly able coming through real comprehensives.

What I meant when I said

"The recent surge in students going to higher education hasnt changed much. While a few more bright, but poor, students might have reached the final stage.."

I meant that the proportion added to my comprehensive schooled very able category is small, and the signal available is degraded by grade inflation in A levels. The same type of rich middle class people make up the majority of students at Oxbridge that they always did and the same group of able comprehensive schooled kids also go to Oxbridge as they did in the past. So I dont think the pool has changed that much in the past couple of decades. The fact that a majority of entrants in the 1960s were at Grammar schools and the proportion from independents rose in the 1980s suggests that this characterisation is correct. The students dont change much.

If you think AEAs are non-coachable, you are sadly deluded. Even IQ tests are coachable and subject to cultural/social norms (which is why non whites score poorly on IQ tests). Any exam that tests knowledge (even if it isnt an area outside that of the standard A level) is going to be subject to coaching. Also the proportion of students taking AEAs is so low and the pass rate so low that in the pre-result phase, the information content is low. Most state applicants didnt seem to be taking them and given that we now face 10 candidates for any place, all of whom are predicted As and who mainly got A*s at GCSE, I stand by my view that the signal quality is poorer now than in the past. My experience is recent.

You dont seem to understand Mary's post. She is saying that they no longer mark with a maximum of 145/200. This will result in more firsts. This also means that relative to past recipients of a degree, todays Cambridge undergrads are receiving a slightly inflated degree. Neither a bad nor a good thing in and of itself, but de facto evidence of inflation. I dont see where she says Cambridge is trying to reduce the number of firsts? I think she means that in the past they didnt award enough first. Indeed she says

"If we’re being any more generous in our marking now (and we do try), that’s because it’s fair, and long overdue."

The words "generous...long overdue" give the game away a bit dont you think?

[Newn1]

Lucy...

I agree with you about the theory of the paper. I just don't think it's the case in practice. You say:

"Also, I'm not sure coaching *would* help for some subjects. Certainly the English paper I was set asked for no more specialist skills than GCSE - in fact, at least one of the questions (inc. 'rewrite text a, b, or c in the style of text x, y, or z') was a direct extension of the old GCSE favourite 'write in the style of Lizzy Bennet's diary'."

Coaching certainly does make a difference for the English AEA. It's not so much to do with teaching 'stuff', as teaching how to approach texts in a more interesting (i.e. non-A Level) way. Having pointers here and there, and guidance on writing coherent, thoughtful (i.e. non-A Level) essays will only help. And this is simply more likely to happen in schools where (a) teachers have a little more time/fewer students, and (b) teachers are lightly more clued-up (probably, though not excusively, as a result of putting more students through these AEAs and, previously, S Levels). Practice makes perfect.

The paper doesn't ask for more 'skills' (once literate, always literate) -- it just asks for more sophisticated thought. Some students will already have this, and will be fine with the paper without coaching. But others, who might not quite be there, can be VERY much advanced by coaching.

The AEA is a good exam, much better than A Levels; but it's by no means a flawless test of raw intelligence or innovation. They're good fun, though (and I'm sure that's not just a rose-tinted view!), and perhaps that's what actually matters.

[Lucy]

NEWN1: You're right about coaching, of course. But, unlike Ken, I don't think that, overall, it is harder now than ever before for bright state-school pupils to stand out. I don't think the inflation of A-level grades does anyone (except perhaps schools who like league tables) any favours, and I don't think universities should use this inflation as an excuse for not attracting state-schooled pupils. Ken rather casually links 'the middle class and clever', and presents the 'poor but brilliant' as a group so insignificant as to have little effect on the statistics ('Note that this doesnt mean that it is not mainly the same students who would have attended, as the most academically able shifted from grammar schools to independent schools'). This is a pretty depressing attitude, and I have to say that if any exam board cares to offer a paper that at least tries to be open to students from all backgrounds, that is to be applauded. The AEA papers are lot better than the system my parents' generation went through, having to prove they had Latin, etc. At least here there's a chance for everyone.

Also, I'm not sure coaching *would* help for some subjects. Certainly the English paper I was set asked for no more specialist skills than GCSE - in fact, at least one of the questions (inc. 'rewrite text a, b, or c in the style of text x, y, or z') was a direct extension of the old GCSE favourite 'write in the style of Lizzy Bennet's diary'.

