Tuesday, February 22, 2011

Non words nonsense

I see the UK Literacy Association have intervened in the debate about the use of non-words in the government’s proposed new reading test for six-year-olds in England.
Testing children on non-words, the proviso being that it is made clear to the child that the words being presented are NOT real words, is a very good way of making sure that children are not using their visual memories alone and that they are using their decoding skills: i.e. their ability to transform what they see to into what they hear, put the sounds together and produce the word. Such a test also presents the teacher with a clear idea of how good a child’s code knowledge is.
Sounds-Write have been using the Nonsense Word Sound Swap game with children from YR upwards for years and, in all that time, not a single one of the nearly eight thousand trained teaching practitioners has ever got back to us to say that it confuses children or that they can’t do it. In actual fact, both teachers and children love the game, and, used in the right way, it is regarded as being great fun.
The game improves children’s skills in blending, segmenting and manipulating sound-spelling correspondences in words extremely well. Research shows that the combination of these skills correlates very highly to good reading and spelling. I wonder if colleagues at the UKLA have ever conducted a non-word assessment or played a sound swap-type game?
David Reedy from the UKLA is quoted as saying: "The test is trying to control all the different variables so that things like meaning don't get in the way. We think that seems a bit bonkers when the whole purpose of reading is to understand words.”
He’s right, of course: the purpose of reading is to understand words. However, in order to understand words, one must first be able to decode them - something that an appallingly high number of children in this country are unable to do.
The UKLA are always and quite rightly banging on about promoting a love of reading. We agree! Phonics advocates are just as passionate that children should develop a love of reading. But, you can’t love reading if you can’t read!

Thursday, February 17, 2011

Huffington Post firing on blanks

Yesterday’s article in the Huffington Post demonstrates perfectly why reading and spelling are taught so appallingly in the USA and the UK.
In her article ‘The Crisis in Education: Let’s Not Wait for Superman’, Dr Marion Blank talks about the problem of illiteracy in schools in the USA and reports a failure rate of between thirty-five and forty percent. She’s absolutely right! So far, so good!  However, this is where it all starts to go downhill. Quoting a line from a Dr Seuss’s The Cat in the Hat - ‘The sun did not shine . It was too wet to play. So we sat in the house on that cold, cold wet day.’ - she claims that only the words ‘sun’, ‘did’, ‘not’, ‘shine’, ‘wet’, ‘sat’, ‘in’ and ‘wet’ can be ‘sounded out’. This is, of course, not at all accurate (I’m being polite!). All of the words in the text can be ’sounded out’ or segmented into their constituent sounds. Thus:
Th e  s u n  d i d   n o t  sh i ne.  I t  w a s  t oo  w e t  t o  p l ay.  S o  w e  s a t  i n  th e  h ou se  th a t  c o l d,        c o l d  w e t  d ay.
There is clearly a problem here and Dr Blank has gone some way to highlighting what it is. The problem with phonics, as it is traditionally mediated, is that it is taught back to front. The words ‘sun’, ‘did’, ‘not’, ‘sat’, ‘in’ and ‘wet’ are easily segmented (and blended). This is because they all combine a simple structure (CVC or VC) with transparent one sound/one-letter spelling - the easy bit, when it’s not so important whether the code is taught from sound-to-print or from print-to-sound. Beyond this level of simplicity, the writing system is far less transparent and more complex to teach. Nonetheless, it doesn’t mean that it can’t be done.
The other words in the passage are not difficult to teach if teachers are using a good quality phonics programme and they have an excellent understanding of the way the writing and the sound systems of the language are related. The most complex words in the three sentences ‘the’, ‘play’ and ‘house’ are easily segmented as indicated above.
We need to teach children from simple to complex and ensure they master the following, which they will need to understand if they are going to be able to read and spell English:
  • that spellings represent the sounds in English;
  • that a spelling can contain more than one letter: the sound /th/ in ‘the’ is represented by two letters, the sound /i-e/ in ‘night’ is spelled with three letters, and the sound /ae/ in ‘eight’ is represented by four letters;
  • that there is frequently more than one way of spelling a sound: the sound /f/ in ‘fun’ can be spelled in 'fun' as well as in ‘off’, in ‘physics’ and in ‘rough’; or, to give another example, the sound /ee/ can be spelled as in ‘meet’, as in ‘tea’, as in ‘happy’, as in ‘key’, as in ‘brief’, as in ‘ski’,  as in ‘he’ as well as as in ‘receive’);
  • and, finally, that a spelling can represent more than one sound, so that the spelling , in Dr Seuss's sentences, can spell the sound /oe/ as in 'so', or /o/ as in 'not'.
If we also teach the sound to spelling correspondences in English, starting with the simple one-to-one instances, and the skills they need to perform in order to use this knowledge, it is possible to teach all children successfully.
Most importantly, if you don’t teach the way the writing system relates to the sounds in the language, you end up teaching phonics backwards and falling foul of the kinds of confusions displayed in Dr Blank’s piece.

