Measuring educational success using standardized tests suffers from Goodhart's law: you incentivize teaching towards the test. What the goal of education should be is to maximize the amount that students retain 3, 5, 10 years later
This is a statistically very difficult goal. It is also dismal, pedagogically. You student will forget 80-90% of everything you teach them; which 10-20% do you think really will stick, in the long run?
This also means that our testing frameworks dramatically miss a set of students who are really good at learning "deeply", but don't bother learning "broadly" enough to do well at exams. Personally, I think these students exist and many go on to do very well
1.75K