Actually, It Was About Cheating
August 24, 2012
A bar chart summing up the debate on results day
As I said in my last post, English grades C and above were only down 1.5%, less than they normally go up by. Overall GCSE grades C and above were only down by 0.4%, and this was before the appeals have been processed. No major change and exactly what would be expected if there was to be no grade inflation.
However, the Twitterverse was still outraged yesterday and I couldn’t understand why. It was seen as utterly unfair that grade boundaries had moved between January and June. One maths teacher pointed out that this often happens to maths A-level, so why should English be any different? From January to June the context has changed, and the exam has changed, so why shouldn’t the boundaries change?
But I was eventually corrected. The exam hadn’t necessarily changed since January. People were reluctant to say it explicitly, but 60% of the English language GCSE from AQA (I haven’t had time to check the other exam boards but I assume it was similar) was controlled assessment. This meant that controlled assessment could be used as the final exam. Students were able to finish their GCSE with an exam that was released months in advance, for which they had time to prepare their answers, and for which the teachers had already seen the mark scheme. Also, as it was modular, the teachers also thought they knew exactly how many marks were necessary for a grade C.
This is what I hadn’t realised. English teachers were in a position to cheat, bend rules, or legitimately provide extra, helpful advice to get (almost) every student to grade C. This was thwarted at the last moment by a change in the boundaries that meant that what they thought was good enough for grade C (and had been good enough for grade C in January) no longer was. This is why they felt robbed. They had been given a chance to hand out grade Cs like they were prizes (that all must have). They thought that if they followed the right strategy then the grades were in the bag, but then they were told: “actually that’s not good enough”. Having set an exam that almost anyone could pass, the exam boards would have been on the verge of a national scandal and could only get out of it by stabbing everyone in the back at the last minute. No wonder some schools were down massively, the more you gamed the system the more kids you would have who were expecting grade C. No surprise some English teachers felt betrayed, they had done everything possible to get dozens of barely literate kids a grade C and it still didn’t work. So that’s what it was all about, not a 1.5% fall in passes since last year, but a massive drop from the expected situation where almost everybody could pass.
Of course, it is at this point that the outrage starts. How dare I suggest that controlled assessments aren’t fair? Yes, they are exams for which the exam papers and the mark schemes were released long before they were sat, but they were only released to English teachers. Am I daring to suggest that English teachers would cheat, bend rules, or find technically permissible but ultimately unfair ways to get everyone up to what they thought was the required grade. What a terrible accusation. I should name names or shut up. No doubt I hate teachers if I think they aren’t scrupulously fair and honest with their assessments. I must provide conclusive evidence.
Of course, I cannot provide conclusive evidence. Sure I’ve heard the stories but I won’t grass up the teachers involved, they were put in an impossible position. Clearly this is just a smear on my part. But let’s review the details here:
1) The basic principle that final exams and their mark schemes are secret and not distributed to teachers beforehand is the norm for a reason. To ignore it for English seems to amount to a claim that English teachers are more honest than everyone else.
2) Teachers, particularly in core subjects, are under ridiculous pressure to get results. Schools have been told they will be forced to be academies if the number of students with GCSEs including maths and English falls below a certain threshold. Teachers are “performance managed” with pay and career progression dependent on getting certain grades from their students.
3) Teachers working for students from deprived backgrounds can view the extra help they give, even when against the rules, as simply providing the help middle class students could get from parents and private tutors. It is easy to view it as being fairer than not helping.
4) The fine line between cheating and not cheating is sometimes hard to draw. It is normal for kids to practise an exam before they do the real one. If the teacher knows what will be in the real one, it can be difficult to avoid making the practice one very similar.
5) Teachers, when they have the anonymity of an internet forum, openly admit that cheating occurs and is widespread. http://community.tes.co.uk/forums/p/565453/7374692.aspx#7374692 and http://community.tes.co.uk/forums/t/565479.aspx?PageIndex=1 are just a couple of examples.
6) There is a history of cheating in the system. Controlled assessment replaces coursework. Anyone who knows anything about teaching knows teachers cheat on coursework. I’m not the only blogger to point this out but you don’t need to rely on anecdote, it all came out in the press a few years back with official reports and exam boards conceding that it was going on.
The claim that teachers don’t cheat, bend rules or give technically legal but blatantly unfair help to target students is absurd. People will no doubt be outraged that I don’t believe the claim, but they are insisting that all personal experience, past history, knowledge of human nature and good practice in examining is irrelevant to the question. They are simply arguing that we must believe the absurd rather than believe something that could put their results, or teachers in general, in a bad light.
Scenes From The Battleground 
An excellent post and further evidence as to why a linear course (and the removal of coursework) can’t happen soon enough
About halfway through, while the thread of controlled assessment is retained, this post pivots from a discussion of what happened yesterday, to a discussion of controlled assessment and its attendant perverse incentives, which increase when they are part of the terminal unit. These should be kept separate.
