h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

Having 25% of current students in a subject achieving grade A in a subject means two things:
a) Having a grade A means someone is in the top 25% of people taking that exam that year.
b) 25% have reached what is claimed to be a particular objective standard in their subject.

In 1983, achieving grade A meant *one* thing:
a) Having a grade A meant someone was in the top 10% of people taking that exam that year.

25 years ago, if you knew someone had an A or a B, you knew they were somewhere in the top 25% of their year in that subject. In *that* limited sense, an A or B then is undeniably the equivalent of a modern A.

Whatever else, it's pretty clear that as a means for distinguishing between good students from a certain academic year, A-levels are less useful than they previously were.
Especially as distinguishing between students from a given year is about the only practical *use* of A-levels, that does seem to be a loss.

I reckon it'd be far better to have a year-relative grading system as before (10% get As, etc), and then some indication of how the examiners actually thought average student abilities had changed from previous years.
That way, there just wouldn't be the annual b*****ks about how pass rates had changed year to year, and the few people actually bothered what examiners thought of year-to-year changes could still find out.

I very much doubt there'd be as much media coverage as present if it was along the lines of:
"10% of students get As (as they do every year) but examiners reckon that an A this year indicates a slightly better level of competence to last year.

In any case, it seems to me that the best people to decide on year-to-year changes in the quality of students are the people who end up with them next - universities and employers - if *they* say that standards are continually rising, I'm perfectly prepared to believe them.

That is interesting,
they showed one university who is having to do their own entrance exams now as they said there are too many people with 3 grade A in A Level and they don't know who the best students are anymore. They showed the test paper they now use to find the best entrants. One question was naming the parts of a heart and the order in which blood travels through heart. Another question was around maths, and looked quite complicated, it involved working out some kind of formula and converting the answer into several differant units of measurement.
.
The other day one university who runs science degree courses said it has to run remedial courses for some of its students.

There is another aspect to this dilemma from the university side. There is a real problem with very popular university courses such as English etc. Here at the University of Manchester English requires 3 A grades, while a very difficult subject like Maths or Physics variants require AAB or ABB. These science courses are much harder than English or American Studies (I work in the system, so I know this to be true), Medicine certainly so.

If we assumed grades just went on the difficulty of the course, English at the UoM would require BBB or thereabouts, American Studies slightly less, meaning every student capable of getting 3Bs would cope on an English degree. It simply isn't fair that such students would miss out on courses they could do well on, and this accounts for some of the grade inflation.

The A Level system does need to be changed, but so does the Uni admissions procedure

>>"If we assumed grades just went on the difficulty of the course, English at the UoM would require BBB or thereabouts, American Studies slightly less, meaning every student capable of getting 3Bs would cope on an English degree. It simply isn't fair that such students would miss out on courses they could do well on, and this accounts for some of the grade inflation."

Surely, if there are N places available on course X, all a university wants to know is:
"Who are the best N students out of the applicants for course X?"
allowing some correction factor for the students who would end up choosing to go elsewhere.

If the only people turned away from a course would be those that couldn't have hacked it anyway, I don't think 'oversubscribed' would be a suitable description.
With a truly oversubscribed course, people who could have done well *will* end up being turned away whatever is done to their A-level grades.

The only thing that having large numbers of people getting 'A's will achieve is to make it harder for universities to work out who the best students are, and *that* simply isn't fair. It means that someone who would once have got 3 Bs could end up taking a place from someone who would once have got 3 As.

>>Surely, if there are N places available on course X, all a university wants to know is:
"Who are the best N students out of the applicants for course X?"<<

While that's certainly true for Maths, sciences, engineering etc, it's not really the case for humanities subjects. In these subjects the uni (though I can only speak for Manchester Uni) a mix is preferred, and a BBB student would be able to do the work and may bring things to a course that an AAA student won't.

There is also the problem of social diversity - very, very few students for poorer backgrounds with no family history of Uni get three or four As, it's disproportionately independent school students who achieve this sort of level. This adds some impetus to give students slightly higher grades in popular humanities subjects.

I know for a fact that many humanities subjects are over subscribed - I've been turning away perfectly well qualified students all day because their desired course is too popular, rather than being too difficult.

>>"While that's certainly true for Maths, sciences, engineering etc, it's not really the case for humanities subjects. In these subjects the uni (though I can only speak for Manchester Uni) a mix is preferred, and a BBB student would be able to do the work and may bring things to a course that an AAA student won't."

The thing is, the way things are now, a course limiting its intake to AAA students actually going to get essentially the same ability-spread of students that they would have done if requiring BBB 25 years ago.

>>"I know for a fact that many humanities subjects are over subscribed - I've been turning away perfectly well qualified students all day because their desired course is too popular, rather than being too difficult."

