Tag Archives: Doug Lemov

What I think about…reading

Moving schools and with more than an eye on headship is sure to get you reflecting. The following posts are what I think about various things, in no particular order. Previous posts were about displays, learning generally and maths. Next up – reading.

I’m proposing a model for teaching reading grounded in the various books that I’ve read. The examples will be for a fiction text but I think the principles apply to teaching non fiction too.

Reading model

Some principles

The first principle to be mindful of is that the teaching of reading is not the asking and answering of questions about a text: that’s testing comprehension.  Sure, asking and answering questions is an important part of developing comprehension – it’s one way we get children to think hard about what they have heard or read – but there is much more to it than that.  Any reader constructs a mental model of the content of what they have read – we don’t usually remember text verbatim without rereading many times and deliberately trying to remember it word for word. Poor comprehenders construct weaker, less detailed and perhaps outright inaccurate mental models whereas good comprehenders construct more accurate and elaborate ones.  One goal of teaching reading then is to ensure children construct good mental models of what they have read. I’m making the assumption here that children can decode fluently and focusing solely on the development of language comprehension.

Simple view of reading

Good readers combine word recognition with language comprehension to be able to decode the print and understand the language it yields. Once fluent in decoding, it is depth and breadth of vocabulary and general knowledge that contribute to comprehension and so the teaching of reading must develop vocabulary and background knowledge.

Developing reading comprehension

Poor comprehenders share many similar characteristics which we need to understand and use to drive the teaching of reading.  Poor comprehenders:

  • have limited general knowledge
  • have a limited knowledge of story structure or don’t relate events in a story to its general structure
  • have a narrow vocabulary and don’t know the meaning of important words
  • read too slowly, without fluency or enough prosody to understand the content
  • focus on word reading without focusing on content
  • make incorrect pronoun references
  • don’t make links between events in the text
  • don’t monitor their own understanding of what they’ve read
  • don’t see the wider context in which the text is set
  • don’t build up a secure understanding of the main events in a story
  • misunderstand figurative language

When it comes to vocabulary, we can’t teach every word or phrase that children might not know and neither should we. If we do, not only would it be incredibly time consuming but we’d also greatly reduce the experience that children have of deciphering meaning from contextual cues. Some words and phrases need to be taught explicitly before or during reading while others can be learned implicitly during reading.  Either way, if children are to master the language, they must think hard over time about its use.  Put the dictionaries away and don’t start off with ‘Who knows what x means?’  These are both particularly inefficient uses of time and are ineffective.  Instead:

  • Model the use of the word in its most common form
  • Use an image (this post from Phil Stock is excellent)
  • Act it out
  • Model other common uses
  • Explain word partners (for example, if teaching the word announce you often see make an announcement together)
  • Show various forms of words including prefixes and suffixes
  • Show words that are similar to and different from the focus word

Lemov (Reading Reconsidered)

That last bullet point is not the same as using the synonym model for teaching word meaning.  Telling  a child that melancholy means sad robs them of the beauty of shades of meaning because it is similar to, not the same as sad.

Memory is key. We remember what we think about, so part of teaching reading needs to be giving children plenty of spaced practice in remembering word meanings, general knowledge, events from the text and details of the characters that are crucial to developing a sufficient mental model of the text. It could well be the case that a child who has shown poor understanding of a text is not unable to comprehend it, they just can’t remember what’s necessary to comprehend. Regular low stakes testing of key knowledge from the text is a strategy to ensure this retention and readiness to mind.  Joe Kirby’s knowledge organisers are very useful for this and here’s one I made for Philip Pullman’s Northern Lights. 

Stage 1 – oral comprehension

Prepared reading, or providing a brief structural overview, ensures that no child hears the story without some prior knowledge.  In the first instance, read aloud or tell children the story. Capture their interest. Retell it, perhaps in different ways.   Lemov, in Reading Reconsideredidentifies different types of reading and here I’d go for what he calls contiguous reading – reading without interruption from start to finish, experiencing the text as a whole.  It may be sensible to teach the meaning of some words that are crucial for overall understanding of the text but not too many at this stage.  I’ve compiled some thoughts on introducing texts and teaching vocabulary here.

What have children understood?

