Unseen Extracts: The Last Barrier to Reading Comprehension?
This is the fifth and final blog in the series – the first covers the overall issues with the KS2 Reading SAT, the second whether it tests reading, the third how it embeds the teaching of generic comprehension reading skills, the fourth on why the reading strands are meaningless and what the test actually measures.
In this blog I examine the use of unseen extracts in the KS2 Reading SAT and what alternatives there are to the SAT itself.
Why Do We Use Unseen Extracts?
It’s a truth universally acknowledged that unseen extracts with (largely) unknown content should be used to test reading comprehension at KS2.
I never questioned it before I learnt about the role of knowledge in reading. My experience as a Year 6 teacher trying to prepare children for the SAT made me realise that there was no clear knowledge base to be taught for the Reading SAT unlike the Maths or SPaG equivalents.
Unseen Extracts
In a conversation with David Didau, he stated that unseen extracts have been used since the 1920s to encourage close reading. This certainly seems to be behind their continued use.
Our understanding of reading has moved on significantly since that time, especially the role that knowledge plays in reading and why there is a vocabulary and knowledge gap between advantaged and disadvantaged pupils.
Apart from one secondary teacher I spoke to, all the others agreed that unseen extracts advantage the already advantaged, unlike the seen extracts in the Literature GCSE. Even the teacher who did defend them did so on the basis of what was being assessed – the writer’s craft.
However, this defense can’t be made for the KS2 Reading SAT. Only 5.5% of the marks awarded between 2016 and 2019 were equivalent to what is being assessed at GCSE level that could be deemed writer’s craft (see blog 3)
If one believes that reading is a set of generic reading skills applied then it does not matter what the content of the test is. But we know that reading is not skills-based and requires background and domain knowledge.
If the unseen extracts used for the KS2 Reading SAT relied on a knowledge base that we know all children could reasonably have then it would not disadvantage some pupils however it is not clear that they do.
The Reading SAT and the National Curriculum
The KS2 Reading SAT consists of three unseen extracts – two fiction and one non-fiction. Only in one year was a poetry extract used as a fiction text.
Analysing the KS2 Reading SAT papers for the last four years, it’s clear the extracts are not linked to the National Curriculum. Only 12% of the answers for questions asked required knowledge that would have been taught to pupils on the National Curriculum.
The following chart shows the links to the national curriculum links from the 2016 to 2019 Reading SATs: 1
| 2016 | 2017 | 2018 | 2019 | |
| Science | Plants
Evolution and Inheritance Habitats Food chains |
Earth and Space | Living things and their habitats
Food chains |
Plants
Habitats |
| History | Invaders: (Stone Age, Romans, Anglo-Saxons, Vikings, Normans) | |||
| Geography | Rivers
Land use |
Rivers |
Non-fiction extracts are not directly based on content from the National Curriculum or any links there are, are tenuous. Certainly there is little link to knowledge or concepts that children are learning across KS1 or KS2.
For the fiction extracts there is no link to the National Curriculum. This is understandable up to a point as we don’t have set texts but they don’t even aim to reflect the range of genres we are expected to teach that are in the attainment targets. In some cases this is understandable as some categories – books from other cultures and traditions or modern fiction – span genres but for others there is no reason – e.g. myths.
It’s hard to pinpoint Literary genres but thanks to Jasmine Lane’s help (@MsJasmineMN) I was able to create a more accurate picture.2 The classification of the fiction extracts is as follows (with the categories from the National Curriculum in brackets):
2016
The Lost Queen – Realistic Fiction (modern fiction)
The Wild Ride – Realistic Fiction (stories from other cultures and traditions)
2017
Gaby to the Rescue – Realistic Fiction (modern fiction)
An Encounter at Sea – Realistic Fiction (stories from other cultures and traditions)
2018
Grannie – Narrative Poem – lyric, verse
Albion’s Dream – Fantasy (modern fiction)
2019
The Park – Realistic Fiction (modern fiction)
Music Box – Fantasy (modern fiction)
I don’t know if the emphasis on realistic fiction is deliberate or not. However, it is clear that a range of genres is not covered. Neither is there an attempt to include the range of specific genres that are outlined in the National Curriculum. What was required from the National Curriculum was some joining up with secondary but this did not happen and greater specification of the genres to be taught so that writer’s craft could be questioned.