[Newn1]

Lucy: "[...] AEA papers, which are deliberately designed such that students should not receive (and should not find beneficial!) coaching."

That's a bit naive, isn't it? Unless you don't actually mean to imply that the AEAs are coaching-proof. Of course good AEA performance can be coached, if the student is reasonably bright in the first place. The first year of the exam (and the first year of the exam is different in each school and subject, depending on how widespread they are) is the only year which is reasonably coaching-resistant. After that, it's fairly easy.

Don't think that the top schools, or the top teachers (in any school), are not offering serious help with these. They are. And the inequality of exam preparation is just as prevalent in these exams as in all others. Only consolation is you have to be that litle it brighter to pass them.

[DAVID VINTER]

I am now well retired, yes I have a deree, but here is another factor to allow for. My late wife was a school secretary, with only a few O levels but was an excellent typist. The school headmistress was taking an external MA. She asked my wife to type a dissertation. Examining the work, she refused to turn out such dreadful spelling and english. And consequently, corrected and re-phrased the lot. Of course the MA was awarded. What would you have done?

[Lucy]

@Ken, I'm baffled.

'more students should mean lower average ability' - how?! We have, quite evidently, failed to come to a consensus about the ability of applicants, so how can you possibly get to this generalisation? Statistics need context.

@Michael, I do see what you mean, but I'm not sure that's what Ken's saying - or not all he's saying. But cheers for explaining.
'Unless you believe that the system didnt work well in choosing the percentage who tried for university in the old days, which is unlikely since this group is still overwhelmingly made up of pupils of independent & selective schools and always has been, I really dont believe that the ability level has changed much.'

Unless I believe ... well, since it's exactly what I said I believed, please assume I do believe the system didn't work. Where does this assumption that selective/independent schooling automatically denotes fitness to do well at Oxbridge? I find it bizarre, and so I imagine do the half of undergraduates who attended state schools.

As it happens, I too have some familiarity with Oxbridge admissions procedures - more, perhaps, than you, since you refer to 'S' level papers that were withdrawn over five years ago. They were replaced by AEA papers, which are deliberately designed such that students should not receive (and should not find beneficial!) coaching.

'So a good Upper Second from Oxbridge was the equivalent of a first from many other good universities, and with a bit more generous marking, people who used to get good Upper seconds will now receive firsts. BUT that is'nt the point.'

I think it's precisely the point. I don't deny that grade inflation exists, but I do object if Cambridge is, as Mary says, trying half-heartedly to even out the picture by reducing the number of firsts. I agree entirely that there should be more low degrees in evidence from the less good universities (but not necessarily the new ones - come on, a little bit of examination of snobbery wouldn't go amiss).

Finally, 'The point is that grade inflation has taken place across the board'. My point is that it hasn't. Not if Cambridge marks too hard at its top end, as you admit it may do. In a way, I agree with you, because I also find it worrying that 2.2s and 3rds - let alone Ordinary degrees - are vanishing out of the picture. I agree that it's suspect when a university with a lousy track record turns out a year of students with very similar marks.

But Mary wrote the article about her experience of the issue, at Cambridge, and I don't see how you can blend her claim that dons fail to award ENOUGH firsts with a picture of 'grade inflation across the board'.

[Ken]

@Lucy.

Not sure that the numbers have risen that much (and cet. paribus if anything more students should mean lower average ability). The numbers entering higher education have changed, but that doesnt mean much. In the old days - 1970s, the A level system was designed to funnel the best 10-15% through to university, of whom the best 1 and a bit % to 2% went to Oxbridge. Now the much inflated A level system funnels 40% into university, of whom the best 2% go to Oxbridge. Unless you believe that the system didnt work well in choosing the percentage who tried for university in the old days, which is unlikely since this group is still overwhelmingly made up of pupils of independent & selective schools and always has been, I really dont believe that the ability level has changed much.