Clearly it is tricky for teachers to know what children have understood and by asking questions all we really know is whether they are capable of comprehending, not whether they actually comprehend independent of us. Before any specific questioning, it would be useful to get an idea of what they have understood by asking them to tell you about what they’ve just read. The decisions they make about what they say (or write)  reveal what they think is important and you can also judge the accuracy of their literal and inferential comprehension. Aidan Chambers’ Tell me gives advice on developing this in a slightly more structured way whilst still retaining the importance of open questioning.

The key to this stage of reading is the focus on oral language comprehension.  Difficulty decoding should not be a barrier to children experiencing and understanding age appropriate texts.  Lemov puts this beautifully:

Low readers are often balkanised to reading only lower level texts, fed on a diet of only what’s accessbile to them – they’re consigned to lower standards from the outset by our very efforts to help them.

Lemov (Reading Reconsidered)

This is one of the reasons why I’m in favour of the whole class teaching of reading and not the carousel type ‘guided reading’.  Listening to texts and using open questions to prompt discussions ensures that the focus in on language development in a way that is not restricted by poor decoding.  Having said that, those children who are not decoding to the standard expected will still need some sort of intervention running concurrently to this so that they catch up.  The benefits of focusing on oral language comprehension have been shown in the results of the York Reading for Meaning Project, written about in Developing Reading Comprehension by Clarke, Truelove, Hulme and Snowling and here.

Stage 2 – modelling the reader’s thought processes and shared reading 

The information that teachers can gather from the open questioning in stage 1 then focuses modelled and shared reading on specific parts of the text. The teacher can model the reader’s thought processes, and get children thinking about the tricky bits. This isn’t simply reading the text from beginning to end; reading will be interspersed with commentary, explanation or making links to general knowledge.  Lemov calls this line by line reading, with frequent pauses for analysis and allowing the teacher to show children that good readers think while they read in order to achieve an acceptable standard of coherence.  As children get older and texts get longer, teachers can’t lead shared reading of the whole text, so by initially earmarking sections that children are likely to misunderstand and by using information gathered from stage 1, shared reading can be focused on addressing misconceptions.  Again, Lemov puts it succinctly:

Shared reading mitigates the risk of misreading.

Lemov (Reading Reconsidered)

I’d expect children to then read the text independently, drawing on what they’ve heard from the teacher’s modelling and all the oral language work. Children should have the opportunities for multiple readings of at least the tricky bits.  These bouts of reading become iterative: children build layers of understating with each reading.  For those children whose decoding is weak, they can be directed to smaller extracts, practising decoding and fluency with a text that they should have a decent understanding of following all of the language work.  It’s important to continue to get children thinking about new words that were taught in stage 1.  If that vocabulary is to be reliably internalised, they’ll need multiple interactions.

This is also an ideal point to make some links to non-fiction that can supplement understanding of the fiction. Questioning that involves deliberate comparison between the fiction and non fiction complements understanding of both.  For example, if reading Robert Louis Stevenson’s Treasure Island, spending some time on books or extracts such as below will significantly aid comprehension.

Non fiction links

Written responses

Writing is thinking, and to paraphrase Lemov in Reading Reconsidered, not being able to record their thoughts about what they’ve read on paper does not make them invalid, but children are at a significant disadvantage if they are unable to craft an articulate, effective sentence explaining what they have understood.  To this end, returning to those original open questions and working with children to refine their responses and write them effectively is a valuable use of time.  The teacher can model scanning the text for the part needed to refine an idea, or to check a detail, and then children should also be expected to behave in that way.  This post by Lemov makes very interesting reading on that topic.

Stage 3 – targeted questioning

It’s standard practice to ask questions of a text after it’s been read but a great deal of care needs to be taken in choosing or discarding already written questions, or in writing them ourselves. Questions need to be text dependent, otherwise what we’re really doing is getting children to activate general knowledge. An example of this, from Understanding and teaching reading comprehension by Oakhill, Cain and Elbro, is:

Where does Linda’s pet hamster live?

  1. In a bed
  2. In a cage
  3. In a bag
  4. In a hat

The possibility of guessing the right answer here would tell the teacher very little of the child’s ability to comprehend text and so asking questions where understanding is dependent on what’s written or what must be inferred from the text is a must. Doug Lemov espouses the importance of text dependent questions in Reading Reconsidered.

