4 ms·

I’d love to see more linkage between teacher pay and student success. Unfortunately there aren’t any public schools in my area that factor performance into pay,

I’d love to see more linkage between teacher pay and student success. Unfortunately there aren’t any public schools in my area that factor performance into pay, presumably because the unions are very strong here (CA). Instead, we have a mix of teachers, including some who throw things and yell at students, but can’t be fired.

jayd16 2y ago
It would be a perverse incentive to get struggling kids out of your class and off your record. Same flawed logic behind No Child Left Behind. Underfunding problem areas is not a recipe for success.
- gnicholas 2y ago
  Sure, you’d want to use students’ incoming skills and trajectory to make sure you’re not penalizing teachers who have lots of low-performing kids.
  - permo-w 2y ago
    that doesn't solve the stated problem
    - gnicholas 2y ago
      There isn’t an incentive to get rid of a student who is low-performing if that aspect is taken into account when calculating the teachers’ impact on student success.
      - permo-w 2y ago
        a student who is low-performing is not necessarily a student whose grades are low. they could also be a student who struggles to improve. and let's not forget the other perverse incentive here: if growth is all that's taken into account what happens to the students whose grades are already good?
        gnicholas 2y ago
        It wouldn’t just be looking at grades, of course. If you have a 5th grader getting A’s that turns into a 6th grader getting A’s, no one would assume they haven’t learned anything. You could look at scores on the many standardized tests that kids already take, in combination with grades.
        permo-w 2y ago
        scores on standardised tests != grades?
        gnicholas 2y ago
        No, of course they don’t! This is especially true with the adaptive tests that are now used, which allow students to demonstrate competency well above grade level.
cloverich 2y ago
I'd imagine you'd end up under paying in poorer areas where performance is generally lower and needs are higher. AFAIK performance is generally more tied to parents (involvement, class, etc) than to teachers. Interesting anecdote on teachers throwing things at students, here in Oregon its usually the opposite case we hear people complaining about.
- gnicholas 2y ago
  It would be much more appropriate to implement the system at a local level to avoid the issues you describe. That way you’re not comparing teachers in wildly disparate environments.
  - permo-w 2y ago
    if you leave the specifics to be implemented at local level then you're just going to create perverse incentives in a variable manner based on how competent local administrators are. the whole idea is poor. there's little benefit in putting pressure on teachers like this. all you're doing is unnaturally dragging their motivations away from helping kids grow into well-rounded individuals for the sake of them and society to earning more money based on some arbitrary goal dreamed up by a local bureaucrat
    - gnicholas 2y ago
      Funny, it works great at the districts I’m aware of that use it. You seem to be assuming that teachers whose pay is tied only to seniority care more about kids growing into well-rounded individuals. Why would you assume that?
      - permo-w 2y ago
        does it? based on what metric? data please why would I assume that? because their motivations are not being skewed by an additional perverse financial incentive?
        2y ago
psychlops 2y ago
Pretty sure every teacher would simply move to wealthy area where they didn't have to teach anything and they would still get amazing successful students.
- gnicholas 2y ago
  Not if they’re measured against the other teachers at that school! We see too much of what you describe in our public school, which takes credit for all the student achievement, even though they effectively ignore 30% of students for all academic purposes.
HPsquared 2y ago
It could be an "umbrellas cause rain" though, they may well increase spending in areas with poorly-performing students.
permo-w 2y ago
factoring performance into pay will simply create perverse incentives that will damage students in other ways. it's obvious that paying teachers better will result in greater student success because of supply and demand in the labour market.
mdasen 2y ago
I think that's simply hard to do. What is student success? Test scores? Teachers will teach to the test or even help students cheat on standardized tests if their pay is at stake. If the evaluation is known ahead of time, we'll just see Goodhart's Law in action ("When a measure becomes a target, it ceases to be a good measure" https://en.wikipedia.org/wiki/Goodhart's_law) https://en.wikipedia.org/wiki/Goodhart's_law). Not only that, but a huge part of "student success" isn't based on the teacher's performance. For example, if you give me a class of wealthy students whose parents have graduate degrees, it'll be a lot easier to teach them than a classroom of low-income students without access to resources who aren't set up for success. In fact, school administrators could tip the scales toward teachers they like and retaliate against those they don't like even with "objective" measures. Let's say that we have a perfect test of student advancement - a perfect measure, can't be taught to, and can't be cheated on. I dislike you as a teacher and load up your class with all the known disruptive students. You spend your time on classroom management (controlling student outbursts, keeping students in their seats, etc) rather than teaching. This perfect, objective measure comes around and it turns out you're a terrible teacher! Your students aren't doing better than before your class. Except the fault of that is mine - I manufactured a classroom that was doomed from the start. That's not to say these things don't happen in industry as well. I've seen managers give people they didn't like projects they knew would fail or gave them a team of under-performers to drag them down. However, it's kinda worse when kid's futures are in the balance. Yes, there are certainly better and worse teachers, but it can be hard to determine who is doing well.
- gnicholas 2y ago
  > However, it's kinda worse when kid's futures are in the balance. Considering the alternative is paying people simply based on how many years based on seniority alone, we should endeavor to use a system that at least attempts to take performance into account.
RobotToaster 2y ago
That would be a terrible idea for multiple reasons, mostly because of Goodhart's law https://en.wikipedia.org/wiki/Goodhart%27s_law https://en.wikipedia.org/wiki/Goodhart%27s_law
- gnicholas 2y ago
  So is it a bad idea to have metrics for promotion and pay in other industries, where they are very common? The unique aspect about teaching is that the people who could evaluate them subjectively are children, and we wouldn’t want to put too much weight on their opinions. Seems like there should be some objective measures, instead of just throwing up our hands and paying based on seniority (which creates other bad incentives).

