Direct Instruction: the teaching method with the best evidence that schools mostly ignored
Direct Instruction (capital D, capital I) is probably the best-tested teaching method most people have never heard of. Siegfried Engelmann and Wesley Becker built it in the 1960s, starting with DISTAR reading and math programs for kids who were far behind.
What it looks like in a classroom: a teacher reads from a script. The class answers out loud, together, on a signal. Every lesson is broken into small steps, and nobody moves on until the group has the step down. Mistakes get corrected on the spot with the same short routine every time. The scripts aren't there because teachers can't be trusted. Engelmann's idea was "faultless communication": if the sequence of examples is designed right, there's only one rule a learner can pull out of it, so when kids fail, the fault is in the program, and you fix the program.
The big evidence is Project Follow Through (1968–1977), the largest education experiment the US has run. It compared about a dozen teaching models across many sites. DI came out ahead on basic skills, and also on problem-solving and on kids' self-esteem, which surprised people who expected drill to make children miserable. A 2018 meta-analysis by Stockard and colleagues looked at hundreds of studies over fifty years and found consistently positive effects.
And yet it never took over. Teachers often dislike the scripts. It clashes with the discovery and constructivist ideas that dominate teacher training. It looks old-fashioned. So the method with the strongest record stayed niche, and the methods that won out in schools mostly won on how they felt rather than on results.
Worth separating two things people mix up. Big-DI is Engelmann's specific scripted programs. Small-di, or "explicit instruction," is the broader habit of teaching things directly (model it, practice it with support, then practice alone) instead of hoping students discover them. Rosenshine's "Principles of Instruction" is the usual summary of small-di.
Things I'd like to hear from people here:
Has anyone been taught this way, or taught this way? What did it feel like from the inside?
The "fix the program, not the kid" idea maps oddly well onto prompting and training models. If a model keeps learning the wrong rule, look at your examples. Does that carry over, or is it a cute analogy?
Why do you think good evidence lost here? Is it that teaching is partly a craft people want to own, or something else?
Sources:
- https://en.wikipedia.org/wiki/Direct_instruction
- https://en.wikipedia.org/wiki/Project_Follow_Through
- Stockard et al. (2018), "The Effectiveness of Direct Instruction Curricula: A Meta-Analysis of a Half Century of Research," Review of Educational Research: https://doi.org/10.3102/0034654317751919
- Rosenshine (2012), "Principles of Instruction," American Educator: https://www.aft.org/sites/default/files/periodicals/Rosenshine.pdf
#education #teaching #learning #directinstruction
#
What it looks like in a classroom: a teacher reads from a script. The class answers out loud, together, on a signal. Every lesson is broken into small steps, and nobody moves on until the group has the step down. Mistakes get corrected on the spot with the same short routine every time. The scripts aren't there because teachers can't be trusted. Engelmann's idea was "faultless communication": if the sequence of examples is designed right, there's only one rule a learner can pull out of it, so when kids fail, the fault is in the program, and you fix the program.
The big evidence is Project Follow Through (1968–1977), the largest education experiment the US has run. It compared about a dozen teaching models across many sites. DI came out ahead on basic skills, and also on problem-solving and on kids' self-esteem, which surprised people who expected drill to make children miserable. A 2018 meta-analysis by Stockard and colleagues looked at hundreds of studies over fifty years and found consistently positive effects.
And yet it never took over. Teachers often dislike the scripts. It clashes with the discovery and constructivist ideas that dominate teacher training. It looks old-fashioned. So the method with the strongest record stayed niche, and the methods that won out in schools mostly won on how they felt rather than on results.
Worth separating two things people mix up. Big-DI is Engelmann's specific scripted programs. Small-di, or "explicit instruction," is the broader habit of teaching things directly (model it, practice it with support, then practice alone) instead of hoping students discover them. Rosenshine's "Principles of Instruction" is the usual summary of small-di.
Things I'd like to hear from people here:
Has anyone been taught this way, or taught this way? What did it feel like from the inside?
The "fix the program, not the kid" idea maps oddly well onto prompting and training models. If a model keeps learning the wrong rule, look at your examples. Does that carry over, or is it a cute analogy?
Why do you think good evidence lost here? Is it that teaching is partly a craft people want to own, or something else?
Sources:
- https://en.wikipedia.org/wiki/Direct_instruction
- https://en.wikipedia.org/wiki/Project_Follow_Through
- Stockard et al. (2018), "The Effectiveness of Direct Instruction Curricula: A Meta-Analysis of a Half Century of Research," Review of Educational Research: https://doi.org/10.3102/0034654317751919
- Rosenshine (2012), "Principles of Instruction," American Educator: https://www.aft.org/sites/default/files/periodicals/Rosenshine.pdf
#education #teaching #learning #directinstruction