Fairlands 065 _.jpg

Kirsten Snook  is an English  Teaching and Learning Adviser for  HfL

originally published here November 2015

And this year’s  [2015]  Phonics Check threshold will be…32!

It has not gone unnoticed that the mystery threshold has remained the same for a 4th year running. Why? Well, some investigation into the scope of the test framework (design) has unearthed some interesting areas that we felt you should be aware of; these mean that, even if the numerical threshold remains the same, the difficulty level can easily be ramped up without breaching their own design. This gives STA the ability to gradually increase the complexity via how much a child is being asked to do at once and has implications for pace and progression of phonics teaching, reading, careful use of reading assessment, subject knowledge etc. So, what does this actually have to do with this newsletter’s theme of vocabulary and comprehension? Well, quite a lot.

A couple of words from the phonics screening check that schools have asked us about are used to illustrate some points. Many thanks to those schools who, through asking searching questions, have helped us to further support you.


“It should be noted that where items contain a number of the different features listed above, decoding will become more difficult. It will become less likely that a child working at the minimum expected standard will be able to decode such items appropriately. For example, a child will be less likely to decode an item containing both a consonant string and a less frequent vowel digraph, than an item with a consonant string but a frequent, consistent vowel digraph.”

(p6, para 5 Assessment Framework)

They acknowledge that words containing two characteristics (e.g. split digraph ‘i_e’ and soft ‘c’) at the same time are more challenging and that children are less likely to accurately decode these words. They don’t rule out including two target concepts at once though, in the guidance. Teaching children about analogy can be a useful strategy to build into the diet, as can ensuring they try out different pronunciations when decoding real words to ensure it makes sense (e.g. short and long ‘i’). Schools are also noticing children coming through with insecure knowledge of letter names (see Development Matters, 40-60months) which also helps equip them with more alternative pronunciations – linking these things can be a helpful prompt (“Does it look right and sound right?”).

The main point

here is about focusing more on lifting above the minimum expected standard, and moving towards securing, which does have other implications for diet, progression, pedagogy etc.


“It is necessary to start with easier words in section 1 to make the phonics screening check accessible and to provide some information to teachers if their children are unable to decode relatively simple words. However, the words at the end of the phonics screening check are around the level of difficulty we expect children to reach by the end of Year 1. These items will provide more information on whether children are working above or below the expected standard.”

(p8, section 3.1 para 4, Assessment Framework)

The perception that use of the suffix ‘-ing’ is a phase 6 item is slightly erroneous – phase 6 was always more about the shift into orthographic spelling strategies (for accuracy), than reading.  Adding ‘-ing’ is something that typically occurs in age-related books from early Y1 onwards (at least – and often before).  The new NC has a statement for Y1: ‘read words containing taught GPCs and -s, -es, -ing, -ed, -er and -est endings’. As the Y1 spelling specifies ‘where no change is needed to the root word’, one must infer that the reading expectation should include words where change both is and is not needed to the root, which indicates a progression of reading being slightly easier than spelling.  We must remember  that this is the more challenging end of the check though, and by this point we are usually finding that only those children beyond the minimum expected standard for the age accurately decode these words.

Some words that came up towards the end of the check also were felt to be unrealistic for children’s reading books at this stage. The link between the two parts of the Simple View of Reading (SVoR) becomes apparent as even if they have not read these words before, they can use their decoding together with previously-encountered oral vocabulary to make sensible adjustments. This seems to be a subtle way of encouraging schools to broaden children’s vocabulary, make links between the SVoR dimensions and – ultimately – have children reading more widely and often than currently. Scheme books lend a good platform but leave the wider reading diet a bit undernourished:

“All real words will be checked for frequency in the Children’s Printed Word Database maintained by the Department of Psychology, University of Essex. All real words in the check will be found in the database and the check will contain between 40 per cent and 60 per cent of real words that are low frequency, defined as fewer than 20 occurrences per million words in the database.”

(p17, section 4.1 para 3, Assessment Framework)

Bearing in mind that 20 of the 40 words will be real words, this means that 8-12 of the 20 real words could be low-frequency… a real test of sheer reading mileage, re-reading for fluency and NOFAN best-practice assessment principles (e.g. ensuring guided and independent reading books really are closely-matched to their current decoding ability, 90%+ accuracy/ no more than one word in ten left as an error, and any new words can be scooped up with support). New National Curriculum end-of-Key-Stage tests show a real emphasis on the need for increasing children’s vocabulary, e.g. through the use of rich shared texts and the read-aloud programme.

It will become more essential than ever to ensure that the ‘end of the day story’ doesn’t get squeezed out of the day at KS1, and continues at least 3 times per week at KS2. (I absolutely love the phrase “3 o-clock stop”!) Without this, and whole-class shared reading texts exposing children to higher-level vocabulary, they are limited to the vocabulary they can largely decode already. In turn, this can a) stunt their oral vocabulary and b) let children think that words just have to be decoded and not then checked for understanding/making sense (passive reading behaviour). Throughout the new NC there are references to understanding what has been read to/by them. The draft performance descriptors made reference to this and the ability to self-correct, and our new set of Guided Reading objectives for KS1 will also lace this theme and the new heightened expectations throughout.

In short, we have to stuff them full of words and expand their oral vocabulary, so that when they encounter new words they use their initial decoding (phonics first) but then also use the other half of the SVoR (language comprehension) to check and change pronunciations if the word doesn’t make sense. ‘Diving’ is a perfect example of this. Picture the child who initially reads it as ‘divving’, quickly realises that it doesn’t make sense, sees there is no alien next to it and so it should make sense, and then starts to try alternative pronunciations (letter names = Development Matters 40-60 months). That is the very sort of active reading behaviour we need to see happening in tandem with decoding, as a fleshed-out view of Y1 end-of-year expectations, if Year 2 are to appropriately handle the c.45% direct retrieval and c.30% inference weightings in the 2016 key stage assessments.


For the Children’s Printed Word Database, see: http://bit.ly/1Oyt3WF

For banks of example words that feature alternative pronunciations and graphemes, see p134 onwards of ‘Letters and Sounds’: http://bit.ly/1JRCiK7

Phonics Assessment Framework, the STA’s published check design booklet (note: all emboldened phrases in the quotes above taken from here are the author’s own emphasis): http://bit.ly/1jGEnmk