The Exam Hysteria Continues…

August 26, 2012

This has been written on my phone, so apologies for the lack of links etc. I may add a few bits and pieces in a few days time when I have internet access.

The backlash to the end of grade inflation has continued. Politicians; newspapers of all stripes, and many different types of teacher have labelled it a crisis. Most have obvious agendas. Politicians need to score points. The media needs stories. Teachers, particularly headteachers, need excuses now that grade inflation has ended and a bad year will look like a bad year. All want some excuse to say that there is a problem, without actually saying exams should get easier forever.

The desperate hunt for an anomaly in the exam results (beyond the mere fact that this year grades didn’t go up) has focused on AQA English language GCSE. As I explained last time, this exam was about as dodgy as it could get. Thanks to a new style of exam where 60% was “controlled assessment” English departments were in a situation where it seemed there were few obstacles to getting kids to grade C and they were only prevented by a last minute movement of boundaries. Now a moment’s reflection would make one think: “hang on, does anyone really believe that almost two thirds of school leavers have a good standard of reading and writing is an underestimate?” It should be a source of national shame that English GCSE had reached that point (and let’s not forget English exams this time round included studying subjects as complex as celebrity interviews and reality TV) . But we have become so used to lowering the bar in English that when this time the standard was raised to a level of difficulty beyond anything we have seen since 2010, it came as an enormous blow to schools, particularly those who had manipulated their scores only to discover they hadn’t manipulated them far enough.

The claims of unfairness I have seen have focused on the following:

1) Controlled assessment boundaries moved between January and June. These were small moves, in fact if I have got this correct they were within the “tolerance” levels for controlled assessments (i.e. the amount schools can overmark by without anyone caring). However, these do mean that the same performance would be worth more in January than in June. Of course, the ridiculous situation where you can do the same assessment at different times of the year is a problem. However, once it’s been accepted we have to admit that once some students have done it for January, and it’s been marked, it is going to be a lot easier to do it in June. If you think that exams should peg grades to particular performances rather than to level of difficulty you might object. But if you do think that, then you are arguing for something that will inevitably result in grade inflation. It’s also worth adding that AQA had warned schools that these boundaries could move in the early years of a new course, although they didn’t specify that it could change between the January and June submission.

2) Grade boundaries on the formal exam went up by 10 marks. This is apparently shocking. How could two different exams have different grade boundaries? Well the clue is in the word “different”. This is not unprecedented; the example I keep hearing about on twitter is A-level maths exams which have had boundaries move by this sort of amount between January and June. There is no real ground for complaint here but going on about it has confused a lot of people who have assumed that it was the controlled assessments where the boundaries changed by 10 marks, rather than the rather more understandable situation of different exams having different grade boundaries.

3) The change occurred between January and June. I have covered above the way this has happened and why it is not cause for concern. However, one conspiracy theory has it that for such changes to occur then it must be the case that lots and lots of students must have got grade C in January and the only way AQA could stop grade inflation was to mark really harshly in June so that every excess grade C from January was compensated for by taking a grade C from students in June. However, AQA have now said that 94% of controlled assessments were submitted in June. This is not the clearest statement about how many students submitted at different times (some could have submitted both times) but it would make it highly unlikely that there were so many grade Cs in January that students in June had to take a fall. It also means that even if there was insufficient rigour in the January exam, it would hardly justify replicating it in June to avoid disparity.

4) Politicisation. It is an inevitable fact that whenever anything changes in education, somebody complains that it is “political”. Nobody ever explains why that makes it wrong. Nobody ever explains why the status quo is politically neutral. Personally I want education to be political. I want people to argue about the principles involved; object to the injustices, and appeal to the public for support for their ideas. Removing politics from governance is an incoherent idea, and can only really be interpreted as removing democracy.

5) Won’t somebody please think of the children?

I have lost count of how many appeals there have been to the suffering of students who have had to endure the ordeal of an exam slightly tougher than last year’s. Some have even implied that actually looking at the statistics to see what happened, rather than being outraged at the fate of any child who failed, shows my heartlessness. I am afraid that I have yet to find a reply to this argument which isn’t rude and angry, so I’ll save it for the comments if anyone needs it.

