Yr 1 reading/phonics

[RunsWithScissors]

Hello,

DD (5.5) seems to be doing pretty well. Nearer the top end of reading in her class (on orange band, I know not stunning based on MN standards ;-) but she's moved up leaps and bounds from the beginning of the year.

The phonics test is this week, and her teacher caught me yesterday to say she doesn't think she'll pass it. I know it's for the school to see how she's doing, etc. she's moved her into a different phonics group to help her out.

I'd noticed she doesn't tend to sound things out much, I think she remembers words/word recognition?

I didn't learn phonics growing up, but can't recall the learning process of reading that I went through. I've always loved reading, as does DD.

So, my questions are:

Is the lack of ability/knowledge going to make it harder for her? She seems to be progressing really well with her reading, and has wonderful comprehension of what she reads. Very expressive when she reads a book for the first time, so I know she is understanding it. I'm just wondering if a better grasp of phonics would make it easier for her, or do some children naturally read in a different way?

Secondly, although her spelling is also progressing really well I do notice that some misspelled words reflect her speech (which we are having assessed) eg. 'Wiv' for 'with'. Her hearing test was fine last year, she has a great vocabulary and can explain things really well.

I am a bit confused tjough, as she seems to use sounding out to spell. Is this not a similar skill to reading by sounding out?

I know the school will do a great job to support her, and we are thrilled with her progress this year. I just want to ensure we are doing what we can to support her, and that we aren't missing out on things that might make it easier for her/be a more natural fit for her style of learning.

Thanks if you've read this far!

Thanks Mrz. I'll look into that. She does love to read, which makes me very happy. But it would be really awful if she hit a ceiling in terms of her free reading and lost her passion for books.

Using context to work out what a word might mean is fine (though not always particularly reliable) but teaching the use of context to work out what a word says is really damaging, as Itshouldntmatter is discovering . It isn't 'whole word' reading but it is one of those 'other strategies' which the phonics sceptics are so upset about not being allowed to include in their teaching repertoire.

The preference for Chiff & Bip may have unwittingly contributed to her preference for contextual guessing as they are written expressly to encourage the strategies which phonics teaching bans! The strategies are: guessing words from the pictures, guessing words from context and guessing words from the initial letter. Phonics teaching bans them because not only are they completely unnecessary when using decoding and blending for word ID but they are actually damaging. It is far easier to 'guess' a word than to decode and blend it and is quite a successful strategy when reading books written specifically to support guessing strategies! It all falls apart when vocabulary becomes more complex and multisyllable words more frequent. The child might even be put off reading at this stage as it becomes far more difficult...

Advice so far has been to encourage her to work out multisyllable words one syllable at a time, then blend the syllables together (that's how most grownups do it!). I'd just add to cover the word and disclose it syllable at a time if the child persists in trying to guess instead of decode.

P.S Once a child can decode competently and knows most of the letter/sound correspondences the ORT books are perfectly decodable for them. All books are decodable once you know the code

Of course homophone should be homographs and heteronyms

Mrz, it seems to me that you are conflating research on effective teaching with that on how students actually learn to read
.
I've been doing a little fun research into models of reading. One name that was cited more than any other was Mark Seidenberg. From his Wikipedia entry:

Mark Seidenberg... is a specialist in psycholinguistics, focusing specifically on the cognitive and neurological bases of language and reading. ...Seidenberg has published over a hundred scientific articles and has been recently honored as one of the 250 most-cited researchers in the areas of psychology and psychiatry.

This is from a 2007 paper:

"Consider learning to pronounce a word out loud. Assume the child generates a pronunciation based on the current state of his/her knowledge. The child's own output (on the production side) is also an input (on the comprehension side). If the word is pronounced correctly, then it should also produce coherent patterns if passed through the comprehension system. If it is pronounced incorrectly, then it cannot be comprehended, the failure to activate semantics would itself provide a strong error signal... The child may attempt another pronunciation that succeeds, or may require an explicit teaching signal. Jorm and Share described a similar self-teaching mechanism: the child learns by pronouncing a letter string and matching the computed output to a word that the individual has learned from using speech."

I do not imagine for a moment that this field of linguistic research is without its controversies, but certainly one of the most eminent minds on the subject entertains ideas similar to those I was proposing upthread.

I have some more very interesting excerpts, which I will post in a moment.

