It must be terribly constraining to use explicit instruction, right? You can’t just throw in innovative activities or cool problems because you have to follow principles such as the gradual release of control. Barak Rosenshine clearly articulates these constraints in this article.
Instead, we might wish for more freedom.
Do other teaching methods afford this freedom? Problem based learning, project based learning, inquiry learning, the maker movement and other constructivist approaches seem promising.
But perhaps we should be careful what we wish for.
Firstly, constructivist methods constrain the use of ‘telling’ – which is the pejorative term that is often deployed to mean ‘explaining things’. Either we don’t ‘tell’, we limit ‘telling’ or we give it a role after problem solving.
And there are also constraints on content.
If you want your lessons to be effective then you probably need to limit constructivist instruction to areas where students have lots of prior knowledge. Not all advocates will state this as a prerequisite but it does seem consistent with the cognitive science.
Proponents of constructivism will, however, offer other constraints. Perhaps they might suggest topics for instruction must be authentic, relevant and/or real-world. That’s pretty limiting, especially in maths education. Perhaps they will argue that topics must invoke genuine curiosity and questions must engender ‘perplexity‘. It might be hard to teach a standard course in this way so perhaps the curriculum should change. I have argued before that it’s relatively easy to think of funky questions but not ones that will also deliver an entire, systematic program of study.
What if you want to try ‘productive failure’? This is the current favourite constructivist method. It’s constructivism-lite. Following a period of open-ended problem solving, students are given explicit instruction. It seems to have the strongest evidence of any constructivist approach although I would question how some of this research has been done and the controls that have been used.
But here’s the thing. Manu Kapur, chief productive failure researcher, sets out the following constraints:
(a) The initial problem-solving task should be challenging enough to engage the learner in the exploration, but not so challenging that the learner gives up;
(b) it must admit multiple solutions, strategies, and representations, that is, afford sufficient problem and solution spaces for exploration;
(c) the problem should activate learner’s prior knowledge— formal as well as intuitive—to solve the problem;
(d) a teacher or an expert should build upon the student-gener- ated solutions by comparing and contrasting them with the correct solution, thereby directing attention to and aiding encoding of the critical features of the targeted concept.
Some of these seem hard to define and might even lead to the idea being unfalsifiable i.e. if it works then it met these conditions and if it didn’t then prior knowledge wasn’t sufficiently activated or the task was too challenging.
So what of explicit instruction? It turns out that whereas it constrains methods to some degree, it may not constrain content at all.
As Rosenshine points out, explicit instruction seems to be the best approach to teach pretty much anything, including rarefied and potentially reified concepts such as reading comprehension and problem solving.
So perhaps explicit instruction offers the most freedom for teachers. Who would have thought it?