Why would it be ineffective? Suppose you're a teacher at a school where the kids are all below grade level. Sounds like a much larger opportunity to get those bonuses than a school where all kids are above average.
> or maybe your should think a bit before blaming teachers
Teachers are only human, and humans respond to positive incentives. The current system has no incentives.
You claimed that the fix to getting lucky with a batch of good students (or unlucky with bad) is random assignment. Now it seems that you're claiming that getting a bad batch is a good deal as it will be easier to get them to improve...
I think you have baked into your plan an incorrect assumption: You are massively overestimating the effect a teacher's input has on student output. Of all the things that lead to student education performance, the quality/performance of the teacher is very low on the list. Teachers are not factory workers, where if they are more skilled, or faster, or better trained, they'll produce more widgets faster.
Most teachers can predict each of their student's year-end educational performance by the end of the first parent-teacher meeting week. Students whose parents who are not involved or where there is no culture valuing education at home are pretty much screwed, no matter how much effort is spent on them, and students whose parents are dialed in and taking an active role in their educations are going to succeed regardless of whether the teacher is even there.
Basing a teacher's bonus on student performance will have one effect: Teachers will be incentivized to move to schools or districts with better students.
There are teachers who are better than other teachers, but it's not generally measurable in "student outcomes". Just like there are better programmers than other programmers, but it's not measurable in "company revenue".
> You are massively overestimating the effect a teacher's input has on student output
Who needs teachers, then, if they are so ineffectual? Might as well replace them with a canned video course.
BTW, every school knows who the good teachers are and who the useless ones are. They get paid exactly the same. Do you think that's a good system?
> There are teachers who are better than other teachers, but it's not generally measurable in "student outcomes".
Of course they are.
> Just like there are better programmers than other programmers, but it's not measurable in "company revenue".
Company revenue is the sum of all the contributions of its workers. A student outcome is not the sum of the other student's outcomes, and is measurable independently.
> Teachers will be incentivized to move to schools or districts with better students.
Sure. And there are only so many of those positions available.
My parents and step-parents are all retired teachers so I've seen the system from both the student and teacher's sides (and now also as a parent). School is not like a factory where raw material (students) come in, teachers apply work onto the raw material, and then finished product (educated students) come out. You can measure the students, but you are not measuring teaching quality. Student success is probably close to 95% parents/culture/homelife/nutrition and 5% some result of teacher input. If you have a reliable way to isolate and measure that 5% independently, by all means, suggest it to your local school board. They would absolutely love it.
> Who needs teachers, then, if they are so ineffectual? Might as well replace them with a canned video course.
Replacing teachers by canned video courses does not a priori sound like a bad idea.
The central reason why this is not done is that school also serves as daycare, so you need some employees to supervise the children, i.e. removing the teachers will hardly decrease the employee count. All together, if we consider the additinal cost of creating the video courses, this measure would hardly decrease the cost of schooling. So politicians think "never change a running system" and "avoid the trouble with the teacher's unions" and leave everything as it is.
I think it’s the reverse. The baseline remedial student growth is say .8 grades per year. Therefore to get 1.0 grades of progress you would need 125% baseline effectiveness. To get the same with a 1.2 growth rate student it would only take 83%. Students also can’t be judged in isolation, remedial students adversely affect other students so that needs to be taken into account somehow.
I’m not saying it couldn’t be done but you would need a pretty sophisticated model to try to figure out who is or isn’t effective. Then once you turned on the model you would need to constantly tweek it to handle metrics based tampering.
That has always been a very popular and well received idea (or maybe your should think a bit before blaming teachers).