One step beyond – assessing what we value

/, English/One step beyond – assessing what we value

Hey you, don’t teach that. Teach this!

Do we always teach what we value? it seems to me that when push comes to shove, we end up teaching what is assessed. The urgency of accountability results, inexorably, in teaching to the test. And this, sadly, ends up with teachers teaching stuff that they don’t particularly value. I’m not in any way a mathematician, but one of the problems with maths at GCSE is that the knowledge students are taught is atomised: they are rarely shown the links and connections between, say, vectors and averages. Why not? Because the examination doesn’t require them to know this. But would it help their conceptual understanding of maths? I think so. I’m on less shaky ground when discussing English – the ‘skills’ and ‘knowledge’ need to pass the AQA English Language GCSE are fatuous. A pupil’s ability to think about language is not valued; the strictures of the mark scheme insist that questions must be answered formulaically with little room for individuality, and none for passion.

But what if we changed what we assessed? Now that National Curriculum levels will no longer be statutory for September 2014 we have a golden opportunity to design assessment systems that will, in turn, allow us to come up with a curriculum  bursting with that which we value. Life after levels could be a thrilling prospect.

I love these images from Tom Sherrington’s presentation at the SSAT National Conference in 2012:

Screen Shot 2014-04-05 at 12.14.44

This so clearly illustrates the effect assessment has on outcomes. As long as there’s a ceiling, children will be cribbed cabinned and constrained into merely meeting our expectations. But what if we lifted the lid?

I’m also really excited by opportunities offered by Harpaz’s taxonomy of Performances of Understanding

Screen Shot 2014-04-02 at 01.18.42

Most assessment rubrics stop mid way down the second column and never enter into the territory of criticising or creating knowledge until way after GCSE. So what if we were to design an assessment system that deliberately waded in to this murky, unmapped territory? In Making Learning Whole: How Seven Principles of Teaching Can Transform Education, David Perkins talks about ‘playing the whole game’. Learning should not just be about learning content,  but getting better at things.  It should be about thinking with what you know to go further; finding explanations and justifications, and involve curiosity, discovery and creativity. Whoa there! Wind back a bit – did I really just write that? Learning should involve curiosity, discovery and creativity? Doesn’t this besmirch and contradict everything I’ve been saying for the past year or so? Well, happily, as long as we acknowledge the very real dichotomy between teaching knowledge and questioning knowledge, I’m more than willing to embrace the contradiction.

Perkins suggests that we should consider what ‘playing the whole game’ in our subject might look like – what do academics, or experts in our domain do? And if we work backwards from this, maybe we will be able to see how ‘junior versions’ or ‘threshold activities’ might ‘make the game worth playing’. In my subject, experts are either professional writers or academic literary critics – if we were interested in ‘playing the whole game’ of English then it would be worth thinking about what it is these people do and to develop a curriculum and an assessment system that allowed pupils to play a ‘junior version’ of these games. Of course all this must be sensibly filtered through what we know about the science of learning and we must never forget that experts and novices have quite distinct brain architecture – they  think in very different ways. (See this post for a description.) We must therefore be mindful not to overload working memory when playing these ‘junior versions’ of our subjects’ games.

So, with it that buzzing about in my brain, I made the long journey through the mist shrouded cobbled streets of Durham to work with the quite wonderful English department at Belmont Community School to help them design a new Key Stage 3 curriculum and a new post-levels assessment system. It goes without saying that I’m heavily indebted to the work of Alex QuigleyJoe Kirby and Phil Bagstock –  their thinking has certainly made it easier for those of us following in their footsteps.

That said, here’s what we came up with:

I wouldn’t want to claim that this model is perfect; it’s a product of its context and was very much led by what Belmont’s English department wanted, and only partly shaped by what I thought it should or could be. One thing that they were very clear about was that reading and writing should not be assessed separately, so the ‘organising concepts’ that we decided upon have not been explicitly divided, although it’s pretty clear that some relate more closely to either reading or writing skills. But what we were all particularly proud of was the fact that assessment allowed, even encouraged, pupils to go beyond where levels left off and will, we hope, allow the lid to be lifted on the assessment tasks they will go on to develop. Although this model is still focussed on generic skills, it is designed to be a template that can be adapted when drawing up rubrics for specific assessment tasks.

