Learning 'exception words' as wholes.

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

Post Reply
User avatar
maizie
Administrator
Posts: 3121
Joined: Sun Mar 14, 2004 10:38 pm
Location: N.E England

Learning 'exception words' as wholes.

Post by maizie » Mon May 30, 2016 12:23 pm

Learning words as ‘wholes’.

I am getting a bit worried about the tendency of some academics to support the teaching of some English words as ‘wholes’ while teaching systematic synthetic phonics and it being advocated in books about teaching phonics. I am worried about this on two levels.

Firstly, from my experience of working with KS3 struggling readers who were taught by the old UK National Literacy Strategy (NLS). Being ‘strugglers’ at primary school their phonic knowledge was always very incomplete and they had spent a great deal of time (as evidenced from their primary Individual Education Plans [IEPs]) learning to ‘sight read’ the first 45 High Frequency Words (HFWs) as set out by the NLS for Reception and Year 1 children. Although, on the whole, these pupils were very secure on their reading of most of these words (and not much else) I often found that they confused very similar words, such as some/same, come/came and were prone to reversals such as was/saw, of/for. I tended to attribute these errors to the fact that teaching a word as a ‘whole’ diminishes the importance of the internal detail of the letters and letter order within the word and that it does not place any emphasis on the need to ‘read’ words from left to right. I may have been wrong to do so but it seemed like a very logical conclusion to draw.

I would also surmise that the very fact that this requirement (to learn the first 45 HFWS) appeared year after year on pupils’ IEPs indicated that they found these words much more difficult than did their peers.

These pupils usually represented between 15 - 20% of our intake (from a large number of feeder schools) so they were not an insignificant proportion of our intake. My contention here is that learning even a few words as ‘wholes’ is difficult for a significant number of children and that it seems to set them up for inaccurate reading of a number of common words; thus impairing their ability to comprehend what they read or causing them to misunderstand what they read because they have substituted similar looking but not synonymous words.

Secondly, there is no real consensus on what is an ‘irregular’, or 'exception', word. I have seen them defined by academics as words which do not conform to the main spelling rules of English. This immediately prompts questions such as: What are the main spelling rules of English? Who decides them? Is there a definitive list?

I have seen estimates of the ‘regularity’ of English spelling which range from 85% to 95%. Given that the English lexicon is well in excess of 250,000 words (up to 1,000,000 by some estimates) that 10% difference represents an awful lot of words. Even at the higher estimate that leaves some 12,500 words out of 250,000 potentially to be learned as ‘wholes’ if the academic definition of ‘irregularity’ is the deciding factor ( if my maths is correct!).

Given the rather fluid understanding of what constitutes ‘irregular’ it seems to give permission to teachers, who may already be sceptical of systematic synthetic phonics (SSP) instruction to teach an unnecessarily large number of words as ‘wholes’; thus disadvantaging a significant number of slower to learn pupils.

On the other hand, it may be argued that, over all, the greater attention to sounding out and blending all through the word from left to right which is at the core of SSP instruction may help these children to generalise these skills to words taught as wholes. As I have not worked with children who have been taught under the latest guidance I have no way of telling that this might be so.

I cannot help feeling, though, that the method of teaching such words which is advocated in UK SP/LP programmes, of still teaching these words through the decoding and blending route (most of the word being easily decodable but with an unusual or unique letter/sound correspondence) may be the safest way of ensuring that more children learn to read words accurately and there is no confusion over which words have to be taught/learned/read as ‘wholes.

P.S Comments welcome.
Should you wish to respond registering is easy, free, and doesn't put you on a mailing list

chew8
Posts: 4150
Joined: Sun Nov 02, 2003 6:26 pm

Re: Learning 'exception words' as wholes.

Post by chew8 » Mon May 30, 2016 9:32 pm

I can only go on what I see in my voluntary work – that has been with 80-odd Y3 children and 30 Y1s this year.

By and large, I think these children have received the kind of teaching which is now fairly typical – there’s much more emphasis on sounding-out-and-blending, but ‘sight’-words and guessing from pictures and context haven’t gone away completely. Nevertheless, I think it’s much more usual than it used to be for children to know, without being reminded, that they should tackle unfamiliar words by sounding and blending. This is now true of even the weakest Y3 children I work with, whereas it wasn’t in the past.

I have to be a bit tentative in what I say about words often taught as ‘sight’ words, as I’m relying on subjective impressions rather than objective checks. I think children are now better than they used to be at reading these words – e.g. I don’t think the ‘was’/’saw’ type of confusion arises so much. If this is indeed the case, the reason may be that although ‘was’ may still be taught as a ‘sight’ word, children are now more used to processing the letters in other words from left to right, so there’s probably an element of that even in their approach to ‘sight’ words.

