GCSE grades fall for the first time: were they manipulated?

[Truronian]

http://www.bbc.co.uk/news/education-19349444
http://www.guardian.co.uk/education/2012/aug/23/gcse-results-fall-first-time?newsfeed=true

A few titbits from the articles:

This year's results show 69.4% of entries earned grades A*-C, compared with 69.8% last year.

In English language and combined English literature and language exams, results went down from 65.5% getting A*-C to 64.2%.

And head teachers representing dozens of schools say some students have been marked down by an entire grade in English compared with the results that teachers had predicted.

Brian Lightman, general secretary of the Association of School and College Leaders, said: "The grade boundaries in English have been moved up on the C/D boundary, moved up substantially by about 10 or 18 marks. This has been done at the end of the course. A number of schools have had a serious effect on their 5 A*-C indicator. But it appears standards of achievement haven't changed one iota, it's just the way the standards of marking have changed. English is the key to all the professions, it's the key to university."

And a horrible graph:

(I seriously have no idea how the BBC can get away with publishing such a naff graph)

So GCSEs are out, and results have dropped in all categories for the first time ever. Despite OFQUAL being and independent organisation, there is little doubt in my mind that this is down to governmental pressure.

I've highlighted the English results as they are a subject which has been particularly effected. The 1.3% percent average drop is translating into 10-20% drops in some schools that contain a large number of C-D borderline kids, indeed it's the C-D boundary that has been hit the hardest. The manner in which it has been done is also very underhand... there exam boards have merely shifted the grade boundaries by several marks at the end of the course without warning. This means schools have been planning intervention, setting predicted grades and budgeting based on predicted 5 A*-C results that they had no way of knowing were too high. It's also worth noting that the altered grade boundaries did not effect January entries, so students who entered in January have ended up with a better grade than those that entered in June despite being of similar abilities in English.

I'm really quite annoyed about this. It feels as though the outgoing Year 11 students have become victims of the 'stop dumbing down' movement and have been unfairly subject to harsh grade boundaries that will adversely effect their future prospects. I have no problem with improving standards, but merely marking exams more harshly is not going to do this, and in fact it's going to have the opposite effect when you don't tell teachers that the grade boundaries are going to change dramatically.

EDIT: Just to clarify, my annoyance is with what's happened in English.

 

[filli_noctus]

Considering it's also the first year A-level grades have dropped it seems plausible that artificial grade depression is at work. Theoretically (but not certainly) A-level grades should have a similar curve to GCSE's on a two year lag.

 

[MrCynical]

I'm not convinced there's much evidence that the exams have been made unfairly harsh or otherwise manipulated here. There's only been a very small decrease in the percentage of top grades. Unless either intelligence or teaching quality has been increasing steadily this graph ought to be a flat line. Even with a steady improvement you'd expect some oscillation about the trend line rather than an increase every single year.

The one bit of evidence that might point to deliberate intervention is that the A-levels also show a similarly miniscule decrease in top grades this year. The fact this has happened in both sets of exams in the same year, after a few decades of steady increase is probably the most convincing sign of deliberate intervention. A decrease in either one alone would not be surprising (and should really have been seen plenty of times over the last few decades unless grade boundaries have been eased year on year).

I'm really quite annoyed about this. It feels as though the outgoing Year 11 students have become victims of the 'stop dumbing down' movement and have been unfairly subject to harsh grade boundaries that will adversely effect their future prospects. I have no problem with improving standards, but merely marking exams more harshly is not going to do this, and in fact it's going to have the opposite effect when you don't tell teachers that the grade boundaries are going to change dramatically.

Well if you look at that graph there can be no arguing that there's a problem - extend the trend ignoring 2012 and you hit the 100% mark rather soon. Even if it tails off the closer the average gets to the maximum, the less useful the exam is for actually distinguishing between students. Whether it's playing with the grade boundaries or bringing in a more rigorous exam, grade inflation has to be curtailed somehow. Best way to do that might be to introduce an exam with a different name and much harsher grade boundaries, just to make it clear grades before and after are not equivalent.

I can't see this year's drop having much of an impact on students - their results are still the highest except for 2011. Should we regard everyone from 2010 and earlier as being subjected to unfairly harsh grade boundaries as well?