A final point: it has now been claimed that thousands of (probably very able) students switched from English GCSE to English iGCSE this year. If true, and I have understood this correctly, then we are no longer talking about English exams that toughened enough to cause the 1.5% fall in GCSE English passes (before appeals). We are likely to be talking about an exam that, given the change in intake, was even closer to last year’s in terms of grades given than even the 1.5% figure suggests. Unless some new evidence turns up, this is still looking like a fuss about nothing; a complaint based on innumeracy, politics and a desperate effort to avoid responsibility for the scandal of dumbing-down. If you don’t support dumbing-down and grade inflation, and you don’t have some sensational new piee of evidence, then there really is no excuse to join in.


  1. Before some wider thoughts I must address some areas where there may be technical innaccuracy/confusion:

    a) Highest ever English A*-C percentage was 65%, so ‘7 out of 10’ is a little misleading

    b) The controlled assessment boundary changes were not “within tolerance”. ‘Tolerance’ is a term which applies to the level to which the exam board will allow teachers’ raw scores to stray from the examiners’ raw scores for their pupil’s sampled controlled assessments. It does not make sense to apply it to exam boards’ changes, aside from which the changes would certainly have been ‘beyond tolerance’ if you did; that is to say, were the exam board assessing themselves they would have found their own marking diverted too much from their own professed benchmarks to be acceptable!

    c) Percentage fall of 1.5% is a measurement of state schools. the iGCSE shift was done by independent schools. Therefore while that shift caused a problem for the exam board, who had a different scaling and weighting to deal with when tackling the scaling of converting marks to grades, you cannot attribute the 1.5% results drop to this shift as the schools that shifted were never part of the previous measurement.

    • 1) I could claim a rounding up here, but this has more to do with the fact I was blogging from my phone and it wasn’t easy to double-check things. It was, of course, all GCSEs that have a pass rate of 70% not English alone. I have now corrected this.

      2) Your first sentence here seems to contest the claim that changes were within tolerance, but then the rest of the paragraph is arguing that we should ignore the claim, not that the fact is untrue. I’d be grateful if you could clarify. I don’t accept the argument that it is irrelevant. The point is that if teachers have been consistently marking at the high end of the tolerance levels, examiners cannot remark, but there is every reason to move the boundaries to compensate for this.

      3) This seems to be simply wrong. (If it were true then your point 1 would have been wrong.)

  2. In answer to your question on politics and teaching Andrew, “No body ever explains why that is wrong” I think I’ve consistently argued why for a long time on my own blog, but here goes.

    Strategies are formulated and implemented as policies in schools for two major reasons. The first is because the originators believe they will successfully and effectively educate a nation’s children. The second is because the originators have a political goal. The introduction of comprehensive schools and the demise of the terminal examination are prime examples of the latter.

    The moment you stand up to teach in front of someone else’s child, you lose any right you may normally possess as a political animal, precisely because you embody something far more important than politics: to that child you exemplify impartiality and reason. And I would argue that anyone who doesn’t appreciate that shouldn’t be teaching.

    • Sorry, you are just begging the question here. We can define “politicisation” as having a political goal, but we are still no closer to identifying why it is wrong and why some goals are political and some aren’t. There’s certainly no obvious reason why we should see the tripartite system as apolitical and the comprehensive system as political.

      • Andrew, I think it’s wrong when it privileges a political aim over an educational one…as in the case of the introduction of comprehensive schools where the motive was equity, not excellence in education; or when argument over the use of phonics strays from serious, academic research about cause and effect into discussions about whole texts and “relevance;” or when the introduction of the EMA was about trying to reduce the burgeoning numbers of NEETS rather than improve the educational performance of the same children. Just imagine how the £545 million a year the EMA cost (money we didn’t have) could have been spent, if the goal was to improve those same teens’ educational performance.

        The latest posts here exemplify that it’s become the norm to think that politics and education are inseparable.

