5 ms·

> Unfortunately the effects of a single teacher are so noisy and indirect that it's basically impossible to measure their performance. The lack of firm data is

> Unfortunately the effects of a single teacher are so noisy and indirect that it's basically impossible to measure their performance. The lack of firm data is why the debate about teacher compensation is so fraught. The problem is not that it's hard to measure. The people is that teachers actually have very little ability to make a marginal impact on their students though their performance, except at the very tail ends of the distribution. In other words, structural factors that teachers can't directly control, like family support or district curriculum standards, have a much larger impact on student performance than teaching quality does. (Except, of course, for the exceptionally bad teachers, but that's better addressed by firing them than by adjusting the compensation strategy).

camelite 9y ago
That's a myth. The real problem is that the average teacher (relative performance) is so far away from being close to maximally performant (as proved in studies which rigourously control teacher behaviour) that the impact of variation about that mean is swamped by environmental factors. People conclude that since swapping teachers doesn't make an appreciable difference, teacher performance can't make a difference. When in reality the problem is that teaching is uniformly bad (compared to maximal possible performance, which is very difficult. this is why the answer is not to look for better teachers but to look for better, uniformly implementable, curriculums)
- asdfgadsfgasfdg 9y ago
  If teaching is indeed universally bad (which I doubt but lets go with the idea as a thought experiment) - then surely the issue must lie with training/the accepted teaching methods?
  - lr4444lr 9y ago
    I disagree with the parent post and do not dismiss your suggestion, but no, that needn't be the only cause. The profession itself may be disincentivized in a number of ways against bringing in capable people.
  - marcosdumay 9y ago
    It's not necessarily training and methods. Any environmental issue would create the same situation (curriculum selection, normative bias into testable results, etc), as would a self reinforcing cycle of undervaluing the professionals and competent people running away from the undervalued profession.
  - camelite 9y ago
    Yes, I agree completely. Though I would emphasise that teaching universally bad only when compared to what is achieveable from tightly scripted presentations. I think the actual task of teaching flawlessly while "winging it" is extremely extremely difficult, and beyond at least 98% of the general population, even positing a good working environment.
- Ciao_a_tutti 9y ago
  I might become a teacher and I don't want to be a bad one. I want to be exceptionally good. Do you have any suggestions?
  - lr4444lr 9y ago
    Pick an area with a strong union that will cover your ass when you have to use your experience, skill, and knowledge to do what has to be done when the admin bureaucrats not only don't help, but even become your chronic obstacle. Make sure you don't try to be too heroic before you get tenure, though. Also, try to get a subbing or student teaching position that will put you with the same group of kids regularly for at least a month to be more sure you can handle the stress of the job.
  - camelite 9y ago
    I would reccommend looking into everything written by Siegfried Engelmann, particularly "Theory of Instruction". Honestly though, that might not be enough. Maybe you have the brains and drive to do it, for me the only unquestionab success has been using scripted presentations written by people who are smarter than I am.
smallnamespace 9y ago
> The problem is not that it's hard to measure. The people is that teachers actually have very little ability to make a marginal impact on their students though their performance, except at the very tail ends of the distribution. That's essentially true for most sales people and traders as well. There are VERY heavy tails -- that one guy at the car dealership who sells 3x as much as his peers, or the trader at Citadel whose trades made over $1bn in a year. But imagine how the teaching profession would change if 1) the gains could be accurately measured and 2) we assigned payment or prestige to teachers based on (1).
- stevenwoo 9y ago
  Based on national test scores, what we know works in a small homogeneous country at least is professionalizing teaching - raising the standards to become a teacher. They didn't raise salaries to the same level as other professionals like lawyers and doctors, it's about the same pay as the USA, but Finland managed to go from worst to the top in Europe in one generation. https://www.theatlantic.com/national/archive/2013/01/the-case-for-a-teacher-bar-exam/267030/ https://www.theatlantic.com/national/archive/2013/01/the-cas...
  - mickronome 9y ago
    It's also interesting to note that children in Finland got a shorter school day, and mostly no homework at all. The shortening of the school day was especially prominent in for the youngest children.
- chimeracoder 9y ago
  > That's essentially true for most sales people and traders as well. It's definitely not true in most sales positions, particularly the ones that are highly commissions-based. > But imagine how the teaching profession would change if 1) the gains could be accurately measured and 2) we assigned payment or prestige to teachers based on (1). We can, and we do. The fact is: the gains just aren't that large. If you want to improve student performance, the best place to spend the money isn't by raising teacher salaries.
lr4444lr 9y ago
Disagree. The tails of the distribution are where the teacher effect is probably the least, because there are significant obstacles on the low end that make focusing on learning impossible and immense innate talent at the high end where the kid would learn if you put him in a closet with the textbook and a flashlight. The real problem with teacher success metrics is that they are outcome-based. It is no more useful to measure the success of a world class surgeon by including the outcomes of people who completely ignore the post-op care directions and instructions for lifestyle changes.