 

[Truronian]

I'm not convinced there's much evidence that the exams have been made unfairly harsh or otherwise manipulated here. There's only been a very small decrease in the percentage of top grades. Unless either intelligence or teaching quality has been increasing steadily this graph ought to be a flat line. Even with a steady improvement you'd expect some oscillation about the trend line rather than an increase every single year.

Let me elaborate more on the English controlled assessment grade boundaries. English controlled assessments are papers sat in school, marked by school teachers and than sampled for moderation. They are selected from a pool provided by the exam boards. They can be completed at any point in the year and submitted either in January or June. AQA (maybe other exam boards as well) have made the unprecedented decision to up the grade boundaries on these controlled assessments in June: pupils need 46 marks for a C rather than the 43 they needed in January. Nothing has changed about the work, nothing has changed about the mark scheme. This means that the same piece of work submitted at different times of the year would get different grades, which is nothing short of scandalous in my eyes. It's causing distorted statistics aswell: schools that submitted all controlled assessment in January are looking comparatively better than those that submitted in June. This is one of the major reasons that I can only conclude that manipulation of the grade boundaries to get a palatable results spread is responsible for this dip.

The one bit of evidence that might point to deliberate intervention is that the A-levels also show a similarly miniscule decrease in top grades this year. The fact this has happened in both sets of exams in the same year, after a few decades of steady increase is probably the most convincing sign of deliberate intervention. A decrease in either one alone would not be surprising (and should really have been seen plenty of times over the last few decades unless grade boundaries have been eased year on year).

Without trying to defeat my own argument, a drop in A-Level results could also be down to new exam syllabuses. The real issue with the English GCSE is that the grade boundaries have been altered at the very end of the course.

Well if you look at that graph there can be no arguing that there's a problem - extend the trend ignoring 2012 and you hit the 100% mark rather soon. Even if it tails off the closer the average gets to the maximum, the less useful the exam is for actually distinguishing between students. Whether it's playing with the grade boundaries or bringing in a more rigorous exam, grade inflation has to be curtailed somehow. Best way to do that might be to introduce an exam with a different name and much harsher grade boundaries, just to make it clear grades before and after are not equivalent.

I don't disagree. I think the rise in exam results is primarily down to improving standards of teaching, but competition for the most passable exam is also no doubt partly to blame. It is not the drop that concerns me here so much as the questionable decisions from the exam boards. My own subject (Maths) has dropped this year as well and that doesn't concern me: there are no sinister behind the scenes actions at play.

I'm undecided on the idea of a new examination. The GCSE has its flaws (grade inflation being one of them), but I'm concerned that any examination introduced by Michael Gove would contain topics that Michael Gove wants rather than topics that would actually be useful for students.

I can't see this year's drop having much of an impact on students - their results are still the highest except for 2011. Should we regard everyone from 2010 and earlier as being subjected to unfairly harsh grade boundaries as well?

Well, the English results have had a major impact on some students in my school. They were told they were on target for a C and so focused on other subjects where they were below target, and then have ended up with a D. It's devastating for the English teachers as well: officially this 10% drop (in schools with a large numberof C/D borderline kids) is on their heads.

It's not just students I'm concerned about. The government has raised floor targets from 35% to 40% (5 A*-C inc Eng and Maths) this year. Schools that fail to meet these targets are being threatened with closure of being turned into academies. It's not fair to tell schools to work towards getting 40% and then at the end of the year change the requirements for getting a C: it undermines the works schools do in improving standards. Of course, it does allow Gove to turn more schools into academies... convenient, eh?

 

[MrCynical]

This means that the same piece of work submitted at different times of the year would get different grades, which is nothing short of scandalous in my eyes. It's causing distorted statistics aswell: schools that submitted all controlled assessment in January are looking comparatively better than those that submitted in June. This is one of the major reasons that I can only conclude that manipulation of the grade boundaries to get a palatable results spread is responsible for this dip.

I can see your point there (controlled assessment is what they're calling the coursework component at the moment isn't it?) On reflection though is this actually that different from the usual normalisation process used to calculate UMS? The same actual number of marks on an exam one year won't get you the same score in UMS (and so not necessarily the same grade) as the next year. That can be the case even if the mark scheme doesn't change. I think these kind of drifting grade boundaries are a large part of the problem, but I can't get too scandalised that a boundary has drifted up for once against the usual trend. Bit of a mistake to do it in the middle of the year I'll grant you - what if it had been done in September instead?.

