After Stuart Kime’s excellent talk at Saturday’s Third Space event, the most significant input for us as a school was the excellent exposition of why formative assessment so often runs into the buffers, and what to do about it, from Nikki Booth. A PowerPoint of his talk can be found here, but you can also read the full article from the CCT journal, Impact. You have to be a member to get the whole thing, so the presentation is a bit more accessible. Nikki’s basic thesis was that there has not been enough clear definition of what formative assessment actually is, to enable educators to use it confidently to advance pupil learning. In the process, he moved the ground away from formative assessment as a concept onto the accurate use of feedback and success criteria within a Shirley Clarke-inspired growth-mindset approach. His telling of the story was masterly and as all his experience is from the secondary school sector, I found that I had learnt masses by the end, even though it was material I already was very familiar with.
- I liked Nikki’s definition of formative assessment as being the use of information to adjust teaching and learning, during the learning process, a dialogue between the student and the teacher. This seems to me to be fundamental if we are to extract from the assessment process its full potential. We are using our children as research resources to progress learning for all in the class.
- The requirement that the information be used and acted upon is also central to its use, the lack of such maybe being the reason why Rob Coe (2013, Improving Education: a triumph of hope over experience, CEM) says that formative assessment has little or limited impact on learning outcomes. Surprising? Challenging to teachers? I think it is. Part of it is that formative assessment has turned into “mini-summative assessments” in which we constantly grade pupils and then try and move them on in the lesson.
- The work of Ruth Butler in Israel was the first (but definitely not the last) to evaluate the response of students to formative assessment carried out in terms of grades and in terms of comments. The way Butler worked was to give students a task to do, then had it independently marked, and fed back two days later in different ways: comments, comments with grades, grades only. Marks were kept in the teacher’s planners. They were then given some similar work to do and told that they would be marked in the same way as previously. The startling outcome (then!) was that those who had been given grades only made no subsequent gain, whilst those receiving comments only made a gain in marks averaging 30%. So far, so interesting. The students who got a high grade felt good about this, and those that got a low mark felt rubbish, but neither made any progress at all. Where the fascinating outcome is in those who got given both grades AND comments. These students also failed to make progress, with the same attitudes – high grades felt OK, low grades did not. But still, despite the comments, it seemed to Butler that the grades “fixed the mindset” of students and stopped them progressing. Subsequent work by many others has reinforced this finding. Alfie Kohn (OK, he is on the left wing of this debate) says therefore that that we should never grade students whilst they are still learning; as soon as they receive a grade, the learning stops as students focus far more on the consequences of the grades. So grading students in the lesson directly inhibits learning.
- The problem often lies in our confusion around what constitutes good data. Grades, levels, marks – these always are an approximation to the truth about what a child can or cannot do. They are shorthand for a bunch of skills – at best – and at worst are a backward looking view of what a child can do. We know this, and for various spurious reasons we continue to pay homage to the approximate data rather than focusing on the exactitude of what children can and cannot do. The insistence by Stuart Kime that we refer to information is a helpful and liberating one. Data means numbers in our minds, though the Latin root simply means that which is given.
- The way to progress is to “close the circle” of learning within each lesson and start thinking about clear learning intentions, strong and helpful success criteria, and feedback that is task centred. The problem with grades is that they feed the ego, as Butler originally wrote. And the strength of the ego so fixes our own view of learning that no matter how much we think we have a growth mindset, the success represented by grades enhances the feeling of self-worth (whether negatively or positively) to the extent that comments on work fade to the background. Arriving at the work of Shirley Clarke is the right place for this debate to end up, and a thorough study of the different types of success criteria (such as the one that Tracey Feil presented here) will help us move children on without the inhibition of the self-worth issues.
- The use of focused multiple choice questions that elicit an understanding of common misconceptions is an area that Dan Marshall at Christ the Sower has explored very competently with the use of the Plickers tool, but you can do it with mini-whiteboards (or slates!) just as well. Such “hinge point” multiple choice questions check student understanding at a given point in the lesson and tell the teacher immediately where there are gaps in understanding.
There is much more in Nikki’s presentation. I will be using some of this material – particularly the Butler research, and Tracey’s work on feedback – with NQTs at MKTSA this afternoon.
Final reflections. We are well down this road at Christ the Sower. We need to keep talking about it, sharing it, challenging each other to keep focused on it, knowing that it has such a thorough evidence base.