According to some reports, the government in England is planning to introduce a new reading test for Year 8 students.
Currently, there are no national tests for students in England in between SATs at the end of Year 6 and GCSEs in Year 11. Plenty of people have argued that Years 7 to 9 in particular are therefore “wasted years” where students - and schools - can find it too easy to drift.
A new national test in Year 8 will presumably be designed to fix this, and, according to the reports, the aim of the test is to provide schools with useful diagnostic information too.
I can see a Year 8 reading test being a genuinely positive force that drives improvement across secondary schools. I can equally see it being a complete disaster that undoes many of the gains English secondaries have made over the last decade.
As ever, the devil will be in the picky, picky detail.
What we don’t need: another unseen reading comprehension
What we absolutely don’t need is another unseen reading comprehension test of the type we currently use in Year 6 and Year 11.
The good thing about these tests is that they tell you who the good and bad readers are.
The bad thing about them is that they obscure the reasons for good and bad performance. This means that it’s very difficult to prepare for them in advance, or to gain useful information from them about how to become a better reader.
If the government is serious about wanting a test to provide useful diagnostic information, then this kind of test would be a terrible idea.
The problem with unseen reading tests
As Dan Willingham says, reading tests are knowledge tests in disguise. Here’s a great example, again from Willingham.
I believed him when he said he had a lake house, until he said it was forty feet from the water at high tide.
To understand this sentence, you need to know a crucial fact: that lakes don’t have appreciable tides.
If you know that crucial fact, you can make all kinds of interesting inferences and character analysis. You can infer that the character referred to as “him” is someone who likes to show off, and exaggerates and lies in the process. But he is not very good at lying, and other people catch him out and may laugh at him, talk about him behind his back, and mistrust him.
If you don’t know that crucial fact, you are stuck trying to “make inferences” that end up making little sense.
Most reading is like this. Without even realising it, we use our background knowledge to make sense of text, and when we don’t have background knowledge, what we read doesn’t make sense.
My personal experience of this, which I have written about before, is of giving two classes of Year 11s a GCSE past paper which had an unseen text from an Arthur C Clarke short story. The story was set in a futuristic ice age where London was being threatened by a glacier. Not one of the students knew what a glacier was. (The examiners’ report noted that this was the case nationally too.)
Once the students knew what a glacier was, it unlocked the entire text for them.
But of course, explaining what a glacier is to them after the exam is not going to help them when it comes to the next unseen comprehension. That test is not going to be about glaciers! It is going to be about dodos, or mountains, or marathon running, or something else completely unguessable.
This is what I mean about unseen reading tests not driving good behaviour. Because you have no idea what the test will be on, you end up trying to teach reading as though it is a generic skill.
This generally involves reading lots of short and simple texts and practising skills like “making inferences” and “finding the main idea”. The research shows that some short instruction in such strategies might have some benefit, but spending a lot of time on them is counter-productive, and they cannot compensate if you just don’t know what the words mean.
The problem is that unseen reading tests encourage schools and students to spend too much time on these limited strategies, and not enough on long-term vocabulary and knowledge building that will make a much bigger difference.
If we want to encourage better teaching, we need a better test.
A better idea
Fortunately, there are better tests out there – and there are places that have trialled them and show they work.
For decades, the educationalist E.D. Hirsch has been pointing out the flaws with the traditional unseen reading test, and has suggested a simple and elegant solution. The texts on reading tests should be based on the content in the curriculum. Teachers and students would know what topics the reading test would be on, and could prepare accordingly. The test would start to drive sensible behaviour that would lead to long-term reading gains.
Some American states have been listening to Hirsch. In the last few weeks, there’s been a lot of talk about the “Southern Surge” – a group of southern US states whose scores on the National Assessment of Educational Progress have risen dramatically. Louisiana is one of them – and Louisiana is a state that has introduced far-reaching curriculum and assessment reforms influenced by the cognitive science research on reading. You can view some of the materials here. They’ve also integrated the work of The Writing Revolution, a brilliant writing instruction programme.
In the UK, we could do something similar. For example, the government could commission a series of high-quality non-fiction texts based on some of the key topics in the national curriculum - maybe 25 or 30 in total. They could issue these texts as an anthology for every student. The Year 8 reading test would be on one of these texts.
The best way to prepare for the test would be to read and understand all the texts in the anthology. The best response to poor performance in the test would be to teach the content students struggled with more thoroughly. Both these activities would help students become better readers.
Learning from others
I have written a lot on this Substack about some of the elements of England’s education reforms that others could learn from, but this is one area where England could do with copying others.
Our phonics test in Year 2 is influencing many other jurisdictions, but our reading tests at Year 6 and Year 11 leave a lot to be desired. We also don’t test any subjects other than literacy and numeracy until Year 11. To quote Dan Willingham again: teaching history is teaching reading. Teaching science is teaching reading. Teaching geography is teaching reading.
And yet our assessment system inadvertently encourages schools to reduce the time spent on teaching history, science and geography in order to increase the time spent on preparing for reading tests. It’s short-sighted and counter-productive.
The wrong kind of Year 8 reading assessment would make this problem much worse.
The right kind would make it much better.
We will wait and see what our government come up with!
The way you explain why reading comprehension tests aren’t that helpful is so clear and concrete. Thank you!
When I first taught in the 80s,
I would give my students a simple vocab or comprehension diagnostic just to gauge my students and identify weaknesses and strengths.
Then a computer program came along that did a similar quick diagnostic and gave my students a reading level and a list of recommended books and took around 30 min. My students liked this test, especially the recommended reading. When standard testing came in to the US in the 90s it all went downhill. Meaningless stats that no one was either qualified or had the time to sift through. Students were disengaged from their reading score that seemed more punitive than encouraging. Not to mention the incredible time testing consumed and took away from lesson time. Highly damaging with little to recommend it.