I don't disagree. I think the rise in exam results is primarily down to improving standards of teaching, but competition for the most passable exam is also no doubt partly to blame. It is not the drop that concerns me here so much as the questionable decisions from the exam boards. My own subject (Maths) has dropped this year as well and that doesn't concern me: there are no sinister behind the scenes actions at play.

You've highlighted a 3 mark change in the grade boundary for the English coursework. Is this a one off problem, or are there other changes you consider sinister? That isn't a large enough alteration to account for the difference in results averaged over all subjects. BBC also notes a decrease for science for example - anything odd happening there?

Well, the English results have had a major impact on some students in my school. They were told they were on target for a C and so focused on other subjects where they were below target, and then have ended up with a D. It's devastating for the English teachers as well: officially this 10% drop (in schools with a large numberof C/D borderline kids) is on their heads.

It's indisputable that there are borderline candidates where those 3 marks did in the end make the difference between a C and a D. However I don't think much of this "they thought they were on target" argument with the implication that if they'd known of the boundary change they'd have put in more time and effort to reach the higher target. You're talking a three mark change in one component of the English GCSE, in which the marking is highly subjective. I'd expect more than a three mark variation simply depending on who is marking it. The idea that a given student could run that close to the boundary and still sensibly be confident of a C is ridiculous. As is the idea that you could know to that degree of precision whether they were "on target" or not. I agree some people did lose out to that grade boundary change. I don't agree that them knowing about it in advance would have made a difference in their actual marks.

It's not just students I'm concerned about. The government has raised floor targets from 35% to 40% (5 A*-C inc Eng and Maths) this year. Schools that fail to meet these targets are being threatened with closure of being turned into academies. It's not fair to tell schools to work towards getting 40% and then at the end of the year change the requirements for getting a C: it undermines the works schools do in improving standards. Of course, it does allow Gove to turn more schools into academies... convenient, eh?

Again, I have my doubts that you can predict English results to anywhere near enough precision for this change to make the difference between "confident we'll pass" and failing. Some schools at the very margin may have ended up on the wrong side of the line because of it though. I would also say that GCSEs are much more important to the schools than to the individual students. I'm not going to be too convinced this is the result of politics without rather more evidence of unreasonable changes in grade boundaries.

 

[Truronian]

I can see your point there (controlled assessment is what they're calling the coursework component at the moment isn't it?) On reflection though is this actually that different from the usual normalisation process used to calculate UMS? The same actual number of marks on an exam one year won't get you the same score in UMS (and so not necessarily the same grade) as the next year. That can be the case even if the mark scheme doesn't change. I think these kind of drifting grade boundaries are a large part of the problem, but I can't get too scandalised that a boundary has drifted up for once against the usual trend. Bit of a mistake to do it in the middle of the year I'll grant you - what if it had been done in September instead?.

Yes but year on year the assessments and exams change. We're talking about exactly the same assessments (right down to the letter) being done at different times of the year being marked to different standards. So if Johnny Johnson did his controlled assessment in November and then submitted in January, he got a better mark that he would have if he submitted that same piece of work in June.

You've highlighted a 3 mark change in the grade boundary for the English coursework. Is this a one off problem, or are there other changes you consider sinister? That isn't a large enough alteration to account for the difference in results averaged over all subjects. BBC also notes a decrease for science for example - anything odd happening there?

That (IMO) is the major one. There is also the AQA Foundation Paper 1. The C boundary for that has risen 10 marks (from 43 to 53) from the January paper. As I said, my major beef with these GCSEs is what's happened to English. The overall fall in GCSE results is not especially suspicious.

It's indisputable that there are borderline candidates where those 3 marks did in the end make the difference between a C and a D. However I don't think much of this "they thought they were on target" argument with the implication that if they'd known of the boundary change they'd have put in more time and effort to reach the higher target. You're talking a three mark change in one component of the English GCSE, in which the marking is highly subjective. I'd expect more than a three mark variation simply depending on who is marking it. The idea that a given student could run that close to the boundary and still sensibly be confident of a C is ridiculous. As is the idea that you could know to that degree of precision whether they were "on target" or not. I agree some people did lose out to that grade boundary change. I don't agree that them knowing about it in advance would have made a difference in their actual marks.