Your comments with regard to entrance requirements show the problem of anecdotal evidence. I am familiar with Oxford admissions. Yes, 3 As are now no longer rare. Yes, some people are asked for higher results. But the point is that for the vast majority of people interviewing in December before they take their A levels, the predictions are for all As. Taking the able state school pupil who needs to show ability and for whom taking 4 A levels and a couple of S levels is impossible as no one in his school is available to teach, the difficulty of standing out has gotten worse not better. I stand by the view that dumbing down A levels has made the signal about ability much noisier (and thus less effective).

Yes, the gentleman's third has disappeared for good reasons at Oxbridge. But this doesnt explain the lack of thirds in the new universities (but grade inflation does).

You seem to be saying that Cambridge marks too hard. That may be true. Oxbridge has always marked hard. So a good Upper Second from Oxbridge was the equivalent of a first from many other good universities, and with a bit more generous marking, people who used to get good Upper seconds will now receive firsts. BUT that is'nt the point. You are saying a bit of grade inflation is good. I don't disagree. What I am saying is that grade inflation is a reality, yet you seem to be unable to acknowledge the point.

The point is that grade inflation has taken place across the board. What is disturbing is that it represents a loosening of standards. If a first from any university should be of the same standard, say 10% of Oxbridge students should receive them and 0.01% of students from new universities should receive them. Clearly not happening. What is disturbing is that as the percentage of those receiving 1st/Upper 2nds in lower ranked universities has risen, the percentage at Oxbridge has also risen. And we have Mary saying that she is participating in grade inflation. A perfectly reasonable reaction even for Oxbridge to maintain their relative position by giving out a few higher marks given the grade inflation elsewhere.

I am not denying the decline of the lazy student, the sports playing gentlemen's third. (I noted that in my very first post.) I dont disagree that an element of grade inflation can be a good thing. What is worrying is that we have started down the slippery slope that the Americans started down in the 50s, to the point that for the able a B is treated like a fail and an A is normal. 4.0 cumulative GPA...

[Michael Bulley]

To Lucy: the "sophisticated interpretation" I referred to has been, in part, taken up by Ken, who says that some pupils at private schools in the 80s might have been at a state grammar if it had been the 60s. We shall never know the numbers. I'd say, even so, that a reasonable conclusion to draw from the crude statistics is that in the 80s there was a greater proportion of school pupils who did not go to Oxbridge who were better than the worst who did than was the case in the 60s. It's a crude way of putting it, but I hope you see what I mean.

[Lucy]

@ Ken. I was under the impression that the number of undergraduates at Oxbridge had risen quite a lot over the last few decades - which doesn't mean of course that the number per academic has risen with it. Certainly, the number of undergraduates overall, across the country, has risen hugely. So too have numbers applying to university. The Oxbridge application process has also changed, and among other things, it is no longer actively encouraged that colleges carry a share of those who are no so interested in the narrow academic confines of their registered degree course. I reckon that there are, therefore, fewer people who genuinely deserve a third at Oxbridge than there were, say, in the fifties or sixties. Many of those who do seem to be in line for a third these days are likely to find that, despite other aptitudes they may possess, they are required or pressured to leave their courses. I'm not sure how much of any of this is progress, but I'm pretty sure it is the case. Comparing the contemporary student body with the student body of thirty or fifty years ago just isn't comparing apples with apples.

And while A-level grades have been hugely inflated, it is also true that expectations have rised with them. In my parents' generation, AAA was considered really quite something, and not all Oxbridge undergraduates achieved such high marks. Now, it's not uncommon for far more specific and higher achievements - eg., 'you must have AAAA, we want you to sit the AEA (new-style Step papers) in subjects x and y, and send us a 2000 word essay on z'. I know someone who was asked to pass the Maths and Physics olympiads to a certain level. Now that's not easy stuff! By contrast, I am well aware that some years ago some colleges used to take on students they knew were there for the Oxbridge experience, for the sport, etc. To have graduated with a third and played in the first XI, or rowed in the college first boat, is something that quite a lot of ex-Cambridge old boys from, say, Pembroke, were happy to choose. This just isn't the case any more. Far, far more applicants are academically inclined, because it is hammered home that you can't get in on a closed scholarship, daddy's money, and a willingness to do well at sports.