This lack of clarity drives the teaching of extracts with little known content. Even when the extracts are linked to the curriculum, the tests used to measure reading comprehension across all year groups don’t. This mismatch means we are left in the dark about how well a child is progressing and what impact the methods or curriculum is having on comprehension. What this way of measuring and tracking progress in reading does is focus our efforts away from reading fluency in the middle years of primary to practicing reading comprehension.
Advantaging the Advantaged?
The purpose of the KS2 Reading SAT is to assess the reading ability of all children and it seems clear to me that the extracts are chosen to try and ensure as many children can access the test.
Rightly this means children from all groups in society. However selecting texts on this basis means it indirectly becomes a test of how middle class a child is. If we select texts based on what the average Year 6 child should be able to read and what might be challenging to them then all we do is reveal the vocabulary and knowledge gap between the advantaged and disadvantaged pupils. As the content of the non-fiction and the genre of the fiction is not linked to the curriculum we can’t start to address these gaps on the curriculum or in our teaching. Continuously highlighting the gap with no means of addressing it just means we carry on recreating it.
Attempts to dumb down content don’t help. While the link to the National Curriculum is weak and content drawn from a wider source than the curriculum – any vocabulary or knowledge gaps identified can’t be fed back into the curriculum. When we dumb down the content all we are doing is avoiding the issue of the gap. Moreover we are lowering standards of the content to be learnt but without any real benefits to the reader themselves. Sure the results look better but the struggling reader is no better off in terms of being able to access more complex material in primary or secondary.
There is a belief that using known extracts makes the test ‘easier’ somehow. Yet we don’t make this case for the other primary tests so why reading? All an unknown extract with unknown content reveals is the lack of knowledge and there is a gap but no means of addressing it on the curriculum as it’s not linked to the curriculum. So we continue to highlight a gap but not address it.
Even though the test generally advantages the advantaged, it doesn’t help us to gain an understanding of their real reading ability either. Even for advantaged children it’s chance and coincidence that the books they’ve read and knowledge they’ve gained outside school happens to be what is needed during that particular test. In their case, the same issue applies, what can be done about the knowledge they don’t have?
Reading Assessment Categories
What do the assessment categories really mean when we acknowledge the role of knowledge? In the SPaG and Maths the categories tell us who has the most knowledge and can apply it. In Reading it’s supposed to mean who can apply generic reading comprehension skills best. In reality the categories are indicators of who has the most vocabulary and knowledge both in and out of school. The test doesn’t tell us anything that we don’t already know from the background of the child, our general knowledge of the child and their reading records.
We can’t change much beyond the school gates in terms of the knowledge the pupils gain. Even if we make parents less financially poor we aren’t going to change the knowledge they have. We can encourage but not enforce activities such as going to the library or museums, etc. We do try to do as much of this as part of school as we can but there comes a point where trying to make the children superficially middle class hits a limit. We do only have so much time in the day and in the school year to teach. If we spend more of it outside the classroom then children will miss out the knowledge we teach on the curriculum. Even if it’s not in the KS2 Reading test, the National Curriculum does provide a foundational knowledge needed for secondary and beyond.
In terms of using knowledge gaps from the test in our teaching there are a few options but none of them are ideal. There is little point in teaching the precise vocabulary or knowledge they didn’t know because it might not come up again for years, if ever again. We can look more generally at the types of vocabulary – e.g. synonyms of words but with no idea of what specifically to teach. This drives more breadth at the expense of depth. If we try teaching as many as words as possible – which some are trying to do, they gain in breadth but not necessarily depth. The question remains which words? If not linked to a specific curriculum, this becomes unmanageable pretty quickly. Also this does not address issues struggling readers face as it’s not just the amount of vocabulary they have, it’s the lack of depth of understanding of those words and connections too. Lack of re-encountering the vocabulary and practice in using it and applying it means the forgetting curve kicks in meaning efforts are wasted long term.
We could try teaching as much domain and background knowledge as possible. This would mean reducing the time we spend on the national curriculum units for non-fiction but with no clear idea of what to add. For fiction, this would mean resorting to extracts to cover as many different genres as possible as there is a limit to the speed with which we can read a book. This may not have a detrimental effect one the stronger readers among the disadvantaged who are doing lots of reading too but breadth at the expense of depth exacerbates the problem for struggling readers. They need us to teach whole texts the most so they can acquire more complex vocabulary they wouldn’t come across otherwise in their own reading. Plus all children gain better knowledge of characters, themes, etc in an in depth way that enables better inferences compared to reading extracts which are divorced from the wider narrative. It’s not always clear why characters act the way they do without the backstory. This means children are guessing when answering inference questions and one of the reasons why they get the answer wrong. Only a better understanding of the story and characters will lead to correct inferences being made in this instance, not general background knowledge.