As I have mentioned above, the 3 mark difference in CAs is only on element of the grade boundaries rise. There is the above mentioned 10 mark rise for Paper 1 Foundation. There has also been a 3 mark rise in the Unit2 (which is only out of 45) C grade boundary.

The grade boundaries are below if you want more information. The English GCSE in question is made up of three modules: ENG1F, ENG02 and ENL03. The C grade boundaries in those modules have risen 10 marks, 3 marks and 3 marks respectively. That's nearly a whole grade in ENG1F, about half a grade in ENG02 and a minor (but unjustifiable) rise in ENL03. That works out at slightly more than half a grade. That is a very significant difference: if you input the minimum marks for a C on the January papers into the June UMS calculator you end up with 162 which is closer to the D/E boundary (150) than the C/D boundary (180).

January Grade boundaries
June Grade boundaries

Again, I have my doubts that you can predict English results to anywhere near enough precision for this change to make the difference between "confident we'll pass" and failing. Some schools at the very margin may have ended up on the wrong side of the line because of it though. I would also say that GCSEs are much more important to the schools than to the individual students. I'm not going to be too convinced this is the result of politics without rather more evidence of unreasonable changes in grade boundaries.

Schools that have a lot of C/D borderline students have seen many of the pupils they were predicting low Cs ed up getting high Ds. That will come as a major shock, as schools tend to be very good at predicting grades. For example, out of my 21 Y11 students in maths I predicted all but one of my students' grades correctly, and that individual only failed as she didn't turn up to the exam! Those good prediction rely on the exam board not making any major changes to the grade boundary structure mid way through the year. Those predictions are also relied upon when planing intervention and budgeting. Furthermore, OFSTED expect schools to be able to predict their grades to a certain degree of accuracy.

 

[~Corsair#01~]

Most of the rise in grades over the past 20 years has just been a result of lowering standards. Why is this any less fair than what has gone before?

In the past you could say that someone who got a high D would have got a low C if they'd been tested a year or two later. Now it's the reverse and you're acting as if some sacred principle has been violated.

 

[Truronian]

Most of the rise in grades over the past 20 years has just been a result of lowering standards.

Evidence?

Why is this any less fair than what has gone before?

Because it's been done in an underhand way. I'm talking specifically about English, where the grade boundaries were changed dramatically just before the exam. This included changing the grade boundaries for one controlled assessment (coursework) so that exactly the same piece of work submitted as part of the same GCSE would score more in January than in June. If you want more information, reads my responses to Mr. Cynical.

 

[Truronian]

Incidently, ENG02 (speaking and listening) is also assessed internally. So that's another module which in which the same piece of work would score differently in January and June.

 

[~Corsair#01~]

Evidence?

It's a struggle to find anything to base this on other than anecdotal or subjective evidence (just as I'm sure you would struggle to find strong evidence to refute what I said), but I'll have a look and let you know if I find anything.

 

[Truronian]

It's a struggle to find anything to base this on other than anecdotal or subjective evidence (just as I'm sure you would struggle to find strong evidence to refute what I said), but I'll have a look and let you know if I find anything.

Speaking for my own subject here:

I've little doubt that maths papers have got easier in the last few decades but I'm not sure this alone points to reduced standards... after all, you could fill GCSEs with quantum mechanics questions and watch results plummet, but it would be had to argue that would raise standards. I think teaching has got better in the past decades, and the resources available to students have got significantly better. The internet, interactive whiteboards, tablets and smartphones are all wonderful resources for learning that simple weren't around twenty or thirty years ago.

Maybe you're correct about exam inflation being down to slipping standards, maybe you're not. Regardless, what's happened in the English exams this June is not the way to go about solving the problem. It's unfair on students, teachers and schools.

See here for a slightly more eloquent explanation of what's happened

Once again, I'm not annoyed by the overall drop in results, if the system was set up well it wouldn't have taken 20-odd years to see the first decrease. What I'm incensed by is the manipulation that's taken place in the English Language exam.

 

[Truronian]

Well, the media is doing a very good job of letting OFQUAL and the exam boards obfuscate and get away with this. Endless questions about the 1.5% drop from last year despite no-one complaining about that. No-one asking about the massively shifted goalpost between January and June this year which is the real problem. OFQUAL claim that some heads have seen massive improvement in their English pass-rates; no excrement Sherlock! Those schools submitted in January when the grade boundaries were much lower. One HoD on TES reported a 20% improvement after he had pupils sit and submit all their components in January.

