Writing: The Case for Deliberate Practice

One of the main goals of teaching English is to help pupils become better at writing. Expert writers seem to be able to write with minimal effort. They write quickly, accurately and effectively, their sentences well-constructed, their ideas expressed with clarity, producing texts that appeal to the reader whilst also being suited perfectly to the task at hand. This is, however, no easy feat: expert writing involves the control of cognitive, perceptual and motor processes, each of which has become relatively effortless, most likely through extended and deliberate practice.

Deliberate practice can help to reduce the attentional demands of particular processes. Simple perceptual-motor processes can, through deliberate practice, become completely effortless. If you have ever trained in a particular sport, you will recognize this process of automisation. I play football and, as a result of sustained deliberate practice, do not have to think about the position of my feet, the angle of my leg or where my foot will meet the ball when kicking. Someone who has never kicked a ball before will need to concentrate on all of these things. The sheer number of things to consider (and we haven’t even talked about positioning or anything to do with other players) will mean that this may be a difficult task to execute proficiently and consistently. Writing, however, is not a simple perceptual-motor process and in my last post, I attempted to outline its complexity. Because of this complexity, effective writing instruction should seek to reduce rather than entirely eliminate the relative effort required for each of the components and processes involved.

Deliberate practice can do exactly that. By reducing the attentional demands on the components, it can free up pupil’s attention so that they can concentrate on the whole performance, allowing them to respond flexibly and effectively to the task at hand.

Real Life Case Studies

Isaac Asimov, one of the most distinguished and prolific writers of science fiction, wrote almost 500 books in a career that spanned more than 40 years. A study (Ohlsson 1992) explored Asimov’s writing career in an attempt to draw inferences regarding the effects of practice. The study split his oeuvre into groups of 100 books. Although the books would obviously vary in length and complexity, the researchers assumed that this would roughly average out. Asimov completed his first 100 books in 237 months; his second 100 books in 113 months; his third 100 books in 69 months and his fourth in 42 months. His writing sped up significantly as his career progressed and it seems reasonable to attribute this improvement to practice and increasing expertise.

A second study (Raskin 1936) found that scientists and authors produced their best work in their mid-30s, ten years or so after their first publications. In a study of poets (Wishbow 1998), researchers found that over 80% of 66 poets that were listed in the Norton Anthology of Poetry were first published ten years after starting reading and writing poetry. Both of these studies can also be seen as lending support to the idea that practice is beneficial.

Defining Practice

We are all familiar with the general meaning of practice, perhaps equating it to sustained effort or repetition, but what else is required to ensure that practice is maximally effective. In his brilliant book ‘Peak’, Anders Ericsson, the world’s leading expert on deliberate practice, explains that if practice is to be an effective means of developing expertise, it needs to ‘purposive’ or ideally ‘deliberate’. Let’s have a look at each one in turn.

Purposive Practice

  1. Well defined, specific goals

Teachers should choose goals that are suited to their student’s current level of expertise. If their pupils are struggling with transcription, then asking them produce extended pieces of writing will not be the best instructional choice. It will be more useful to build their transcriptional fluency, perhaps through systematic spelling instruction, dictation activities or short, timed writing practice.

A pupil’s level of writing expertise is unlikely to be the same across genres or different types of tasks. Expertise is often very narrow and highly domain specific. When pupils enter year 7, they are often pretty good at writing stories, having learned and practiced this text type a lot during primary school. Although they are adept at writing narratives, they are usually relative beginners or even complete novices at writing analytically. Because of this, it will probably not be a good idea to expect them to write essays that are of a comparable length to their creative pieces. Instead, instruction and practice should be focused at the sentence and paragraph level, building fluency with these components with the eventual aim of creating fluency in the wider, composite task of writing text response essays.

2. Putting a bunch of baby steps together to reach a longer term goal

Although the goal may be fluent and proficient extended compositions, the best way to achieve this is likely to involve practicing the components that build up the composite whole. Designing restrictive practice activities that focus on isolated components can be really helpful.

3. Focus and Attention

If pupils aren’t concentrating or paying attention, then it doesn’t matter how good we think we our lesson is, they won’t get better at writing. Mike Hobbis explains some of the research into attention in this post

4. Feedback

In Peak, Ericsson points out that ‘meaningful positive feedback is crucial for motivation……..this can be internal feedback such as the satisfaction of seeing yourself improve at something, or external feedback provided by others’

To prevent errors from becoming ingrained, in the initial stages of instruction, corrective feedback should be immediate. Effective feedback requires the giver to have a well-developed mental representation of what is being performed. If a task is vague, then this can makes things difficult. Feedback involves comparing these expert mental representations to the performance: the difference is what the student needs to do to improve.

