Opinion about the Year One Phonics Screening Check

Moderators: Debbie Hepplewhite, maizie, Lesley Drake, Susan Godsland

Post Reply
User avatar
Debbie Hepplewhite
Administrator
Posts: 3663
Joined: Mon Aug 29, 2005 4:13 pm
Location: Berkshire
Contact:

Opinion about the Year One Phonics Screening Check

Post by Debbie Hepplewhite » Thu Oct 11, 2012 10:37 pm

I think it is important to discuss in some depth the advent of the Year One Phonics Screening Check in England including the media coverage of responses to it.

In the latest edition of the 'teach PRIMARY' magazine, on side 5, there is an item taking up half a side with this:
Phonics check checked

NO

If you're a Y1 teacher, how useful did you find the Phonics Screening Check administered at the end of last term? Following the pilot scheme, the Government announced that 43 per cent of teachers said they had been able to pinpoint six-year old pupils with reading problems of which they were not previously aware. But following national roll-out, a survey conducted by three of the main teaching unions found nine in 10 year one teachers said the phonics check did not tell them anything new about the reading ability of their pupils. So what happened?

Feedback from the NUT, NAHT and ATL survey revealed that many teachers found fluent readers were confused by made-up words such as 'strom' as they are so close to real words ('storm') and children assumed they were misprints and tried to make sense of them. A significant proportion, 43%, also commented they had felt under additional pressure to teach synthetic phonics in the week before the checks to the detriment of other literacy activity.

Given teachers frustration, Christine Blower, general secretary of the NUT has called for the phonics check to be scrapped. For her, the survey has provided "stark" evidence that schools are being made to squander money on what they know to be an unreliable 'progress report'.
I cannot see any named author to this piece.

I've emailed the editor, Joe Carter, about it.

Please add your comments, whatever they may be, so that we can have a full and proper discussion about any and all aspects the Year One Phonics Screening Check.

chew8
Posts: 4187
Joined: Sun Nov 02, 2003 6:26 pm

Re: Opinion about the Year One Phonics Screening Check

Post by chew8 » Thu Oct 11, 2012 10:59 pm

If all teachers already teach phonics, as the unions often claim, why would 43% feel 'under additional pressure' to teach synthetic phonics (sounding-out-and-blending phonics) in the week before the screening check?

Jenny C.

Toots

Re: Opinion about the Year One Phonics Screening Check

Post by Toots » Thu Oct 11, 2012 11:45 pm

OK. I am opposed to the phonics screening check for the following reasons:
1. It encourages teaching to the test with children learning to decode lists of nonwords using time when children should be learning to decode real words, from which they can move onto reading real words in quality texts in order to learn.
2. Linked to the above, it over-emphasises the phonics aspect of literacy and risks a neglect of other aspects which it is important to cover alongside phonics teaching, such as listening and responding to story, reading for purpose, developing oral language and vocabulary.
3. The test is not fit for purpose as it uses a combination of nonwords and real words. Judging from the content of the sample tests the nonwords section tests phonic decoding skills, but in the real words section phonically plausible but incorrect responses are not allowed. It is therefore not a test of phonic decoding.
4. Practice sessions given to children using nonwords risks exposing pupils to illegal spellings and letter strings, by teachers unaware of such subtleties, when they should be developing an instinct for genuine English spelling patterns.
5. Children who have developed instant recognition of words and a mature reading style may fail the test by misreading nonwords as real words, and then their time will be wasted with inappropriate teaching of phonic decoding which would represent a backwards step.
6. The fact that 58% passed the phonics test this year, when much higher percentages pass the primary school reading assessments, would indicate that children's ability to pass the test does not correlate with their success in reading. That would, in turn, indicate that measures to increase time spent on phonics in order to achieve a higher pass rate has no guarantee of increasing success in reading.