Hopefully as press releases and complaints start coming in from schools across the country the media will actually start to correctly publish what the indignation is about. That 1.5% drop only exists because the exam boards massively changed their grade boundaries when thy realized after January they were not going to get the drop they (presumably) were supposed to. They have exploited the fact that the January and June results are part of the same year to mask a rise in January's results with a massive drop in June's.

 

[MrCynical]

I've little doubt that maths papers have got easier in the last few decades but I'm not sure this alone points to reduced standards... after all, you could fill GCSEs with quantum mechanics questions and watch results plummet, but it would be had to argue that would raise standards.

It is possible though to compare syllabuses today with those of twenty years ago and see what a student is expected to know for the exam. It's pretty unambiguous that the content has been reduced. Unless you're arguing that the bits that have been cut were stuff that everyone got wrong anyway (i.e. your quantum mechanics example) then students are coming out with less knowledge and skill than they used to. Certainly in my own subject (chemistry) students less than ten years below me are alarmingly bad at moles calculations and simple organic structures, which were the easy bits when I took the GCSE. They don't appear to have any additional knowledge or skills either, except maybe some rather vague ideas about "green chemistry" so you can't argue it's a broader but shallower education. It's hard to claim that they have been educated to the same standard than I was, never mind someone taught twenty to thirty years ago. There's enough maths in my field for it to be fairly obvious in that subject as well (it was also a bit of a giveaway at A-level when the exam board split the content from three modules into four for the following year without expanding the syllabus or changing the number of modules you're required to take).

I think teaching has got better in the past decades, and the resources available to students have got significantly better. The internet, interactive whiteboards, tablets and smartphones are all wonderful resources for learning that simple weren't around twenty or thirty years ago

Serious question - do you find these things actually help? When I was taking GCSEs interactive whiteboards were the latest thing, but if I'm honest I don't think we learnt anything new because of them. Well, except how to recalibrate the board so all the writing would come out reversed and freak out the less tech savvy teachers. They always gave the impression of being an expensive toy to show off rather than being of practical use.

 

[Truronian]

It is possible though to compare syllabuses today with those of twenty years ago and see what a student is expected to know for the exam. It's pretty unambiguous that the content has been reduced. Unless you're arguing that the bits that have been cut were stuff that everyone got wrong anyway (i.e. your quantum mechanics example) then students are coming out with less knowledge and skill than they used to. Certainly in my own subject (chemistry) students less than ten years below me are alarmingly bad at moles calculations and simple organic structures, which were the easy bits when I took the GCSE. They don't appear to have any additional knowledge or skills either, except maybe some rather vague ideas about "green chemistry" so you can't argue it's a broader but shallower education. It's hard to claim that they have been educated to the same standard than I was, never mind someone taught twenty to thirty years ago. There's enough maths in my field for it to be fairly obvious in that subject as well (it was also a bit of a giveaway at A-level when the exam board split the content from three modules into four for the following year without expanding the syllabus or changing the number of modules you're required to take).

As I said, I'm new to teaching. Most of my colleagues who have been doing it for decades don't think it's got any harder, and that topics have generally been replaced rather than omitted, but they may have just not noticed the creep. My point with the quantum mechanics thing was that some topics may have been removed which were quite hard and quite useless (large amounts of imperial metric conversion, using log tables, circle theorems for some more realistic examples) and been replaced with easier but more important (for the average school leaver) content. I'm not sure if that's the case, but if it were then I don't think that would be slipping standards even though it would lead to easier exams.

Serious question - do you find these things actually help? When I was taking GCSEs interactive whiteboards were the latest thing, but if I'm honest I don't think we learnt anything new because of them. Well, except how to recalibrate the board so all the writing would come out reversed and freak out the less tech savvy teachers. They always gave the impression of being an expensive toy to show off rather than being of practical use.

They help so long as you know how to use them. I had the same experience as you when they first came in; teachers didn't utilise them very well. The current crop of teachers coming through (in maths in particular) are making very good use of them.

