researchED 2017 workshop 1: exploring the complexities of reading comprehension difficulties

    Published: 17 September 2017

    In the first of a two part blog, Martin Galway shares his thoughts on presenting a pair of workshops (one on grammar and, here, a late-notice addition on reading comprehension) at this year’s researchED  conference.

    Just over a week ago, I had the pleasure and privilege of presenting at this year’s conference in Stratford London.  I can’t lie.  It was somewhat nerve-wracking.  My colleagues on the maths team had been due to present, too, on fluency in KS1 maths.  Circumstances outside of their control forced them to cancel a few days before the event and so I offered to double up and added a presentation on reading comprehension difficulties. ( You can read about my other session here).

    My session straddled lunch: reading comprehension difficulties just before, and primary grammar and transition just after.  Which effectively meant no time to eat. I must at this point thank the genuine hero who turned up early for my grammar session and kindly fetched me an all-important bottle of water.   I got the chance to meet and talk to lots of lovely, engaging people who helped me to further develop my own thoughts and ideas. Thank you in particular to the poor unfortunates who  seriously constrained their own lunch thanks to their commitment to the joys of talking about reading. Further compensation came when I had the chance to watch Alex Quigley speak powerfully on vocabulary, and Oliver Caviglioli (for the second year running) putting his work on dual coding into practice.

    Session 1: Deceptively Simple Views of Reading: exploring the complexities of reading comprehension difficulties

    This session reflected the past two summers’ reading – last year on fluency; this year on what lies beneath the complex act of reading for meaning.  Why was it called “deceptively simple views of reading”? I was playing on the name of the conceptual model of reading that underpins our curriculum, but this was not a critique of  Gough and Tumner’s Simple View of Reading  (SVOR). What I looked at was  some of the problems that have arisen in the interpretation of such models and related research in the primary classroom.  The SVOR allowed me to frame the session so that we were focused on a sometimes overlooked minority group: children that have what appear to be strong word recognition but that (not-so-simply) struggle to comprehend what they read. Essentially, children that sit within the bottom right-hand quadrant of the Simple View:


    I’ve already pointed out that I have some concerns with how this model has been interpreted on occasion (essentially when the two continua have been treated as though they were quite distinct when it comes to framing teaching,). This concern sits within a bigger picture of wider concerns about how we talk about reading online – and in the media – and the potential to overlook the more diverse needs of the primary classroom.

    Tread with care

    One of the privileges of my job is that when the summer holiday hits, we are able to use the time freed from school support to catch up more fully on research, and write resources or training.  Last summer, I had worked with my colleague Alison on a KS2 Guided Reading Toolkit and part of that was a section on reading fluency.  Inspired by the insights and reading suggestions of my colleague Kirsten, I had spent the summer moving from one study to another, following up references, exploring challenges.  The net result of months of reading was a five-page chapter, heavily made up of visuals. I remember  thinking at the time : I wonder how many people will realise how much reading informed this? And this leads me to a quandary of sorts that I used to bookend my session.

    I opened the session with a sort of caveat that warned of certain reading practices that superficially appear to be  research-informed but that are in fact somewhat mutated and in some cases either have been seen to lessen the impact seen in studies (I am thinking of reciprocal teaching here) or have wound up so streamlined that they cannot hope to account for the spectrum of needs that you will find in a properly inclusive classroom. Perhaps oddly, I warned against taking summaries of complex areas of research or simplified views of complex areas at face value.  That, therefore, included the talk I was in the process of delivering.  Very meta, eh?


    In the session I drew upon recent reading of digests of reading research.  I drew particularly from the work of Jane Oakhill, Kate Cain and Carsten Elbro; Morag Stuart and Rhona Stainthorp,  and Wayne Tennent’s recent book, to consider some of the central concerns in reading comprehension.  I also referred to Daniel Willingham’s more recent The Reading Mind (because you are almost legally obliged to refer to Willingham in some way nowadays – and it is a very good – digestible – book).

    In doing so I hoped to highlight, once again, the complexity of what goes on in the act of proficient reading.  I explored a few conceptual models that helped to move us towards a more complex view of the constituent knowledge, and then the processes, that make up the proficient reading act.  I lingered on fluency for a short while, partly because of its proven importance and partly to wring the reading of summer 2016 that little bit further.

    Inference making

    I moved on to inference-making which emerges as a central component in making meaning in much of the literature on poor comprehension. Yet in this arena, messages abound that confuse and confound.  Let me just start with a document that sits apart from the the surrounding research.  The 2016-17 Interim assessment framework for reading – a document that all year 6 teachers – and many of their colleagues – would be familiar with.   Here is a fragment of it as presented in my powerpoint:


    Here we see how understanding (and then the teaching) of inference might be somewhat skewed. Bullet 4 explicitly refers to inference making and it is typically this that is spoken of in schools in relation to inference. Largely because this is what is tested too and accountability regimes  – to different but inevitable degrees – influence practice.