When you write “having set an exam that almost anyone could pass, the exam boards would have been on the verge of a national scandal” it presumes the absolute inability of students to meet an objective performance standard en masse. If this is what the exam board, and the government, intends, they should say so explicitly. Instead, they have said that a work which contains the following a,b,c qualities relating to standards of English should receive grade x, and that is what they should mark to until they have stated otherwise.
Then follows the idea that the results taking an upwards jump is IN ITSELF a smoking gun to prove cheating, so much so as to justify a blanket mark change which affects every single student whether their teacher ‘played the system’ or not. This is nonsense. The exam board have a system by which to verify results, if that system is not fit for practice the exam board should take the hit for that. I think it a fair statement to say that children have the right to be presumed competent until proven thick on the basis of work produced, and should be the last to suffer the effects of poor management by adults.
But even if you’re prepared to go along with that, it’s ineffective. IF we assume widespread cheating, how would the results have looked yesterday without any cheating from any quarter? erm, horrific. Let’s imagine you are a ‘cheating school’. Turns out getting students low C’s wasn’t enough. What’s your next step? Let’s imagine you are a ‘non-cheating school’. You played by the rules (though you know they can be bent) and the exam board changed them. What might be your next step?
One can agree with everything you’ve written about from the paragraph beginning “of course” downwards without falsely bundling that up with what happened yesterday. Controlled Assessments may be “about cheating”; “It”, actually wasn’t.
It pivots to deal with the most likely response to the claim the exam structure encouraged manipulation, namely “how dare you make that accusation!” – a response that turned up more than once on Twitter.
You mean it assumes that not everybody can be a C (or above) student by the end of the course? Doesn’t seem unreasonable.
Assessment criteria have always been a nonsense. They can always be met in easy or difficult ways. They can never be the minimum, nor the maximum, required for a grade. Criteria based assessment never works. The criteria are too vague to judge or too detailed to apply to real people.
If you design an exam which is easy to manipulate, and then suddenly grades jump through the roof, then it is hardly a leap of imagination to think the two are connected.
So we give out Cs for nothing, until such time as the exam is competently designed? Shifting the boundaries is not ideal, but it is hard to see another alternative. It has always been accepted that kids don’t have a right to higher grades to compensate for teacher/school/exam board incompetence. Grades have to reflect their ability, even if a better system/school/teacher would have done more to improve their ability.
I’m the first to acknowledge that this ridiculous exam would, if left unreformed, encourage more cheating year-on-year. Fortunately the tide has turned on modular exams, early entry and (with any luck) controlled assessment.
Cheating may be a strong term to use; I would use gaming the system. The system was badly designed and, due to the pressure they are under, teachers will try and manipulate the system to the greatest advantage for their students.
I can not blame the teachers for trying but equally, the exam boards have to control grades being produced and can not have massive jumps. Change in grade boundaries like this has happened before with AQA, when core and additional science were first introduced only a few schools cashed in their core course at the end of Yr 10 with the majority waiting until the end of Yr 11. Unfortunately for them the grade boundaries changed significantly when they had a much larger cohort of marks submitted.
The proposed reduction in complexity of the courses can only be a good thing and will ensure that teachers push their students to achieve the highest possible marks, not a mark that will just pull them over the C grade boundary.
When I used the word “cheating” in the previous post I was blogging late at night, and might have been better advised to say “manipulated”.
However, here I think it is important to be blunt. The manipulation includes a lot of cheating, not always deliberate but definitely activity in violation of the rules. It also includes activities that, if not against the rules, would still be seen as cheating by the public. Perhaps if we see ourselves as honest it is hard to admit that this final category is cheating, but if held up to scrutiny we’d look like MPs in the expenses scandal saying “well I didn’t break the rules in claiming all that public money for my own benefit”.
An exam system needs to look clean as well as be clean, and it is that (not the teachers who often have no choice) that I am condemning. Not as if we are free to whistleblow and then continue our careers.
I call a spade a spade. I have witnessed coursework cheating, I know of teachers that have admitted it and I have even taken action to prevent it in younger staff.
I know of a TA who saw regular cheating during the old ks2 stats tests too. The teachers wandered around the exam hall giving hints and help.
I myself when younger probably used to ‘over help’ weaker kids with coursework and so on- until I wisened up and thought ‘hang on- its not my GCSE! They should get what they are capable of as individuals!
I feel back in the day I probably crossed the line from ‘support’ into ‘leading’ at times. My excuse is immaturity, over keenness, wanting the best for my kids, slight worry that my kids grades would reflect on me etc.
But what I did was nothing compared to some other stuff I have seen or heard about.