 

[Samson]

From the PA

It admitted that the grade boundaries were higher in June than they were in January. Students who gained their English GCSE this summer will be given an extra chance to re-sit, Ofqual said.
...
"The issue is not the June but the January boundaries. Again, examiners used their best judgment in setting these boundaries, but they had less data and information to work with.

Full text:

Spoiler :

January's GCSE English exams were "graded generously" but the June boundaries were properly set and candidates' work properly graded, Ofqual has said.

In its initial report into this summer's English results, the regulator said it had concluded that when exam boards came to set grade boundaries in June, they were "better equipped to make judgments as there was more information available".

It admitted that the grade boundaries were higher in June than they were in January. Students who gained their English GCSE this summer will be given an extra chance to re-sit, Ofqual said.

Ofqual chief regulator Glenys Stacey said: "We are grateful to schools and colleges for bringing their concerns to our attention so quickly. In response, we have looked carefully at how the exam boards have managed the awarding of all GCSE English qualifications this year.

"People were particularly concerned about the June grade boundaries. We have found that examiners acted properly, and set the boundaries using their best professional judgment, taking into account all of the evidence available to them. The June boundaries have been properly set, and candidates' work properly graded.

"The issue is not the June but the January boundaries. Again, examiners used their best judgment in setting these boundaries, but they had less data and information to work with.

"Most candidates were not sitting at the time, they were waiting for June, and because they were new qualifications, examiners could not rely so much on direct comparisons with the past. As a result, those grade boundaries were set generously.

"We have spoken to exam boards and they have been very responsive. Recognising the strength of feeling, they will be offering early resits for students who sat the June units. We will also talk with schools, exam boards and assessment experts to see what lessons can be learnt and what can be done better in the future."

Ofqual began an investigation into this summer's GCSE English results following an outcry from schools. As national GCSE results were published last week, angry headteachers claimed that exam boards had raised grade boundaries in English halfway through the year amid fears that too many children were going to get a C.

The regulator said that it will not be revisiting the June grade boundaries as they are right and it would contradict its responsibility to maintain standards. Pupils will be given a one-off re-sit chance this November.

Do you agree (that the June ones were OK but January's were too easy)? Is the extra chance to resit much help? From my (rather distant) recollection you could resit anyway, and you wanted to neither give up your summer to revision or delay your further eduction.

 

[Truronian]

My English teacher colleagues were of the opinion that the (Unit 1) exams were fairly similar in difficulty (for our exam board, AQA); if anything June's was harder than January's (and so should have had lower grade boundaries) as the June exam called upon the students to write a script, something that there had been little preparation for. The other two units were exactly the same and there has been no explanation for the 3 marks rise in both of their grade boundaries. Both Unit 1 exams seemed in line with exams of recent years, a huge shift in grade boundaries was not expected which is why it came as a major shock. GCSE are supposed to be criterion based, you get a C if you can demonstrate certain skills in your work. This means that in theory everyone in a cohort could get a C if they all possess good English skills. What OFQUAL have been pressuring the exam boards to do is set their grade boundaries on percentage: a certain percentage should be getting this grade; don't let the percentage get too high. When they saw that January's result implied that percentage would rise for the 25th consecutive year it was decided to massively raise the grade boundaries for to manipulate the results so that the whole year results went down slightly.

OFQUAL were a big part of engineering this grade boundaries change, so it's not surprising that they've given it the all clear. I expect one of the other investigations or the potential legal action from heads will result in justice for the students and schools involved, and could potentially be pretty damning for OFQUAL and the boards.

 

[MrCynical]

A hypothetical question Truonian: If the grade boundaries had been increased prior to the Januray exams rather than the June exams, would you have had a problem with this year's exam results?

GCSE are supposed to be criterion based, you get a C if you can demonstrate certain skills in your work. This means that in theory everyone in a cohort could get a C if they all possess good English skills.

Well, again I would dispute how objective those criteria are in subjects like English but I do see your point. The problem is that if content is reduced or grade boundaries are relaxed then that C grade no longer indicates the same skills year on year. A C from a later year then indicates a smaller skillset or lower competance in each skill than one from an earlier year. That is the whole declining standards issue. To restore comparibility you have to either raise the grade boundaries or restore the syllabus content, and that will reduce the number of top grades - there's no way round this.

What OFQUAL have been pressuring the exam boards to do is set their grade boundaries on percentage: a certain percentage should be getting this grade; don't let the percentage get too high.