Even correct answers on the test don’t always reveal that the child knows the content we have taught. We don’t know for sure what the extent of guessing is in the test. This is because the content tested is not linked to the curriculum and the way the questions are formatted. Take multiple choice questions if they were based on what we have taught then a correct answer would arise mainly as the child knows the content and has understood it. We would know if a child has guessed it correctly based on our knowledge of the child. I agree there will always be grey areas but the test as it stands leaves us in the dark on the whole. It doesn’t give us an accurate picture of the child’s learning at all except for what they don’t know.
Again the problem is highlighted with no means of addressing it without broadening the curriculum so much that the learning is shallow. We need to break out of the circle of identifying the gap and but no realistic way to address it without making it harder for the struggling reader to gain what they need to become better readers. The way to do this is to link the test to the curriculum.
Accountability
As I said in the first blog in the series – I believe in both being held accountable for my teaching and that schools should be held accountable for outcomes. But this can only be achieved fairly if the tests used are fit for purpose.
The KS2 Reading test is one of the three tests which together are used to hold schools to account. In its current format, it does not do this. Those of us who believe in the knowledge curriculum and a fair accountability system need to be honest. Just as there are no magic transferable skills, there is no magic transferable knowledge. Fluctuations in test results will drive the kind of practices we think are harmful and act as a barrier to the introduction of ones that we know will do the most good – including reading whole texts and focusing on reading fluency.
The test ultimately hurts traditionalists and progressives alike as it drives an intervention culture which means that resources are directed towards Year 6 and not more fairly distributed or placed where they could more usefully address the issue – namely from Reception to Year 3.
The complaints about narrowing the Year 6 curriculum and too much time testing in Year 6 can’t just be dismissed, they arise for a reason. The endless practising of tests is not helpful to teacher or pupil, it does not lead to an improved curriculum or better long term outcomes, especially for those pupils who need the most support to learn to read. Realistically, we will always have some revision before any kind of test but at least this should focus on the revision of knowledge that will hold the pupils in good stead in secondary, as we do for Maths and SPaG.
Solutions
There are a number of potential solutions. Since my first blog I have thought of a few variations. The table below outlines them:
| Solution | Option 1: Known Extracts:Non-fiction – knowledge from the National Curriculum 3Fiction – set texts |
Option 2:
Known Extracts: Non-fiction – knowledge from the National Curriculum 3 Fiction – taken from an anthology |
Option 3:
Scrap the Reading SATs and introduce fluency tests in Years 2 and 6. |
| Strengths |
|
Same as option 1
Anthology allows for a greater range of texts. |
|
| Weaknesses |
|
|
An alternative would be to have reading fluency checks in Years 2 and 5 alongside a Reading SAT in Year 6. |
My personal preference would be a combination of Options 1 and 3.
There will never be a perfect solution because there never is but if we are to truly transform our society then we need to make as many children literate as possible. We have for a long time made huge changes for little gain. What I’m suggesting is a small number of changes for maximum gain for all children but which will benefit those who are disadvantaged the most.
The very fact that I’m the one writing this should give some hope. We can take a child whose parents speak little English, one of whom left school at 14 with little education and one who is illiterate, and through our collective efforts in the education system transform that child into a literate adult.
We owe it to our most disadvantaged pupils to ensure that this is not an anecdote or an accident or a success story to be celebrated. This should be the living reality of even the poorest child no matter what school they attend, no matter where they attend it in this country.
We can do this.
References:
1 – I accept that this is a subjective analysis and any revisions anyone has to add are welcome.
2 – All of this is thanks to Jasmine Lane whose brain I picked for around 3 hours on a zoom call. To say that she had the patience of a saint is to underestimate her completely.