One exam board asks for its ‘random sample’ (for inspection purposes) to be sent to them BEFORE the centre has to submit its marks. So if a school wanted to they could give inflated marks to all the other kids but give fair marks to the sampled kids. Im quite sure many schools abuse this although I don’t think my current school does.
Anyway coursework is tricky on this issue-
When does a ‘hint’ become unfair help?
How explicit can the initial teacher instructions be?
Are writing frames ever acceptable?
Can you answer a direct student question?
Can a teacher point out a silly error that would ruin otherwise good coursework?
Can you issue ‘ideal’ exemplars for training?
Can a teacher ever look at a draft essay/coursework?
We are taught to do this as part of AfL and feedback…
Can the kids confer?- we have taught them to group participate after all…
I think Gove is right to go for terminal exams and I also think there should one board per subject. We are a tiny country after all.
I think a good model for an individual subject would be:
Terminal Exams 80%
Coursework During Course 10%
Conduct/Attitude/Attendance/Reliability during Course 10%
And the combined 20% moderated by software so that it cannot be more than 10% above the average of the terminal exams %. Or a sliding scale for schools that have appeared too generous with coursework compared to the terminal grades.
This would mean that kids would have to do the coursework or it would likely cost a grade or even 2.
But the fact that its only 20% in total would limit the ability of schools to use coursework to artificially inflate grades.
ps I make no apology of including the american style addition of conduct etc into the equation for the final academic grade.
Brilliant. KB
When you make teachers responsible for individual children’s exam results, then cheating is of course one obvious outcome. (If it walks like a duck and quacks like a duck…it’s a duck!) The authors of Freakonomics spotted this some years ago when they looked at some 700,000 exam scripts in Chicago.
This has nailed it. Furthermore, there are companies out there who actively promote gaming the system. Pixel were actually name checked in Newsnight!
I used to ‘help’ shedloads of kids get decent grades on AQA Science ISAs. Otherwise they’d get few marks for drawing the graphs or tables etc. Even when I put the graphs / tables on the board, a good % of the pupils would still copy them so badly – to the point of missing a C grade.
I didn’t do this for the pupils. I did it for my benefit. If I hadn’t got the pupils’ grades up I would have been told to do another ISA, and another, and another, until the management drone in charge reluctantly accepted / realised the pupils just weren’t going to get there.
I have noticed that several lower-achieving schools in my area, have seen results increase over the c-d area again, boosting their overall gold standard etc often markedly, whereas schools that traditionally perform well have seen results stagnate, or higher grades have reduced by a small amount. I’d be very suspicious of the grade boundaries being concertina-d to get more pupils over the c-d hurdle at the expense of higher grades. This is what seems to have happened almost across the board in this totally corrupt unfit-for-purpose system we are running. When the spokeswoman for the standards of assessment comes across on Newsnight as shiftier and more evasive / professionally coached to say effectively nothing than the slipperiest politician, you just know something is up.
This was a great nation that helped win a world war at great cost, and built a nation fit for heroes complete with a welfare state out of the ashes of an awful conflict. When did these ignorant careerist monkeys get to the top of the ladder at the expense of people who can actually do things and talk straight?
I also marked for Edexcel and the answers that were acceptable as correct were mind-boggling, but that’s another story. Nice bit of cash though.
Difficult one Judge N,
Your drone would have had someone behind him/her insisting grades go up.
That someone will have had the Head teacher on their ass.
The Head teacher has the Board of Governors on his ass.
The governors have to get grades up or the reputation will suffer and y7 roll with suffer and/or get higher number of undesirables and less aspirational kids.
Then the rot will really set in. Exclusions go up, teachers leave, results go down more, spiral in place.
pity the poor drone!
Solution? Ban the publishing of school results?
Rebuttal to solution? Freeedom of Information, Parents Right to Know & Choose.
I cant see this ever ending…. its a catch 22.
Thanks for a great explanation of the ins-and-outs of the English exams and the impact of controlled assessment, for those of us who didn’t know. Your explanation of what has happened sounds spot-on.
What this all says to me is that schools have to find new ways to get children to want to be highly literate and numerate, because the pressure to hit literacy and numeracy targets, both at primary and secondary levels is just going to grow and grow. This is not just Gove it is a cross-party realisation (e.g. Michael Barber in last week’s TES), so floor targets are going to keep rising and massive pressure will be put on schools to find new ways to raise achievement. The gaming and the perverse incentives have been sussed and will be progressively closed down around league tables, while the league tables will stay.