camelite 9y ago

That's a myth. The real problem is that the average teacher (relative performance) is so far away from being close to maximally performant (as proved in studies which rigourously control teacher behaviour) that the impact of variation about that mean is swamped by environmental factors. People conclude that since swapping teachers doesn't make an appreciable difference, teacher performance can't make a difference. When in reality the problem is that teaching is uniformly bad (compared to maximal possible performance, which is very difficult. this is why the answer is not to look for better teachers but to look for better, uniformly implementable, curriculums)

asdfgadsfgasfdg 9y ago
If teaching is indeed universally bad (which I doubt but lets go with the idea as a thought experiment) - then surely the issue must lie with training/the accepted teaching methods?
- lr4444lr 9y ago
  I disagree with the parent post and do not dismiss your suggestion, but no, that needn't be the only cause. The profession itself may be disincentivized in a number of ways against bringing in capable people.
- marcosdumay 9y ago
  It's not necessarily training and methods. Any environmental issue would create the same situation (curriculum selection, normative bias into testable results, etc), as would a self reinforcing cycle of undervaluing the professionals and competent people running away from the undervalued profession.
- camelite 9y ago
  Yes, I agree completely. Though I would emphasise that teaching universally bad only when compared to what is achieveable from tightly scripted presentations. I think the actual task of teaching flawlessly while "winging it" is extremely extremely difficult, and beyond at least 98% of the general population, even positing a good working environment.
Ciao_a_tutti 9y ago
I might become a teacher and I don't want to be a bad one. I want to be exceptionally good. Do you have any suggestions?
- lr4444lr 9y ago
  Pick an area with a strong union that will cover your ass when you have to use your experience, skill, and knowledge to do what has to be done when the admin bureaucrats not only don't help, but even become your chronic obstacle. Make sure you don't try to be too heroic before you get tenure, though. Also, try to get a subbing or student teaching position that will put you with the same group of kids regularly for at least a month to be more sure you can handle the stress of the job.
- camelite 9y ago
  I would reccommend looking into everything written by Siegfried Engelmann, particularly "Theory of Instruction". Honestly though, that might not be enough. Maybe you have the brains and drive to do it, for me the only unquestionab success has been using scripted presentations written by people who are smarter than I am.

smallnamespace 9y ago

> The problem is not that it's hard to measure. The people is that teachers actually have very little ability to make a marginal impact on their students though their performance, except at the very tail ends of the distribution. That's essentially true for most sales people and traders as well. There are VERY heavy tails -- that one guy at the car dealership who sells 3x as much as his peers, or the trader at Citadel whose trades made over $1bn in a year. But imagine how the teaching profession would change if 1) the gains could be accurately measured and 2) we assigned payment or prestige to teachers based on (1).

stevenwoo 9y ago
Based on national test scores, what we know works in a small homogeneous country at least is professionalizing teaching - raising the standards to become a teacher. They didn't raise salaries to the same level as other professionals like lawyers and doctors, it's about the same pay as the USA, but Finland managed to go from worst to the top in Europe in one generation. https://www.theatlantic.com/national/archive/2013/01/the-case-for-a-teacher-bar-exam/267030/ https://www.theatlantic.com/national/archive/2013/01/the-cas...
- mickronome 9y ago
  It's also interesting to note that children in Finland got a shorter school day, and mostly no homework at all. The shortening of the school day was especially prominent in for the youngest children.
chimeracoder 9y ago
> That's essentially true for most sales people and traders as well. It's definitely not true in most sales positions, particularly the ones that are highly commissions-based. > But imagine how the teaching profession would change if 1) the gains could be accurately measured and 2) we assigned payment or prestige to teachers based on (1). We can, and we do. The fact is: the gains just aren't that large. If you want to improve student performance, the best place to spend the money isn't by raising teacher salaries.

lr4444lr 9y ago

Disagree. The tails of the distribution are where the teacher effect is probably the least, because there are significant obstacles on the low end that make focusing on learning impossible and immense innate talent at the high end where the kid would learn if you put him in a closet with the textbook and a flashlight. The real problem with teacher success metrics is that they are outcome-based. It is no more useful to measure the success of a world class surgeon by including the outcomes of people who completely ignore the post-op care directions and instructions for lifestyle changes.