The thing that bothers me is that, while you say that Mary admits to grade inflation, I'm reading the bit where she says 'Our problem is that we don’t give high enough marks, not that we mark too generously'. In some subjects, Cambridge is pretty chary about handing out firsts. That matters if, like quite a lot of Cambridge grads, you want to go on to further study, because there is a perception that you need a very good degree to do a postgrad at a good university. In actual fact, it's pretty clear that a good 2:1 from Cambridge may be as well-received as a 1st from somewhere else. Obv. there should be some overlap, because exams aren't a perfect way of testing fitness for postgrad study (after all, you may never sit another exam again), but probably not as much overlap as there currently seems to be. I think Cambridge is trying too hard to ward off claims that they've devalued high grades – that’s why I was glad to hear what Mary had to say.

[apologies for long post!]

[Lucy]

Michael, I don't see how proportions of state vs private-schooled undergraduates can tell you anything about their 'overall quality'.

[Ken]

@Lucy. I think that you will find that the number of students per academic probably hasnt changed much in the past few decades, except perhaps rise a bit. The number of undergraduates in 2005 was 11,500 and in 1996 was 11,200. Before that I would guess there has been some increase in the number of students, but this will be in the form of new courses. Oxbridge between them include 2% of the 1.1 million students going to university. So it remains a small high ability group.
What about the applicant pool? Well, the bottom line is that entry into Oxbridge over the years is a self selecting sample. Only those who think they could succeed would apply. We know that a big portion of the shift from 60%+ State (grammar) to 60% independent in the 80s is due to the shift away from grammar schools. Note that this doesnt mean that it is not mainly the same students who would have attended, as the most academically able shifted from grammar schools to independent schools. (The people really stuffed by the abolition of grammars are the poor but brilliant.But the middle class and clever simply started paying fees.)
There is some anecdotal evidence that clever middle class children are moved from independent to state schools in their final years to benefit from the slight bias to state schools at Oxbridge. Anyway, Oxbridge has always been choosy and has always chosen from the top of the pool of candidates. The composition of the top end of the pool of candidates hasnt changed much (overwhelmingly independent or selective). The recent surge in students going to higher education hasnt changed much. While a few more bright, but poor, students might have reached the final stage, note that the horrible dumbing down of GCE O levels into GCSEs and the wholesale inflation of A levels means that the information contained in such exams has been degraded, so universities now have much greater difficulty in working out who are the best candidates.
So overall, I dont see that I should have to concede anything on the shift in numbers of students or the nmber of applicants or the quality of applicants.

Finally let me note that in classics, the pool was and is overwhelmingly from independent schools and that unlike 30 years ago Greats is no longer considered the best of degrees. (more's the pity.) Best in the sense that the career civil servant was a classicist a la Sir Humphrey of Yes Minister.

Mary has admitted to grade inflation (marking with a higher ceiling) and being more generous. This means that in comparison to students in previous years who were equally able, students are receiving higher marks. This isnt necessarily a bad thing, but I am not disagreeing with that. What I am saying is that the timing matches too well with the rise of new universities and the gradual inflation of degrees in other places and that thus Alderman is correct. At lower level universities, at least some of this was probably to boost positions in surveys. This devalues the currency of a degree. If taken too far, it can be ruinous.

[Richard]

Newn1: there is of course a statistical question, to which an answer must be available, though I don't know what it is. I'd be very surprised if any statistics showed that Oxford or Cambridge graduates were disadvantaged in the labour market (not, in any case, if the stats are sufficiently complex to distinguish between graduate level jobs and macdonalds)...

Of course, such stats would not strictly speaking prove whether an individual was disadvantaged by having been at Oxford or Cambridge: it might just reflect the high quality of Oxbridge intake all over again (i.e., it might be that an Oxbridge graduate was likely to do well because s/he was the kind of person likely to be admitted to Oxford or Cambridge in the first place, but would in fact have been at even more of an advantage graduating from somewhere else - but to tell the truth I very much doubt this).

Presumably in particular cases these things will always depend on the prejudices of individuals responsible for hiring decisions: one might think "Pah, Oxbridge, I never got in, and I don't want some ivory tower type" where another might think "good, I want a smart and ambitious high-flier". But I'd be surprised if Oxbridge graduates ended up at a disadvantage on average...