5. Getting out of your comfort zone

Purposive practice should involve a level of challenge that is just above pupils’ current performance levels. It should be achievable yet challenging and this is often a hard balance to achieve. Rosenshine’s principles of Instruction, based upon the product-process research, would point to 80% as being an optimum success rate for pupils: on average, pupils should be getting 80% of the answers correct in a lesson. Too easy and pupils may lose interest; too hard and pupils may give up. They need to see that if they devote sustained attention to the task, that the goal is achievable. Asking pupils to write answers to GCSE exam questions in year 7 is unlikely to strike this balance.

Deliberate Practice

Deliberate practice requires that teachers ‘set practice activities designed to help pupils improve’. In order to create such activities, the teacher will need to follow a number of guidelines:

  1. Teachers must know what makes up expert writing

Deliberate practice is different from other forms of practice because it requires a fairly well established domain where ‘the best performers have attained a level of performance that clearly sets them apart from people who are just entering the field.’ Writing seems to fit this description. Although there will be huge variations in style, tone, register and form across expert writing, even within a specific genre or text type, expert writers will likely have utilized similar cognitive processes in order to create their pieces. Deftly juggling mental representations that span author, text and prospective readers, they will have recursively planned, composed and reviewed their writing. These processes can be isolated and practiced with students and these mental representations can be incrementally developed too, through extensive reading and studying models. Pupils should also be taught and asked to apply specific sentence structures or literary techniques.

How can this be achieved:

  • Practice, practice practice! Teachers should regularly be practicing writing themselves by completing the same tasks that pupils are asked to do. Almost every time pupils write, I write as well.
  • Reading, analyzing and discussing student work with colleagues

2. They must understand the relationship between components and wholes

Extended writing, like most other complex cognitive or physical skills, can be split up into different components, each of which can be taught and practiced in isolation before being combined into increasingly complex wholes. There are many advantages to this approach and it is no surprise that drills and restrictive practice tasks are standard in sports training and music teaching. Drills allow initial massed practice so that pupils, through immediate corrective feedback, are able to perform accurately. These drills can then be distributed over time so that pupils retain the skill and, once the desired level of accuracy has been reached, pupils can begin to work on their fluency by attempting to complete trials within a specific time limit. If pupils become fluent in these components, they are far more likely to be able to use them in wider writing.

Practising sentence structures on their own, however, is unlikely to be very successful. Pupils need to eventually use these components in their wider writing and an effective instructional sequence will gradually shift from restrictive practice to wider application. Something like this:

How can this be achieved:

  • Components should be chosen and sequenced based on their utility-anaphora may well be more useful to pupils than anastrophe.
  • Components should be chosen that are easily combined with others: specific high utility sentence structures are ideal here.
  • Particular attention should be paid to the shift from drills to wider application so that pupils are able to apply what they have practiced in extended writing. If not, pupils will not transfer their knowledge to wider writing. Transfer is notoriously difficult to achieve, although one method of achieving it is to make pupils  aware of the similarity between different contexts or tasks. This can be achieved through prompts, success criteria and asking pupils to underline the components that they have included in their writing.
  • Ensure that there is a suitable balance between practicing components and appropriate extended writing. Most of the time, classes should be doing both: drills to develop accuracy and fluency; extended writing to push pupils thinking about the content as well as to combine and apply fluent components. If pupils only complete drills, this will only help them get better at drills and they may not transfer this knowledge to extended writing. If pupils only do extended writing, this is unlikely to accelerate their progress as much as balancing the two.

If you are interested, many of the ideas in this post are based on ‘Training Advanced Writing Skills: The Case for Deliberate Practice‘ by Ronald T Kellogg and Alison T Whiteford.

Next Post: Reading and Writing: A Reciprocal Relationship


The Development of Writing Expertise: From Telling to Crafting.

Although there are lots of models for describing writing, many split it into three main cognitive processes: planning, writing (often called ‘translating’ in research) and reviewing. These processes should not be seen as sequential or separate and the act of writing will involve recursive interaction between all three:

But what does expertise mean in writing? How can we describe the development of expertise? One particularly influential model (Bereiter and Scardamalia 1987) splits the development of expertise in writing into three stages:

  1. Knowledge Telling
  2. Knowledge Transforming
  3. Knowledge Crafting

The diagram above attempts to describe how writers develop in expertise by focusing on the wider processes of writing (planning, translating and reviewing) as well as the mental representations that writer’s rely upon. Cognitive Load Theory and research into expertise would suggest that an important difference between experts and novices is the depth and breadth of their background knowledge. Experts have more accurate and comprehensive mental representations of whatever it is they are thinking about and are able to retrieve and apply these from their long term memories, thereby circumventing the seemingly unalterable limitations of working memory. If you look at the diagram above, you will see that this model of writing development is in agreement with these ideas.

Knowledge Telling

At this novice stage of writing, pupils are focused on retrieving what it is they want to say and then turning these ideas into a written form. Writing at this stage will be a simplistic restatement of a pupils’ thoughts. While they are unlikely to be focused on how an imagined reader may interact with what they have written, this is not to say that they are entirely solipsistic in their approach: even very young children recognise that other people think differently and have different perspectives.