So - what's new?
Unless universities could interview and test every single applicant, and give a place to everyone they thought might benefit, however great or small that number might be, then people are going to be turned away. The more popular the course, the more will be turned away.
If universities were going to interview and fairly test *everyone* who applied, A levels would be one huge waste of time and money.

If exam results are (for good or ill) one major set of criteria that are being used for selection, if courses are oversubscribed, supply and demand would tend to suggest that the exam-result criteria would be tightened to where they reduce applications or rule out applicants sufficiently that further selection by interview, etc can happen without the university having to waste time and money interviewing every single applicant.

Effectively, the requirement may be set to be somewhere between the two seperate limits defined by:
a) How qualified do people need to be to tackle the course?
b) How high can the requirement be and still allow us to fill the course?
An 'easy' course may end up with a higher threshold than a 'harder' one, even if limit a) would actually be far lower for the easy course.

>>"There is also the problem of social diversity - very, very few students for poorer backgrounds with no family history of Uni get three or four As, it's disproportionately independent school students who achieve this sort of level."

Without wanting to sound elitist, that's possibly an effect that might have been *exacerbated* by social mobility in the last few decades. Many of the brightest working-class kids from the 60s, 70s and early 80s (from a time when there were liveable student grants) are now (at least economically speaking) middle-class parents, and even if not using independent schools, are often likely to have done their damndest to avoid their kid going to the least-well-regarded school in town.

>>"This adds some impetus to give students slightly higher grades in popular humanities subjects."

Why?
Surely, if the A-band widens, that will also mean that the relatively advantaged kids who would have got BBB 25 years ago now also get AAA, since they'll be taking the same exams as the disadvantaged kids.

At best, someone seeking social justice might hope that when it gets to the point where a university can afford to ask for AAA for popular courses, so many people will *still* achieve that that further selection must be random, or based on some other system that would prove more advantageous to the disadvantaged student who would previously only have got a mix of As and Bs despite being talented.

That's at the top end of the requirements scale, but further down, the option of continually raising the entry requirements still exists, so in any kind of 'arms race' between examining boards and universities, the universities retain the upper hand, should they wish to wield it.

Basically, in a system where universities use grades as a filter, they're going to be smart enough to adapt to increasing pass-rates at the various grades. At the top end, where they *can't* use grades, one might *hope* they'd then have to resort to maybe a more socially-fair system of narrowing down applicants, but such a system could be implemented just as well without grade inflation, and even if thought desirable, could presumably benefit from some standardisation and openness.

What the current system *does* presumably do is produce very many more people than previously with extremely high grades who *still* don't get the course they want, and who may well therefore feel that the system has been unfair to them even if it has been assigning places purely on merit.

I think that is some terrible snobbishness towards non-science subjects HonestIago. It has always been the case that popular courses in whatever subject have a high grade threshold. And it has always been easier to get on unpopular courses as they tend to give lower offers.

I remember chatting to one of my lecturers about the medicine degree at my uni - he said that three As at A-level wasn't really necessary to be able to do the work, but our school of medicine was so over-subscribed that they could ask for whatever they wanted. The vet courses were similarly over-subscribed.

*wonders what the average offer is for Physics or Chemistry these days* I remember that the engineering courses at my uni tended to give much lower offers than say physics or maths because they were less popular courses, not necessarily easier though.

I worked at Birmingham Uni chemistry department until last year. Their official line is that they make offers in the realms of ABB (possibly BBC or BBC now I can't quite remember),

However, the reality is that they can't get students of this standard at the moment and a fairly whopping proportion come in with D and E grades with no maths skills.

What the official university line is and the reality of the situation may be two different things.

Exactly. At my son's sixth form college, they were pointing out that by choosing a less popular subject such as chemistry, anglo-saxon studies etc it would be easier to get a place at uni.

When I went to uni (34 years ago ) my unpopular subject required Ds & Es. Now there are far more applicants & As & Bs are needed. I'm not convinced that the students are any brighter, though.

Nothing's chnaged there then.

When I was looking at Universities (in the year 19 Hundred and Frozen to death) it was always reckoned that Russina Studies was a shoo in for Cambridge. A fact proven when a guy from my year got in with 2 E's. Of course, he transferred to do something more useful once he was there.

>>The thing is, the way things are now, a course limiting its intake to AAA students actually going to get essentially the same ability-spread of students that they would have done if requiring BBB 25 years ago<<

Yep. But I think you'll find that the numbers of people getting A grades in Maths, Physics and Chemistry don't really vary that much, and don't experience much, if any, grade inflation. It's in the humanities and social sciences where you see true grade inflation.