Somebody might make a case that Oxford or Cambridge 2.2 graduates ought to be disadvantaged by comparison with other 2.2 graduates, since they got a below average degree even with teaching at a level of intensity far greater than anywhere else (i.e., it might be that somebody who missed the 2.1 elsewhere would have got it with the intimate attention afforded by the tutorial system). But I think this would probably be unfair.

Best,
R

[Michael Bulley]

Lucy's comments (28 June) need to be set against some statistics about the intake of Oxford and Cambridge students over the last 40 years. Those statistics need sophisticated interpretation, but in their crude form they are as follows: in the mid 60s, about 63% of undergraduates at Oxford and Cambridge came from state schools, the rest coming from private schools. Now, in 2008, after seven or eight years of successive increases in the proportion of state school students at those universities, more than 50% of undergraduates there are from state schools, but I think it is not as much as 55%. Those statistics seem to indicate something about the overall quality of undergraduates at Oxford and Cambridge between about 1970 and 2000. Whether that has any bearing on the quality of work deemed sufficient for this or that level of degree at those universities during that time is something we perhaps will never be able to decide.

[Lucy]

Ken, is it possible that the greater number of applicants and undergraduates at Mary's university has had an impact on grades awarded?

At the moment, a far larger proportion of the population apply to universities than did in, say, the fifties or the seventies. It seems to me that it is possible, perhaps even probable, that a greater number of bright and capable people make their way into Oxbridge each year. If that is true, then it is obvious that the number of people achieving high grades will be larger now than it was formerly. In addition, if the number of potential applicants increases hugely (as I believe it has), then the university will be better able to select the most capable and talented undergraduates. This number will be even larger if we take into account factors such as the fairly modern pressure to reject (either at entrance level or in the first year) those students who are likely to struggle.

I fail to see how, given these rather obvious changed circumstances, you can be so sure that Mary is describing grade inflation?

[Newn1]

Richard: "On the other hand, since the word "Oxford" or "Cambridge" carries its own kudos, even where the graduate concerned is relatively mediocre, I suppose there is an element of swings and roundabouts..."

Interesting, this, because there is a significant proportion of Cambridge students who genuinely feel the word "Cambridge" causes and will continue to cause them problems with employment & employability. (Partiuclarly if they get a 2.2 rather than the mass-2.1). There is a real feeling amoungst many that having this university on their CV will actively count against them in several areas. Whether that's true or not in real terms, I don't know. But it's food for thought, anyway.

[Tony Francis]

http://www.cbsnews.com/stories/2008/06/27/ap/national/main4214448.shtml

[Ken]

Mary

Your argument comes in two parts. The first portion is with regard to thirds, the bottom half. Here you have perhaps a beta minus. With regard to the thirds, you may well be correct that some of the change is caused by the disappearance of "gentlemen's thirds", those individuals who excelled at sport or at being gentlemen so that it wasnt "desirable" to send them down, but no degree class above third was possibly defensible. But, this only looks at Oxbridge (and perhaps a few other top universities). What about all the new universities? With entrance requirements so very low (and they have to be, given the proportion of the population cohort being allowed in) by definition, many of the students will deserve no more than a third. But grade inflation at the bottom has reduced the number of thirds.
The second portion of your argument applies to the top degree classes. Here you have admitted to marking using a higher ceiling. This, by definition, given the fixed hurdle rate for firsts and upper seconds, means there is grade inflation. Your argument is that this was a desirable thing, which is different from saying that grade inflation had not happened. (Take a student in 1985 who scores 66. He gets an upper second. Under your new mark scheme in 2008, the same student with the same effort/ability and gets 70). I dont deny that in some cases this form of grade inflation is desirable. Certainly it makes UK university results more explicable to the Americans with their ridiculous 4.0 cumulative GPAs. However, at the new universities I suspect more than a little of the Alderman analysis is correct. Given the pressure put on students at one former poly to write positive comments on a survey used as part of one of the league tables, there is pressure to perform - and it need not be in the form of powerful administrators and weak lecturers, although this no doubt occurs, but it can be a conspiracy of silence as everyone conforms to a lower standard. A higher survey position can be achieved through a little grade inflation. Any currency that suffers from inflation will be devalued, but it allows the universities to claim, in the short run, they are doing better (in the surveys) and this puts pressure on universities slightly higher up the food chain and so on. In the end, this inflation reaches Oxbridge.
As I understand it, this grade inflation process occurred in pretty much the same way in the US, with a creeping up of average marks infecting every department in every university. You need to consider the "when" and "why" you began to mark with a higher maximum, rather than witter about how a litte inflation is a good thing. This is where your analysis is most flawed.