At the ‘knowledge telling’ stage, it is assumed that the pupil is likely to have an impoverished mental representation of what the text they have written actually says as well as how a reader may interpret it.  Young children can sometimes struggle to understand what they have just written and this may be because when they are writing, they are almost entirely focused on the process of converting their thoughts into writing and not on how the text actually reads. At this stage, pupils’ transcription skills will be far from fluent: their handwriting may be laborious and their spelling may require sustained concentration and focus. Because pupils have to direct their attention to these component skills, they may be unable to concentrate fully on the meaning of what they have written.

A second assumption here is that it is impossible to focus on reader interpretation in the absence of fully comprehending what has been written. Additionally, because these mental representations are not stable or clear within a pupils’ mind, they are unable to use them when planning and reviewing and this explains the limited use of planning and reviewing by writers who are within this early stage.

Knowledge Transforming

This second stage involves a constant interaction between the writer and the text that they are writing, with the writer changing what they want to say as a result of the process of writing. As they read what they have already written, they will not only strengthen their mental representation of their writing, but they may also trigger additional planning or sentence generation. For example, if the writer was writing a paragraph about Macbeth’s dagger soliloquy, they may begin by writing about his introspective questioning; this may then help them to decide what would come next in their analysis.

At this stage, the act of writing has become a process of actively constituting knowledge. Whereas knowledge telling involves the simplistic translation of thought into text, proficient writing (and the ‘knowledge transforming’ stage could be seen as the beginnings of proficiency) often involves a process of discovery in that writers only discover exactly what it is they want to say during the actual process of writing.

Flower and Hayes (1980) describe the shift from’ knowledge telling’ to ‘knowledge transforming’ like this:

‘At one end of the spectrum, writers are merely trying to express a network of ideas already formed and available in memory; at the other, writers are consciously attempting to probe for analogues and contradictions, to form new concepts, and perhaps even to restructure their knowledge of the subject.’

Knowledge Crafting

This third stage represents true proficiency and expertise; writers who are operating at this level are constantly thinking about their readers and how what they have written will be interpreted. This will likely involve review and adjustment during the writing process, perhaps changing words or rephrasing sentences in order to maintain a specific tone or perspective. Writers operating at this level of expertise are more likely to make extensive structural adjustments and will be concerned with the form and shape of their argument. At this stage, writers are able to devote attention to what they have written, the ideas in their head that are yet to be translated into text as well as the needs or reactions of a potential reader.

Instructional Implications of this Model of Expertise

Knowledge Crafting requires a writer to do multiple things at once as they juggle different mental representations (text, author and reader) whilst simultaneously generating ideas, spelling accurately, forming letters properly and writing coherent sentences. Writers will also be recursively planning and reviewing as they write.

Writing is a composite skill made up of multiple interacting components. If pupils are to develop in expertise and therefore devote attention to the text, author and reader, they will need to reduce the load on their working memories. Beginners are likely to only focus on ‘telling’ because their dysfluent spelling and handwriting will take up a lot of their working memory capacity. Once their spelling and handwriting has become fluent and therefore automatized, they are more likely to be able to devote attention to the wider processes of planning and reviewing.

So what can we do to help pupils develop in expertise? How can we ensure that pupils move from telling to transforming to crafting?

  1. Reading and Writing: connecting the two.

One possible avenue for instruction is through harnessing the reciprocal relationship between reading and writing, both of which share a common knowledge base. Reading can improve writing and writing can improve reading.

2. Knowledge

Effective composition requires an extensive range of knowledge, including vocabulary, grammatical know-how, discourse structures and domain specific knowledge. If writers are to be successful, they need to be able to rapidly retrieve this knowledge when writing. If it is not accessible from long term memory, perhaps through lack of practice or because too much time has passed since they original learned it, then they will be unable to use it when writing.

3. Attentional Funneling

It can be helpful to explicitly teach pupils strategies for each of the three main cognitive processes of writing (planning, translating and reviewing). Initially, and in order to reduce unwanted cognitive load, each process should be taught in isolation. Strategy instruction, like almost everything else, will be made more efficient if it is taught through the I-We-YOU continuum, beginning with lots of modelling and moving gradually towards independent practice. While splitting up the writing process and teaching each part separately may be very different to how experts write, this is to be expected: effective instruction will often look different to the final performance.

4. Deliberate Practice

If pupils receive sufficient practice in the components and wider cognitive processes of writing, then each will require gradually less attention and effort. Like attention funneling, the goal of deliberate practice is to bring pupils to an acceptable level of writing fluency so that they can then successfully use and juggle multiple representations (text, author, reader) whilst writing.

Deliberate practice requires:

  1. Effortful exertion to improve performance
  2. Intrinsic Motivation to engage in the task
  3. Tasks that are within the reach of an individual’s current level of ability
  4. Effective feedback
  5. High levels of repetition

If you are interested, many of the ideas in this post are based on ‘Training writing skills: A cognitive development perspective‘ by Ronald T Kellogg

Next post: Writing Expertise: The Case For Deliberate Practice