Measuring improvement is even easier to cheat. One merely has to purposefully give wrong answers in the early parts of the class, and then answer more earnestly in the latter half of the class.
You'd need to have continuous tracking of performance across grades. Low performance at the beginning of a school year would translate to a drop in progress compared to the end of the previous year.
You could still game it by aiming for the average, but really the average is the expectation. And in a world where subject matter difficulty can scale with a child irrespective of the skill level of peers their age, getting an average result on a test is ok.
You could still game it by aiming for the average, but really the average is the expectation. And in a world where subject matter difficulty can scale with a child irrespective of the skill level of peers their age, getting an average result on a test is ok.