So, what did you get up to in the summer?
Working in curriculum development, the year has a different shape. As a teacher for twenty years my least favourite month was always September because … well, you’ll know. August was synonymous with being able to get away from thinking about school, at least until the results arrived. I bristled, as many teachers do, about the occasional jibe about long holidays. Any teacher working hard needs and deserves every day of those breaks to recharge the batteries and redevelop the capacity to deal with the constant barrage of challenges and opportunities that a typical school day brings.
Part of my summer this year was spent editing KS3 test questions. I’m not going for the sympathy vote here (it was only part of the summer, after all) and although it doesn’t sound like much fun, writing good test items is an art as well as a science. Good questions should engage as well as challenge.
We’re still at a cross roads, it seems to me, regarding the form and function of testing at KS3. We know (and largely accept) that progress needs to be tracked and reported upon – the question is how.
There seem to be three schools of thought here. The first is the ‘status quo’ approach that some schools are taking, with retaining the use of levels. The advantage with this is that it’s a system that is known and understood. Even though we have a new Programme of Study that has no level descriptors, the old ones can be made to fit. Well, sort of. Bit of tweaking and twisting, but well enough to show who’s making better progress. The thing to watch with this approach is how to set targets – it’s easy to say “Well done, you’ve got a good Level 5, now see if you can get to Level 6” but not so easy to suggest how. Some schools have used ‘level ladders’, but of course these will need re-working as well.
The second approach is the hard numbers one. Adopting this approach, the translation into levels is dropped and tests are used that are reported in scores. As long as the tests are progressive (i.e. the Y8 ones are harder than the Y7 ones but easier than the Y9 ones) this works and saves having to apply old level descriptors to new statements. Oh, and Excel is happier with 52/60 than with Level 6a. Same limitation though – how do you set curricular targets?
The third option is to develop outcomes for each topic, and then judge success against those. This is harder to develop but is more powerful in terms of setting targets. You might say to a student, for example, that they’re good at identifying how sound waves are generated and can describe how they are reflected, and that their next step is to make better use of the wave model in explaining these characteristics.
The questions we’re developing will support any of these of course. Whether you want to work in ‘old money’, raw scores or meeting outcomes, good test items will help. Fundamental question is, of course, not how you assess but why. Is it to report where students are or to identify how to improve? This really needs to be the starting point – once this is resolved the form and function fall into place.