When designing questions, teachers must also use knowledge of the characteristics of poor comprehenders in order to model corrective thought processes and to ensure children think in a way that helps them to comprehend more reliably.  For example, we should give them plenty of practice in working out to what or whom pronouns refer.

The education system we work within requires examinations to be passed which then provides opportunities.  Preparing children for success is morally imperative. Write questions in the style of SATs questions about the text, model the thinking process behind successful responses and give children practice doing just that.

Stage 4 – fluency and prosody

Don’t misunderstand – children should be supported continually to read fluently with appropriate intonation and expression. It’s just that to do that well, a reader needs to understand the text. At this stage, that should be the case. Reading for fluency and intonation using a text that children know very well should yield great results and not only that, it provides another opportunity to glean previously missed understanding.

So there it is. A model for teaching a text that moves from oral to printed comprehension; general to specific questioning; and oral to written responses, all the while practising fluency and developing language.


Filed under Reading

…for who will coach the coaches? Part 2

In a previous post, I explained the importance of a school’s vision for coaching matching the already established culture. There are a couple of reasons why I think that talking about coaching in those early weeks of the new school, with that carefully planned language, is necessary:

  • It gives leaders time to match up staff with coaches while giving them time to settle into a new school / room / role / year group etc.
  • It generates a little momentum for when coaching is launched. Think of the busker who scatters his guitar case with a few notes and coins, rather than start with an empty case. This isn’t such a big change – we’ve already started…
  • It creates time to work with coaches on their repertoire of strategies before meeting with their colleagues.

Managing Change

At this point, any potential barriers need calling out and possible solutions shared. In Dan and Chip Heath’s book ‘Switch’, they talk about the elephant and rider analogy in managing change. The elephant represents emotion and the rider rationality. The elephant will only go where the rider wants it to if it so chooses. The rider cannot force the elephant to do anything. Unless the rider knows where he wants the elephant to go, they will not end up at the desired destination. To manage change, we need to motivate the elephant and direct the rider. We need to understand what it is about coaching that will motivate our colleagues’ elephants. Selling the expectation that it will lead to being a better teacher is a good starting point. We are motivated by the drive to acquire, to bond, to comprehend and to defend. Coaching can lead us to acquiring skills and knowledge about teaching which can make us better teachers. It can lead us to have quality conversations with our colleagues, helping each other to improve and bond along the way. It can lead us to deeper comprehension about effective teaching. It can reinforce the principle that we do the best that we can for the children that we teach, defending their present and their future.

Motivation without direction is useless so we need the attention to detail that the rider provides. What aspect of teaching are we practising? When will it happen? How will we practise the strategies? Which particular coaching strategies are most approproate for this teacher at this time? These details need planning for carefully because our time is valuable and clarity is what the rider needs. Each coach will need a coaching plan and part of the work with the coaches in September will be putting those together.

Along with motivating the elephant and directing the rider, if we want to arrive at a certain destination, the path needs to be clear. We need to remove any barriers so that we can get there. Time and the various other commitments that teachers have are the metaphorical logs blocking the path and must be removed for coaching to work. To start with, the time issue can be addressed by only asking a small time commitment per week – say half an hour. This half an hour cannot be at lunchtime or at 4.30pm on a Friday as that would diminish the status that we want to create for coaching. We have to provide release time from teaching responsibilities, where appropriate, for this to happen. The half an hour could involve twenty minutes of deliberately practising a strategy in class, followed by a ten minute conversation while someone else covers the class. Short and managaeable.