Literature Genres
| Fiction | Drama
(Plays) |
Comedy | ||
| Tragedy | ||||
| Historical | ||||
| Other | ||||
| Poetry
Narrative – tells a story Non-narrative – single speaker presents a state of mind or emotional state |
Verse | Haiku | ||
| Lyric – short, emotional | Limerick
Cinquain Tanka |
|||
| Song
Ode Ballad Elegy Sonnet Riddle |
||||
| Epic – long, emotional | ||||
| Free Verse | ||||
| Nonsense poem | ||||
| Prose | Narrative Fiction | Horror
Mystery/Crime Romance Action Adventure Suspense/Thriller Magical Realism – 100 years of solitude Gabriel Garcia Marquez, Midnight’s Children |
Plot types:
Quest Voyage and Return Rags to Riches Rebirth Comedy Tragedy/Tragic Hero (Greek myths) Overcoming the Monster Explorer Ruler Creator Caregiver Magician Hero Outlaw Lover Jester Regular Person Themes – dependent on text Literary Periods |
|
| Fantasy | ||||
| Science Fiction | ||||
| Historical Fiction | ||||
| Realistic Fiction | ||||
| Gothic Fiction | ||||
| Speculative Fiction | Dystopian/
Utopian |
|||
| Folklore/
Folktale: |
Fable – short story with personified animals and a moral | |||
| Mythology – gods/goddesses, creation stories | ||||
| Tall tale – hyperbole/exaggeration | ||||
| Legend – based on real person or place, exaggerated | ||||
| Fairytale | ||||
| Parable | ||||
| Non-
Fiction |
Narrative | Diary
Journal Letter Travelogue |
||
| Speech | ||||
| Autobiography | Memoirs | |||
| Biography | ||||
| Essay | Formal writing
Magazine articles Newspaper articles Opinion pieces |
|||
| Informational | Popular science
Science book Literary criticism Natural history Philosophy Theology |
|||
| Technical writing | Textbook
Almanac Dictionary Thesaurus Encyclopedia Guides and manuals Handbook |
|||
| Diagrams | Blueprint
Design document |
3 –
Science and Foundation Units for Reading SATs
KS2 Science Units
| Content Suitable | Potentially the content is not specific | Content not suitable | |
| Year 3 | Plants – functions of parts of a flower; seed formation and dispersion; water transportation
Animals, including humans – skeletons and muscles Forces and magnets – magnetic force, attract/repel, magnetic poles Rocks – grouping, fossils, soil Light – light, dark, shadows |
Animals, including humans – nutrition | |
| Year 4 | Living things and their habitats – classification, environments
Animals, including humans – digestive system, teeth States of matter – solid, liquids, gases, water cycle Sound – vibrations, pitch, volume Electricity – uses, electrical circuits, switches, conductors and insulators |
Animals, including humans – food chainsStates of matter – changing states | |
| Year 5 | Properties and changes of materials – classification, dissolving, mixing and reversible/ irreversible changes
Earth and space – Sun, Earth, Moon, day and night Forces – gravity, water resistance, friction, mechanisms. |
Animals, including humans – Development to old age | Living things and their habitats – life cycles, reproduction |
| Year 6 | Living things and their habitats – classification, water and nutrient transportation
Animals including humans – circulatory system Evolution and inheritance – fossils, adaptation Light – travel Electricity – voltage, functions of components, symbols in diagrams |
Living things and their habitats – diet/exercise/drugs/lifestyle in humans
Evolution and inheritance – offspring |
KS2 History Units
| Content Suitable | Content not suitable | |
| Key Stage 2 | changes in Britain from the Stone Age to the Iron Age
Examples (non-statutory) This could include:
the Roman Empire and its impact on Britain Examples (non-statutory) This could include:
Britain’s settlement by Anglo-Saxons and Scots the Viking and Anglo-Saxon struggle for the Kingdom of England to the time of Edward the Confessor Examples (non-statutory) This could include:
|
|
KS2 Geography Units
| Content Suitable | Content not suitable | |
| Key Stage 2 | Locational knowledge
identifying key topographical features (including hills, mountains, coasts and rivers) identify the position and significance of latitude, longitude, Equator, Northern Hemisphere, Southern Hemisphere, the Tropics of Cancer and Capricorn, Arctic and Antarctic Circle, the Prime/Greenwich Meridian and time zones (including day and night)
Human and physical geography |
Locational knowledge
locate the world’s countries, using maps to focus on Europe (including the location of Russia) and North and South America, concentrating on their environmental regions, key physical and human characteristics, countries, and major cities name and locate counties and cities of the United Kingdom, geographical regions and their identifying human and physical characteristics, and land-use patterns; and understand how some of these aspects have changed over time Place knowledge Human and physical geography |