The brief for the curriculum was that it would allow the extended study of culturally rich texts and would attempt place English literature within a sequential context. We decided to study only 3 broad topics, or schemes of learning per year, and to have two assessment points per scheme of learning. This still looks a little sketchy in certain areas but this document gives you the gist of it:

Screen Shot 2014-04-05 at 16.15.43

We then went into more detail to plan out what some of the schemes of working would look like using two excellent planning tools taken from Creating Outstanding Classrooms: A whole-school approach. Here some planning examples from the Year 7 ‘Story of English’ scheme:

These schemes are based on rich enquiry questions, packed full of interesting and challenging content, and assessed against pupils’ mastery of the organising concepts of English. They are designed to ensure that pupils are getting better at the subjects and, as Harry Webb says, “there is nothing more motivating than becoming better at something.”

The pragmatist in me suggests that the overwhelming majority of schools will stick with levels out of fear and inertia. We’ll carry on with the same old same old like pit ponies released from the mines only to continue wandering aimlessly in the dark.  I’d like to think this post offers two points:

1) Although this process isn’t easy, it is certainly possible. It took two days with the whole department off timetable and me there to guide, shape, question and suggest. This was a considerable investment by the school, but everyone involved left feeling excited about the future and satisfied with the process. If you’d like to talk to me about doing something similar, please get in touch.

2) Daisy Christodoulou has written persuasively about why we should rid ourselves of the shackles of National Curriculum Levels, but more importantly this opportunity presents us with the ability to build something that might make a real difference to our pupils. This is a precious opportunity: don’t squander it.

Thanks for the title, and credit for getting this whole process rolling, must go to the inimitable Dan Brinton, Deputy Head at Belmont.

UPDATE: I’ve begun the process of trying to assess the efficacy of this assessment system here. Please feel free to add comments.

Related posts

Principled curriculum design: the English curriculum
Is there a way to avoid teaching rubbish in English?
Redesigning a curriculum



  1. Ian Lynch April 5, 2014 at 1:27 pm - Reply

    Shows that the people with real power in the curriculum are the Awarding Organisations.

  2. […] Read more on The Learning Spy… […]

  3. discreetteacher April 5, 2014 at 2:24 pm - Reply

    I have been thinking about this too personally for a while. I abuse the hell out of the ‘testing effect’ and the ideas about both ‘interleaving’ and ‘desirable difficulties’ so I know that the base of knowledge that my students have is increasing at a rate that should be sustainable. However, I know that there is a pressing need to get them to build upon the knowledge that I have given them, so once they have amassed knowledge should we then be giving them ‘open tasks’? Should I be spending some time in lessons explaining the links between different areas of mathematics (I always use a known area to move towards and unknown, I don’t want to overload the working memory do I?) but should I make all links explicit? Basically, I am really happy using a ‘mangled’ version of Direct Instruction (as I can’t afford the real thing) but I don’t know how to take it further without open tasks.

    • David Didau April 5, 2014 at 4:05 pm - Reply

      A solution might work using Robinson’s Trivium model:

      1. The ‘grammar’ of the subject taught be direct transmission
      2. This knowledge is then critiqued through a dialectic process
      3. New understanding is expressed as beautiful work through the process of rhetoric.

      Does that work?

      • discreetteacher April 5, 2014 at 5:24 pm - Reply

        So let me get this straight there should be a running theme whose pinnacle is ‘rhetoric’ as in:

        1 – give knowledge, transmit
        2 – group discussion about what the knowledge means, how it fits into the worldview, how it helps to solve problems.
        3 – find problem requiring grasp of knowledge, ideally a problem that allows the student to show off their new understanding as opposed to be a question that uses the new knowledge in a different way (as we would be explaining that because it is always more efficient to explain than discover)?

  4. mikercameron April 5, 2014 at 3:34 pm - Reply

    This is interesting for a number of reasons.

    Firstly, the assessment grid looks a lot like Level Descriptors to me (with, as David suggests, unnecessary adjectives and adverbs removed). I know I’m a bit of an old stick in the mud but I can’t see how any assessment system can be constructed without descriptors of some kind. I do fear that in trying to avoid them some will create unworkable systems, so its good to see them here.