Practising being a coach

A coach will have a repertoire of support strategies to draw upon to support colleagues. Like any other domain of expertise though, coaches will need to practise their wares in order to be as effective as they can be. There will be a few strategies that a school could identify early on that would yield the best results. Pareto’s principle, or the law of the vital few, is that 80% of the output comes from 20% of the input, that is, a few key strategies could provide the greatest return on improving teacher quality. These key coaching strategies could be:

  • Demonstration lessons
  • Team teaching
  • Quality and timing of feedback
  • Coaching conversations
  • Shared planning and marking

These strategies will need to be practised, with other coaches playing the role of the teaching colleague. So in the first few weeks of term, coaches could meet regularly to practise getting demonstration lessons as clear as possible. When coaches can do this with automaticity, they can focus more upon the reactions of their colleagues, tailoring what they’re doing to meet their needs better. They can practise the subtleties of team teaching – when to step back, when to model a particular strategy. They can practise giving quality feedback in those brief lulls in lessons that would enable their colleague to listen and act immediately by repeating the focus teaching strategy. They can practise the skillful listening and questioning needed to help a colleague solve a problem. If after 3 or 4 weeks back in the new term, coaches have met and practised these strategies, then they are prepared for doing so for real. These strategies need a context to be practised within though. In my school it will include some teaching practice that we deem to be of highest value in terms of outcomes for children:

  • Modelled and shared writing
  • Oral and written feedback on children’s work
  • Co-constructing a writers’ toolkit
  • Modelling mathematical strategies
  • Explicitly addressing misconceptions

Working with coaches in this way enables them to act their way into thinking, and gives them a sound experience in which to frame the language they use to share the vision for coaching and CPD with their colleagues. Also, spacing out the sessions over a few weeks will contribute to maximised retention of the strategies by the coaches. Interspersed with these practices, I’d expect the coaches to be reading in order to build their knowledge. Books like Practice Perfect by Doug Lemov and The Perfect Teacher Coach by Jackie Beere are essential reading, along with great blog posts like these from Alex Quigley @huntingenglish (here, here and here and Shaun Allison @shaunallison (here and here).

In the final post in this series, I’ll be thinking about how coaching can fit into a wider CPD program.

1 Comment

Filed under Coaching, CPD

Making deliberate practice work

Previously, I have written about the need to turn ideas about deliberate practice as part of CPD into firm actions. I have, in the last week, attempted to make this work in two different situations, with varied success. This post is my analysis of why one situation worked better than the other.

The first session I tried out was for a group of 30-40 teachers and head teachers. It was part of a training day in Talk for Writing, specifically around developing writers’ toolkits with children, which I have written about here. After a demonstration of a couple of variations of how it could be done, there was some discussion about what the teacher actually does to generate a writers’ toolkit with children. This was fine and understood well by the delegates, so the next stage of the plan was to set up some deliberate practice, which I explained using the following model.


And it didn’t work. I can think of a few reasons. In Doug Lemov’s book, Practice Perfect, his rule number 24 is: Apply first, then reflect. The delegates, either through having a lot to talk about following what they had seen and heard, or through avoidance of potentially awkward acting in role, did a lot of reflecting. Having started the process in earnest, the first moment that arose that caused a discussion was grabbed and this disrupted the intended deliberate practice. Alas, I was not skilful enough to redirect such a large group and the moment passed. Plus it was lunchtime.

Lesson learned. It’s very difficult to keep a large group focused on deliberate practice.

I had an altogether different opportunity to try out setting up some deliberate practice later in the week. Having observed a colleague’s maths lesson (Year 1, division), I selected an aspect of what the teacher did, that if tweaked, could have been great. It was to do with explicitly addressing a common misconception. Here’s what I wrote as a plan:




I talked through each part with the teacher and asked them to talk through each bit, writing / drawing on a prepared IWB file as necessary. When the teacher presented effectively, I said so and asked them to repeat that bit. When the teacher presented something ineffectively, I referred to my plan, suggested or modelled an improvement, then they did that bit again. Some bits worked better than others. I thought I knew the content very well, but clearly not well enough for all the subtleties of what the teacher could have said. As such, my feedback could have been clearer at points.

Next, I’ll find 5 minutes a couple of times over the next week to run that practice drill again. Then, the teacher will have a go with a small group of children, with me coaching and offering live advice which can be acted upon immediately. Finally, having experienced the process, I intend to use that teacher in a larger group with other colleagues who teach the same year group, getting then to deliberately practise the same explanation frequently over a period of time then applying to their own classes.

So, my advice to anyone in a similar situation would be to start working with one person, in order to manage distractions. Keep the content simple and plan for the most common responses that would need feedback.

Leave a comment

Filed under CPD