jayd16 2y ago

It would be a perverse incentive to get struggling kids out of your class and off your record. Same flawed logic behind No Child Left Behind. Underfunding problem areas is not a recipe for success.

gnicholas 2y ago
Sure, you’d want to use students’ incoming skills and trajectory to make sure you’re not penalizing teachers who have lots of low-performing kids.
- permo-w 2y ago
  that doesn't solve the stated problem
  - gnicholas 2y ago
    There isn’t an incentive to get rid of a student who is low-performing if that aspect is taken into account when calculating the teachers’ impact on student success.
    - permo-w 2y ago
      a student who is low-performing is not necessarily a student whose grades are low. they could also be a student who struggles to improve. and let's not forget the other perverse incentive here: if growth is all that's taken into account what happens to the students whose grades are already good?
      - gnicholas 2y ago
        It wouldn’t just be looking at grades, of course. If you have a 5th grader getting A’s that turns into a 6th grader getting A’s, no one would assume they haven’t learned anything. You could look at scores on the many standardized tests that kids already take, in combination with grades.
        permo-w 2y ago
        scores on standardised tests != grades?
        gnicholas 2y ago
        No, of course they don’t! This is especially true with the adaptive tests that are now used, which allow students to demonstrate competency well above grade level.

cloverich 2y ago

I'd imagine you'd end up under paying in poorer areas where performance is generally lower and needs are higher. AFAIK performance is generally more tied to parents (involvement, class, etc) than to teachers. Interesting anecdote on teachers throwing things at students, here in Oregon its usually the opposite case we hear people complaining about.

gnicholas 2y ago
It would be much more appropriate to implement the system at a local level to avoid the issues you describe. That way you’re not comparing teachers in wildly disparate environments.
- permo-w 2y ago
  if you leave the specifics to be implemented at local level then you're just going to create perverse incentives in a variable manner based on how competent local administrators are. the whole idea is poor. there's little benefit in putting pressure on teachers like this. all you're doing is unnaturally dragging their motivations away from helping kids grow into well-rounded individuals for the sake of them and society to earning more money based on some arbitrary goal dreamed up by a local bureaucrat
  - gnicholas 2y ago
    Funny, it works great at the districts I’m aware of that use it. You seem to be assuming that teachers whose pay is tied only to seniority care more about kids growing into well-rounded individuals. Why would you assume that?
    - permo-w 2y ago
      does it? based on what metric? data please why would I assume that? because their motivations are not being skewed by an additional perverse financial incentive?
      - 2y ago

psychlops 2y ago

Pretty sure every teacher would simply move to wealthy area where they didn't have to teach anything and they would still get amazing successful students.

gnicholas 2y ago
Not if they’re measured against the other teachers at that school! We see too much of what you describe in our public school, which takes credit for all the student achievement, even though they effectively ignore 30% of students for all academic purposes.

HPsquared 2y ago

It could be an "umbrellas cause rain" though, they may well increase spending in areas with poorly-performing students.

permo-w 2y ago

factoring performance into pay will simply create perverse incentives that will damage students in other ways. it's obvious that paying teachers better will result in greater student success because of supply and demand in the labour market.

mdasen 2y ago

I think that's simply hard to do. What is student success? Test scores? Teachers will teach to the test or even help students cheat on standardized tests if their pay is at stake. If the evaluation is known ahead of time, we'll just see Goodhart's Law in action ("When a measure becomes a target, it ceases to be a good measure" https://en.wikipedia.org/wiki/Goodhart's_law) https://en.wikipedia.org/wiki/Goodhart's_law). Not only that, but a huge part of "student success" isn't based on the teacher's performance. For example, if you give me a class of wealthy students whose parents have graduate degrees, it'll be a lot easier to teach them than a classroom of low-income students without access to resources who aren't set up for success. In fact, school administrators could tip the scales toward teachers they like and retaliate against those they don't like even with "objective" measures. Let's say that we have a perfect test of student advancement - a perfect measure, can't be taught to, and can't be cheated on. I dislike you as a teacher and load up your class with all the known disruptive students. You spend your time on classroom management (controlling student outbursts, keeping students in their seats, etc) rather than teaching. This perfect, objective measure comes around and it turns out you're a terrible teacher! Your students aren't doing better than before your class. Except the fault of that is mine - I manufactured a classroom that was doomed from the start. That's not to say these things don't happen in industry as well. I've seen managers give people they didn't like projects they knew would fail or gave them a team of under-performers to drag them down. However, it's kinda worse when kid's futures are in the balance. Yes, there are certainly better and worse teachers, but it can be hard to determine who is doing well.

gnicholas 2y ago
> However, it's kinda worse when kid's futures are in the balance. Considering the alternative is paying people simply based on how many years based on seniority alone, we should endeavor to use a system that at least attempts to take performance into account.

RobotToaster 2y ago

That would be a terrible idea for multiple reasons, mostly because of Goodhart's law https://en.wikipedia.org/wiki/Goodhart%27s_law https://en.wikipedia.org/wiki/Goodhart%27s_law

gnicholas 2y ago
So is it a bad idea to have metrics for promotion and pay in other industries, where they are very common? The unique aspect about teaching is that the people who could evaluate them subjectively are children, and we wouldn’t want to put too much weight on their opinions. Seems like there should be some objective measures, instead of just throwing up our hands and paying based on seniority (which creates other bad incentives).