        Try this mind experiment. Imagine your own child is being taught by a teacher who believes their job not only allows them to discuss politics with the pupils they teach, but to actively influence their nascent political thinking. Imagine then that same teacher espouses an extreme political viewpoint you utterly reject. Then contrast that with another teacher who resists every attempt by pupils to elicit their personal (never mind political) views… but instead teaches your child to think for themselves.

  3. The basic problem with the exams is the complexity of the modular system. This allows teachers to play a probability game focused at the C grade boundary without actually aiming to teach their subject in such a way that all their students make the maximum possible progress. A simple system (like IGCSE) would avoid situations like this which can only have a detrimental effect on the public’s view of the examination system

  4. Point 1. I totally agree that anyone who believes that education is above politics is not thinking straight. There are not many things which are so entirely political as education. Who shall go to university, how will it be funded. Will there be a national curriculum, what will be in it. Will we teach farming and practical subjects in all state schools. Will we have morning prayers compulsory for all etc etc etc. every decision made regarding education is political and a tool of government that can be used to shape the kind of society we want.

    2. Results. Sometimes we read articles telling us it’s that when only n% of 16 yr olds can get 5 good GCSEs this is evidence of a failing education system. This comes from many different interest groups. If the results increase, then this is decried as grade inflation. If the % pass rates are static, then all targets for improving can be binned immediately as impossible to achieve. I am in the camp that believe that grade inflation in many, but not all, cases has been caused by the use of inferior qualifications and widespread gaming. Summed up, either game the GCSE or switch to a DIDA/BTEC. Hidden amongst the grade inflated figures of the past years has been a particular casualty in the higher grades where students capable of A/B have been left to struggle to a C with little support. This was a lazy practice, brought about by poor leadership in schools, where the headline published figure has been deemed more important than the real quality of the education in the school.

  5. For an illuminating analysis of what was wrong try

  6. Frank it wasn’t very illuminating. Most was an emotional response and seemed to be arguing that students with clearly poor English skills (they found English exams hard) deserved C’s because they worked hard. For example an illuminating personal account would be one that gave us some hard percentages. What percentage predicted C’s failed to achieve them? The implication that the whole cohort was sitting on low C’s that turned to D’s comes across as rather ridiculous and blunts my compassion, despite myself being entirely used to an exam system that does your students over on one paper or another each year.

    • Hi, Heather. I see what you mean about the emotional content of much of the post.

      I think that regarding the % you are interested in, the author wouldn’t have access to the national picture. If he gave the figures for his own school that would only be anecdotal.

      The part I found illuminating was the section that concentrated on what happened in English. From the reports I’d read I’d assumed that there had been a drop in all grades whereas it seems it was just the D/C grade that was moved. If there is a feeling that there has been grade inflation (and I’m not arguing that there hasn’t) then why not rein in all the grades at C and above?

      If an exam is perceived as being easier one year than another, the Boards can adjust the grade boundaries to take this not account. This maintains some parity between different cohorts.

      The impression I am being given from much of what I read is that this year there has been a decision to ‘move the goal posts’ irrespective of how the candidates had performed. It wasn’t as if they looked at papers and said, ‘Well this has got the correct number of marks for grade C but clearly isn’t worth it so we’ll change the boundaries.’ The decision seems to have been made without any regard to the quality of work produced in the exams this Summer.

      Based on past experience and the vast amount of material now available from exam boards, teachers led the pupils to believe the work they were producing was at Grade C level. And indeed it was until Ofqual changed the boundary. They now find that it isn’t; rather like the ‘Good Lesson’ that is now deemed only ‘Satisfactory’.

      If they want to make it harder to gain a grade C in English then surely they should change the grade descriptors and the exemplar material accordingly, at the outset of the series, so that everybody knows what they have to achieve.

      I thought this section of his post was pretty compelling.

      • or… alternatively… children and teachers simply try their very best… and let the chips fall where they may.

        If a child (or teacher) is content to settle for a low C, they can hardly whine if they end up with a high D.

        They sailed close to the wind and got burnt (ha- Del Trotter Mixed Metaphors at your service).