Then you are wrong

Here is another excerpt:

"A second issue is this: how does the experience of the child in an instructional setting correspond to what happens in a model - or more importantly, in the child's brain? Consider the situation in which a teacher provides explicit instruction about how spellings are pronounced... Its interesting to observe that there may be significant gaps between what a teacher thinks he/she is teaching (e.g. a rule about how a vowel is pronounced, which is an explicit type of knowledge) and what is occurring at these other levels (which are implicit). Would a teaching method that is more closely modeled on what we think is occurring at these other levels be more effective? I don't know the answer or think it is by any means obvious. Here I would just note that there are differences between learning (in the neural network model sense) and instruction (in the pedagogical sense), and that a complete theory of how children learn would explain how the effects of instruction are mediated."

And one last one, for now, for Mashabell:

"The nature of the correspondences between input and output codes varies. In alphabetic orthographies, orthography and phonology are highly correlated: letters and letter patterns represent sounds. The degree to which they are correlated (the consistency of the mapping across words) varies in alphabetic writing systems. For example, English contains more irregular correspondences than in the writing system for Serbian, in which letter-sound correspondences are almost entirely predictable....."

".. The interesting observation is that the correspondences in English are not completely consistent; there are many words whose pronunciations deviate from what would be expect if the system were strictly rule-governed. In standard approaches, these words are treated as exceptions that must be learned by rote. This is the core idea underlying dual-route theories.

Note however that the exceptions are not arbitrary. HAVE is not pronounced "gloop;" it overlaps with many other words including HAT, HAS and HIVE. Thus the spelling-sound correspondences of English can be said to be rule-governed only if the rules are not obliged to apply in all cases; the system admits many forms that deviate from these central tendencies in differing degrees. Seidenberg and McClelland introduced the term "quasi-regular" to describe bodies of knowledge that have this character.

Connectionist networks are intrinsically well suited to the problem of learning in quasi-regular domains...The models challenge the deep seated intuition that behaviour is rule-governed, by demonstrating that a wholly different type of mechanism can account for the phenomena, one that is consistent with other facts about learning and its brain basis. ... the characterisation of language as rule-governed is taken as an informal characterisation of some aspects of the underlying processing system, convenient perhaps, but not accurate in detail."

Did you miss the bit where I said we don't teach rules because there aren't any?

No, I didn't miss that part. As I said, the last post was for Mashabell, and wasn't really on topic here. The first two quotes were more pertinent, regarding children who correct alien words to real ones. Apparently, this is an intrinsic learning mechanism, which was my earlier intuition, that we were debating previously. The second post was regarding the difference between how teachers teach and how learners learn.
In general, I found the entire article to be quite at odds with the phonics purists stance in this forum.

In general, I found the entire article to be quite at odds with the phonics purists stance in this forum.

If you trawl around the internet you will find lots of articles which are at odds with what the 'phonics purists' are saying here.

Consider two things:

1. When the children take the phonics check they are explicitly told that the 'alien words' are not real words so why should they try and make them into 'real' words that they know?

2. A 6 y old has a very limited reading vocabulary. Any 'new' word that they encounter is the equivalent of an 'alien word' as they have not seen it before. If they insist on turning 'new' words into words which are familiar to them how are they ever going to extend their reading vocabulary?

The process which is being described in your extracts is more one of trying known alternative sounds for particular graphemes, not disregarding the graphemes and the order in which they come in the word in order to turn an unknown word into a known one.

"Mark Seidenberg, a new psychology professor at UW-Madison, is ready to lead the phonics troops. He was one of five experts commissioned by the American Psychological Association to write a scholarly paper assessing what psychology and linguistics research says about reading and instruction. He's also written an article for the March issue of Scientific American on how reading should be taught.

His message is simple: Whole-language instruction is a failure, an experiment dreamed up in an ivory tower, peddled by celebrity educators and inflicted on unwitting children and parents. Who says so? "An overwhelming pile of research favoring phonics," according to Seidenberg.

Beginning readers already know how sounds relate to meaning. When they learn that alphabet symbols represent the sounds of language, they learn to read, Seidenberg said. Despite the research, reading instruction "became very politicized and ideological," he said, with conservatives lined up behind phonics and liberals espousing whole-language approaches. The whole-language camp doesn't trust the science, he said. "You can't base these things on ideas about intuition," Seidenberg said.

Legislation like the federal bill and similar state efforts will be helpful, he said, to ensure "at least a reasonable emphasis on phonics everywhere."

Seidenberg doesn't blame teachers; he blames schools of education. When he talked to educators in California, where he taught until last summer, "they thought I was from Mars. They thought I was the enemy. They were interested in teaching literature. It's two cultures. The science goes one way and the ed schools tend to go another way."