JAC
Posts: 517
Joined: Tue Nov 15, 2005 1:51 am

Re: Opinion about the Year One Phonics Screening Check

Post by JAC » Fri Oct 12, 2012 12:27 am

Feedback from the NUT, NAHT and ATL survey revealed that many teachers found fluent readers were confused by made-up words such as 'strom' as they are so close to real words ('storm') and children assumed they were misprints and tried to make sense of them.
To suppose that an error on this particular word is not a decoding error, but rather the result of the child's presupposition that the word is a misprint, I find highly improbable. To have an understanding that government documents contain misprints and need correction - I don't think so in 6 year olds. Much more likely that they simply confused similar-looking words; easy to do, and the point of the test is to discriminate which children do not/ can not accurately decode, epecially similar looking words.
Adult, expert readers may well consider the possibility of adjusting for misprints, but we cannot assume children, in this test situation, have done the same thing. It is better to make a conservative judgement and assume it is an error, which can be addressed.

geraldinecarter
Posts: 993
Joined: Thu Jan 06, 2011 6:40 pm

Re: Opinion about the Year One Phonics Screening Check

Post by geraldinecarter » Fri Oct 12, 2012 12:34 am

Toots - are you sanguine about using mixed methods/multi-cueing etc? Do you ever consider that since the mid to late 1960s there has been an enormous tail of underachievement- 20%+ of children unable to access the secondary curriculum. It is very concerning that unions never speak up for this enormous number of children whose lives are shattered by their loss of confidence and limited life chances.
It's interesting reading comments from teachers who understand how to teach all young children to read - ie no stress, no big deal, no need to 'teach to the test etc' - children are taught how to link sound to letter right at the beginning of their reading instruction. It seems very patronizing to children when it is stated that they can't understand what is expected when they see a large 'pretend' figure alongside a word . Not too much to ask a teacher to spend 2-3 minutes a day giving children a picture against an (orthographically correct) pretend word during the weeks leading to the check, I would have thought.
Teachers who have been well trained and understand how to teach children to read don't need to make a meal of synthetic phonics. Intelligent teachers should, of course, make sure that fluent readers have appropriate literature to read.

Jenny observed a while ago:
in order to be sure that you are measuring decoding ability (rather than the recognition of words already familiar in their printed form) you have to be sure that the items can’t be familiar to any of the testees. In fact when children are in their first year or two of school, even real words are often unfamiliar in their printed form, but you can’t be sure which children have already encountered which words so it’s safer to level the playing-field by using non-words, hence the use of non-words in the Year 1 decoding test. If children have been well taught, they shouldn’t be fazed by these as they should be used to sounding out and blending whenever they encounter an unfamiliar string of letters.

john walker
Posts: 369
Joined: Sun Nov 16, 2003 10:46 am
Location: Buckingham
Contact:

