The scalability of academic success

In a preview of his upcoming book, We are Doomed: Reclaiming Conservative Pessimism, John Derbyshire writes:

As the Thernstroms point out, a lot of these prescriptions for school reform assume an unlimited supply of “saints and masochists” — teachers like those in the KIPPS schools, who, Mr. Tough tells us, work 15 to 16 hours a day. I am sure there are some people who enter the teaching profession with the desire to crunch their way daily across the crack-vial-littered streets of crime-wrecked inner-city neighborhoods in order to put in 15-hour working days, but I doubt there are many.

That’s ed theorists for you. They love to talk about how the top 1 percent of superbly excellent and inspired teachers can lift up the bottom 5 percent of students. That’s interesting in its own way, but not very important as an issue in education. An example of an important issue would be: What can the average or mediocre teacher do for the average or mediocre student? Why is that question never asked? Because here in the Republic of Happy Talk, there is no such thing as an average — let alone mediocre! — student. “Given the opportunity, most people could do most anything.”

Derbyshire is correct. It's easy to point to a successful teacher (e.g. Jaime Escalante), school (KIPP academies), or movement (charter schools) and conclude that the solution to education is to replicate that success on a mass scale. But rarely do we stop and consider whether these things are realistically scalable.

This is what I was getting at yesterday when I posted on the $125,000 teacher experiment. Suppose the experiment is a smashing success. Then what? The reality is that there is limited number of bright, dedicated teachers, and there is certain a limit to how much we afford to spend on them. There are other practical considerations as well, like the enormous resistance from teachers unions for meaningful reform. The charter school movement has hardly taken off, and already the pressure is on to unionize them and force them back into a place that looks very much like where they came from.

So what's the solution? Derbyshire doesn't hint at one, and I admit that I'm also at a loss.

Lately, I've been doing a lot of reading at an education blog called D-Ed Reckoning. KDeRosa, the main blogger there, has a lot of strong and interesting opinions on all of this. From what I gather so far, his (her?) view is that it all comes down to proper instruction techniques. All the other things we hear about, says he -- parental involvement, socio-economic status, class size, teacher dedication, student responsibility, etc. -- play a part, but pale in comparison to whether we are using reliably effective teaching techniques.

In particular, KDeRosa is partial to a program called Direct Instruction, which professes to outperform other approaches in academic studies. Unfortunately, I don't know enough to tell a good study from a bad study, but I'm cautiously impressed by what I've read about DI so far.

If I'm understanding KDeRosa, he believes that DI doesn't require elite teachers and highly dedicated students in an optimal environment. That's intriguing, because if we can get significant results simply by putting proven instruction methods in the hands of average teachers who are teaching average kids, we've solve the scalability problem.

Too good to be true? Yeah, I'm wondering that too.