I had an conversation with a just-retired teacher this weekend. In addition to the days (a week?) of testing the students at her school had to endure, they were a site for testing a new test and so were subjected to another week of testing.
It's possible for students to opt-out of standardized testing but of course the school is required to have a very high participation rate (90+%?) or else they'll be penalized. One of the teachers had been informing students of their right to opt-out and making the necessary forms available to them.
As soon as the administration saw a number of students opting out, they brought them in one-by-one and interrogated them as to which of their teachers was doing this. He was admonished (or worse).
The whole system of criteria and incentives are so broken. This is analytics and big-data gone awry. "You can't improve what you don't measure," sure, but the wrong things are being measured and the incentives are wrong and the consequences backwards. Sigh.
When you start measuring things in education, what you get collapses onto what you are measuring, like a wave function collapsing in quantum mechanics. Everything else goes out the window.
In the UK, we have a Parliamentary Committee for Education, the leader of which is Graham Stuart, a Conservative MP who has shown on a number of occasions that he is prepared to challenge the education minister, Michael Gove, when the committee sees an issue developing.
One morning on the radio he said something like 'if you set targets for schools, don't be surprised if schools do what they have to do to achieve the targets'. I didn't have time to get the quote down verbatim.
In the UK context this particular incident would be impossible for a variety of reasons, but we do see very heavy resources being concentrated on students expected to get a C (pass) on GCSE exams to the detriment (arguably) of weaker students and students at the top of the ability range. Triage may help manage casualties on a battle field but it is not so good for school children.
Most schools call the students they focus on 'DtoC' students. Gove is trying to hack the measurement system to remove the incentives for schools doing this. Same as OA, it might be better to have a system where percentage pass scores are not mechanistically used in this way.
My understanding of the OA is that the school where the teacher cheated was under threat of closure unless test scores reached some arbitrary percentage. The existence of tests themselves is not the issue as you say.
> "You can't improve what you don't measure," sure, but the wrong things are being measured and the incentives are wrong and the consequences backwards.
A direct corollary: you get what you measure. Nothing more, and nothing less.
In education and humans therein lies the problem. When all effort is expended towards hitting metrics, only the chosen metrics will be improved. Everything else is left to degrade. If you add the incentives to punish for not hitting the targets, of course you get cheating. A little at first, but when the improved goals are now set based on false data, you encourage a more systematic cheating.
One aspect of problem has been already witnessed in Finland, the country of supposedly superior education system. University staff are well aware of this but can do nothing about it. If I had to guess, I would say that the administators are simply divorced from reality.
I don't know if there is much cheating going on. We may not be that far down in the drain; what we do have, is a constant pressure to over-achieve. There are teachers and professors who all say same the same thing.
I wonder what happens when the researchers and teachers will get punished for not meeting their >100% performance targets year-over-year.
"I wonder what happens when the researchers and teachers will get punished for not meeting their >100% performance targets year-over-year."
This is not considered a bug in the system, if we graduate more of them than we need, and as seen in the article cronyism and gang/mafia mentality is rampant.
So in a system built on cheating, the "cooperative" will have their performance targets fixed, and the "non-team players" will be removed.
Superficially the goal of standardized testing is to produce a populace thats real good at taking standardized tests, which is of course a totally useless goal, but its easy to measure. Slightly deeper, the goal of standardized testing is to remove autonomy from teachers as a punishment by forcing them to teach to the test. A bit deeper and the purpose of standardized testing is to encourage corruption. If one party has a political goal of fighting the teachers union, what better long term plan than to force them all to become thugs or leave? The other party likes it also, because who doesn't want an army of thugs?
Nobody wants the kids to learn anything, unfortunately. Sometimes it happens accidentally, but it doesn't matter.
> This is analytics and big-data gone awry. "You can't improve what you don't measure," sure, but the wrong things are being measured and the incentives are wrong and the consequences backwards. Sigh.
This works well when it's factory spitting out Chevy sedans. You measure the parts, if they don't fit you fix them or replace them.
But it can't work for people. Human beings are not cogs in a machine. They're not easily measured, and the act of measuring can actually change the values... for the worse.
Heh. I've heard from other teachers that we have the opposite issue: some schools encouraging students who will perform poorly to be absent whenever testing happens.
(Australia, not US, so not sure how you lot handle absentees for testing purposes.)
It's possible for students to opt-out of standardized testing but of course the school is required to have a very high participation rate (90+%?) or else they'll be penalized. One of the teachers had been informing students of their right to opt-out and making the necessary forms available to them.
As soon as the administration saw a number of students opting out, they brought them in one-by-one and interrogated them as to which of their teachers was doing this. He was admonished (or worse).
The whole system of criteria and incentives are so broken. This is analytics and big-data gone awry. "You can't improve what you don't measure," sure, but the wrong things are being measured and the incentives are wrong and the consequences backwards. Sigh.