        It seems painfully obvious to me why we should never have embarked on a modular system or a coursework heavy system as a nation.

  7. Heather,
    I have to concur- Im NUT but that letter sent to Gove is an embarrassment.

    Looked like it had been written by an irksome 16yo old.

    And even if one were to consider the English grades unjust (I don’t particularly) the conflation of other points in the letter would make it impossible for Gove to regrade the boundaries.

    If he was wavering on the issue, that letter (if it becomes well known) will mean he wont be able to!

    It’s ironic the letter concerned standards of English and alluded to conduct and fairness.

    They may as well have started the letter with the salutation “Dear Toffy nosed Dickhead”…

  8. I can see that if the same coursework got one one grade in Jan and another in June teachers will be up in arms. Because this happened in English not an option subject this made the issue high profile. However…. What does an exam board do when it is faced with mountains of students all just over the C boundary because of rampant gaming of the system? In practice there has to be a bit of norm referencing and there always has been. The evidence is there in the claims of schools that so many of their candidates suffered. Why on earth were so many candidates all low C’s? Perhaps this should be investigated as actively as the boundary changes?
    Why is this scenario suddenly so mightily unjust when changes in grade boundaries often cause outcomes that are unjust to individuals? One GCSE subject in my school this summer have been totally done over by changes in the coursework boundaries. Around 25 A stars per year is average for this subject and this year they got three (sorry don’t remember the other stats off hand, just A star as it caused most stir.) The same thing happened to my subject last year. I’m baffled that the English situation can cause so much protest when outcomes that are highly ‘unjust’ are a normal part of ‘exam roulette’.
    Anyone on an assumed low C must know they are quite likely to end up with a D.
    As a fairly experienced examiner I also find it odd that such a fuss can be made about a difference in marks that is within the margin of error allowed for any examiner. The naive assumption that English papers are marked with such accuracy that this change in grade boundary is more significant than what happens every year to a large minority of papers – marked a little strictly but within tolerance, is also just ignorant. Who you happened to get as a marker is more significant than the boundary change.
    It also makes me feel angry that year in and year out I have to put up with dodgy marking on one paper or another (mainly A Level) but the teacher unions have shown no interest whatsoever in investigating what is causing the problems and trying to get the exam boards to make marking more consistent. I know they are cutting corners at standardisation and many colleagues are finding that the marking of our papers is becoming more and more erratic. The impact on students is enormous.

    Teaching unions please actually represent your members and students by doing something about this, instead of trying to score political points against Gove.

  9. When you realise WHY the board changed the boundary (a mountain of kids with coursework over the C boundary, ruining the normal distribution of marks) the letter seems even a bit more ridiculous.
    The complaint that there is some form of elitism because only the C boundary was changed is a bit comic. No, there was a lot of anti-elitism – in schools- which meant only the C boundary was viewed as worth very actively ‘massaging’.
    I assume inflated coursework marks is also why grade boundaries for coursework went up so high in other subjects. However, as those subjects didnt already know where the grade boundaries would be there was not great unnatural bunching – just inflated marks. Schools like mine where the subject involved was relatively law abiding got totally shafted by rising grade boundaries because too many other schools were less scrupulous.
    Publishing grade boundaries in January was the big mistake and all this self righteous indignation is particularly nauseating because it abuses the ignorance of the public that has no idea how the system works. I feel very sorry for schools that were quite honest in the way they ran the coursework. They will be sitting on rotten marks now. Their kids really have been done over.

  10. Heather, again you are correct which in turn is a strong case for

    1. lowering the % contribution of coursework to the final GCSE grade and

    2. having a formula limiting the ability of courswork marks to inflate a grade. ie ensuring any CW marks are ‘adjusted’ to within a certain range of an individuals terminal paper marks.

    This eliminates the cheating effect but still requires kids to complete work during the year. Its a lot cheaper too.

  11. Cheers Rob – sensible solutions!

  12. […] been a lot written over the past week about the GCSE English results and the alleged unfairness in this year’s […]

  13. […] Old’s dissenting view here and […]

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 )

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: