Suppose we do this. How will we know if it failed or succeeded?
The best bridge I've got so far is that testing is a stop gap between here and wherever we want to get. (I assume the end goal is education over just simple literacy.)
Testing has 2 negative aspects to it
1) Its analogous to the effect nicotine or other habit forming drug/medicine has on its subject/user. After a while they have more than just an alleviating effect, the subject is itself altered to depend and operate under the influence of the chemical.
2) For any sufficiently complex task, tests create narrow fields of focus and reduce creativity and mastery. A movement towards the letter of the law, but not the spirit. If you want non educational examples, take everything from CEO performance, to tax evasion, to behavior in video games. It makes an argument for indirect measurement of outcomes.
At the core of the issue is that absolute terribleness of testing to actually measure anything other than itself, and how there is only a small range of measurable tests that can be conducted which are also applicable at mass scales.
Now I can't see ways to ameliorate or reduce these effects. One attempt has been to 'improve tests' - what with more complex tests and so on. All I've seen it do is create a push for more rote learning and often creates a side effect of whimsical marking.
I'm definitely, not going to argue that we should find a new or improved test. That would be as hand-wavy as saying "testing/measurables will solve it".
I suspect the answer will lie here, and at the intersection of what Panos learned with his cheating scandal. Likely a move towards more peer/group oriented work, which has a real world test which needs to be performed. Which of course would require teachers to have more time to work with and test the final result.
I'll respond and see if I can flesh the idea a bit later. Thoughts?
You guys (Americans) had a great system which failed your weakest but managed to get you'll to the moon.
And then for some reason your country decided it wanted to test like India and China.
This is lunacy from where I stand. Having walked through the wringer in another country I know exactly what it becomes. Learning divorced from understanding.
Stop testing, break class size in half, increase teacher pay and double the number of teachers. But no one wants to do that because of the daft political climate, fund crunch, difficulty and the alternative of tests.
So a non human solution.
I've seen the end state of the path you have begun walking down, and it is flat out teach the test land.
Students by guides to the material with the shortest most efficient answers written down for memorization.
People realize that multiple choice tests are too easy, so more difficult tests are devised, in the hope of ensuring that it makes sure students are better prepared.
This results in yet more teach the test.
Tests are a poor proxy for checking if someone understands the material.
The argument is that"tests are all you have and better than nothing.".
What people assume is that testing is static, it's not. The testing process is dynamic and exerts a force on the behavior of the students and teachers, and over time over the recruitment and measures of students and teachers themselves.