Phonics never disappeared, he said, it just got pushed out of the classroom. Parents turned to private phonics programs, tutors and computer software. (Critics say phonics is boring; Seidenberg says watch how long children will play computer phonics games.)

Parents need to demand phonics instruction, he said, and ask exactly what it means if their school says it offers a balanced approach. "I personally think it isn't a political issue," he said. "It's a question of what's the best way to teach and making sure educators don't just try out ideas because they seem clever to them in their ivory towers."

phonics is simple to teach - easy to measure..

nice for the gov to judge teacher by.. easy to number crunch

it is not about getting pupils to 'think'.. or 'question'

At the end of the day the gov wants.. people who don't question what they are doing.. compliant non thinking individuals are far easier to control.

Phonics teaching works better for most children than previously used methods. This much is true.
What doesnot follow from this are all the other assumptions made by Mrz: that the phonics screening test never fails strong readers, that children should not use certain strategies (which is an entirely different question to them being taught them) or that there are no individual children that benefit from alternative methods if phonics isn't working for them.
What I have read so far undermines the notion that there is good phonics reading and bad sight word reading, only that there is a gradual building up of recognised units of varying sizes, with no reference in the brain to the manner in which it was learned. It does not support the horror story of good readers failing to be able to learn new words but insists that this will be a natural by product of learning to read any words, even with no additional formal teaching at all, due to the proposed model of how the brain actually builds its reading knowledge base. Note that i am not proposing that the opposite of these assumptions is true, only that these questions do not necessarily have simple answers, and no-one is stupid simply for asking them.
Seidenbergs writing also came across as having lots of humility: he speaks often of the usefulness of alternative models, of the complexity of the issues discussed and of how there are no simple answers to many of the big questions. This is in huge contrast to the "you're wrong and I'm right" views we all read on here.
Now, if Mrz would like to debate the implications of this research on the topic of good readers failing to read alien words due to correcting them to real words, (which it seems is an intrinsic part of how the brain works, and not an inefficient tactic bound to cause later failures), rather than seek out unconnected quotes from interviews that support a very general pro phonics position, that would be great. I'm already a supporter of pure phonics teaching in school, anyway.

Maizie, the problem with children being told not to convert into real words is that they are then really being treated on how well they can follow this instruction, not just how well they have learned the different phonics blocks.
You've also not read the new words bit the same way I did. Children compare new written words against known spoken words, exactly as I guessed much earlier in the thread (I'll admit to feeling a bit smug as to how well seiderbergs model backed up that idea) . It makes much more sense that way. Remember, one of the main ways children improve their reading skills is not by being explicitly taught, but by exposure through their own reading in order to build up their implicit knowledge base. They need to have a mechanism of self correction.

Maizie asked,
1) When the children take the phonics check they are explicitly told that the 'alien words' are not real words so why should they try and make them into 'real' words that they know?

Because learning to read English (outside the phonics test) is all about making sense of words. That is the only way children learn to pronounce words like 'child, said, was ...' correctly.

In English there is no 'correct' pronunciation of letters or letter strings outside words. Pronunciations of letters are right or wrong only in the context of words:
who/when, house/hour, an/any, on/only, tough/cough...

That's why learning to read English takes such a long time (3 years v 3 months in Finnish).

Tested, not treated, sorry.

Mickeys

Micksy have you administered the check?

Nope. I've read the instructions and looked at some past tests. Would my anecdotal experience be illuminating even if i had? I've already cited a reference about the gulf between teacher intuition of what is occurring versus what is actually happening in a child's brain.
I'm very aware that your school is excellent in teaching children to read alien words. However it does not follow that in schools placing less emphasis on this aspect, some good readers may do well at the real and fail on the alien. Indeed, since your school does so well at preparing the alien words, I would say you have less experience than most in testing these children.

No I wondered if you'd had experience of good readers who don't understand simple spoken language or if you just imagine all 6 year olds are incapable of understanding " all the words on this page are made up so you won't have seen or heard them before" or that you imagine they are all so arrogant they assume they know every single word in the English language.

I'm a secondary maths teacher and I'm acutely aware of how possible it is to spend a week teaching students the correct method of doing something, only to have them apply a previously learned misconception in a test. In many classes, half of them fail on "underline the date and title with a ruler".

Mrz, don't light any matches. There's so many straw men around you, I would fear for your life.

-You seem to be very ill informed - my Y1s have been reading a story version of The Tempest which gives them plenty of practise reading unfamiliar words Prospero, Caliban, Sycorax, Trinculo, Juno you should try it ??

Sorry, Mrz, is that aimed at me, or someone who is proposing children never meet new words?