[FatherOfCurses]

It's not a claim, there is psychological data supporting the negative impact stress can have on recall and that for some students a test is not the best way to identify their retained knowledge.

Your single example identifies a situation where we would want someone resilient to stress to pass the testing process.

However if you are evaluating someone's ability to identify characters in Shakespeare who are most closely representative of the Bard himself, a proctored exam may not be the only environment where that could be demonstrated.

Anyone in software knows as well that for a test to be effective it has to be written properly. In the history of academic and standardized testing however there has been little rigor in the construction of tests, and those who pass are ones who give the "right" answers regardless of whether those answers are true proofs of knowledge.

[WalterBright]

I've been around tests all my life. I don't know anyone who bombed the test but mastered the material. And the ones who aced the tests - turns out that they actually had mastered the material.

As for people who cannot perform under the stress of a test, how are they going to perform otherwise? Anyhow, the solution to test anxiety is to keep taking tests - the anxiety will recede.

For example, the first time I tried public speaking I was paralyzed. But I kept trying it, and the anxiety went away, and now I do a fair amount of public speaking and enjoy it very much.

[mangodrunk]

That still doesn’t seem like a reasonable reason to not do tests. In the software analogy you have a unit test that passes but the actual software fails when used in the real world.

What is the point of “knowledge” that can’t be demonstrated. How will that person demonstrate their knowledge of Shakespeare?