Re: Opinion about the Year One Phonics Screening Check

Post by john walker » Fri Oct 12, 2012 9:15 am

OK. I am opposed to the phonics screening check for the following reasons:
1. It encourages teaching to the test with children learning to decode lists of nonwords using time when children should be learning to decode real words, from which they can move onto reading real words in quality texts in order to learn.
This is a complete red-herring which, I believe, stems from a fundamental misunderstanding of what is happening when children learn to read.
Whether it’s a real word or a nonsense word, children should be able to link sounds together accurately and produce a word. In the past, many people on this forum and elsewhere have made the point that non-words pop up in all sorts of written texts. Notwithstanding the fact that most children, by the age of six years will have a vocabulary of around 10,000 words, there are still words which crop up in everyday texts that they will not have come across before: longer, polysyllabic words (many of which contain ‘non-word’ syllables), less frequently encountered words, more technical words, and so on.
It seems obvious to me that if a child cannot decode, then they are not going to be able to read efficiently.
2. Linked to the above, it over-emphasises the phonics aspect of literacy and risks a neglect of other aspects which it is important to cover alongside phonics teaching, such as listening and responding to story, reading for purpose, developing oral language and vocabulary.
There’s absolutely no reason at all why attention to the teaching of the essential ability to decode successfully should precipitate a neglect of ‘other aspects of literacy’. This is the kind of calumny propagated by anti-phonics extremists like Rosen, Blower and company.
3. The test is not fit for purpose as it uses a combination of nonwords and real words. Judging from the content of the sample tests the nonwords section tests phonic decoding skills, but in the real words section phonically plausible but incorrect responses are not allowed. It is therefore not a test of phonic decoding.
I don’t believe for a minute that children who could decode successfully were ‘thrown’ by real words and read them as non-words. In order to make this claim, you need to provide evidence and, as far as I have heard, no one has produced any such suggestion before.
Where you might be on stronger ground is in claiming that children ‘normalise’ non-words. This is certainly the case, though there are various explanations for why this might happen. One is that, for example, children who read ‘storm’ for ‘strom’, simply can’t decode accurately! Another is that young children make mistakes. Another is that some children who entered school with low mental ages and were not ready to begin formal learning haven’t had time to catch up.
4. Practice sessions given to children using nonwords risks exposing pupils to illegal spellings and letter strings, by teachers unaware of such subtleties, when they should be developing an instinct for genuine English spelling patterns.
Again, where’s the evidence for this? Where are the illegal letters strings and spellings in the test? As for the much maligned ‘strom’, you’d have to be able to read it accurately to read the word ‘strombus’ or ‘stromboid’, or ‘strombuliform’ – not much good telling someone that they saw a snail with a stormbus in its back, is it? Or that one is going to Stormboli for our holiday this year.
5. Children who have developed instant recognition of words and a mature reading style may fail the test by misreading nonwords as real words, and then their time will be wasted with inappropriate teaching of phonic decoding which would represent a backwards step.
In my opinion, there is no such thing as ‘instant recognition of words’ in the sense that you, Toots, seem to be using the term. To me, instant recognition of words means automaticity, i.e. when the decoding process is so good it takes place faster than the conscious brain is able to register it – under the level of conscious attention. This is the view of someone who has worked for most of her professional life on studying reading and the brain – Diane McGuinness, by the way. I’m also not claiming that to be able to reach automaticity one must be taught phonics but that’s a different question.
There is a second answer to the point you make though: Yes, there are children being taught phonics who can read really very well. I hesitate to say perfectly because most, in KS1 at least, can’t read Dickens yet. However, what good quality phonics teaching can do for them is to demonstrate conceptually how the alphabet works so that they can generalise it. It can also have a highly providential effect on such children’s spelling, as I happened to see yesterday in a school.
6. The fact that 58% passed the phonics test this year, when much higher percentages pass the primary school reading assessments, would indicate that children's ability to pass the test does not correlate with their success in reading. That would, in turn, indicate that measures to increase time spent on phonics in order to achieve a higher pass rate has no guarantee of increasing success in reading.
Why anyone would want to draw firm conclusions from SATs tests, I really have no idea. In terms of teaching to the test, I would venture that in some schools they do nothing else but for almost a whole year before the tests are taken, quite apart from some of the dubious practices that go on in preparing for the tests. Look at the astonishing results we’ve just had at KS2. Then look at the screening tests being conducted in secondary schools!
Oh, and one more thing! I have never taught phonics to a single child with the aim of getting them to pass a test. My purpose has always been to enable them to access any text without exception.
John Walker
Sounds-Write
www.sounds-write.co.uk
http://literacyblog.blogspot.com