    Secondly, for some reason I am instinctively drawn to multidimensional taxonomies. I suspect it is because they are harder to then use to pin a single number on a child. There has to be some descriptive or visual indicator of performance. Herein lies a difficulty. I assume that the intention isn’t to go to all this trouble to then condense the assessment information down to a number. So a question is, how will the assessment be fed back to the child and their parents and how will it be used strategically within the department/school (if at all).

    Thirdly it does look as if it could be replicable across subjects (and even phases). Some would say that isn’t important. My view is that it will be bad enough having different assessment systems in different schools. It would be a higher form of madness to run several different ones in the same school. There is little in this system to offend anyone (I see that as a positive) but it can be used at different levels (pardon me) by different people.

    Finally, a weakness. If the starting point is Harpaz, then I’m not sure the model properly enables the use of both dimensions of the taxonomy. Yes, the “Beyond” ‘level’ brings in some of this, but it would be helpful to see an explicit breaking out of the ‘presenting’, operating on’ and ‘criticise and create’ dimension. One way to do this would be to perhaps assign each of the vertical strands to one or more of those dimensions. This would be a bit of a fudge but would IMO extend the model. Alternate ways could be found but would be technology dependent to both record and display due to 2D limitations of paper. Moving forward this may be the answer – anything that gets away from using a single number to define a child would be better.

    • David Didau April 5, 2014 at 4:01 pm - Reply

      Whoah! Are you proposing a 3D assessment model! I would LOVE to see what that looked like!

      In terms of feeding back to children & parents I suspect the dept I worked with to create this will revert to communicating via some sort of alpha-numeric device but I don’t think it has to be that way – couldn’t we just use the descriptors?

      Also, the Harpaz taxonomy was implicit throughout our construction of the model – it would also have to be tried to the way individual assessment tasks were designed. One example on the Story of English planning gives pupils the option to pastiche – this would need to be developed to allow them to ‘go beyond’.

      Thanks for the comments – very constructive

      • mikercameron April 5, 2014 at 4:57 pm - Reply

        I think I would play around with something like this –

        The imagery is one of growth rather than just linear progression. Also the expansion can continue outwards so there is no limitation on the students possible progression, which was where you started. You might also start to see some common shapes appearing linked to different types of students.

        Consideration could also be given to coloured shading of segments to bring in the horizontal Harpaz dimension.

        As I say, requires technology, but I think this is the direction we are going to be moving. Personally I like the visual depiction, as long as the underlying descriptive info is available as well (technology again).

    • Ian Lynch April 5, 2014 at 4:17 pm - Reply

      Mike is correct in that you do need some sort of criteria. At least to specify the basic competences that should be learned. Use the minimum complexity to do the job. A simple system based on internationally agreed standards but with flexibility to locally refine it to specific needs. Back it with free and open source management tools specifically designed for the job working from classroom practice. That was my starting point and after 80,000+ registered learners I don’t have much evidence against continuing in this way. Don’t reinvent yet another unconnected system with complex layers of administrative bureaucracy. Simple is often better. Teachers can too easily get dragged into a whole raft of admin. bureaucracy that distracts from teaching and adds very little real value to raising standards.

  5. heatherfblog April 5, 2014 at 5:15 pm - Reply

    I do like your curriculum! You wanted feedback from English teachers but I hope as a history teacher my points are relevant. I think that this approach suffers from the same problem as current levels. Whenever the apparent difficulty is judged generically, separate from the content you run into problems. This is because it is how specific subject matter is used that creates difficulty, the generic hierarchies can’t take account of this.
    In the end you need specific mark schemes for your task that take account of content and that you can refine in the light of student responses. You can then use the same task each year and thus get some form of bench marking. The kid’s grasp of the content on a module will dictate the quality of their analysis. Are they acquiring generic analytical skills as they go? Possibly but hard to measure independent of content. That content then can change the difficulty of the task – so of course, reaching a certain level in one task won’t mean you can in another.