One thing that may be relevant is that the National Curriculum now presents 'exception words' as words which should be learnt for spelling purposes. This means that children have to learn the letters and their order - the words can't be learnt as 'wholes'.

Jenny C.

User avatar
Debbie Hepplewhite
Administrator
Posts: 3648
Joined: Mon Aug 29, 2005 4:13 pm
Location: Berkshire
Contact:

Re: Learning 'exception words' as wholes.

Post by Debbie Hepplewhite » Tue May 31, 2016 3:31 pm

Maggie - I have had the same experience as you when it comes to older learners who have had a diet of the '45 Reception words' on their individual education plans and have wondered myself why teachers didn't appreciate that if they couldn't learn those words in Reception or the infants, then clearly the sight word approach of such a small number of words was not working!

I have also had Jenny's experience of children who have had an increase of phonics provision being more able to tackle reading a wide range of word-structures, including those with more unusual spellings, as a consequence of the greater emphasis on phonics provision nowadays (in England).

There are still some children, however, who find it very difficult to read new words without a more thorough knowledge of specific letter/s-sound correspondences of the alphabetic code - in other words, the profile of child who does not show signs of 'self-teaching' - of being able to transfer the skill of blending to words with spellings they have not yet encountered.

Maggie, you raised the issue of 'generalisation' (below) but not all children manage this equally:
On the other hand, it may be argued that, over all, the greater attention to sounding out and blending all through the word from left to right which is at the core of SSP instruction may help these children to generalise these skills to words taught as wholes.
My worry is that such children will continue to have special needs when it comes to reading because they are, nowadays, amongst other children who have fared very well from whatever phonics provision they have had and many children are more able to self-teach and tackle new words and texts.

The danger of this is that rigorous, content-rich phonics provision is not available, or as available, for the slower-to-read children as they get older. So we should still be vigilant for 'mixed methods' and 'multi-cueing reading strategies' prevailing or setting-in by default.

Good systematic synthetic phonics programmes should indeed include the high-frequency words with more unusual spelling patterns drip-fed into the body of work sensibly.

User avatar
Debbie Hepplewhite
Administrator
Posts: 3648
Joined: Mon Aug 29, 2005 4:13 pm
Location: Berkshire
Contact:

Re: Learning 'exception words' as wholes.

Post by Debbie Hepplewhite » Tue May 31, 2016 3:38 pm

Maggie, I've added your thread with your important concerns onto the IFERI message forum here:

http://www.iferi.org/iferi_forum/viewto ... p=978#p978

chew8
Posts: 4150
Joined: Sun Nov 02, 2003 6:26 pm

Re: Learning 'exception words' as wholes.

Post by chew8 » Fri Jun 03, 2016 11:00 am

Maisie makes the point that there is no real consensus about what is an ‘irregular’ or ‘exception’ word. I think this is understandable. How one defines ‘regular’ is one thing that can make a difference, and another is whether one is considering the question from a reading or spelling perspective.

Diane McGuinness gives her definition of regular spelling on p. 45 of Early Reading Instruction and also mentions the spelling/reading distinction:
She wrote:Nevertheless, some spellings are obviously more common than others, and in the absence of any definition for “regular spelling” I will provide one here. Regular spelling is the only spelling for a particular phoneme (including letter doublings), or the most probable or least ambiguous spelling. For example, the sound /b/ is always spelled b or bb. The spellings ee and ea are the most probable (common) spellings for the sound /ee/ and represent this sound equally often. However, ea is ambiguous (bead, head) and ee is not.

Note that “regular spelling” means regular spelling, not to be confused with regular decoding, which is often how this term is used. For instance, the digraph oa is supposed to be “regular” because it Is usually decoded as /oe/ (broad and oasis are exceptions). But oa is not the most common (regular) spelling for the sound /oe/, one of the many problems that arise when the code is analysed the wrong way round. The most common spelling of /oe/ is o-e, as in home.
By her definition, a very large number of spellings would count as irregular – all those where a phoneme is not spelt in the most common way, even if that spelling occurs in many words, as with ‘oa’ for /oe/. I myself would count ‘oa’ as one of several regular spellings for that sound. Similarly, I would count spellings of /ee/ as regular if they occurred in a reasonable number of words, but would regard the ‘eo’ spelling as in ‘people’ as irregular.