chew8
Posts: 4187
Joined: Sun Nov 02, 2003 6:26 pm

Re: Opinion about the Year One Phonics Screening Check

Post by chew8 » Fri Oct 12, 2012 9:40 am

I drafted the following before seeing John's posting above. Sorry if there's overlap, but I haven't time to redraft!
Toots wrote:1. It encourages teaching to the test with children learning to decode lists of nonwords using time when children should be learning to decode real words, from which they can move onto reading real words in quality texts in order to learn.
The DfE document ‘The phonics screening check: responding to the results’ attempts to counteract this, stating that ‘It is not necessary to practise pseudo-words, if your pupils did not do well on them. The knowledge and skills pupils need to decode them are exactly the same as they need for any unfamiliar word’ (p. 7). This document goes on to point out that it is for assessment purposes that non-words are useful: ‘Pseudo-words, however, are useful for assessment which is why you will find them in systematic, synthetic phonics programmes, as well as in the screening check’.
Toots wrote:2. Linked to the above, it over-emphasises the phonics aspect of literacy and risks a neglect of other aspects which it is important to cover alongside phonics teaching, such as listening and responding to story, reading for purpose, developing oral language and vocabulary.
Yes, but phonics is an aspect of literacy which has been seriously neglected hitherto and that problem really does need to be addressed. It is also an aspect which is particularly crucial in Reception and Y1.
Toots wrote:3. The test is not fit for purpose as it uses a combination of nonwords and real words. Judging from the content of the sample tests the nonwords section tests phonic decoding skills, but in the real words section phonically plausible but incorrect responses are not allowed. It is therefore not a test of phonic decoding.
One solution would be to have the whole thing consisting of non-words, and some teachers have actually suggested this. The fact that ‘phonically plausible but incorrect responses’ to real words are not allowed is a slight problem but not, in my experience, a major one – it's unlikely to make the difference between reaching the threshold and not doing so.
Toots wrote:4. Practice sessions given to children using nonwords risks exposing pupils to illegal spellings and letter strings, by teachers unaware of such subtleties, when they should be developing an instinct for genuine English spelling patterns.
Teachers need training about this. On the other hand, though, even real children’s books sometimes contain spelling patterns which are illegal in English, as also do foreign names which children encounter. Although I agree that children should be exposed mainly to legal English spelling patterns, I would not want to send the message that only patterns which are legal in English are decodable.
Toots wrote:5. Children who have developed instant recognition of words and a mature reading style may fail the test by misreading nonwords as real words, and then their time will be wasted with inappropriate teaching of phonic decoding which would represent a backwards step.
Have you had practical experience of good readers misreading nonwords as real words Toots? If so, what percentage of good readers did this? I ask because I’ve observed various versions of the screening check being administered to several hundred children or have administered them myself, and found this problem to occur only rarely.
Toots wrote:6. The fact that 58% passed the phonics test this year, when much higher percentages pass the primary school reading assessments, would indicate that children's ability to pass the test does not correlate with their success in reading. That would, in turn, indicate that measures to increase time spent on phonics in order to achieve a higher pass rate has no guarantee of increasing success in reading.
Research is being done on the correlation, specifically on the correlation between performance on the screening check and performance in the Key Stage 1 SATs a year later so we should have some better information in due course.

Jenny C.

SLloyd
Posts: 27
Joined: Tue Dec 02, 2003 6:05 pm

Re: Opinion about the Year One Phonics Screening Check

Post by SLloyd » Fri Oct 12, 2012 9:46 am

Many years ago I was asked by an Austrian researcher to test 90 school children on a non-word reading test. I told the children that the words did not make sense and that they must work them out. There was no practicing of non-words. I gave one word as a demonstration and that was all. None of the children were bothered by the task. These children had been taught from the first week of school to always work words out by blending the letter sounds.

In my experience, a child who reads ‘strom’ as ‘storm’ is used to memorizing and guessing words from some letter clues and is not sufficiently used to always processing the words by blending the letter sounds from left to right. I am certain that this particular mistake would not have happened with a child who had been taught with pure synthetic phonics from the beginning.

When you look at older struggling children their reading problems are nearly always decoding problems. These children struggle to read the words. Then when you test them on their letter-sound knowledge you find that there are enormous gaps, especially the digraphs.

