Building a Better Teacher, by Elizabeth Green. p. 281:
infrastructure had three elements: a common curriculum suggesting what students should study; common examinations to test how much of that curriculum they learned; and finally, teacher education to help teachers learn to teach exactly what students are supposed to learn.
She argues that
1. Good teachers make a difference.
2. Teaching itself is a skill that can be taught.
I remain skeptical on both points. On (1), why do researchers like Heckman consistently find support for what I call the null hypothesis, which is that no educational interventions make a large, reliable, long-term difference?
On (2), suppose that there are 50 habits that a great teacher has, and each of these habits can only be learned with intensive practice and immediate feedback. Suppose that it takes two months to learn each habit. If a natural teacher starts with 40 of these habits, it will be a lot less costly to train that teacher than to train a teacher that starts out with just 5 of these habits.
As the author pointed out in a live talk at a local bookstore, there are inevitable tensions in the teaching process. When some students get a concept and others do not, when do you move on?
Also, students respond to a teacher’s authenticity and love. How much rote technique can a teacher use before you lose that?
Montgomery County, Maryland, where I live, smothers its teachers in the common curriculum and common examinations components of infrastructure. The result is that teachers feel stifled by the requirement to be on lesson x on day y. I would add that whenever I have looked at the data, Montgomery County test scores are mediocre. The county spends much more per pupil than other counties in the state, but its test scores are in the middle of the pack. One consequence of the infrastructure is that the student-teacher ratio is high even though the student-staff ratio is low. Actual classroom teachers work very long days and have very little time to receive and reflect on feedback.
I would note that higher education in America has even less of the infrastructure components than does K-12 education, yet higher education is said by some to work well here.
The strength of the book is that it gives us a picture of what better teaching looks like. The author’s descriptions of quality lessons and of schools that develop and guide their teachers are inspiring. If she is correct, and what works idiosyncratically can be made to work systematically, then reading the book would motivate educational leaders to try.
A lot hinges on what you consider an “intervention”.
If you limit yourself to experiments in a public school system, then a lot of them are driven by researchers who frankly have very limited experience and expertise about education. They have some research topic they are peddling, and they obtain funding to do an experiment. When the funding goes away, so does the experiment, and people write some papers and move on.
Regarding teachers, I find it very hard to believe they don’t matter. I can believe they are hard to *detect*. I’ve seen teachers, though, where the students aren’t even listening to what they say for the entire hour–they spend the whole time doing homework for another class, or passing notes to each other.
Your objection to point 1) doesn’t, in fact, address point 1) – it’s closer to addressing point 2).
Good teachers DO make a difference – Jaime Escalante being one of the more famous examples, and anecdotally I am familiar with a school where top performers through the junior school can be traced back to a single early teacher.
And, it’s pretty clear that educational interventions DO make a large, reliable, long-term difference. I did not study Italian in high school, whereas some of my peers did. Today, most if not all of my peers who studied Italian retain a considerably greater familiarity with Italian than I have (which is practically none). Teaching Italian is an educational intervention that makes a large, reliable, long-term difference in the knowledgebase of the students who study it.
So, average teaching IS more effective than no teaching, and some teachers ARE notoriously more effective than other teachers. Why is it, then, that academic studies fail to violate the null hypothesis? Why does no academically-tested intervention appear to systematically improve the quality of teaching?
I don’t know, but I can propose an alternative hypothesis. Let us separate the differences between good teachers and bad teachers into two kinds: those that represent teaching techniques that can be reliably trained; and those that require personal attributes that are difficult – maybe impossible – to train. Let us then hypothesize that any trainable attribute has already been widely trained, possibly because teachers voluntarily adopt new techniques that work.
That’s not to say that teachers will always adopt new techniques that work for some students – because those techniques may not work for other students. And, teachers may not adopt techniques that are difficult or impossible to train – techniques that require, for example, unusually strong social skills (most of us have met teachers who have an uncanny ability to “read” students).
But the fairly simple hypothesis is that if a technique works for most students, and is reasonably trainable, then it will be quickly and widely adopted. The only interventions then likely to be available for educational researchers to study will be those that either don’t work (at least, not for most students) or can’t be trained (which may be why we don’t see more Escalantes). And, in that circumstance, those educational studies will almost always prove consistent with the null hypothesis.
This was interesting re: effective teaching and measurement of same.
http://www.texasobserver.org/walter-stroup-standardized-testing-pearson/
Good teachers, like good coaches, are the ones who recruit the best students/players.