The root meaning of ‘regularity’ is associated with the idea of a rule, but because almost all phonemes in English can be spelt in more than one way, hard-and-fast rules are few and far between. Some people may accept D.McG’s ‘most probable’ definition, which means that very large numbers of words must be regarded as containing irregularly spelt phonemes, but others may prefer to regard all common spellings for a phoneme, not just the most probable, as regular, which means that far fewer words have to be regarded as containing irregularities from a spelling perspective.

In the same chapter, on pp. 57-8, she talks about ‘sight’ words (a concept which she accepts), but the fact that she says there are very few of these shows that she can’t be using her criteria for spelling regularity. She is clearly talking now about DEcoding, not spelling, and must be using more lenient criteria – e.g. she treats ‘broad’ as a ‘sight’ word, but not ‘boat’, ‘soap’, ‘goal’, ‘coast’ etc. Even if the ‘oa’ spelling for /oe/ doesn’t meet her criterion for spelling regularity, it presumably occurs in enough common words to count as regular for reading purposes.

When we consider regularity from the reading perspective, perhaps analysing the code ‘the wrong way round’ is actually the right thing to do rather than being ‘one of the many problems that arise’!

Jenny C.

User avatar
Debbie Hepplewhite
Administrator
Posts: 3648
Joined: Mon Aug 29, 2005 4:13 pm
Location: Berkshire
Contact:

Re: Learning 'exception words' as wholes.

Post by Debbie Hepplewhite » Sun Jun 05, 2016 11:08 am

I think it works really well to ORGANISE the alphabetic code from sound to print because this can work relatively simply.

This means basing the organising of the code on the 'sounds', not the spellings - as there are fewer sounds by far, than spelling.

So, broadly speaking, there are around 44 to 50 units of sound around which to organise the code (around 44 phonemes but some bits of code are two phonemes combined so print IS taken into account for the extra units of sound on the alphabetic code charts that I've compiled such as letter 'x' being code for two phonemes combined: /ks/). From sound to print involves organising the alphabetic code around the sounds of the language and the letters or letter groups which are code for the sounds - a bit like this:

http://www.alphabeticcodecharts.com/One ... ymbols.pdf


If one tries to organise the alphabetic code from print to sound, however, not only are there large numbers of graphemes (letters and letter groups) and their pronunciation alternatives to consider, there are also certain 'letter patterns to take into account. By this I mean, it is not only graphemes at the level of the phoneme in the main that is identifiable to the 'reader' (print to sound), there are other considerations.

I tried to illustrate this through a chart which is print to sound in contrast to the Alphabetic Code Charts which are organised on the basis of sound to print.

See this - don't bother to read the blurb if in a hurry, just look at the content of the chart to see how complex print can be for the reader:

http://www.phonicsinternational.com/Pro ... nglish.pdf

In other words, I think it is too simplistic to think of teaching reading by a sound to print approach without fully appreciating the complex nature of English spelling from a print to sound perspective.

Whereas I highly recommend the first type of Alphabetic Code Chart for teacher-training information, for informing parents and the general public, for organising a body of work as a phonics programme and for both teaching and learning purposes, I also think that teachers should be fully aware of the second type of chart to know what a reader is faced with.

When teaching children phonics based on introducing the sounds systematically with some spelling alternatives, and matching reading books and texts to the code taught already, this gets both teaching and learning off to a flying start - and addresses both reading and spelling.

But, like Jenny has pointed out in her posting above, you have to think what teaching and learning really 'looks like' from sound to print and from print to sound - and not just from one perspective or another. I really appreciate Diane McGuinness's logic, but I also really appreciate Jenny's logic too.

chew8
Posts: 4150
Joined: Sun Nov 02, 2003 6:26 pm

Re: Learning 'exception words' as wholes.

Post by chew8 » Mon Jun 06, 2016 8:50 am

What I was trying to do in my last post was to get to grips with the issue maizie had raised about regularity. Diane McGuinness’s way of organising the alphabetic code from sound to print, together with her decision to regard only the most common spelling of a phoneme as ‘regular’, leaves us having to regard a very large number of words as containing irregularities. I‘m not happy with that – I don’t think the case for code-based teaching is helped by definitions which mean that there is a high proportion of irregularity.

I myself did not know how many phonemes there were in English until I was 50, and I found out then only because I had to start teaching a course which involved some phonetics. Not knowing about this, however, had not prevented my contemporaries and me from becoming competent readers and spellers. It’s true that I complained a lot about the spelling of the students aged 16+ whom I taught from 1978-2000, but most seemed to know how sound-symbol correspondences worked even if they chose wrong spelling alternatives – and this was despite the fact that they had been through primary school at a time when phonics was seriously under-emphasised. I still have spelling-test scripts produced by 6,000+ of these students, and will try to analyse a sample of these so that I can give some facts and figures.
Debbie wrote:I think it works really well to ORGANISE the alphabetic code from sound to print because this can work relatively simply.

