Presumably all that has been compared is the percentage of pupils achieving certain levels at age 11 and then at 14. No account has been taken of the view that comparisons between levels in science at KS2 and KS3 cannot be made in such a simplistic way.
A glance at the programmes of study, and the relevant national tests, shows that conceptual demand and factual content of KS2 tests are much less than those at KS3. Add the gossly different teaching environment of a Year 6 and a Year 8 class, and it is clear how unreliable are such glib comparisons. Measuring a child's improvement is not as simple as comparing numbers in different types of exam taken under different sets of circumstances.
1, Bridge View
Selby, North Yorkshire
The editor writes: The contrast in science performance at key stage 2 and 3 is not based on different percentages achieving as expected, but the fact that this proportion has risen by a quarter at age 11 since 1996 with no equivalent improvement at 14.