Thursday, May 05, 2005

Testing is a diagnostic tool

I came across an anti-testing site and found this stunning revelation:

1. There is no convincing evidence that annual testing actually improves learning.
The logic seems odd. It's like proclaiming that taking the temperature of a patient does not cure his fever. And yet, a whole anti-testing cottage industry is built on this transparent fallacy.

1 comment:

Chris C. said...

That's true.

On the other hand, I've often wondered what's so special about *annual* testing.

Should a state be able to build a better (but more expensive) test and administer it biennially?
Should a state be allowed to to test each student four times a year? That would take away much time from instruction, but could lead to more refined diagnostic data at the student- or teacher-level.

What approach will actually lead to more gains? Is there any evidence to support one approach over another? Does this evidence actually play a role in NCLB's policy?