This week, researchers from IOE, UCL’s Faculty of Education and Society published a new paper about reading.
Reading wars or reading reconciliation? A critical examination of robust research evidence, curriculum policy and teachers’ practices for teaching phonics and reading’ is an interesting and ambitious paper, but not everyone will be convinced by its findings.
It begins with the differentiation of three approaches to literacy - systematic phonics, whole language and balanced literacy.
“Balanced” is a slippery term - all teaching involves different aspects and no doubt the majority of teachers believe that they are striking a “balance”, while in practice they may be doing very different things from each other.
However, the paper’s authors, Dominic Wyse and Alice Bradbury set up “balanced literacy” as something more specific: an approach that involves careful and systematic linking between whole texts and phonics. Approaches that don’t make this link are not classed as balanced literacy.
This leads to the surprising argument in the paper that Ireland uses a whole language approach - whereas the Irish Department of Education view the same approach as “balanced”.
New phonics research: interpreting the findings
There are a couple of other general areas that I think it is worth picking up on when considering how we interpret the paper’s findings.
Firstly, the authors don’t make a clear distinction between early literacy teaching (in Year 1 and Year 2) and later literacy teaching (Year 3 onwards). A key part of the “phonics first and foremost” approach is that most learners should be able to move on from phonics tuition during Year 2 or Year 3. That is when the focus on whole texts and comprehension begins (though, of course, there should be book reading before that).
Focusing research only on Year 1 and Year 2 does focus on a point in literacy teaching when there is a lot of phonics teaching, to the exclusion of comprehension to a certain extent, because learners need to be able to read individual words to read whole texts. We need to talk a lot more about what happens after Year 2.
A second point to make is that the authors do not address struggling readers or “at-risk” readers at all. In fact, in their qualitative review, they exclude papers dealing with at-risk readers. This is disappointing because these are exactly the group who are particularly affected by the approach to literacy teaching they receive.
Systematic phonics approaches are particularly useful to help at-risk readers. Typical learners can often “pick up” phonics quite easily, while at-risk readers particularly benefit from clear, sequenced, systematic learning as in the synthetic phonics approach.
So, we have an approach that does not harm anyone’s reading outcomes and helps those most at risk. And bear in mind that at least 20 per cent of each class could be considered at-risk readers, whether because of learning difficulties, English as an additional language or myriad other factors.
That is not to say that struggling readers should continue to be taught phonics indefinitely. No teaching approach works for every child all of the time. Phonics works for most children most of the time, but often other approaches are needed.
Finally, the authors argue that adopting the “Simple View of Reading” (a theory that a student’s reading comprehension can be predicted by multiplying their skill in decoding by their oral language comprehension) has led to a separation between phonics and language comprehension, whereas these skills need to be combined in successful reading.
I agree that yes, they do need to be combined, but only after each has been established in its own right. If you give young children all the relevant information all at once, they are likely to be overwhelmed.
Practising skills in isolation so that they become familiar and automatic, then combining them later, is more likely to lead to success in the long term.
Why we need to examine the evidence base
In addition to these points, I would also raise questions about two sections of the evidence base that has been drawn upon.
The authors include an analysis of data from the Programme for International Student Assessment (Pisa) tests, comparing the UK to six other English-dominant nations that perform highly in the Pisa league tables.
However, in most cases, the differences in average scores are not significant between different countries. In 2018, Ireland and Canada outperformed England, and England, Australia, the US and New Zealand did not differ significantly.
Meanwhile, in England, our ranking in English has risen significantly over the past four cycles.
There’s also an issue with assuming that whole country differences in reading outcomes at age 15 can be attributed to changes in teaching practice 10 years earlier. This idea is tenuous at best. If anything, Pisa data shows the importance of broader factors such as a country’s GDP, good school attendance, well-qualified teaching staff and so on.
In terms of the paper’s qualitative review, the authors review a meta-analysis of 55 different studies. Meta-analyses aim to create more powerful analysis by combining similar studies. The idea is that combining findings from a range of different contexts provides an opportunity to draw generalised conclusions that can be applied (with caveats) to a wide variety of situations.
The qualitative review in this paper, in contrast, seems to aim to select only papers that describe the situation faced by class teachers in England. There are no papers that meet these exact criteria, but there are some that meet some of the criteria. I didn’t really feel that this qualitative review added to the several meta-analyses already available, including the original meta-analysis, conducted in 2016, which concluded that phonics and comprehension based interventions both led to long term improvements in literacy.
From the qualitative review, the authors draw the following conclusions:
“Phonics teaching is likely to be effective if:
1. implemented with children aged 5 to 6
2. carefully connected with the reading of whole texts, both decodable and real books, including a focus on reading for meaning, in all lessons.
3. undertaken during the course of not more than one whole school year featuring several lessons a week…”
It’s not clear to me where the evidence for these claims have come from. Maybe if the authors had included the studies on at-risk readers they would have found examples where it is useful to teach phonics for more than one school year.
Overall, the paper is an interesting read, but I don’t feel the authors successfully backed up their claims with evidence. They dismiss the extensive research with at-risk readers as not relevant to mainstream classrooms, whereas I believe this research is hugely valuable. In most cases, effective teaching for at-risk children is effective teaching for everyone.
On the other hand, I think the paper does highlight how much we do agree on in literacy research. We agree that reading comprehension is the long term goal of literacy tuition. We agree that children need good quality phonics teaching and that they also need access to high-quality children’s literature. Our disagreements are more about how and when these skills should be integrated.
Julia Carroll is a professor of child development and education at Coventry University
