"No factor is more important to the quality of education than the quality of the teacher."

We know the above statement is blatantly false. It's not even a close call.

Studies are clear. About 80% of student school performance is predicted by factors outside of the school, such as the student him/herself, home, family income, parents, neighborhood, society in general. About 13% of student school performance is predicted by the teachers, such as experience, training, discipline skills. The remaining 7% of student school performance is predicted by the school itself, such as the facilities, class size, textbooks.

This is not the same as saying that changes in quality of teachers and schools can't have profound effects. If there is little substantive difference in the curricula across all schools in the US, then one should expect the little variation in student performance attributed to schools, stated above as 7%. Another way of stating this is, if we we allow or force more diversity in the curricula, rather than using say, the same 4 textbook series across all schools, we would likely see an increase in variation of student performance attributed to the school increase from the 7%, and a concomitant decrease in variation due to the student and teacher.

The author refers to the "quality of education", not "quality of performance" or overall outcome as Larry states. There is a big difference between the two, so I don't think Larry's changing of the authors meaning makes the author's premise, "blatantly false." Had the author said, "quality of overall student achievement or performance, then I would have to agree with Larry. But she didn't.
I do applaud Larry, though, for stating that only 13% of a child's school performance is based on the quality of the teacher. If true, it is a powerful reason to abolish teacher unions and the entire DPI teacher accreditation process. ie., if teacher quality has such a minimal effect, then a good business person should be able to teach business, econ and perhaps math. A fit and active person in good health should be able to teach phy ed. An accomplished musician should be able to teach music and so on. All without being required to attend school and get a teaching degree. Even if this were to result in a loss of 30% effectiveness, it would still only be 30% of 13% or a net loss of performance of only 3.9%. This could easily be made up by going after the remaining variables such as the 7% school effect and 80% rotten parents and society effect. Far cheaper too.

Okay, Reed. I give. What is the difference between quality of education and performance of students, and what metrics are used to measure each?

And, further Reed jumps to unjustified conclusions regarding teacher impact. Of course, Reed has never needed a reason to his diatribes against teachers. It's like hearing a broken record again and again.

Instead of teacher and students, say dentists and patients. It doesn't matter how good dentists are, if patients do not brush or floss, eat a lousy diet, do not have access to fluoridated water, has bad genes, play rugby without teeth guards, don't get regular dental checkups, etc, then the dental outcome for the patients will not be good. Not the dentists' fault!

If the hygienists and dentists are unionized? Doesn't matter. Breaking up their union will make Reed feel good, but of course, it will not better the outcome for the patients. But, of course, Reed actually doesn't care, because his agenda has nothing to do with patient (student) outcome.

As another example, if you have pancreatic cancer, it doesn't matter one whit whether your doctor is good, or bad, or even unlicensed, you're still going to die within 3 to 6 months. In this case, regardless of how variable doctor quality is, it can have no effect on patient outcome.

Getting back to teachers again. If teacher quality is generally uniform in quality, say good, with generally a few who are not adequate (say either burned out or a newbie), then one should not expect much variation on student outcome based on teacher quality.

You likely get the same results if the teachers are mostly uniformly inadequate.

That is, there needs to be significant variation on the teacher side for teacher quality to have a chance to predict student outcome variation, when there is significant student variation.

What about the musician example? I looked up Gallaudet University and found that they do not offer a course much less a major in music. Fancy that. Somehow, I don't think a good musician would be able to teach music at Gallaudet.

Generally, Reed, your views simply have no merit. It would be useful if you even had a child's understanding of probability and statistics. You seem to be quite innumerate.

Well, there you go again. I'll speak slower this time.
1) Quality of education = What they are served up in the schools. How it is delivered, by whom and in what kind of school environment.
2) Quality of performance = The end result when the other factors such as society and rotten parents are taken into consideration.
3) Re my unjustified conclusion: You started off your second paragraph with..."Studies are clear". Sorry Larry, that just don't work here.
4) Pancreatic cancer: Jobs hung on for over 5 years.
5) Gallaudet University: We're talking about high schools here Larry. Stay on topic please.
6) My "diatribe" against teachers: Larry, you were the one who said their effect on student performance was minimal. You demeaned them. I have more respect for teachers than you and have never criticized "teachers." Only admins and others who force them to teach junk.
7) My poor understanding of prob and stat: I tested out of calculus undergrad and took 300 & 400 level calculus classes. Also, several grad level stat classes at UW-Madison. Careful who who call innumerate. My analysis of your 7, 13, 80 example was spot on. Pure math and logic seems to discombobulate you.
8) And as usual, you top everything off with an ad hominem attack. None, btw, in any of my posts.

Quality of education is just a sequence of words. The author says that whatever it is, it is important for teacher success, and student success. Sorry, but I don't think it is relevant to measure teacher success by anything other than student success, which cannot be measured in any other way but by student performance. One can measure education quality by teacher pay, teacher experience in years, teacher professional development, etc. or some checklist filled out by principals observing the teacher in the classroom. But, so what. Unless what we are measuring are good predictors of student learning, they are a waste of time and effort. If resources need to be allocated to improve student learning, they need to be allocated to areas which will have the most success.

We can start clearly with the agreement that student success is unacceptably low and highly variable by any number of measures: state tests, reports from colleges, ACT and SAT scores, international comparisons. These are the outcomes we want to improve; are there really any others that are more relevant? The 13% for teacher quality says the measures being used to evaluate teacher quality do not measure those characteristics of teachers, assuming teachers are important, which would allow us to allocate resources to teachers to become more effective.

Or, the 13% says teachers, like dentists, are important, and are doing the best that can be done under the circumstances, but teacher accountability measures, and other demonizing of teachers will not make the results much better, and one must look elsewhere for significant outcome improvements.

Where do we look? Well, we don't look at public vs private vs charter vs online schools vs school choice vs vouchers, because these differences only account for 7%.

Well, how about the 80% allocated to the student and the environment generally? We can try kindergarten. Maybe some social programs. But that is about it for interventions on which the educational establishment can have an impact. You can blame the student a la The Bell Curve, a convenient and often used excuse. Or you can look at poverty, housing, homelessness, crime, general lack of intelligent and considered positions of "leaders", the pervasive acceptance of bullying and cruelty in public discourse, the thorough acceptance of new and old age myths.

So, why not look at the 80%? Because a discussion and solution in the 80% area gets awfully close to blaming us, rather than them.