    But look at the other bullets – they all rest on inferential processes:

    Bullet 1: Reading fluently at KS2 would require prosodic attention to the text – to attend to this, inferences have to be made. Prosody makes understanding audible.

    Bullet 2: Intonation that shows understanding would, I think it is fair to say, sits within fluency – see Bullet 1.

    Bullet 3: working out the meaning of words form context – it’s got to be inferential !  We are using our knowledge of other words, concepts , phrases, syntax etc to derive meaning.

    Bullet 5: Prediction – guess what – unless a wild guess based on pure imagination – inference.

    The sort of inference making privileged in end of KS2 statutory testing – and in the Interim teacher assessment framework –  prioritises making inferences based on what has been read to establish performance in relation to statutory age-related expectations (ARE).  These inference questions are part of the question: how well are the children able to do this? The trouble is, if  a child does not do this well enough, all too often they are asked to do more of this type of work.  Conventional comprehension exercises seem to be multiplying at quite a rate, seemingly in response to the rude awakening challenges of last year’s KS1 and KS2 reading tests.

    However,  the research that informed my talk points to other kinds of inference making that need to be attended to within the act of reading in order to bolster weaker comprehension.  It is these ‘in the act’ inferences that we need to monitor, model, encourage and develop. We need, in teaching for reading comprehension, to ask not ‘how well are they able to…?’ but ‘what is it that they need to be able to do in order to do this?‘  in order to gain meaning from the texts that they read.

    The session further unpacked what sorts of difficulties might be encountered by less-sure comprehenders, for example difficulties in knowing the ‘who’ or the ‘what’ that  pronouns and other noun phrases are referring to earlier in the text (anaphoric inference),  or  difficulties in the handling of conjunctive inferences which leads to a tenuous, weak or absent grasp on  the relationships between the ideas presented, both of which might be exacerbated by the additional strains that are felt when an inference requires links to be made across intervening text (Oakhill, Cain and Elbro explore this in detail in their 2014 book Understanding and Teaching Reading Comprehension).  We saw that at play for many children in the 2016 test.

    Comprehension monitoring

    Core strategies to foster comprehension monitoring as they read include predicting  (the subject of some recent debate), questioning, clarifying and summarising. Once again, do get in touch for further related reading suggestions.  Those familiar with the method will recognise these as the core reciprocal teaching strategies.  The key issue is how we use this knowledge in class. The length of the session meant that we were not able to explore this too deeply.  From experience and reading, I can say this at least: it is not about isolating the roles and having children practise them on rotation; it is about helping children to have a greater awareness of what proficient readers do – often simultaneously, usually unconsciously.  It is about modeling, guiding and encouraging children to deploy these strategies with increasing ease so that the desired skills develop.  By ‘skills’, we are talking about these processes occurring automatically, seemingly naturally.  In short: active, thinking reading. (For greater insight, see Cain and Oakhill, 2015, and Tennent 2014)

    The importance of dialogue and articulated thinking drew the talk to a close (with Aidan Chambers once again cited as key influence).  Across my reading, attention has been drawn to the need to know what children understand from a text and build from there – not to see how well or not the children can determine what I – or a mark scheme – happen to understand.  In looking at this, it becomes apparent that subtle shifts in language can make all the difference.  In a test, we might ask “Why did X do Y?”.  In a bid to prepare children for the test, many resources, including some of our own, might ask a question framed in exactly the same way.  Test practice is something we do.  But in a routine classroom discussion, in the course of teaching reading, how much better , and more helpful, to ask “why do you think that X did Y?” because within that question we are not only asking “do you think…?”,  we are also  gaining hugely valuable insight into just where we might need to invest our teaching energies.

    I shall leave it there for the reading session.  Once again, I am worried that I have oversimplified the already simplified.  I’ve embedded a sample set of slides below.  If you would like a PDF of the slides for this session, drop me a line at :

    Please note that the final slide here shares a small but very helpful reading list of recent books that look at the topics covered in this session.  The first three books on this list are extremely helpful digests of learning about reading. I highly recommend them as additions to your school’s CPD library in order to support further development of reading expertise across all phases. I can also provide a fuller reading list, again at the e mail address listed above.

    In the next blog I will look at Session 2: Primary Grammar – what really matters in transition.

    For now, here are a few sample slides to be getting on with.

    Thank you for reading.  Feel free to get in touch if you would like to know more.  You can use the e mail above or the @HertsEnglish twitter account.

    Contact details

    Latest blogs

    Receive our latest posts direct to your inbox...