I’ve just been marking some Year 7 end of term tests and I wanted to get people’s views on tests split up in to tiers.
I did two tests from different tiers with my class, and inevitably the levels reported by the tests don’t vary wildly.
To me, the whole purpose of summative testing is to get robust data. Why then do we not use as a matter of course one test that covers all possible levels, make it a bit longer and have everyone in a cohort sit the same exam paper. This approach would at the very least get rid of the “which paper will my students be able to get a C on more easily” debate at GCSE level. I also think that doing this throughout will lead to more robust data which is actually useful.
I’ve heard some arguments for having different tiers, such as students having lots of questions they can’t do being bad for self esteem.
What does everyone think? Do you have a preference?