This means basing the organising of the code on the 'sounds', not the spellings - as there are fewer sounds by far, than spelling.
I understand this logic. What I’m querying is the assumption that the detailed organising of the code (whichever way round it’s done) is necessary for practical teaching and learning purposes. With regard to spelling in particular, I think it may not add anything over and above what children can learn through (a) good teaching which does not depend on such detailed organisation and (b) exposure to correct spellings in their reading.

Jenny C.

User avatar
Debbie Hepplewhite
Administrator
Posts: 3648
Joined: Mon Aug 29, 2005 4:13 pm
Location: Berkshire
Contact:

Re: Learning 'exception words' as wholes.

Post by Debbie Hepplewhite » Mon Jun 06, 2016 11:23 pm

What I’m querying is the assumption that the detailed organising of the code (whichever way round it’s done) is necessary for practical teaching and learning purposes. With regard to spelling in particular, I think it may not add anything over and above what children can learn through (a) good teaching which does not depend on such detailed organisation and (b) exposure to correct spellings in their reading.
Well, we'll have to agree to disagree on this one, Jenny.

I have no research-evidence as such, only anecdotal with regard to how helpful teachers, teacher-trainers, parents, learners and others in various contexts have told me they find the Alphabetic Code Charts.

It's a practical resource which provides information, it can provide a helpful visual explanation (rationale) of the many letter/s-sound (or sound-letter/s) correspondences that we 'mean' when we teach phonics, it provides the 'bigger picture' of the alphabetic code (not the full picture as there are many very unusual and rare spellings as we know), it makes sense when teaching 'a code' to show the key to the code, it can show the organisation and 'order' of specific programmes.

Without a chart, phonics provision can feel like a long (linear) subject with the teacher introducing a correspondence one after another after another. Seeing the alphabetic code in an organised chart makes sense of the teaching and makes it feel less linear.

Also, quick reference to the Alphabetic Code Chart really supports 'incidental' phonics teaching right from the earliest stages of planned phonics teaching. My approach to programme design and guidance is based on both systematic and incidental phonics teaching - both being supported by reference to the chart. This can lead to accelerated self-teaching and addresses differentiation. I firmly believe it is also an intellectual entitlement. Of course I know of children who have been very able readers and spellers with no experience whatsoever of systematic phonics teaching and the use of Alphabetic Code Charts. I can describe first hand, however, how interested such children have been to learn about phonics and the alphabetic code retrospectively. It's very interesting - and important - to have a notion of the code of one's language.

To be honest, why wouldn't we use such a resource?

To me, to avoid its use is no different from avoiding the use of an Alphabet poster or a periodic table, or times tables.

At one point, David and conducted a survey of views and experiences of the potential of Alphabetic Code Charts. This was the result at the time and, interestingly, people responded very quickly:

http://www.phonicsinternational.com/survey_results.pdf

chew8
Posts: 4150
Joined: Sun Nov 02, 2003 6:26 pm

Re: Learning 'exception words' as wholes.

Post by chew8 » Tue Jun 07, 2016 8:42 am

My views are based on what I’ve actually seen by systematically testing the spelling of over 6,000 16-year-olds who had been through primary school before even the phonics of the National Literacy Strategy was introduced – i.e. at a time when look-and-say and whole language were predominant. What got me going on the issue was the fact that their spelling was worse than that of the children I had previously taught in South Africa, where phonics was still routinely taught in primary schools and spelling was considered important - I also have spelling-test scripts from there, so can make objective comparisons. Although my English 16-year-olds were weaker spellers, however, most of their spellings were phonically plausible - what I felt they lacked was word-specific knowledge and a good grasp of morphology.

I’ve kept all 6,000+ spelling-test scripts and have made a start on the analysis I said I’d do, but it’s a mammoth task and not at all straightforward – e.g. I can’t just put words in the ‘phonically implausible’ category as soon as I see an implausibly-spelt sound, as other sounds in the word (sometimes the majority) may be plausibly spelt, so I’m having to count phoneme by phoneme. I’m in schools all day today and at a meeting tonight, so won’t get much more done before tomorrow, but I hope to be able to say a bit more within the next few days.

Re. alphabetic code charts: the issue, for me, is not just whether teachers find them useful, but whether they contribute to accurate spelling over and above good phonics teaching which makes no use of them - e.g. the sort of spelling I was used to seeing in S. Africa.

Jenny C.

chew8
Posts: 4150
Joined: Sun Nov 02, 2003 6:26 pm

Re: Learning 'exception words' as wholes.

Post by chew8 » Fri Jun 10, 2016 10:26 am

Sorry this is long - I've spent a lot of time analysing misspellings over the past few days and trying to think about them in relation to alphabetic code charts.
I wrote:Re. alphabetic code charts: the issue, for me, is not just whether teachers find them useful, but whether they contribute to accurate spelling over and above good phonics teaching which makes no use of them.
I want to stress again that the ‘over and above’ point is a crucial one for me. We know that official policy is now much better than it used to be and that this has had an impact, but we need to know whether alphabetic code charts can give an extra boost.

I’ve now spent many hours looking at spelling-test scripts. The test was the Schonell Graded Word Spelling Test B, numbers 31-100 – I left out numbers 1-30, as they were supposed to be suitable for children below the age of 8, and I was testing 16-year-olds. As I’ve already said, it wasn’t at all straightforward. I started out intending to look at all the errors in the 419 scripts from 1999 (the last year in which I did the testing), but soon found that this could mean analysing about 30,000 misspelt phonemes, so looked in detail just at the six weakest, and at 20 of the better ones, though not just the very best.

The weakest 6 students had scores of 6, 13, 19. 20, 24 and 25 out of 70, and on average they spelt 18% of phonemes implausibly. The better students spelt almost no phonemes implausibly – in spite of having gone through primary school at a time when little or no phonics was taught, they had somehow got to know a good range of spellings for phonemes. In fact, 5 of the 6 weakest spelt ‘island’ correctly; 2 spelt ‘brought’ correctly, and another two spelt it as ‘bought’, so 4 of them knew that unusual vowel spelling. One spelt ‘source’ correctly, one spelt it as ‘sours’, one wrote ‘sour’ then faded out (perhaps didn’t know whether to write ‘s’ or c’ next) and 3 spelt it as ‘sorce’, so there was some knowledge of ‘our’ as well as ‘or’ as a spelling for that sound, and of ‘ce’ for /s/. There were no ‘aw’, ‘au’ or ‘al’ words on the test, but one of the 6 spelt the first syllable of ‘orchestra’ as ‘al’, showing awareness of another possible spelling.

Interestingly, even the WEAKEST South African 16-year-olds to whom I gave exactly the same test in 1987 and 1992 spelt almost no phonemes implausibly, so it seems that the phonics teaching in primary schools there had more or less eliminated that problem, in spite of being less rigorous than I think you would favour, Debbie.

Going back to the ‘over and above’ point: it may help if I give an example. The National Curriculum now says that Y1 children should be taught about adding certain endings to root words, including the regular past-tense ending with its 3 sounds, though at that stage they are expected to produce correct spellings only if no change is needed to the spelling of the root word (landed, jumped, buzzed). In Y2, they are supposed to learn about adding ‘-ed’ and other endings where changes to the root word are necessary – changing y to i (copied), dropping final e (hiked), and doubling a consonant (patted). In the Schonell test, one of the words most often misspelt was ‘equipped’, with ‘equiped’ and ‘equipt’ being very common. From things you’ve said in the past, Debbie, it seems that you would expect a teacher using an alphabetic code chart to say something like ‘In this word, the /p/ is spelt ‘pp’ and the /t/ sound is spelt ‘ed’'. If the children had previously been taught what Y2 children are now supposed to be taught, however, that teacher’s approach would seem much less effective than reminding the children of that earlier teaching: one way implies that word-specific knowledge of ‘equipped’ is necessary for the correct spelling of the phonemes in question – the other way reminds children of very common patterns. I regard the second way as far more useful, and I would like to see all teachers using these reminders wherever possible as I think they are a more efficient and economical way of addressing common errors. Of course word-specific teaching is often necessary, but even there, I wonder whether a chart adds anything over and above what can be done in other ways.

We now have a National Curriculum which, if reasonably well implemented, should result in real improvements in reading and spelling. Research could show that alphabetic code charts make a further contribution, but in the absence of such research, I can’t see any theoretical reason why they should do so, and the nature of typical student errors gives me further cause for doubt. Charts are one way of doing things, but they are not the only way, and I don’t think they should be presented as something that all teachers should use and that the government should promote.

Jenny C.

Post Reply

Who is online

Users browsing this forum: No registered users and 12 guests