By Mike Baker
BBC education correspondent
Well, it could have been worse. A-level results day is never very sane.
Some years there is outrage at the declining standards of A-levels across the board. This year it was psychology which took the flak for being allegedly "easier" than other subjects.
Of course, there was the usual hyperbole from the usual suspects. Someone from the Institute of Directors said A-levels were "meaningless". It is hard to think of a more insulting, and poorly timed, thing to say to students.
Even if you believe A-level standards have declined (and one could have a serious, but probably inconclusive, discussion about this), it is surely absurd to say they are "meaningless".
It remains a simple fact that most of the population find A-levels beyond their ability.
Similarly, those commentators who decry a pass rate of 94.5% should be more careful with their language.
All the careless talk about exams which virtually no-one fails gives a misleading impression. I know from viewers' and readers' emails that many people really believe everyone gets A-levels.
On a radio discussion programme I asked two senior newspaper journalists (neither of them education specialists) what proportion of 18 year olds they thought managed to achieve three A-level passes at the minimum pass-grade of E.
"Sixty per cent," said one. "Eighty per cent," said the other.
What do you think?
The answer is just over 25%. Yes, not much more than 1 in 4 of the age cohort achieve three grade Es or better at A-level.
The confusion arises from the fact that on A-level results day we are not given statistics relating to the whole age cohort but simply to those who entered for the exams.
Since the new AS-level gives everyone a half-time result on their A-level course many - quite sensibly - decide to drop their weaker subjects rather than proceed to the full A-level.
Only one in four 18 year olds get three A-level passes
In other words, the AS-level is a filter which removes most of the potential failures at full A-level. Hence the pass-rate goes up.
An even more significant filter is the GCSE. While the newspaper headlines give many the impression that everyone passes A-level, the reality is that fewer than 40% of the age group go on after GCSEs to attempt A-levels.
In other words, if you imagine examinations as athletics event, the GCSEs are the heats, the AS-levels are the semi-finals, and the full A-levels are the final.
Those who reach the final are the best. We should not be surprised when they do well.
This analogy, though, brings me to another objection made by the critics of A-levels. They say that with so many students getting top grades (and certainly many more do than in the past), how can universities distinguish between the brightest students?
This is a genuine concern. The top universities have to choose between applicants who may all have three A grades or better.
Best of the best
However, we have to ask whether this difficulty (which could be resolved in other ways) is a reason for condemning the whole system.
At the heart of this issue is the question: what are exams for?
There are usually two main definitions of an exam. The first is that it should be a means of identifying winners and losers.
The second is that it should be a test of what students know.
Now, to some extent, the poor old A-level has to try to meet both of these aims.
It was, of course, introduced primarily to sort winners from losers, by identifying who should go on to university.
But back in the 1950s there were university places for fewer than one in 20 school-leavers. Today, just about everyone who takes A-levels will go on to higher education.
Most successful A-level students go on to higher education
Moreover, the old A-level put a quota, or limit, on the number of students who could pass each year. Similarly, the number of A, B, C, D and E grades was fixed in advance (for the technically minded, this is known as a norm-referenced examination).
In other words, no matter how well teachers taught and no matter how hard students studied, the same percentage would get an A grade each year.
This was fine when the sole aim of the A-level was to filter students for a fixed, and very limited, number of university places.
But things are different now. The A-level became a criterion-referenced exam. In essence this means it is less like a rationing process and more like the driving exam: whoever achieves the standard required will pass.
If we believe that the knowledge and skills acquired on A-level courses is something worth having, then we need an examination which tells us whether or not people have mastered that knowledge.
If, on the other hand, we simply regard A-levels as a sorting process for university entrance then we might as well go back to the old norm-referenced exam, with its quotas.
Marks, not grades?
However, the choice between exam types need not be so stark.
The current approach does a good job of showing which candidates have mastered the knowledge and skills required.
It fails primarily in one aim: differentiating between the very bright students, who are getting a sheaf of A grades.
There is a simple solution: use the A-level marks achieved by candidates, not just their final grade.
On their results slip, every candidate is given something called a "uniform mark" for each of their papers. This is more specific than their overall grade.
So, for example, where the maximum mark for an A Level is 600, an A grade requires at least 480 marks, a B grade 420 marks, a C grade 360 and so on.
So, an A grade student may have scored the maximum 600 marks or they may have just scraped in with 480 marks. This is a big range.
Across three A-levels the range between candidates could be even greater.
Someone with three maximum scores would have 1,800 marks, while someone who just achieved their three A grades with nothing to spare would have scored 1,440 marks.
If university admissions officers had access to these mark scores, they would have a further tool to help them distinguish between straight A candidates.
That way we might get an end to the annual ritual of critics saying a 95% pass rate makes A-levels "meaningless".
We welcome your comments at firstname.lastname@example.org although we cannot always answer individual e-mails.