Monday, February 14, 2011

Why Children Can't Read

Over the weekend I saw two sets of parents, both of which are highly educated and middle class and both of which have decided to pay for private tuition in numeracy and literacy for their children.
When I asked them why, their reasons were almost identical: in terms of the children’s literacy, they were not satisfied that their children were able to read with fluency. The problem was NOT one of comprehension: both said that their children understand texts that are read to them without difficulty. However, when it comes to reading similar texts for themselves, it takes the children twenty minutes to struggle through a short paragraph, and by the time they have finished this laborious task, they don’t know what they’ve read.
This is precisely as described by Keith Stanovich in his research: ‘The combination of lack of practice, deficient decoding skills, and difficult materials results in unrewarding early reading experiences that lead to less involvement in reader-related activities. Lack of exposure and practice on the part of the less skilled reader delays the development of automaticity and speed at the word-recognition level. Slow, capacity-draining word-recognition processes require cognitive resources that should be allocated to higher-level processes of text integration and comprehension.’ In other words, the children need to be taught to decode so that they can understand what they are reading.
The parents also said that when they asked the school how their children were doing, they were told that they were ‘fine’. However, they were not given a reading age, nor were they told whether the children’s skills, conceptual understanding and code knowledge was what one would expect for children of their age. In fact, such a breakdown and mode of analysis is completely unknown to most teaching practitioners and parents are almost invariably fobbed with the kind of blandishment mentioned above.
That both sets of parents are highly educated and want the best for their children also gives the lie to the oft-claimed assertion that illiteracy is a problem confined to families who are poorly educated and don’t have a habit of reading to their children.
What the government needs to do is follow Diane McGuinness’s advice given in her book Why Children Can’t Read: 
'First, adopt a method with solid scientific support, one with a comprehensive curriculum. Provide clear, operationally defined goals, and make sure that each sub-goal or sub-skill is developmentally appropriate, so that every child can learn to read, write and spell. Design objective, up-to-date standardized tests normed on a large population over a wide age-range. Get outside testers to administer these tests… Publish the test results of every school at regular intervals.’
But they won’t because if they did it would cause a national outcry when it was discovered how many school are failing to teach their pupils to read and spell.

Wednesday, February 09, 2011

But it's too late, Leighton now, it's too late...

I know I’m a week late on this but it’s taken me a week to be able to compose a temperate response to the news that Leighton Andrews, the Labour education minister for Wales, has suddenly decided that the pupils of Wales ‘deserve better’ and that what is needed now is to raise standards.
After twelve years of calling the shots, this must be one of the most egregious examples of the kind of cynical posturing (the elections are coming up in Wales) that says so much about the way politicians use education and, in particular, literacy. Or should I say illiteracy when, as reported by last week’s Economist, ‘almost a third of schools were deficient, and … two-fifths of pupils entered secondary school in Wales with delayed reading skills’.
What does Andrews say he’s going to do to raise standards that he hasn’t had chance to do in the previous twelve years? Introduce national reading tests. This would, of course, be an improvement on SATs, in the sense that it would actually tell us something about whether children can read or not, although it will be interesting to see what kind of test they decide to employ.
He also intends to discuss with governing bodies of schools their ‘performance data’. Naturally, this won’t mean a return to league tables. Heaven forbid! What performance data he’s talking about is the usual mealy mouthed politician-speak for something he almost certainly knows (at best) very little about. As the Economist put it last week, when the devolved government decide to bin the league tables and replace standardised tests with teacher assessment, ‘robust comparison of schools’ was made ‘impossible’. Quoting researchers at Bristol University, the Economist goes on to maintain that ‘such policies and the relaxing of standards that accompanied them, took almost two points a year off GCSE grades per student’ and that ‘poor students suffered most’. There, it couldn’t be more stark!
What else? When a school is irredeemably failing, he’ll close it. And, there’s going to be a ‘national literacy plan’. Hang on a minute, didn’t we just have the Literacy Strategies, on which over £2 billion was spent and which resulted in the UK slipping further than ever down the international league tables for science, literacy and mathematics?
So Andrews’s answer, just like the former Labour education minister, Ed Balls, is to issue a diktat. Schools must teach children to read! And, having said it, it means that it will happen! Unfortunately, it isn’t going to happen, Mr Andrews, unless you train the teaching practitioners how to teach reading and spelling, using a programme that works (linguistic phonics!). If you don’t, you’ll end up in the same position in another twelve years and that with all its knock-on effects. Not that you care. You’ll probably be out of office or you’ll be able to blame someone else and pretend it didn’t happen on your watch. What a disgrace!