It’s been a long time under discussion, and yet for all the talk in school and out, it seems that many have forgotten the original rationale for the scrapping of levels.
Tim Oates set the explanations out very clearly in the video presented on the DfE YouTube channel (which if you haven’t seen, is worth looking out). The main thrusts of the argument fell into three categories:
- Children were self-labelling
- Undue pace was forced into the curriculum
- The comparability of test scores, best-fit judgements and ‘just in’ measures
But when we look at the majority of systems that have replaced levels, have we really moved on?
My poll of systems being used in primary schools suggests not. The most popular tracking packages appear to make up the majority of school systems, although it’s interesting to see a large proportion of bespoke systems as well as many schools with none. However, despite this variety, one message stands out: from the entirely non-scientific poll I’ve run, it seems that half of schools are still depending on systems that require 3 or 6 points of progress to be measured each year. How does this tackle those initial problems with levels?
Have we simply replaced the self-labelling of “I’m a Level 3”, with “I’m Emerging”? Indeed, in some such systems, might we not run the risk that a child remains permanently as “Emerging”, labelled not only in comparison to his peers, but indeed as a permanent characteristic. Whether the language is “developing”, or “beginning” or “below”, might not the effect be the same or worse than with levels?
As for undue pace, surely by demanding steps of progress again, we’ve simply replicated the same old problems? In fact, I’d argue that we’ve worsened them. Having got so used to the APP model, many schools have now adopted a system that requires the recording of endless theoretically-formative judgements in order to reach a summative point score or category. Once again the risk is that pupils near to thresholds will become the focus, rather than those who most need additional support, and that moving more quickly through the steps will be seen as positive, neglecting the need to secure understanding and skills.
So have we solved the problem of the different meanings of levels? Sadly, the same symptoms are evident: schools and tracking companies have tried to replicate old systems. Rather than focussing on what children can and can’t do, too much time and energy is focussed on predicting the resulting summative judgement. It’s true that the new interim assessment frameworks remove the ‘best-fit’ judgement issue (although I’m not convinced that’s a good thing!), but we still have many systems that focus on using a best-fit approach to summarise judgements using a category label. If you need to get a certain number of ticks to be placed in a particular category, then surely we might as well have stuck with APP?
So what’s the solution?
I’m increasingly coming to the view that our first task should be to separate formative and summative assessment entirely. The current systems just aren’t working.