>>At best, someone seeking social justice might hope that when it gets to the point where a university can afford to ask for AAA for popular courses, so many people will *still* achieve that that further selection must be random, or based on some other system that would prove more advantageous to the disadvantaged student who would previously only have got a mix of As and Bs despite being talented<<

I agree with you entirely, and as I said, the university admissions system does need to be overhauled, and I think this is the exact area they should be looking at.

>>the option of continually raising the entry requirements still exists, so in any kind of 'arms race' between examining boards and universities, the universities retain the upper hand, should they wish to wield it<<

Not really - in a lot of humanities cases the universities have already reached their limits. The big humanities courses like English, History, Drama and Fine Art have reached AAA - they cannot go any higher. Even relatively unpopular courses like philosophy are heading in the triple A direction. I got onto my course in philosophy and politics 3 years ago with ABC (counted as BBB), whereas this year the requirements were ABB, and next year will probably rise to AAB - again for a relatively unpopular course, which most certainly not getting harder.

The universities cannot raise their grades much higher because a lot of courses have reached the 3A limit at the top universities. The talk of bringing in a new A* A Level is only a stop-gap measure at best, because the problems will just carry on and I'd say within 5 - 10 years, we'll be back here.

>>and who may well therefore feel that the system has been unfair to them even if it has been assigning places purely on merit<<

Heck yes. As I say, I've been working on Clearing the past few days and I'd say a good 70-80% of people I've been speaking to have been turned away, despite being capable of doing the courses.

>>I think that is some terrible snobbishness towards non-science subjects HonestIago<<

It's not, that's simply the way it is. As I said in my above post, I did a humanities course for my degree and I've worked extensively in admissions and recruitment for the humanities faculty here, as well as central (Uni-wide) admissions and recruitment. I also did mostly sciences for my A Levels, so I've got a good understanding of both from a students point of view as well as a professional viewpoint.

Maths, sciences and engineering are genuinely fundamentally different from humanities. With most of the humanities, with the exception of languages and 'scientific' humanities like linguistics, pretty much anybody can do them. They usually require the student to learn specific things and learn how to properly debate, but these can be done *relatively* easily, even for a complete newcomer and so very few humanities (again excluding languages and sometimes history).

The sciences and maths are harder because as well as learning of facts and processes, they require true understanding of very difficult and usually very esoteric concepts and this cannot be simply learned, it's usually something thats needs to be developed over a considerable time and so sciences, maths etc will *never* accept someone without a grounding in their specific subject, and usually require a couple of related subjects.

>>*wonders what the average offer is for Physics or Chemistry these days*<<

For the Uni of Manchester and other unis of similar calibre, physics comes in at AAA/AAB with Physics A required and Maths usually preferred or required depending on specific course. Chemistry goes from AAA to ABB depending on the specific course, with an A in Chemistry required and maths or another science preferred or sometimes required. Engineering, depending on type, goes from AAB to BBB, with physics or maths required and often another scientific subject preferred

...so very few humanities (again excluding languages and sometimes history) *require specific subjects to get onto the course...

Hmm, interesting.

I would certainly say that the science degrees are harder. But I would've the rest the other way around.

Anyone can do humanities certainly. But to actually do it well, to push yourself up to a higher level, as far as I'm aware there's no guide for how to do that (learn logical rules of argument perhaps?). You can just read and debate and ask questions and hope it refines your ideas.

Whereas anything mathematically based, you can go out and learn new techniques. At most you might need to ask someone for advice over what method you need to look up. Then again, you need to be a person who understands maths in the first place.

>>"Yep. But I think you'll find that the numbers of people getting A grades in Maths, Physics and Chemistry don't really vary that much, and don't experience much, if any, grade inflation. It's in the humanities and social sciences where you see true grade inflation."

Really?
http://www.bstubbs.co.uk/gender/boys-a.htm
For Boys (so generally lower than Girls' scores) in years up to 2005 (going back a different number of years per subject), the following percentage got As:

Maths: 34.7%, 36.5%, 39%
Maths (including Further): 28.0%, 28.3%, 28.9%, 35.7%, 37.4%
Chemistry: 24.1%, 24.1%, 25.6%, 27.0%, 27.0%, 29.2%, 28.5%
Biology: 16.2%, 16.2%, 16.8%, 19.5%, 19.2%, 20.3%, 22.1%
Physics: 23.6%, 23.6%, 24.1%, 25.3%, 25.7%, 27.0%, 27.0%

English comes out to be largely comparable to Biology - 16% -> 20% in 6 years.

Bear in mind that 25 years ago with the fixed ratios (A:10%, B:15%, C:10%, D:15%, E:20%), the figures would have been 10% for all subjects, every year.
Of the people now getting As in maths, under the old system, one quarter would have got an A, three eighths would have got a B, another quarter would have got a C, and an eighth would have got D. Half the people getting Bs would have ended up with Ds, and the other half with Es.
Of the C-grades, a little over half would have got an E, and the rest would have been given an O-level pass.

