In my subject, MFL, descriptors are everywhere and I’ve always disliked them, for the reasons given by Daisy Christodoulou in “Making Good Progress”. Here are some I descriptors I have come across:
Can write some single words from memory, with plausible spelling
OK – how about this: je salu sack an gomm abite mappell onz
Can write simple words and several short phrases from memory, with understandable spelling
This maybe? je mapple dan mon sack jabite fermay la port eel y a quell nombr
Can write words, phrases and short simple sentences from memory, with understandable spelling
This? je mappel Olivia jai onz an eel y a un livres dan sack mon anniversary cest le 13 may
Yes – you can see “progress” But what is the difference between plausible and understandable? Who decides? And indeed, as a French national said to me, “they are all rubbish anyway.”
However the above examples of descriptors are better than some I have seen. How about “Uses a range of linguistic devices” (what do we mean by a range and what do we mean by a linguistic device?) Or “Can write longer sentences” (longer than what?)
I rejoiced when levels were abolished. But I was aware of the danger of simply replicating them and wanted to avoid lessons where pupils spent time ticking off vague “can do” statements on APP grids. Language learning is too complex to be defined in a list of descriptors. I became attracted by Shaun Allison’s idea of the Growth and Thresholds model of assessment and looked at ways this could be adapted to MFL.
One thing I decided to do was avoid simply dividing each assessment into the 4 skills of listening, speaking, reading and writing. As far as I am aware, other European countries do not use this rigid classification when teaching English. My summative assessments (which normally take one lesson – an hour in my school) therefore consist of two sections: “Linguistic Competence” (which may comprise reading, listening or writing tasks) and “Grammar and Vocabulary,” which usually encompasses translation to and from the target language or gap filling tasks. Both sections have equal weighting. We use textbooks and conduct assessments at the end of every unit in Year 7 and every two units in years 8 and 9. There is a threshold mark in each section for “Above expectations”, “Meeting expectations” or “Below expectations”. This fits in with the system my school uses to track progress since levels were abolished.
But what about speaking, I hear you ask? Well, the problem with speaking tests is they take up a ridiculous amount of time. Moreover, I remember from teaching in other countries that they managed to get their pupils speaking without hauling them out to the front one by one for conversation, or without whole lessons where teachers went round the class listening to pair work. So I leave a formal speaking test until the end of the year. This of course does not mean that pupils we do not get pupils speaking in lessons – far from it – but it does mean we avoid lessons where the teacher’s time is taken up hearing 32 pupils say not very much.
When assessments are given back, pupils write their own paragraphs saying where they did well and where and why they went wrong. These paragraphs are checked by the teacher. Mostly we just tick them off. If a pupil has not analysed their performance sufficiently, we add our own suggestions.
The first challenge I anticipated was to ensure that the questions we asked in our assessments really did encompass as much as possible of the language covered in that unit of work and also contained both easier and more difficult tasks. Fortunately I have a team of experienced teachers and we were able to do this without too much difficulty.
The second challenge is one we are still working on – where to set the thresholds for each section in each assessment. What is the minimum percentage in each section needed for the thresholds “Above expectations”, “Meeting expectations” and “Below expectations?” We still have discussions about this.
Will it work long term? Too early to say. I had hoped that GCSEs would move towards the mixed skill approach that is used at A level. Sadly, this was not to be. My current hope is the work being done on Comparative Judgement will spread to MFL GCSE writing tasks. Having marked to rubrics for an exam board, I have become convinced we need a different system for assessing writing tasks. Comparative Judgement is, in my view, the right way to go.
No assessment system is perfect, but the abolition of levels gives us a chance to change things for the better, provided, of course, your SLT is willing to give some autonomy to departments. In that respect, I have been very fortunate. Most other departments in my school use a descriptor based model. As I say, I wanted something different…..