New key stage 2 reading and mathematics tests, based on a revised national primary curriculum, were introduced in May 2016. In this report, we present findings from research investigating these new testing arrangements. We focused specifically on the approach adopted by the Standards and Testing Agency (STA) to sampling national curriculum learning outcomes within the new tests.
Although we have identified potential areas for improvement and further research, our findings provide support for the robustness of STA’s approach to domain sampling for the new suite of national curriculum tests. Their approach compares favourably with approaches adopted for similar tests, internationally.
Given STA’s interpretation of the national curriculum framework document, the test framework documents appear to translate national curriculum teaching requirements into plausible blueprints for testing. The degree of consistency with which our independent experts rated items from the 2016 tests supports the conclusion that the way in which STA has specified both the content domain and the cognitive domain is plausible. This is particularly important evidence in relation to the cognitive domain because the domain strands were innovative for both maths and reading, having been introduced specifically for the new tests.
Finally, the degree of consistency between STA’s item ratings and those of our independent experts supported the conclusion that STA’s ratings were plausible and, by extension, that the 2016 tests sampled relevantly and representatively. Some important questions remain however. In particular, the representation of problem solving in the maths test framework and, consequently, in the maths test. This area, and others set out in the report, could benefit from further study.