Fourteen Percent and a Change of Direction
Following on from previous posts about sharing lessons (which were actually more about discoverability), this post is yet another re-think, and is based in part on a conversation with François Michonneau.
It’s pretty clear at this point that most people aren’t going to teach someone else’s lesson exactly as written, but are instead going to adapt it for their audience and splice in material of their own. They might keep the explanations but use a different data set, replace exercises with ones they’ve written themselves, reverse the order of a few lectures, or remove some chunks entirely.
I believe adapters will be able to do this faster and better if they understand why a lesson was written the way it was, but that knowledge is usually unwritten. The Carpentries lesson on the Unix shell embodies a lot of experience with topics and orders and examples that didn’t work, or worked less well than what’s in there now. If someone picks it up and starts making changes, there’s a good chance they’ll repeat mistakes we made and learned from years ago, or break a dependency without realizing it until they’re in front of a class (because there’s no equivalent of regression tests for lesson content).
So what if we didn’t share just the lesson (for some value of “just”)? And what if we didn’t share it as if people were going to re-use it as is? What if the unit of sharing was a description of how to build a lesson that met some need? More specifically, what if what we shared described the thought process used to create a lesson on topic X, with one such lesson as a worked example?
- As it happens,
we already teach people a systematic lesson design process
that produces most of this:
- Write or recycle learner personas to figure out who you’re trying to help what they think they want to learn, and how you’re going to help them.
- Brainstorm (e.g., draw a concept map or two).
- Write the summative assessment(s) that will tell you and your learners if the lesson worked.
- Write the formative assessments that will tell you and them if they’re on track.
- Order those.
- Write the bits of lesson that connect them.
- What we don’t do is capture that information, but we could. While some of the instructors I was working with eight months ago pushed back against designing a lesson in such a structured way, having those design docs would make it much easier for someone to figure out where their needs or tastes differed from those of the original author, which I think would help them make changes more quickly and more accurately.
The downside, of course, is that this would require more work from the original author. However, I think that work pays off pretty quickly: I’m certainly building materials better and faster now that I’ve got a process, though I admit that could be due to practice effects rather than the method I’m using. As a next step, I’d like to know if anyone has ever studied the impact that lesson design documentation has on lesson uptake and adaption. If you know of anything along these lines, I’d be grateful for an email.