Well the question here is what you actually want GCSE results to do. Setting the grade boundaries as the X-percentile of all candidates does allow you to tell whether a given person is above or below average which is often what you want to know. The main argument I keep seeing against this is that you could get one year of students being far stronger than a previous one resulting in candidates being undergraded, but I've always felt this argument required a lack of understanding of basic statistics.

Now you can make an argument for an exam which measures a person's ability to do skills A, B and C and passes all who meet a certain bar. With this approach though firstly the bar must be kept at a constant level (in GCSE it's been drifting downwards). Secondly if a qualification is to be valued, by universities/employers/whatever, a significant percentage of people who attempt it must fail to meet that standard. If everyone reaches that threshold the qualification is devalued - I could grab any average joe off the street and they'd all have it. It no longer permits distinction between candidates.

Grade inflation at GCSE has to be stopped if they are to remain of any value. You can argue that this years students have come out with lower grades than they would have without these measures, but they'd also have had lower grades if they'd taken the GCSE even a few years earlier - yes, even with exactly the same skillset they have now. As I hinted in an earlier post: if this year's students are at a disadvantage you could argue that everyone who took them prior to about 2009 was at an equal or larger disadvantage.

 

[Truronian]

A hypothetical question Truonian: If the grade boundaries had been increased prior to the Januray exams rather than the June exams, would you have had a problem with this year's exam results?

I would have a much smaller problem with that, because I think it's a naff way of addressing exam inflation. The major problem is competing exam boards resulting in a race to the bottom, IMO. This would just flick a temporary reset, penalising this years cohort but failing to address the cause of grade inflation.

Well, again I would dispute how objective those criteria are in subjects like English but I do see your point. The problem is that if content is reduced or grade boundaries are relaxed then that C grade no longer indicates the same skills year on year. A C from a later year then indicates a smaller skillset or lower competance in each skill than one from an earlier year. That is the whole declining standards issue. To restore comparibility you have to either raise the grade boundaries or restore the syllabus content, and that will reduce the number of top grades - there's no way round this.

Something does indeed need to be done about it; but the change needs to be communicated to schools and teachers.

Well the question here is what you actually want GCSE results to do. Setting the grade boundaries as the X-percentile of all candidates does allow you to tell whether a given person is above or below average which is often what you want to know. The main argument I keep seeing against this is that you could get one year of students being far stronger than a previous one resulting in candidates being undergraded, but I've always felt this argument required a lack of understanding of basic statistics.

Now you can make an argument for an exam which measures a person's ability to do skills A, B and C and passes all who meet a certain bar. With this approach though firstly the bar must be kept at a constant level (in GCSE it's been drifting downwards). Secondly if a qualification is to be valued, by universities/employers/whatever, a significant percentage of people who attempt it must fail to meet that standard. If everyone reaches that threshold the qualification is devalued - I could grab any average joe off the street and they'd all have it. It no longer permits distinction between candidates.

I don't particularly like the idea of a set percentage getting a given grade, but if we ended up with such a system it wouldn't be the end of the world. However, the current system is specifically not meant to be this; I spent all of last year telling kids that they could get a C if they learnt such and such a skill and had certain knowledge memorised. It's not fair to then give them a D despite learning all the C grade material simply because too many people were coming out as having learnt all the C grade stuff, the equivalent of which has happened in English. This is the fault of the exam boards, but the kids (and possibly the schools) are the one who are paying the price.

Grade inflation at GCSE has to be stopped if they are to remain of any value. You can argue that this years students have come out with lower grades than they would have without these measures, but they'd also have had lower grades if they'd taken the GCSE even a few years earlier - yes, even with exactly the same skillset they have now. As I hinted in an earlier post: if this year's students are at a disadvantage you could argue that everyone who took them prior to about 2009 was at an equal or larger disadvantage.

I'm of the opinion that it would probably be best to move to a new qualification, set by a single exam board. If they want to make this percentage pass based like the O-level, then so be it (though I don't like this idea).

 

[dutchfire]

You have competing exam boards? Who invented that? I would think this would be such a thing where having a single standard would be good.

 

[Truronian]

You have competing exam boards? Who invented that? I would think this would be such a thing where having a single standard would be good.

Indeed, introducing a single exam board is one of the very few plans Michael Gove has up his sleeve that I approve of.