Educational statistics: or how to meaningfully compare apples and bicycles.

We are supposed to give a test in the fall (before we teach the material) and give the same test again in the spring - and then see how much the students have grown. You know, this would show how "effective" the teacher is.

The teachers are not to see the tests (so we cannot teach to the test) (or adequately prepare the students to pass the test).

The test counts as 20% of their grade.

This week we found out the test we will give next week is not exactly the same as the one we gave in the fall. Actually, it is 75% longer than the one in the fall.

Yep, those statistics will be meaningful.