The UK Phonics Check could help reduce teacher workloads

10 Replies

There’s an article in yesterday’s Age newspaper about a proposal from literacy expert Dr Jennifer Buckingham for compulsory use of the UK Phonics check with Australian first grade children. Rather than trying to paraphrase it, I encourage you to read the proposal yourself, it’s in plain English and based on a motza of scientific evidence.

Any teacher, school or interested person can already use the UK Phonics Check. It’s quick, free, simple, downloadable and a useful assessment of early reading skills. Some Australian schools already use it. Click here for the 2016 version.

The test asks children who’ve had about 18 months of literacy instruction to read 20 real words like “chin”, “queen” and “wishing”, plus 20 made-up words like “doil”, “charb” and “barst”.

Since its introduction in England in 2012, the proportion of children passing the Phonics Check has increased each year, and the proportion of children below the expected standard on Year 2 reading tests has fallen by a third. The achievement gap between wealthier and poorer students has also narrowed. Yay to that.

Of course correlation doesn’t imply causation, and testing isn’t teaching. The improvement is probably largely due to the implementation of the 2006 Communication, Language and Literacy Development Strategy, which resourced, trained and required teachers to teach explicit, systematic, synthetic phonics in the early years of schooling.

However, the introduction of the Phonics Check probably helped sharpen everyone’s phonics focus, as well as acting as an accountability measure for government. If you can’t measure it, you can’t manage it etc.

There’s no way to tell whether children who can read real words on a test have memorised them as wholes or are able to sound them out. However, made-up words, which children have never seen before, can only be read successfully by sounding them out.

To successfully read “charb” a child needs to know that it contains three graphemes/spellings, “ch”, “ar” and “b”, match them to the relevant sounds/phonemes and blend them into a word.

Children encounter new vocabulary in books all the time, including nonsense words like “grinch” and “bottersnike” and “muggle”, so children who have grasped the basics of our sound-spelling system don’t blink at being asked to read nonsense words.

To make it absolutely clear which words are made-up ones on the UK Phonics Check, each nonsense word is illustrated with a nice alien, like these:

uk-phonics-check-aliensMost reading problems are the result of poor decoding, and nonsense word reading tests are simply the best way to identify poor decoding skills. Email me if you’d like about a kilometre of supporting references, or see the references in Dr Buckingham’s paper.

NAPLAN starts in Year 3, way too late to identify children needing early intervention. The running record type assessments currently widely used in schools to assess reading skills are far too subjective, not focussed on the things reading research tells us matter most, based on the faulty theory of multi-cueing, and take far too long.

I wish teachers would throw these subjective, time-wasting assessments out the window, and instead use efficient evidence-based assessments like the Castles and Coltheart 2 (Free online! Standardised on Australian children! Writes its own report!) and other free MOTIF tests, the PERA, the YARC or the UK Phonics Check.

Better assessments would immediately reduce teacher workloads. I always want to weep when I see teachers at their desks long after the kids have left and the cleaners are in, pointlessly evaluating error types in order to work out which unscientifically-levelled books to give a child next (don’t start me on levelled books for guided reading), or rushing out the door with a bag full of running records to work on at home.

Better assessments would mean better identification of children in need of extra help, and of what kind of help they need. Fewer kids would fall through the cracks, and intervention resources could be better targeted.

However, the biggest reduction in workloads which could be achieved with better assessments, plus better training and resources in the early years, is for those teaching later in primary school, as well as secondary and even tertiary educators.

The kids who don’t crack the spelling code in their early years generate masses of extra work for other teachers downstream.

They misbehave to distract attention from their difficulties. Of course. Would you want to be the naughty kid, or the dumb kid?

Because they can’t participate effectively in the mainstream curriculum, they generate huge amounts of curriculum differentiation work for teachers.

Anyone who has worked in schools knows that there are many conscientious teachers who put in massive amounts of extra time outside class helping older students whose literacy skills are poor. They come in early, stay back late and generally bust a gut trying to either backfill basic skills that are lacking or compensate for them.

This week I was talking to a young man with serious literacy difficulties who said he got through English in secondary school because one of his teachers would stay back and explain the texts to him, help him understand his homework etc. Perhaps nobody else noticed, and certainly that teacher wasn’t paid any extra. But like so many teachers round the country, she or he was willing to go the extra mile.

I want the proportion of children below the expected standard on NAPLAN to fall by a third. I want the achievement gap between wealthier and poorer students to narrow. If they can do it in the UK, we can surely do it here.

It’s sad that teacher organisations seem to view Dr Buckingham’s proposal as a criticism of teachers, rather than an idea which might help address what literacy statistics tell us is a serious problem, using a quick, inexpensive, evidence-based tool being used effectively elsewhere. It’s sad that they don’t recognise the poor quality of the tests currently widely used, or the amount of teacher time they waste.

There’s nothing more anti-teacher than failing to provide early years teachers with the training and tools they need to teach and assess early literacy according to the best scientific evidence, and to rapidly and successfully identify children requiring extra help.

Today there’s an article in The Conversation by University of Canberra academic Misty Adoniou arguing against the implementation of the Phonics Check, and she was also quoted in the Guardian yesterday in an article called “Researchers warn against further use of phonics testing in schools“.

In The Conversation, Adoniou writes, “The phonics test has been deemed successful because the children get better at doing it over the course of the year”, but my understanding is that most English children only do the test once. Children in Year 2 only do it if they didn’t meet the standard in Year 1, or didn’t do it before, and anyway each year there are different words on the test.

