GCSE Music – Statistics and Grade Boundaries

N.B. This is a post that may get deleted or have amendments later, if it proves that I am being daft, and haven’t read all the relevant small print! Which, as it’s about the new GCSE grading, means there is a lot of small print to read. It’s arisen from a conversation I had with Dr Alison Daubney on twitter, so thanks  for setting this troubling train of thought off! 


On the 15th July 2016, the DfE published revised Grade descriptors for GCSEs graded 9 to 1. (Find them at ‪bit.ly/29EKR0W)  Well, they didn’t, actually, for music. They published them for Grades 8, 5, and 2. (Remember that these grades 9-1 work the same way as music practical exam grades, ie start at 1.) Here they are:

Grade 8

To achieve Grade 8 candidates will be able to:

  • perform challenging music with a high degree of fluency and sensitivity
  • compose using a wide range of musical elements with sophistication, creating effective musical ideas and sustaining interest through their development
  • demonstrate, through aural identification, accurate knowledge of a wide range of musical elements, contexts and language
  • evaluate music to make convincing judgements using musical terminology accurately and effectively

Grade 5

To achieve Grade 5 candidates will be able to:

  • perform music with some technical challenges broadly fluently with some sensitivity
  • compose using a range of musical elements with coherence, creating musical ideas and developing interest with some success
  • demonstrate, through aural identification, mostly accurate knowledge of a range of musical elements, contexts and language
  • evaluate music to make clear judgements using musical terminology appropriately

Grade 2

To achieve Grade 2 candidates will be able to:

  • perform simple pieces with limited fluency and sensitivity
  • compose using a range of musical elements, creating musical ideas with some appeal and limited development
  • demonstrate, through aural identification, some knowledge of musical elements, contexts and language
  • evaluate music to produce simple reflections with inconsistent use of musical terminology


Now, this is what many teachers have been asking me for in emails and in discussion forums for quite a while. But, hold on, I think there’s a problem. Well, lots of problems actually! The first issue is that there seems to be some confusion between criterion-referencing – which is what these grade statements are (well, sort of, but stay with me), and norm-referencing, which is where, in essence, a curve of normal frequency distribution (NFD) is used to allocate the grades within set percentages. If your assessment literacy is a bit ropey on this, detailed explanations are found elsewhere on the net. Here’s a graph of NFD from the NFER (available here)


(Source: http://www.nfer.ac.uk/research/centre-for-assessment/standardised-scores-and-percentile-ranks.cfm)

This means, in this example, that only 2% of entrants can ever get the top grade. If there is an amazingly bright group one year, and an astonishingly dim one the next year, these % figures will remain constant (again, that’s a huge over-simplification, sorry). If, therefore, you have an exam system which is norm-referenced, like the new GCSE will be, then you cannot superimpose criterion-referencing statements on it at the same time; in a similar fashion my petrol car cannot run on diesel. The DfE themselves state here that:

“The descriptors are not designed to be used for awarding purposes, unlike the ‘grade descriptions’ that apply to current GCSEs graded A* to G.”

So, again, what are they for? The DfE again:

“We have developed grade descriptors for the reformed GCSEs graded 9 to 1. These aim to assist teachers by providing an indication of the likely level of performance at grades 2, 5 and 8.

The purpose of these grade descriptors is to give an idea of average performance at the mid-points of grades 2, 5 and 8.”

However, I suspect they will be used in KS3, for assessment, monitoring, and tracking purposes. But I’ll return to that theme later.

What will this look like in practice? We don’t have the breakdown for music, but we do for some subjects, so here are drama and PE:

drama and PE 0716

(Source: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/377771/2014-09-12-board-paper-for-new-gcses-in.pdf)

Now, I’m a tad busy at the moment, but if I get an odd five minutes I’ll try and do one for music. But, TBH, it won’t be much help for anyone in an individual school! Why? Because however big your GCSE cohort is, music is such a small subject it still won’t approximate to the national figures. And if you want to feel really depressed, just remember that the kids at Eton, Chets, and Menuhin are also in there, so they will be (I guess, but I’ll probably get into trouble for saying so) up in the top x%, knocking the kids from bogstandard comp out of the way in the process. (You can’t do footnotes, I don’t think, in this blog app, so – if I’m wrong on this, please shout, and I’m not having a go at specialist music schools, just trying to show how the stats might pan out.)

I’ve written about different assessment types before here. I’m wondering how it would go down with SLT if new grading requirements involve a modified form of comparative assessment, modelled onto a graph of NFD for the pupils in each school. If we can work out the statistical modifiers, after a few years we ought to be able to map the general progress 8 stats onto this for each school, and come up with a reasonable attempt at a progress map. But what this might tell us is a tale for another blog!

Which takes me back to the point that I’m not sure what the point of the published grade descriptors is! But, if I were a hard pressed head of music whose SLT have told them to produce grade descriptors for KS3, I’d be on these like a shot!