Looking at boys and girls combined, for physics and chemistry, basically half the candidates will get an A or B, for Maths, the figure is 60%..

That's what I'd call grade inflation.

>>"Not really - in a lot of humanities cases the universities have already reached their limits."
I *did* make my comments in the context of talking of requirements below AAA, and then proceeded to cover the situation when limits hit the AAA barrier.

I'd agree with that Bouncy, and I think (though I've not got concrete figures for this) that maths and sciences, although having a higher drop-out or tranfer levels than humanities, will have higher proportions of 2.1s and 1sts, meaning those who can hack the course will score highly,

On the other hand, social sciences and humanties, which require real skill and flair to achieve 1sts and 2.1s have lower drop-out rates but also more 2.2s and 3rds. I'll check in work tomorrow to see if I can get any figures that back this up

Wow Potholer - that's me well and truly proven wrong on that point - that's quite a surprise. While I was only guesstimating from what we see at Uni, I'm genuinely surprised we didn't pick up on this. From the Uni's point of view we've seen many more applicants with high A grades in humanities than in the scientific subjects.

With a subject like Further Maths I'd still content that the high A pass rate is because students who aren't capable don't take on the subject.

>>I *did* make my comments in the context of talking of requirements below AAA, and then proceeded to cover the situation when limits hit the AAA barrier.<<

I know, I was just making the point that many institutions are already at that level now in a lot of subjects, that something needs to be done very soon.

Can I just poke my nose in and say that there's no point in trying to do much comparison over the years as the number of grades available has changed, so the spread isn't comparable over time.

There used to be 7 grades to spread the A level marks between, A - E, N(nearly passed- the PC name for Failed)and U for Ungraded or unclassified
From 2002 there was no N grade, so the same mark range needed to be spread over 6 grades.

So you can only really compare like with like. And it's about to change again, we start teaching the new "six into four" from 2007 with first aggregation in 2009, and that's when there will be extension questions for the potential higher grade candidates. This will replace the current separate AEA (advanced extension award).

>>"There used to be 7 grades to spread the A level marks between, A - E, N(nearly passed- the PC name for Failed)and U for Ungraded or unclassified
From 2002 there was no N grade, so the same mark range needed to be spread over 6 grades."

Quite.
And it's *very* instructive looking at what actually happened when the N grade vanised, from the (Boys') figures I linked to (which boards that includes, I'm not sure).

In some subjects the grades seemed to adjust such that the change was generally absorbed across all grades, which seems to at least have some argument in its favour.

What appears to have happened in the case of Maths (including further) (35,000 students) is that the 'suplus' was accomodated by a shuffling-down of boundaries such that the percentages on grades B-U stayed pretty constant, and the A grade absorbed basically the entire amount, which doesn't seem easy to justify from an educational perpective, especially for an exam that already had a pretty high percentage of A grades).

With Classical Studies, in the year after the N grade went, the percentages getting grades D, E, and U all *fell*, whilst grades A,B,C all rose by ~2%.

Communication studies: 12.4% from the N grade was accomodated by the U band *shrinking* by 6%, and the E band losing 3%, whilst C and D grew by 10% and 7.5% respectively.

Economics: D, E and U grades continue their previous shrinkage apparently unperturbed by the 6% coming from the one N grade.

English: The number getting U continued its previous slow decline despite the 5% ex-N-grade students looking for a home. Despite having previously been stable at ~12%, E shrank by 4% and stayed at ~8% thereafter. D basically didn't change.

Geography: U stayed stable, E carried on declining regardless of the loss of N

Physics: E basically stable, but at least U *grew* to some extent and stayed higher - Well Done!

For *all* A-levels, the reallocation to absorb the 7% previously getting N temporarily increased U by <3%, and had no effect at all on the continually shrinking E.

*Personally* speaking, if the grades really are meant to indicate an absolute measure of achievement, which is what government and the examining boards seem to claim, I'd have thought that the N-grade students should really have been accomodated in grades D through U at best - adjustment of the allegedly *absolute* A-grade threshold shouldn't even have been contemplated.

It seems that with quite a few courses, the scrapping of the N grade has meant that practically speaking, 'E' is the new 'N'.

The impression a *cynic* might take is of people using the loss of the N grade as an excuse for a wholesale lowering of supposedly absolute thresholds in many subjects.

PS - a correction to post 115 - the 'Maths (including Further) figures were for the years to 2003, not 2005.

Combined-sex figures including 2006 now available at:
http://www.bstubbs.co.uk/a-lev.htm