I would like teachers to recognize that the phonics check is showing up a problem and to check how well those children know the letter sounds.

SLloyd
Posts: 27
Joined: Tue Dec 02, 2003 6:05 pm

Re: Opinion about the Year One Phonics Screening Check

Post by SLloyd » Fri Oct 12, 2012 9:52 am

Oops - sorry .... no practising of non-words!

Toots

Re: Opinion about the Year One Phonics Screening Check

Post by Toots » Fri Oct 12, 2012 9:56 am

JAC, I agree that the error of reading 'storm' for 'strom' is a decoding error. I don't see how it can be called anything else. And I agree that a 6 year old is unlikely to think in terms of misprints, being novice readers. However, a 6 year old asked to read words will not necessarily go into 'phonic decoding mode' if not in the habit of doing so for their general reading. Once a person knows many words and finds him/herself in the position of being able to read for meaning without stopping to decode, this will be the reading behaviour adopted. As adults we may misread, but we go back to correct our mistakes. This is an aspect of reading for meaning. Why would children who can read most texts they encounter be any different? Of course with the phonic test there is no context, so no imperative to self-correct. Were the phonic test combined with sentence or text reading we might have more interesting results, which would more accurately measure children's reading behaviour and decoding skills. The fact that many teachers found that 'good readers' tended to misread nonwords in this specific way would suggest that my explanation carries more logic than the explanation that good readers are bad decoders. Perhaps it depends on how much credence and respect you give to the teachers' judgements. A lot of dispute on this matter, I find, comes down to that issue.

Yes, Geraldine, children are taught to link sounds to letters right at the beginning of their reading instruction. Surely the aim of this is so that children will be able to sound out words in order to learn and recognise them on repeated viewing. Eventually they will recognise them automatically without linking letters to sounds, instead linking letter strings to words. I believe that the misreading of the nonwords by good readers may have been a symptom of this stage of reading proficiency, not a sign of inability to decode. You bandy around stuff like 'intelligent', 'well-trained' etc. - best to keep off these sort of value judgements, I think. Do you mean 'in agreement with SP theory' and ' trained in SP'? Whether that is the same as 'intelligent' and 'well-trained' has to be a matter for judgement.

John, I will get back to you. Oh heck, and all the others too! :roll:

volunteer
Posts: 755
Joined: Wed Nov 16, 2011 12:46 pm

Re: Opinion about the Year One Phonics Screening Check

Post by volunteer » Fri Oct 12, 2012 10:11 am

I do not object to the principle of a phonic decoding check. I personally would like a more rigorous one. My DD scored 40/40 but in reality when she is reading real books she reads some polysyllabic words inaccurately that she should be able to decode for herself. She is guessing without reading the word in full from left to right.

The phonics work at school does not go far enough to help her with this.

I agree with Toots that children should not be exposed to illegal spellings in non-words. It is sad if teachers produce materials with illegal spellings in them. There are many ancient remedial reading programmes available on the web (e.g. Kirk Legge drills) which only contain real words but many are not in the vocabulary of young children so these would provide "safe practice". I am sure that many current phonics programmes mostly contain enough real and well thought out non-words for teachers not to have to invent their own poor lists containing illegal spellings.

It would seem that the biggest problem still is teacher knowledge rather than the actual test if illegal spellings in non-word practice is a concern.

I still wonder why the DfE produced their own test rather than use a commercially available one e.g. the new GL assessment test of word recognition processes.

I would hope however that over time the national test becomes redundant because the overwhelming majority of children would be receiving thorough phonics tuition at school; the information that the national test gives the school would be inferior to the information they already have on each child through the formative and summative assessment processes that their school phonics teaching programme uses.