    • David Didau April 5, 2014 at 8:59 pm - Reply

      I think you’re right – a specific mark scheme for a specific task is important. The generic assessment we designed could (should?) be the starting point for designing content specific rubrics. It could also be a snapshot description of where pupils are at given points in the year.

  6. […] If you haven’t already read the original post describing the process of design the assessment system and curriculum, please have a look at One step beyond: assessing what we value. […]

  7. thom April 6, 2014 at 1:44 pm - Reply


    I would first of all like to thank you and your colleagues for sharing the material and thoughts behind your work in this post. While in a very different context (international school, primary) we are currently wrestling with the same issues as described regarding ensuring assessment supports what we value in learning, rather than only what will be assessed in an examination. Once you make that leap, assessment becomes more challenging and more interesting.

    I was lucky enough to attend a workshop with David Perkins in 2011 and also spend some time working with him in another workshop at the same conference. His ideas about `playing the whole game` link very closely to Ron Berger`s work, both in the classroom and through the Expeditionary Learning project. There are also powerful examples of these themes in the High Tech High School and the work of Larry Rosenstock, who I think you would find very interesting. And if you start from the premise of, `What do we want students to understand` you do end up looking at curriculum very differently – UbD in some form I guess. I think you are also right to note Martin Robinson`s Trivium model, which is something I am trying to slip into our curriculum development in the future as a tool for framing discussion.

    I am not sure if you are aware but HGSE do an excellent course online on Teaching for Understanding which focuses a great deal on performances for understanding. We encouraged some of our staff to take the course (I am a member of the dreaded SLT!) and now we are going to spread the word about this to all classrooms. In a local culture where grades are legally required and are totally based on a test result this is going to be a long slog, but worth it.

    I enjoy reading your blog and have held of commenting much because I suffer from the pangs of the turmoil that comes with embracing the dichotomy! Keep it up – the cognitive dissonance is good for me.

  8. John Maguire (@StPeterComputes) April 7, 2014 at 11:57 am - Reply

    From a computing curriculum angle much the same is happening at the moment. See where the areas of computing are shown as areas of progression that model increasing levels of complexity.

    What though is the next thing that will happen. Teachers and managers within a school will summarize the tables of progression into stages (or levels) and number them. Now what is needed after that – Oh to standardize levels of difficulty so that we are talking about the same thing in each subject so that we can communicate it.

    I am reminded of the debate about kg and lbs. Being in my 40’s I know that I came from a progressive school that taught kgs at all levels and very little about the imperial alternative. Yet even now it is expected that students must use both formats.

    The first instinct of staff and students and parents will be to say ‘Is this the same as a Level 5?’, ‘What GCSE grade would this be?’. It will take a long time for the audience to move on to one where the value of the descriptor is not one on where it sits within the levels of progression but on where it should take the student next.

    The debate I fear will go on past the value of the fruits of a student’s efforts to the measure of where they stand on the scales of judgement when judged against a sack of potatoes.

  9. teachingbattleground April 7, 2014 at 6:20 pm - Reply

    “Perkins suggests that we should consider what ‘playing the whole game’ in our subject might look like – what do academics, or experts in our domain do? And if we work backwards from this, maybe we will be able to see how ‘junior versions’ or ‘threshold activities’ might ‘make the game worth playing’.”

    I can’t help thinking this is a commonplace idea already, and a bad one. There are no short cuts to thinking like an expert, other than gaining the subject knowledge and fluency of an expert. I think this is something Dan Willingham wrote about.

  10. john gammer (@questionsit) April 8, 2014 at 2:29 pm - Reply

    You can’t teach what is assessed, you can only ‘teach’ it. If you want a pupil to pass GCSE maths, get past papers, drill over and over and Bob is your uncle. Teaching doesn’t come into it. This is what Gove and his mates don’t understand.

    • Ian Lynch April 8, 2014 at 3:31 pm - Reply

      Actually for maths that might work IF the kids were either frightened enough of the consequences of not to doing it or so in love with doing past papers they did it for the craic. That might apply to possibly 10-20% of them? What about the other 80-90%?

      • john gammer (@questionsit) April 8, 2014 at 5:15 pm - Reply

        There are very few kids who do well in exams without cramming. Success in things like GCSE has nothing to do with learning.