But here’s where my next worry lies. Let’s take the Grade 2 words, and do something you never, ever should. Let’s compare it to the National Curriculum. For GCSE (ie by age 16) Grade 2 pupils have to:

  • perform simple pieces with limited fluency and sensitivity

At Key Stage 2 (ie by the age of 11) children are required to be taught to:

  • play and perform in solo and ensemble contexts, using their voices and playing musical instruments with increasing accuracy, fluency, control and expression

Some overlap there, methinks? And what about GCSE Grade 2:

  • compose using a range of musical elements, creating musical ideas with some appeal and limited development

KS2 again:

  • improvise and compose music for a range of purposes using the inter-related dimensions of music

And so on!

Which means that with some cunning wording, it could mean that we see schools producing NC levels redux, but dressed up as new GCSE-related criteria statements. Another worry now emerges: will we see invented sub-divisions of the GCSE grades being used in KS3 to “chart” (in deliberate inverted commas) progress? If so, what do these mean?


All of this blog is very, very speculative, and, as I said, I may be totally wrong, and if so, quite happy to have this pointed out. But this entry has been me thinking out loud, as it were, about the effects of what I have been reading and thinking about recently. As things crystallise I’ll revisit, and issue corrections and clarifications as needed.

Thanks due too to @MissDCox for her excellent blog at http://bit.ly/2a0VI7r which also got me thinking!





About drfautley

Professor Education at Birmingham City University, UK.
This entry was posted in Assessment, GCSE, Grading, Listening, Norm-referencing, Uncategorized and tagged , , , , . Bookmark the permalink.

6 Responses to GCSE Music – Statistics and Grade Boundaries

  1. Ally Daubney says:

    Not speculative at all…I could immediately Point you in the direction if schools using 1-9 through secondary school.

    If I get any find this week I’ll try and do a blog reply. For now…sleeeeeep !


  2. terryloane says:

    I agree, Martin, with your statement that there are “lots of problems actually” with these so-called grade descriptors. The main problem is that, as I will explain, they are inherently meaningless.

    Before I even read the descriptors I said to myself “I bet the only differences between the different grades will be adjectives and adverbs”. And when I did read them I found that I was absolutely right, give or take the occasional difference in a descriptive noun or adjectival phrase. For example the descriptors for grades 8 and 5 seek to differentiate between:

    “evaluate music to make CONVINCING judgements using musical terminology ACCURATELY and EFFECTIVELY”
    “evaluate music to make CLEAR judgements using musical terminology” APPROPRIATELY.

    While the descriptors for grades 5 and 2 seek to define a distinction between:

    “perform music with SOME TECHNICAL challenges BROADLY fluently with SOME sensitivity 2
    “perform SIMPLE pieces with LIMITED fluency and sensitivity”

    [Btw, sorry about the capitals but I was unsure if your blog comment system supports tags for underlining or italics.]

    Now the adjectives and adverbs I have highlighted can have no objective or fixed meaning, no instrumental definition. Their meaning will always be contestable and will depend entirely on the context in which they are used and the expectations of the user. A simple thought experiment should show the stupidity of imagining that adjectives, adverbs, descriptive nouns and adjectival phrases can provide any sort of basis for objective statements about the quality of music-making. Imagine that a well-known concert pianist is giving a recital of Mozart piano sonatas, but is having a bad evening, at least in the opinion of one music critic. The critic writes:

    “Mozart’s sonata in C K545 is a relatively simple piece, yet X managed to play it this evening with a distinct lack of fluency and sensitivity.”

    There would be nothing unusual about this use of language within the context of a concert review, but note how the critic has used virtually the same words as the grade 2 descriptor quoted above. Yet we all know that if this professional concert pianist were to perform the same sonata for a GCSE music exam, they would (even on a bad day) achieve the highest possible grade. Within the context and expectation of the exam they would be deemed to have “perform[ed] challenging music with a high degree of fluency and sensitivity” So our understanding of what it means to “perform simple pieces with limited fluency and sensitivity” depends utterly on whether the performance takes place in the Queen Elizabeth Hall or a secondary school classroom! The words convey no objective meaning in themselves.

    Surely we educators need to have the courage to reject outright such meaningless use of and to support our colleagues in schools in doing so. If music teachers did not feel “hard pressed” to worry about meaningless labels by SLTs who “have told them to produce grade descriptors for KS3” then maybe they would have more energy to devote to the real job – using their professional skill and judgement to provide the most appropriate and enjoyable musical opportunities for their students.

  3. Jane Werry says:

    Thank you Martin for a real eye-opener here. The info about norm referencing is a massive weapon in the war against neo-levels! When you say
    Another worry now emerges: will we see invented sub-divisions of the GCSE grades being used in KS3 to “chart” (in deliberate inverted commas) progress? If so, what do these mean?
    I fear the answer is yes we will, and what they will mean is absolutely nothing. The whole thing will be a massive case of emperor’s new clothes.

    • drfautley says:

      Thanks Jane. What I meant was the thing I have railed against before, where “charting progress” is no such thing, it’s simply proving that the straight line the school’s software has drawn for each pupil is true. So, as you say, emperor’s new clothes.

      Sadly not long after I posted this blog entry, I was contacted by a teacher who told me that they had to write 9X3 new subdivided attainment statements for use from Y7 to do this very thing. I think it’s NC subdivided levels redux! Happy hols!

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 )

Google photo

You are commenting using your Google 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