OFSTED needs to be cannier in their inspection approach too. Looking at APP levels for reading and NC levels for reading is very confusing. To give you an example, my DD is in year 2 and was on teacher-assessed 2a per APP at the start of year 2, and achieved 40/40 in the year 1 test - but I can tell you that she is not as accurate as she could be, and she is never going to receive any further help with this at school. That is my job! APP levels and NC levels are a mish-mash of things and different teachers will give different gradings to the same child. If accurate decoding is important (which it is at all stages in a reader's life) it needs to be measured objectively, and taught thoroughly.

The other way that OFSTED can be misled is by not knowing when they observe a phonics lesson if those children could actually have progressed faster to the stage at which they currently are. You can observe a lovely, well taught phonics lesson, but fail to appreciate that really those children could have done that same work maybe even a year or two ago. The same goes in every subject really. Our school is great at underassessing children on entry so this is not clear.

Perhaps too much measuring does tempt people to cheat too much - -- I await with interest the research into the spike and whether this is because of some massaging of data or to do with the nature of the items in the test, or a bit of both.

chew8
Posts: 4187
Joined: Sun Nov 02, 2003 6:26 pm

Re: Opinion about the Year One Phonics Screening Check

Post by chew8 » Fri Oct 12, 2012 10:23 am

volunteer wrote:I still wonder why the DfE produced their own test rather than use a commercially available one e.g. the new GL assessment test of word recognition processes.


The main reason is probably that schools could get access to commercially-available tests and could coach the children on the items.

My hunch is that the 18 different versions used in the 2011 pilot will enable the DfE to produce different tests of comparable difficulty in future years.

Jenny C.

kenm
Posts: 1495
Joined: Sat Dec 17, 2005 5:19 pm
Location: Berkshire

Re: Opinion about the Year One Phonics Screening Check

Post by kenm » Fri Oct 12, 2012 10:32 am

In 'Phonics check checked', someone wrote:Feedback from the NUT, NAHT and ATL survey revealed that many teachers found fluent readers were confused by made-up words such as 'strom' as they are so close to real words ('storm') and children assumed they were misprints and tried to make sense of them.
Anyone who turns "strom" into "storm", fluent or not, in this context is a poor reader, so the teacher needs to be told that his/her previous assessment was wrong. There are many pairs of mutual anagrams that need to be distinguished; "gaols"/"goals", "palsy"/"plays", "trail"/"trial", "rials"/"rails, "psalm"/"plasm"/"palms", "parts"/"prats" and "sprat"/"strap" all occur on two pages in my anagram dictionary, so that I estimate about 90 similar pairs of five-letter words in English.
Last edited by kenm on Fri Oct 12, 2012 2:15 pm, edited 1 time in total.
"... the innovator has as enemies all those who have done well under the old regime, and only lukewarm allies among those who may do well under the new." Niccolo Macchiavelli, "The Prince", Chapter 6

chew8
Posts: 4187
Joined: Sun Nov 02, 2003 6:26 pm

Re: Opinion about the Year One Phonics Screening Check

Post by chew8 » Fri Oct 12, 2012 10:35 am

Having said what I said above, I should mention that people involved in developing the new GL Assessment Diagnostic Test of Word Reading Processes were also involved in developing the Y1 screening check. It's possible that they will do research on the correlation. I've administered the DTWRP to a Y1 class for whom I also know screening check results and have passed on both sets of results.

Jenny C.

volunteer
Posts: 755
Joined: Wed Nov 16, 2011 12:46 pm

Re: Opinion about the Year One Phonics Screening Check

Post by volunteer » Fri Oct 12, 2012 10:40 am

I've purchase the GL assessment test for myself. I quite like it, but it is a shame that it does only have one version. If they were able to do a UK schools version that changed every year and was procured by the DfE this would be fantastic, but it's probably very difficult to do this under procurement rules ..... and imagine if a decoding check was rolled out for years 1 to 8 or so (can't remember the upper age of that test) ........ just how much hell would break loose!

Post Reply

Who is online

Users browsing this forum: No registered users and 50 guests