        • Ian Lynch April 8, 2014 at 6:07 pm - Reply

          I think it has something to do with it but I think we will agree that its certainly less than optimum – that’s why I set up a new Awarding Organisation and got qualifications accredited that get league table points but without the political constraints of GCSE. Here is an example of what one student produced for their coursework which I think would be very unlikely with GCSE. But then everyone knows vocational alternatives are inferior 😉

  11. […] Hey you, don’t teach that. Teach this! Do we always teach what we value? it seems to me that when push comes to shove, we end up teaching what is assessed. The urgency of accountability results, inexorably, in teaching to the test. And this, sadly, ends up with teachers teaching stuff that they don’t particularly value.  […]

  12. 300000questions April 18, 2014 at 2:08 pm - Reply

    We’re enjoying the freedom to create our own assessment criteria for KS3 and make it more challenging. The ‘exceptional and beyond’ grid will be the most important for us and THANK YOU David and colleagues for sharing this. The wording has really helped me tweak and improve what we had begun. I would echo some of the concerns about needing more specific descriptors. Whilst I like the idea of all skills in English being interdisciplinary and important, we have found that we still need / want separate assessment grids for reading and writing at KS3. However, we have included more rigorous criteria in reading for QWC. I also think we’d need slightly more specific criteria than modelled here. I know my staff like something a bit specific to hang a mark / comment on and we are hoping to use the assessment grids with pupils to help them target set.

    The idea of three broad topics in the curriculum is interesting. I would worry slightly about reliance on extracts rather than whole texts, but I’m sure this could be / is being planned for in sow. We have tried to use broad titles for units of work with ambitious content too. I really like the planning methodology from ‘Creating Outstanding Classrooms.’ We need to think more about this in our team. I think finding out what they already know will be absolutely key, especially as we don’t really know what data we’ll get from Primary Schools now. On the other hand, a less numerical and labelling approach between KS2 and 3 might enable us to actually look at the transition of the child itself rather than worrying about set levels of progress that don’t really take into account the child or external factors.

    Amazingly thought provoking blog post. Thanks as ever David.

  13. Susie Crozier April 22, 2014 at 2:51 pm - Reply

    Don’t know if this helps anyone else, but as the author of the “Story of English” scheme from these sessions, I have begun each lesson or set of 3 lessons with a question. The principle behind this is to unlock thinking and allow ieas freedom without the shackles of the criteria that we’re used to. In fact, our (as yet untried) criteria has beyond at is highest eschelon. There will always be pupils who are put off by the top grades of a subject because they think they’ll never get there, but there is something about the word ‘beyond’ that doesn’t simply smack of high achievement in old defintions, but of direction – beyond doesn’t mean up alone (see Buzz Lightyear for details).

    What I am thinking is that the questions that overarch each series of lessons are the means to encourage pupils to achieve ‘beyond’. What scares me is that it’s a different way of teaching, freer in thought than anything I’ve done before. What terrifies me is that someone is going to want something tangible or restricted, because it’s easily measurable, or, more importantly that I’ll revert to teaching that way … because it’s easily measureable.

    I can say for definite that a I haven’t been this excited about teaching for a long time, a very long time. I can’t say it’ll work well or even work, (although instinct tells me parts of this scheme will fly), but I’m feeling good about it.

  14. David Didau April 25, 2014 at 12:37 pm - Reply

    Thanks so much for commenting Suzie – I’d love to see the latest incarnation of your scheme of learning – I’m delighted you’re so excited about teaching it

  15. […] can also read David’s excellent blog about his work with us One step beyond – assessing what we value as well as his subsequent blog about the efficacy of our assessment system Does it do what it’s […]

  16. […] Hey you, don’t teach that. Teach this! Do we always teach what we value? it seems to me that when push comes to shove, we end up teaching what is assessed. The urgency of accountability results, inexorably, in teaching to the test.  […]

  17. Assessment | Pearltrees September 8, 2014 at 9:08 pm - Reply

    […] manageable and useful for teachers? Will it identify where students are falling behind soon enough? One step beyond – assessing what we value. Hey you, don’t teach […]

Constructive feedback is always appreciated

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: