New GSCEs too difficult?

[Trishtrash]

"In GCSE English it's all exams – there is no coursework – and pupils are not allowed to bring in any of the texts. They effectively have to memorise three texts and 18 poems. The expectation is killing them.'

The above is a quote from today's Daily Mail - sorry!

Am I being unreasonable to think that that is not an unreasonable thing to require of an A-Level Student? I did my A-Levels over 30 years ago in a bog-standard comprehensive and we couldn't do any coursework ahead of the game and we certainly couldn't take any of the texts into the exam (that would have made it so much easier!!).

I remember having to memorise vast swathes of poetry (Keats, Wordsworth, Somerset Maugham etc...) and chunks of text (Doris Lessing, Return of the Native, A Winter's Tale are ones that I vaguely remember) in the expectation that we would need to quote from the poetry/texts to support a variety of themes/ideas that we might be asked questions on.

I have no idea about the rest of the curriculum as I did Art, English and History. I definitely had to memorise tons for the History element (I did modern History so stuff about Russian Revolution, WW1 & 2 and the EEC). I know that kids are under enormous pressure now and I got an A for my English Literature but there was no A* around then from what I remember (it WAS a long time ago!)

Is the problem that the teachers haven't been adequately prepared or supported to teach for this style of exam? If the kids are going in after two years of expecting another style of exam then I really feel for them but is this the case?

If only the top 20-odd percent took O-levels and the top 90% got an A (or whatever), then all that was needed to make it easier to get a grade A was for a large number of less able students to be entered for O-level. More students entered = larger number of As available, but because the extra students weren’t equally distributed in ability, the standard needed to get an A would be dragged downwards.

This actually happened in the first few years of O-levels.

Is this the marking system from now on for the new 1-9s? Ie a curve instead of fixed boundaries? Or is it a mixture of both?

In order to prevent grade inflation, since 2012 the proportion of kids getting each grade has been roughly fixed at the proportion who got that grade the previous year. This continues with the 9-1 - the proportion who passed last year will be about the same this year, the proportion who got an A or higher last year will be getting a 7 or higher this year.

I don't actually believe in "grade inflation", although more people get higher grades, I'd actually put it down to the quality of teaching and the resources that kids get now.

Much better than copying out of a 20 year old text book in a "temporary classroom" or being dictated to.

Teachers are under far more pressure for kids to succeed too, where as back when I got educated if there were bad results one year it was just a "bad year".

But it is grade inflation. Because in the past it didn't matter if everyone did better, you competed against your peers. Only a certain percentage of As etc were awarded.
And I think you have a lack of understanding of what was demanded from students 40 years ago.

You can repeat that you competed against your peers till you are blue in the face, its a facile point because no one could differentiate between an O level in one year and the next because the grades awarded were the same, but the standard needed to achieve each grade could be ( and was) significantly different.

In getting employment, university places etc, this meant standards could vary wildly and very able people missed out on opportunities because of the unfairness of this.

It wasn't a great way of doing things

Although back in the bad old days there wasn’t such a mania about As (obviously no A*s then). As previously noted, hardly anyone got all A grades. It was very rare .

Having “only” a few As certainly didn’t prevent me or my peers from going to top universities, even Oxbridge.

crunchymint

And I think you have a lack of understanding of what was demanded from students 40 years ago.

How are you supposed to know what was demanded of students 40 years ago when there is no data to support any given theory?

Grading exam results using the Bellshaped curve meant you couldnt inflate grades as the same percentage of pupils got As,Bs and Cs every year regardless of whether it was an easy or hard paper or what the highest mark was. The point of this type of grading was you were directly compared to your year group cohort.

AndromedaPerseus

The problem with this type of grading, though, is that it does not compare students to an objective standard. If you get one student who is something of a child prodigy, he or she will set the curve for everyone and that will skew the results.

I am a college instructor and I NEVER grade on a curve. EVER. I feel it's unfair to the students. Now, if you look at the results, I have high enough numbers of students that the results will likely approximate a normal curve, but I didn't specifically engineer that outcome. It creates too many opportunities for successful grade appeals and it's unfair. Moreover, when you're comparing results across several schools, it makes it more difficult to objectively compare the schools.

As I've said, grading on a curve made it tremendously hard for universities and other institutions to measure the strength of applications from students as an A grade in one year was not the same standard as another.

Setting a standard set of marks say 80% for an A is much fairer as you can always see how well a child has done. Across a set of papers any given years exams will be the same level of difficulty. The idea that it causes "grade inflation" meaning the exams are easier is spurious, its actually a far fairer way of marking, and a much more accurate reflection of the ability of the child.

Setting a standard set of marks say 80% for an A is much fairer as you can always see how well a child has done. Across a set of papers any given years exams will be the same level of difficulty.

No they won't. They might try to make the questions the same level but they will never achieve this. I think there is much less likely to be variation in the academic ability of the students taking the exam and the teachers.

The exams will be about the same level of difficulty across a range of papers, having both examined and taught at a high level I can say that in my two subjects the standard did not vary greatly.

However there were some years where we did extremely well in A*-B grades. Had they been graded on a bell curve students who got an A grade in one year would have been a B/C in another, which is hardly fair.

I'm an English teacher. It simply isn't meaningful to be memorising quotes in the hope that you can apply them to whatever question comes up. I really struggled with this when it came to memorising Yeats in my second year at uni - my worst exam result ever, I just couldn't do it.

The exams will be about the same level of difficulty across a range of papers, having both examined and taught at a high level I can say that in my two subjects the standard did not vary greatly.

I have too at (university level and beyond) as does DH and they are never exactly the same. It's impossible. And while it is true that if you have a fairly small number of students ability will vary across a cohort from year to year, if it is a population of 1000s there won't be much variation. IQ doesn't change massively from one year to the next and neither do teaching standards.

They are never exactly the same no, but the standard is comparable.

The idea of being on a bell curve is that you spend all of your life competitng with those in your year group, it doesn't work like that, and yes there used to be quite large variations in what was considered an A when using the bell curve grading.

They are never exactly the same no, but the standard is comparable.

I disagree. There is a fair amount of variation from one year to the next. You are assuming that the marks are much higher one year rather than the next it will be because the students are much better but if there are 1000s of students, this is unlikely.

The idea of being on a bell curve is that you spend all of your life competitng with those in your year group, it doesn't work like that, and yes there used to be quite large variations in what was considered an A when using the bell curve grading.

The large variations were very probably due the questions differing in their difficulty though! You wouldn't expect children to be much taller one year compared with the previous year or anything else so why assume that academic ability will suddenly change across a whole population. If the average score is very different it is probably because the questions are very different .

I disagree about the questions being in varying difficulty across a set of papers. The difficulty of a set of exams from year to year will not be extremely different, and will adhere to a standard.

There were large variations in different years between what was an A and what wasn't. Setting a standard % and then ensuring the standards of the exam are equal is a far fairer way of operating this system.

I disagree about the questions being in varying difficulty across a set of papers. The difficulty of a set of exams from year to year will not be extremely different, and will adhere to a standard.

That's impossible though. Because the questions have to different there will inevitably be some variation. You only have to look through past papers in any subject to see that.

There were large variations in different years between what was an A and what wasn't. Setting a standard % and then ensuring the standards of the exam are equal is a far fairer way of operating this system.

Yes, but as I keep saying the large variations were more likely to be due to differences in the difficulty of the questions than sudden changes in a large population. You can't ensure that the questions are equal. You can assume that in a large population, there won't be huge variation in IQ, teaching ability, height etc etc from one year to the next.

Across a set of papers in one year the standard will be broadly the same. One paper might be "nicer" than another, but when taken as a set of exams the level of questioning will be at the same standard.

The variation can be to do with a whole host of reasons, not just IQ.

Like I said setting a standard for where the grade boundary is is fairer than saying only a certain number can get a certain grade.

Across a set of papers in one year the standard will be broadly the same. One paper might be "nicer" than another, but when taken as a set of exams the level of questioning will be at the same standard.

Whether or not they are "broadly" the same is very subjective. Certainly examiners will try to make them similar when setting the questions but the very fact that marks are higher one year versus the next suggests they are not.

The variation can be to do with a whole host of reasons, not just IQ.

So what things might change greatly across the whole country from one year to the next that may cause marks to be different. I can't think of anything that would change in one year that could cause a big shift in marks apart from the difficulty of questions. Average IQ won't change, teaching ability across the whole country won't suddenly change. What would be the cause of variation over one year if not difficulty of questions?

"Average IQ won't change"

But Averages aren't an accurate measurement of the level of ability. The average might stay the similar but the level of ability of the cohort can change, influencing the grades.

In the end limiting the number of A grades isn't really fair, set a standard and then allow students to aim for it.

Piddling about "grade inflation" is just a lot of rubbish

But Averages aren't an accurate measurement of the level of ability. The average might stay the similar but the level of ability of the cohort can change, influencing the grades.

Perhaps I shouldn't have just said "average". The median, standard deviation, distribution of abilities etc etc also won't change much from year to year at a population level. Obviously, it may be different among a fairly small number of children e.g. a class or even several classes but if you are comparing 1000s across the country there won't be big changes in a year in IQ, ability, teaching or anything I can think of apart from a change in difficult of questions. Can you think of anything else that could cause marks to be much different one year compared with the previous one?

The weather, the arrangement of the exams on the day, a new syllabus/changes to the syllabus, and yes varying ability, it doesn't take many higher results to pull the curve upward.

Remembering that during O levels, only 20% of the school age population took them, and far fewer students took A levels, we can talk about a tens, or maybe a few hundred results influencing the curve.

What is the problem of setting a standard like 80%, is it that too many people get an A? If its deserved it should be rewarded rather than being arbitrarily awarded on % of students achieving high marks

I have a good working memory and would have been fine with memorising texts... but as an employer, in terms of transferable skills I’d much rather see analytical and critical thinking developed. It’s absolute nonsense to put so much emphasis on memorising. I’m just hoping things will have swung back again by the time my infant son is sitting exams.