Where is the evidence for reading comprehension strategies?
Introducing the £1m Kingsnorth Prize 2020
Ok, let me be completely transparent from the outset:
- There is no denying it: I am, in my current thinking, heavily prejudiced against reading comprehension strategies (RCS) when it comes to improving children’s long-term reading comprehension. I’ve written extensively about this prejudice both here and here.
- While I am placing all my chips on a heavily-biased view, I am also ready to publicly repent and change my mind in the face of overwhelming evidence. It will hurt, but I am genuinely ready.
- I don’t actually have £1,000,000. Sorry.
The Problem
Reading comprehension strategies (or skills) are often lumped together and presented as the evidence-based route to improving children’s overall comprehension of text.
This has led many teachers to assume that if a child is struggling to comprehend books independently — the answer is simply to practice these reading comprehension skills…on anything at all.
What we’ve ended up with, as a profession, is this…and this.
There isn’t a major educational publisher out there who isn’t pushing you to use random worksheets and extracts to ‘help the children develop their reading understanding and comprehension skills’.
Implicit in this claim is that children will do this irrespective of the text. If they simply apply these skills to a worksheet, then, we are told they will magically get better at comprehending the next book they read independently.
The evidence, however, does not appear to back this up one jot.
Big Claims
In terms of educational research, the claim that explicit teaching of reading comprehension strategies improves overall reading comprehension is about as uncontroversial as they come.
Here is an excerpt from the EEF’s evidence summary for RCS:
It would seem these conclusions are based on a pretty solid evidence base:
And this is from their guidance report on improving literacy across KS2:
In the US, the National Reading Panel released their widely-cited report in 2000 which is often held up to be the gold-standard proof.
Here is the Wikipedia summary on their findings re RCS:
My Claim
Some RCS can clearly deepen understanding of a specific text, and any skilled teacher will draw on strategies such as summarisation. However, I have seen no convincing evidence at all that they will bring about future comprehension of unseen, independently read texts. I’m also taking aim at the phenomenon of teaching the skill irrespective of the text (eg inference using pics). Our best bet is still to put vocabulary, background knowledge and choice of text firmly in the foreground of reading lessons.
The £1m Kingsnorth Prize
I am offering (but not paying) a £1m cash prize to the first person who can find the following:
- Studies with a combined sample of 200 or more children where the experimental group beat the control group by more than 4 marks in a delayed, standardised reading comprehension test which was not written or administered by the researchers themselves.
I have trawled through as many of the references in the NRP and EEF reports as I can, as well as the studies cited in the numerous meta-analyses, but to no avail.
I am getting bored, am I missing anything?
Why does it matter?
If you teach primary, you will already know how hegemonic reading comprehension strategies are in the classroom.
Virtually every approach or scheme of work puts these strategies at their heart, and has led to a belief (or forced compliance?) amongst teachers that these skills and strategies are content-agnostic. That is to say: practice them on one soul-destroying worksheet and you will magically get better at inferring the next one.
I can’t put it better than Natalie Wexler, author of The Knowledge Gap. Here is an excerpt from the book:
The theory that has shaped the American approach to elementary education goes like this: Reading comprehension is a set of skills that can be taught completely disconnected from content. Teach children to identify captions in a simple text — or find the main idea, or make inferences, or any one of a number of other skills — and eventually they’ll be able to grasp the meaning of any text put in front of them.
But cognitive scientists have known for decades that the most important factor in comprehension isn’t a set of generally applicable skills; it’s how much background knowledge the reader has about the topic. If you don’t have enough knowledge and vocabulary to understand the text, no amount of “skills” practice will help. Given the lack of attention to building knowledge in school, the system ends up further privileging the kids who are already privileged — those who have highly educated parents and are more likely to pick up sophisticated knowledge and vocabulary at home.
In my view, the intense focus on RCS has completely distorted the teaching of reading and has minimised understanding of the greater role of background knowledge and vocabulary in reading comprehension — something which is only just becoming part of the mainstream discourse amongst primary teachers.
Am I wrong?
