Monday, June 2, 2014

Direct Instruction Works, Sort Of

Direct Instruction works, it seems. See Raymond, recapping the latest, greatest chapter of From the Ivory Tower to the Schoolhouse:
Unlike Bloom's Taxonomy, multiple intelligences, and the project method, Englemann's Direct Instruction works (with the research to show it) when teachers are philosophically compatible with the method and they implement it with fidelity. 
Direct Instruction clashes directly with a lot of the teaching principles that I hold dear. It's a scripted curriculum. Everything that the teacher does follows the script, and that's a direct attack on the idea that classroom ought to involve much intellectual work. Englemann's program doles out rewards and punishments to reinforce behaviors in students, and not in the "aren't grades really a reward?" sense but in the real sense of using carrots and sticks to drive learning.

It shouldn't work. It works.

For a nice, balanced take on the implications of Direct Instruction for policy, for teaching, and for progressive teaching notions, I'll send you back to Jack Schneider for "Direct Instruction works. And I’d never send my own child to a school that uses it."

So maybe the effectiveness of Direct Instruction draws blood from progressive teaching principles and maybe it's compatible in some way or maybe it bla bla bla but Raymond cuts to the heart of things when he asks: what happens when there truly is a conflict? What happens when there is a major clash between the teaching principles that I believe most strongly and what research shows to be effective?
But where do we draw the line between philosophical compatibility and the need for teachers to be open minded? To be learners? As professionals, when should our philosophies give way to what we can gain from research, regardless of compatibility?
It seems to me that a major goal of Schneider's book is to teach researchers not to ask that question. The beliefs and values of teachers can't easily be changed, and to make an impact research needs to take this for granted. Should teachers be willing to change their core values in the face of compelling evidence? Absolutely. Will we? That doesn't seem likely.

Despite what the evidence may show, Direct Instruction can, at best, only sort of work because it's only sort of compatible with the profession's philosophical commitments. A truly effective idea that makes all involved want to throw up isn't "truly effective" in any interesting sense. (On the other hand, Direct Instruction only makes some people throw up, and it otherwise has a positive effect so that counts as progress.)

At the end of all this, we end up with a host of pragmatic questions for researchers to consider. Does my research idea attack core values of teachers? Do I think that I can change the core values of teaching? Do I think I can have a limited impact, even without becoming widely accepted? What's the path to impact? 

I feel compelled to close by sharing my excitement at better understanding an idea that I had only previously mocked. Direct Instruction comes out looking much better than Gardner's multiple intelligences in this story. DI has a strong research basis and can reasonably claim to have made a positive impact on schooling. The case for MI is much sketchier. If there's a moral here, it goes something like this: familiar ideas need the most scrutiny, but uncomfortable ideas need a fair shot. Scrutiny too, of course, but uncomfortable ideas are going to get that anyway.

Raymond Johnson and I are reading Jack Schneider's new book, "From the Ivory Tower to the Schoolhouse: How Scholarship Becomes Common Knowledge in Education." This was Chapter Four, in response to Raymond's postPreviously:

Chapter 1: Bloom's Taxonomy (Michael's post, Raymond's reply)
Chapter 2: Multiple Intelligences (Raymond's post, Michael's reply)
Chapter 3: The Project Method (Michael's post, Raymond's reply)


  1. Direct Instruction is certainly "better" than Multiple Intelligences, in that one is an instruction technique and the other is an idea.

    But that video is not terribly convincing. I mean, seriously. Show me they know it in a month.

    1. Yeah, nothing in this post supports my claim that there's evidence for Direct Instruction. I have no desire to make that case here, but Raymond links to a study in his post, and Schneider's chapter traces the empirical record of DI.

      I mostly shared the video because I found it interesting.