I have blogged about this in the past, but I make no apology for doing so again. I am inspired to do so by having attended the Learning First conference in Sheffield, where there were calls for an end to tick box teaching and assessing children against dozens of criteria. I am also currently marking examination papers, which has again brought the whole problem of success criteria into focus. “I like it, because it’s good” is the standard offering from a pupil who has obviously sat down with a list of words to be incorporated into work and ticked off. Even worse are statements which directly contradict each other, but have obviously been inserted into a piece of work to enable more “success criteria” boxes to be ticked.
The DfE itself ran into a mess by producing a list of “can do” statements for primary schools. Instead, they could have saved themselves a lot of trouble by simply publishing exemplar questions for the Year 6 SATs – nothing more. Of course, there would have been initial protest by people who cannot see beyond success criteria, saying “But what do they need to do to pass?” and “Show the sort of work that would be a pass standard!” But once you make it clear there is no set “pass” standard, teachers are then freed up to try to get their pupils to be the best they can possibly be, engaging with the content of a subject, rather than trying to meet vague criteria.
So let’s ditch criteria altogether! Instead, we make it clear that pupils will be assessed against questions, rather than criteria. Instead of publishing a specification, examination boards should simply provide exemplar questions for GCSE and A level. They should not publish examples of “A grade answers” or “C grade answers” – all answers to questions should be norm referenced. Instead of frantically checking whether a pupil has fulfilled some vague, subjective criteria, pupils are encouraged to think how their answers could be improved – no more infuriating “is this enough to get me an A?” questions! Essays could be marked according to comparative judgements.
“Aha!” I hear you say. “To compare two essays, a teacher must have some criteria in his/her head to make the comparison!” This is true to some extent. But I would venture to suggest that attempts to define these criteria in a series of statements are doomed to fail and you end up with what Daisy Christodoulou so memorably termed “adverb soup.” The criteria which make one essay better than another will be varied and complex and will often be very specific to the two essays being compared. Why not just accept that education is complex and does not easily fit into a series of boxes?
“But norm referencing has its own unfairness!” is the next objection. True enough – under my proposed system, an answer which would have been a C grade one year could be a B grade the following year, depending on the results of a particular cohort. However, I would say it is no more unfair than missing out on a grade, because success criteria have been interpreted differently by different examiners. And more importantly, teachers could go back to engaging pupils with the content of their subject, rather than poring over mark schemes.
Perhaps the most controversial aspect of norm referencing will be when a pupil asks “so what do i need to do to improve?” At that point, instead of producing a list of criteria, a teacher will be obliged to find a better piece of work, which might well be from another person in the class. This horrifies some people. Yet when you think about it, why should it? In school sports days up and down the country, pupils will be having their performance norm referenced and they will be comparing their performance with those of their peers. It seems to work for sport.