The DfE has announced its proposals for the regrading of GCSEs. This has prompted a torrent of unfavourable comment from those usually opposed to the education policies of Michael Gove. I believe the changes to be welcome and potentially laying important foundations for further positive reforms. The grade inflation of the last two decades has been dangerously corrosive of the entire basis of secondary schooling and had to addressed. See my posts here and here
This is an opportunity to create a statistically and educationally sound basis for assessing and grading end of KS4 attainment. I believe that such assessment is necessary in order to inform student choices post 16. It does not need to be 'high stakes' for schools. A sound assessment system needs to establish objective standards of 'difficulty' for each grade, comparability between subjects (so far as possible), encourage effective deep learning, be resistant to 'gaming' and to support the individual cognitive development of pupils. It is also necessary to ensure comparability with international standards so that the market driven race to the bottom that has characterised the English system can never happen again.
The first step is to tie grades to criterion referenced statements of attainment. I propose going back to Bloom for this. This is how I think the proposed new system could match Bloom's Taxonomy of the 1950s. In the following system the new numerical grade, which I think is better regarded as a 'level', is matched to a Bloom attainment category.
L9 - Exceptional L5-8
L8 - Synthesis, Evaluation and Creativity
L7 - Analysis
L6 - Application
L5 - Understanding - Piaget formal - Kahneman System 2
L4 - Understanding - Piaget concrete - Kahneman System 1
L3 - Even More Knowledge
L2 - More Knowledge
L1 - Some significant Knowledge
To qualify for the award of a particular level it would be necessary for the pupil to demonstrate some capability at that level. An award made at any given level signifies competence in working at all the preceding levels. It is therefore a developmental approach.
As noted in my grade inflation post, The GCE/GCSE C grade was originally designed to be the minimum threshold for progressing to post-16 academic study. It therefore corresponded to L5 on the new system. The current GCSE C grade matches only L4 on the new system because it can now be achieved without demonstrating any understanding at the formal level. I think this is right, reflecting the degradation that has resulted from grade inflation. The barrier between L4 and L5 reflects what Piagetians like me regard as the significant cognitive hurdle that exists between 'descriptive' and 'abstract' thinking. The DfE proposal implies another cognitive hurdle between L8 (A*) and L9 (exceptional A*). I am not sure about this but it could be a good idea.
If you don't like Piaget, then I suggest you try Kahneman's interpretation of Piaget's distinction between concrete and formal operations. There was an excellent BBC2 'Horizon' programme about Daniel Kahneman broadcast on 24 February called, 'How you really make decisions'.
I discuss the relevance of Kahneman to education here.
I have corresponded occasionally with Michael Shayer. This is what he wrote to me when I asked him about the link between Piaget and Kahneman.
"I would say that Piaget's concrete operational thinking, that I usually think of as descriptive thinking, actually shows in great detail what the agenda of [Kahneman's] System 1 thinking is. Indeed Piaget himself said that formal thinking is of value only when it is asked to do further work [Kahneman's Slow Thinking] on what is already a well-described situation in concrete operational terms."
I regard this as a profound insight. This is where 'metacognition' comes in. Metacognition is 'thinking about thinking'. It is an essential part of both Piaget's 'formal' thinking, and of Kahneman's 'slow, System 2 thinking'.
The new grading system will be first applied to English and maths. In the context of maths, L4 reflects the ability to work with 'concrete' concepts, for example, like manipulating numbers and calculating lengths, areas and volumes. L5 requires the ability to work with abstract 'representations' of numbers using algebra and also to cope with logical processes such as Euclidian geometrical inferences and trigonometry.
In order to meet the 'five good GCSEs including English and maths' high stakes threshold, maths teachers learned how to coach 'concrete' level pupils to get C grades through System 1 approaches applied to carefully selected parts of the syllabus. Thus the GCSE C grade slipped down through the concrete/formal hurdle, and there was no incentive to help pupils progress over it.
There has been a lot of debate about whether it will be L5 or L4 that drives league tables and floor targets.
This is where the new system is likely to become degraded by the Gove free market ideology. As soon as any particular level becomes a high stakes target for schools then corruption and degradation will result. If the maths GCSE results of the 'most improved' schools throughout the period of grade inflation are analysed it can be seen that C grade improvements have usually been obtained as a result of the depression of E, D and B grades. If L5 on the new system becomes a 'high stakes' target then Levels, 3,4 and 6 will suffer in the same way as perverse incentives come into play. This will corrupt the potential of the new grading system to support and encourage high quality developmental teaching across the full ability and grade range.
Therefore for the new grades to be of maximum benefit to a reformed education system there must be no high stakes targets for schools at ANY attainment Level.
This means no crude league tables.