FAQ – Direct Instruction

This series of posts tackles some common issues that keep arising as I write about education.

Question

You have described yourself as an advocate of ‘explicit instruction’. Is this the same thing as ‘direct instruction’ and, if so, why not use that term?

Answer

Discussing definitions is boring but sometimes necessary. Unfortunately, the term ‘direct instruction’ is ambiguous. I am going to follow Barak Rosenshine here in the way that I deal with this issue but, even then, it is worth noting that Zig Engelmann vehemently disagrees with Rosenshine. Basically, the term ‘direct instruction’ is fraught.

Rosenshine identifies no less than five meanings of ‘direct instruction’.

1. Academic instruction that is led by a teacher regardless of the quality of instruction.

2. The instructional procedures that were used by effective teachers in the teacher effects research.

3. Instructional procedures used by teachers when they taught cognitive strategies to students.

4. Instructional procedures used in the Distar (Direct Instruction Systems in Arithmetic and Reading) programs.

5. Instruction where direct instruction is portrayed in negative terms such as settings where the teacher lectures and the students sit passively.

The teacher effects research that Rosenshine identifies in Type 2 is the process-product research conducted from roughly the 1950s through to the end of the 1970s. Many people are unaware of this research because it seems to have gone out of fashion. Rosenshine has written an accessible article here that describes the features of this kind of instruction.

It is broadly this kind of direct instruction that I am referring to when I advocate ‘explicit instruction’. However, I also sometimes extend the meaning into Type 1, particularly when discussing the kind of large-scale correlational study described in this paper.

Type 3 is a specific kind of direct instruction aimed at teaching things like reading comprehension strategies. Although I am sceptical about the benefits of spending extended time on such instruction (see Willingham), I think this demonstrates that, even when the learning objective is somehow ‘higher order’, direct and explicit approaches are superior.

I do not directly advocate Type 4 Distar programs. However, I view these as a particular form of explicit instruction. I am intrigued by their adoption in Cape York and by some of the unique features of these programs – such as the scripting of lessons – and I think that, through Project Follow Through, they provide empirical evidence to support explicit instruction more generally.

Type 5 is the sort of direct instruction that people set-up in experiments when they want to prove that it’s ineffective. Typically, this kind of direct instruction is non-interactive. The problem here is that there are no mechanisms to ensure student attention. Some commentators tend to refer me to studies where traditional university lectures are compared to ‘active’ learning – which is often something as simple as adding the use of ‘clickers’ to the lectures – and then claim this is evidence against direct instruction and/or in favour of constructivism. It is worth noting that, in contrast, Type 2, 3 and 4 direct instruction are all highly interactive.

A good example of Type 5 direct instruction may be found in this paper.