[Richard]

Ken, as Mary suggests, is clearly mistaken. By his argument, it would never be possible to conclude that standards were really improving, or to reflect this in grades. And I think Mary is right that the number of real slackers these days is probably smaller than in the past, and elite universities are drawing talent from a wider and harder working pool than in the past (it's not really very long ago that most universities ruled out fifty percent of the population - the female half - which is obviously not the best way to attract the best and hardest working students...).

But where Ken speaks of different kinds of universities ("the new universities") he may be touching on a real point. I have the anecdotal impression that, in a humanities subject like classics, Cambridge tends to award firsts to about ten percent of candidates. This is about the same as one might expect in any other good quality university. But, along with Oxford, Cambridge gets a better intake than any other university, and then gives them teaching which while not necessarily always better is certainly much more intense. One might expect Cambridge to award about twice as many firsts as e.g. Bristol, Durham, Kings London, UCL, Edinburgh, St Andrews etc. (I give these as examples of good quality universities which all the same don't have such very strong admissions standards as Oxbridge; the list is not meant to be complete!). But this isn't the case.

I suspect that an exam performance which gets a very high 2.1 in Cambridge might well get a first somewhere else. And it is at least possible that universities with rather weaker admissions criteria (often = "new universities") sometimes give firsts to their best students when they would not have got the same grade elsewhere (to put it more positively, these are universities which are willing to teach students who might turn out very well but wouldn't have been given a chance somewhere else; indubitably some who didn't get the best A-levels will turn out excellently a few years later...).

On the other hand, since the word "Oxford" or "Cambridge" carries its own kudos, even where the graduate concerned is relatively mediocre, I suppose there is an element of swings and roundabouts...

All best,
Richard

[Mary]

ken. I am not sure I am quite so logically challenged as you suggest. My argument is that a significant category of students who gave relatively little attention to their academic work has been replaced by those who do work hard. Ergo their exam papers are better and they get given the marks they deserve. This does not mean necessarily that students are getting cleverer, though the wider catchment of applicants may mean that is the case. It does offer a better explanation for the disappearance of the thirds than administrative pressure to inflate grades.

[Ken]

Your analysis is (as is far too often the case) flawed, clearly logical reasoning is not your forte. You are basically admitting that you are giving higher marks than before. This IS GRADE INFLATION. What it means is that in the old days a first went only to the truly brilliant. Now it goes to the very good. This by definition agrees with the Alderman analysis of rising numbers of firsts and Upper seconds.

If you want, you can argue that some grade inflation is a good thing. This is a valid and very plausible argument - both at degree and at lower levels. To give the standard defence, imagine there is a pupil who needs a "C" in GCSE mathematics to receive a job. Perhaps 20 years ago he would not have received a "C", but would have been equally capable of doing the job, then an element of grade inflation is a desirable thing so the individual gets the job (or gets to do the postgraduate work he wants.)

What we dont want is for the standards to slip so much that the student with a "C" in mathematics cannot do the job. This is clearly an issue with degrees from some universities - those getting firsts from some new universities fail when doing postgraduate work in top class universities. This is the correct way to analyse the question. Your analysis is gamma/beta minus at best.

[Richard]

Lidwina: personally, I'd prefer buildings to stay up *and* to be aesthetically pleasing (or at least inoffensive). And of course not falling down doesn't make a building a good building; rather, it is the minimum requirement.

I love the phrase "Engineering - and thus, by extension, the whole of the real existing universe -"! I'm guessing that the way to stop this being as silly as it sounds is to come up with a very narrow definition of "really existing"...

Returning to topic: I still don't think that anything Michael Bulley has said really removes the force of my original objection. Even if a great variety of subjects were covered by the selection of exam questions, the exam papers would still reflect one kind of approach rather than another. And if the setting of national papers were not to be a kind of national curriculum, then the papers would have to be either huge in length or extremely numerous, in which case the complexity of the system would become undesirably great.

Best wishes,
Richard