Let’s start by defining ‘explicit instruction’ as highly structured, interactive teaching where students are explicitly taught everything that they need before being asked to apply it in gradually wider and freer contexts. It can also be understood as the I-We-You model where responsibility is gradually shifted from teacher to student. In even simpler terms: tell them stuff, ask them questions about it and ask them to apply it.
To some, this approach can seem ill-suited to the more complex, higher-order aspects of learning (in English, these are the things like creative writing, essay writing and textual analysis) and it may seem that explicit instruction is more suited to the teaching of basic, fundamental concepts and skills like sentence constructions or vocabulary. There is a certain allure to this mode of thinking: the restrictive nature of explicit instruction seems to clang against the desired features of freer and open-ended tasks. We want students to write with flair, to make perceptive and nuanced arguments and to craft beautiful, creative pieces of work that shimmer with originality and imaginative ideas. Will excessive structure stifle this process? Is explicit instruction the wrong approach here? Should we instead adopt a less structured approach that is more aligned with discovery and inquiry?
I would argue that we are making a form of category error here. The nature of the task or content should not be the determining factor as to which broad instructional approach to use; instead, we should choose our instructional approach based upon the level of expertise of the student.
Let’s look at some evidence in support of the idea that explicit instruction is entirely suitable for complex, higher-order tasks:
In the 1970s, researchers set out to ascertain what it was that made teachers effective. They visited classrooms in an attempt to draw correlations between the teacher’s actions and the resulting academic outcomes. The findings were collated by Brophy and Good in this paper and the now ubiquitous Principles of Instruction is perhaps the most famous summary of this research. The ‘Principles of Instruction’ are seen by many as a list of common, explicit teaching strategies. When summing up their findings towards the end of the paper, Brophy and Good explain:
‘At least two common themes cut across the findings, despite the need for limitations and qualifications. One is that academic learning is influenced by the amount of time that students spend engaged in appropriate academic tasks.’
Perhaps unsurprisingly, we should consistently deliver efficient lessons where students spend maximal time thinking about content. Open ended group work can make this difficult to achieve as the potential for-off task behaviour is greater than classic whole class teaching.
The second is that students learn more efficiently when their teachers first structure new information for them and then help them relate it to what they already know, then monitor performance and provide corrective feedback during recitation, drill, practice or application activities.
This statement seems to implicitly describe how to ensure students engage in Fiorelli and Mayer’s three stages of cognitive processing from Learning as a Generative Activity:
- Select information to attend to.
- Organise the material into a coherent cognitive structure in working memory
- Integrating it with relevant prior knowledge activated from long term memory
In the absence of explicit instruction, a possible concern is that students will fail to ‘select’ the correct information, effectively precluding them from engaging in the second two elements of cognitive processing.
Brophy and Good then continue:
‘For a time, these generalisations seemed confined to the early grades or to basic rather than more advanced skills. However, it now appears that they apply to any body of knowledge or set of skills that has sufficiently well organized and analysed so that it can be presented (explained, modelled) systematically and then practiced or applied during activities that call for student performance that can be evaluated for quality and (when incorrect or imperfect) given corrective feedback.’
So can explicit instruction be used to teach creative writing or analytical essays? If the statement above is to be believed, then yes it can. Extended writing can be explained, modelled and practiced; it can also be evaluated and feedback can be given. In fact, I’m struggling to think of a skill-whether complex or not-that falls outside of this description. Is this because the description is so wide and vague that it is meaningless? Perhaps it is because ‘explicit instruction’ includes so many broad principles, many of which would also be claimed by pedagogical approaches that see themselves in opposition to this school of thought. Perhaps when we say ‘explicit instruction’, it is understood by some in the pejorative sense of being a robotic and excessively didactic approach, instead of an approach that contains many elements of ‘common sense teaching’ and things that most good teachers probably do, whether they proudly label their practice with a specific pedagogical flavour or swear no allegiance to a particular school of thought? All of these questions are worth asking: concepts in education can be nebulous and professionals often seem to be talking past each other, lost in a fog of indeterminate signifiers.
Despite the conceptual wooliness and obvious overlap between supposedly separate pedagogical approaches, the issue remains that the complexity of the task should not be the defining consideration when choosing an instructional approach. If instead, we accept that an instructional approach should be chosen based upon the level of expertise of the learner, then explicit instruction is suitable for teaching complex tasks like extended writing.