Schools could follow the kids of approaches you can see in the schools you will find at http://www.naace.co.uk/thirdmillenniumlearningaward. It’s only by changing the attitude of the kids that schools will get the step-change in achievement, particularly of those living in environments that don’t help. And this is a whole-school thing and hence the responsibility of the school leaders. It must take a great deal of courage for teachers in schools near the floor target to resist gaming the system and giving extra help, when the stakes are so high for the school leaders who pressure the teachers. The leaders of such schools will either have to find ways to get the higher achievement without demanding the impossible of the teachers, or get out of the way and let others do this.
This:
http://www.guardian.co.uk/education/2012/aug/26/gcse-exams-woodside-high-school-joan-mcvittie
is interesting in the context of this accusation. Note the timeline of exams taken in Jan (why?)
The only good thing about the current GCSE fuss is that it does expose where politics masquerades as education. Nice to see something I wrote about 5 years ago is finally being grasped.
You’re like Lance Armstrong’s team mate… :) A good post as always. I have mixed feelings on the issues at hand as delineating clearly exactly what is and what isn’t acceptable is often quite tough. I’ve often analysed mark schemes with pupils prior to exams to ensure that they know what they need to do to gain a good grade (I have, of course, taught them all the underlying concepts previously though). Is this ‘unfair’? Am I swindling the system in some way but ‘teaching to the exam’?
The alternative is that I send kids (often intelligent, motivated and deserving of success) into exams (or controlled assessments) where they’re not quite clear on what they’re being assessed on and run the risk of them not demonstrating the skills required to succeed even though I know that they possess them e.g. even if they are au fait with advanced punctuation, if they don’t drop a couple of semi-colons into their exam response the examiner cannot tick a particular box on the mark scheme and their (deserved) grade is likely to suffer.
Perhaps this is less of a problem in other subjects where questions are more closed in nature?
Ultimately, its a difficult one to solve unless we pass more responsibility for exam results to individual pupils (which would naturally herald a concomitant abdication of responsibility from teachers which would likely bring its own set of problems I suppose).
Controlled assessment is worth 60% ? Are you sure? You might want to check that.
Scrap that. I realise you are considering s&l in that too.
An interesting aspect of this conversation is that the examining board we work with, AQA, did not provide the grade boundaries in advance for the Controlled Assessments at all. They provided clear assessment rubrics and support programmes to allow us to ensure we were marking consistently to these rubrics. This had us placing the students into ‘bands’ of 1-5 or 1-6 and it was clear that these could not be conveniently translated into grades.
Our department used this as an opportunity to explain to our senior management that we were thus unable to provide them with any secure information about what grade these students would eventually get for their GCSE. We were indeed under extreme pressure to provide a list of ‘students on the borderline’, however since we couldn’t provide such a list – we simply didn’t have that information – we instead worked with all the students to try to assist them to develop the skills and knowledge they needed to both be successful in the assessments and also in their loves to come.
The PISA rankings would be useful statistics to use as a reference point to demonstrate that ‘grade inflation’ does occur in this country – as what PISA suggests is that the students in the UK are deteriorating in English literacy in relation to other countries.
New Zealand, where I am from runs a standards-based national assessment scheme. In such an environment, assuming assessment is done rigorously, there is no reason not to aspire to everyone passing in the end. But only if the quality of education has improved, not just to make people feel better.
I have been dismayed about the depth and severity of the cheating in schools in the UK. I have seen shocking evidence of this in schools in the last couple of years. I’m proud to say the school in which I currently work still maintains high standards of professionalism – though we are also living constantly under the threat of extinction as a result.
The obsession with grade boundaries and assessment results as a measure of school effectiveness has completely distorted education in the United Kingdom. Yet, I still see light at the end of the tunnel.
Last year we told our students, their parents and the SLT that we couldn’t confidently predict the students’ grades for GCSE, but that we had worked our level best (as had the students) to be as well-prepared as possible – and our students did us proud. They exceeded the performance of the prior year’s cohort by 14% and blew any thought of our predicting their performance in advance out of the water.
In the end, with these major terminal summative assessments – it all comes down to the play on the day, after all.
[…] It’s as though getting your school higher up the league table is more important than the outcomes for individual children. It’s almost as if achieving the top grade in Ofsted is more important than children being able to independently achieve the top grade possible for each of them in each subject they study. Of course many people would argue that these things are synonymous. A high league table position means that all children achieve their full potential. Or does it show that the leaders within those schools are merely good at ‘playing the game’. Some people have even suggested that there may be cheating going on. […]
[…] However what I want to draw attention to is that today’s headlines come at a time when I believe we have an over-supportive assessment system in this country. Some people have pointed the finger at how the system has been geared to judge the school and not the pupil. Others have expressed their fears that our system is being ‘gamed’ – teachers and school leaders are ‘playing the game’ to make their schools look good, without actually doing the job of raising standards through good teaching. A helpful way of looking at this may be David Weston’s hastily but well-drawn graph of the continuum between teaching and cheating. Or you just might say that actually we’re all cheating. […]