Adoniou also disparages some UK teachers’ practice of teaching children how to read made-up words. I’d like to know how children can be expected to read books like Harry Potter, Dr Seuss, Winnie the Pooh, Lord of the Rings or Lewis Carroll if they don’t have this skill. What about Pokemon cards, road signs, menus, maps, the footy record, train timetables, shop names and brands?

wollongong-train-map greeves-st-nandos
quirkle rogan-josh

I wrote a  2012 blog post about the blurriness of the line between real and nonsense words, if you’re interested. Suffice to say that the point of having an alphabetic code is that you can write anything you can say. You make a new words up and write them down, just like JK Rowling, how cool is that?

Apparently contradicting her earlier suggestion that measurable improvements in children’s literacy in England since the introduction of the Phonics Check are due to test practice effects, Adoniou writes that “six-year-olds in England are getting better at sounding out individual decodable words…What isn’t clear is if they are getting better at reading”.

I find this baffling. Sounding out words is a part of reading. Skilled readers do it quickly and automatically, while novice readers do it slowly and laboriously. What Adoniou says makes about as much sense as saying “six-year-olds are improving their ability to pedal and balance on a bike, but we don’t know whether they are getting any better at cycling”.

Adoniou’s other main arguments against the Phonics Check seem to be:

  • Teachers already collect heaps of data and don’t know what to do with it all (stop requiring teachers to pointlessly collect so much meaningless data, perhaps?)
  • Northern Irish children do well on the international PIRLS reading test, and Canberra children do well on NAPLAN, yet neither of them have “a phonics-only approach to reading”. Whatever that is. There’s precisely nobody who advocates teaching phonemic awareness and phonics without teaching vocabulary, comprehension and fluency. How is the reading performance of Irish or ACT children (where the Phonics Check is not from) an argument against the Phonics Check? Also, I don’t know about Northern Ireland, but Canberra has the second-highest income in the country, and wealthier kids tend to do better on tests.
  • Phonics is only a part of reading. This is true. Everyone agrees. Knowing the road rules is only a part of driving. Does Adoniou propose we stop testing learner drivers’ road rule knowledge?
  • The background to the UK Phonics Check is research done in Scotland, which Adoniou writes was not peer-reviewed, has been critiqued and “the children from the study ultimately did not perform any better than any other school in Scotland’s national standardised reading test in Year 7”. She doesn’t name or link to this research so we can only assume she refers to the 7-year Clackmannanshire study, luckily I have the link here. The study’s report on the Scottish government website says, “The children in Primary 7 comprehended what they read 3.5 months above what would be expected for their chronological age”, which seems inconsistent with Adoniou’s statement that they were not better readers. I can’t check this out for you, because she doesn’t provide a reference.
  • The Phonics Check is a “distraction”, presumably from teaching as usual. Perhaps when we get the PIRLS 2016 results, we might all welcome a distraction.
  • The money would be better spent on Year Four deep comprehension. I’m trying to imagine how children in Year Four who can’t decode text very well will be able to deeply comprehend it, however well they are taught, and I can’t say I’m getting very far.

The teacher unions’ other, more understandable objection to the Phonics Check is that it’s a distraction from the government’s plan to cut Gonski funding in 2017. Luckily those of us who care about equity in education are not going to be so easily distracted, here’s the link to the I Give A Gonski campaign website if you haven’t already seen it.


10 thoughts on “The UK Phonics Check could help reduce teacher workloads

  1. jo Gawne

    Misty Adoniou misses the point of non-word testing. The main purpose of this type of test is to control for words that have been memorized and seen before. It is designed to test whether or not a participant can decode brand new, unseen words. Children need to be able to decode an unseen word before they can connect it to their prior knowledge or the context (see Dahaene, 2009)

  2. Gillian

    thank you Alison – your article is linked in the comments to Misty’s article so hopefully some of those readers will also view it! Effective support teachers are already administering screening assessments but a government endorsed, compulsory screener, normed in Australia, with hopefully a similar in-kind funding model like the UK offered for schools to buy support/resources would be welcome!

  3. Sam

    Thank you Alison for the article.
    I was wondering if you could please clarify when the test would be administered in Australia. Would it be the end of Year One or the end of Foundation? I always advocate for earlier screening, so that intervention can be put in place sooner, but perhaps there is research that suggests reasons for the timing they use in the UK? I would love to hear your thoughts and any links to research that would provide guidance as to best practice in schools. Very much appreciated.

    1. alison Post author

      Sorry for slow reply, I’ve had a nasty virus. The UK Phonics Check is administered in the second half of the second year of schooling, is my understanding. Kids start school very young in the UK so I think most of them are still only six. The test also assumes that all teachers will be using explicit, systematic synthetic phonics with the children since school entry, which is not the case here, but should be. Lots of relevant research is in the three National Inquiries that you’ll find links to in the Research menu item above.

    2. jo Gawne

      The phonics check is administered in Year 1 as students in the UK will have been taught all 44 letter sounds and their basic graphemes by then. It does not assume students come to school with this knowledge. This is the teachers job. If the grapheme phoneme correspondences are not in the curriculum, they may or may not have been taught. The phonics check will highlight this though, and hopefully draw attention to the fact that this needs to be put explicitly in each state’s curriculum (or national curriculum preferably).

    3. jo Gawne

      Apart from the National Inquiries in UK,US and AUST, Hattie (generally accepted in all ed circles now, even whole lang proponents) has clear evidence in using phonics for reading. Louisa Moats, Stanislas Dahaene, Marianne Wolf are among many that have conducted rigorous research on the neuroscience of reading. They all link systematic phonics instruction to success in reading.


Leave a Reply

Your email address will not be published. Required fields are marked *