Research into Cognitive Strategies
Successful extended writing is difficult to define and proficiency will take many guises. An essay will be made up of many different sub-components that, when coherently combined, will result in the perceived proficiency of the final piece. Added to this, judgements of quality will be inherently subjective; after all, the aesthetic and personal nature of writing is what makes it interesting and enjoyable. So does this mean that we cannot explicitly teach extended writing? If there are myriad ways of writing a good essay, then how are we to help students get better at this important skill?
One solution is through ‘cognitive strategies’
In ‘The Use of Scaffolds for Teaching Higher-Level Cognitive Strategies’, cognitive strategies are defined as being ‘more like supports or suggestions than actual step-by-step directives.’
Here’s a summary of how to teach these:
These suggestions share many of the conceptions mentioned by Brophy and Good (modelling, practice etc), and this is unsurprising given the fact that cognitive strategies research is one of the three sources of research that underpin Rosenshine’s Principles of Instruction. You will notice that independent practice is the final stage here, preceded by a gradual shift from teacher to student, and through a clear process of backwards fading, from models to completion problems to the removal of all support when they finally do it for themselves. Again, this is I-We-You.
In the final section of ‘‘The Use of Scaffolds for Teaching Higher-Level Cognitive Strategies’, Rosenshine and Meister explain:
‘Such concepts as modelling, thinking aloud, using cue cards, anticipating errors, and providing expert models, can also be applied to the teaching of well-structured skills’
So cognitive strategies can be used for both basic and more complex skills, lending further support to the idea that the type of content should not be the determining factor in choosing between direct instruction or inquiry methods.
They then continue:
‘This suggests that instead of a dichotomy, there is a continuum from well-structured, explicit skills to cognitive strategies. At all points in the continuum, some instructional practices, such as presenting information in small steps and providing guided practice, are important. Yet, as one moves from well-structured skills to cognitive strategies, the value of providing students with scaffolds-models, concrete prompts, think-alouds, simplified problems, suggestions and hints-increases.
‘The tools that we refer to as scaffolds are at a middle level of specificity. That is they provide support for the student, but they do not specify each and every step to be taken. There is something appealing about this middle level. It lies somewhere beteween the specificity of behavioural objectives that seemed overly demanding to some, and the lack of instruction that many criticized in discovery learning settings. Perhaps it is the beginning of a new synthesis.
And this is where we return to the problem of fuzzy concepts and talking past each other. For those who see direct/explicit instruction as being ‘behavioural objectives that seemed overly demanding’, scaffolding will probably still feature in their practice; equally, for those who see ‘the lack of instruction’ in ‘discovery learning’ as a problem, scaffolding will also almost certainly be a strategy that they use.
Cognitive Load Theory
According to Clark et al, ‘prior knowledge is the one individual difference that has been consistently shown to interact with different instructional methods’. The ‘Worked Example Effect’ and the ‘Expertise Reversal Effect’ would point to the fact that, irrespective of whether the content is simple or ill-structured, novice learners would benefit more from studying worked examples whereas experts would be better attempting to apply their knowledge by solving problems. Sweller et al maintain that ‘we should provide learners with as much relevant information as we are able’ and that ‘assisting learners to obtain needed information during problem solving should be beneficial’ as well as positing that ‘Providing them with that information directly and explicitly should be even more beneficial.’, statements that seem like explicit teaching to me!
So where does this leave us?
Instructional choices should be made based upon the expertise of the student, expertise here referring to their level of prior knowledge:
There is an important addition though. Prior knowledge here should not, however, be read as something akin to ‘general proficiency’. Let me explain what I mean. Even if a student is a phenomenal writer who can skillfully apply a wide range of sentence constructions, analytical components and different essay structures, if you want to teach them something new, the most efficient way of doing this will probably be through explicit instruction. While they may be proficient at lots of useful stuff, if they know nothing about Absolute Phrases, asking them to work them out for themselves or succeed with minimal guidance is probably not the right approach.
When you teach something new to someone, you should begin with explicit instruction. The complexity or structure of the task is probably irrelevant, the important variable being the level of expertise of the student in that specific thing.
In this sense, teaching reminds me of a fractal. If you zoom into an essay, whatever component you zoom into should have been initially taught via explicit instruction before support was gradually removed and students were asked to practice it independently. The structure of an efficient instructional sequence will look almost identical for teaching essay construction, paragraph composition, sentence creation or vocabulary acquisition: I-We-Yous all the way down.