John Bald/language and literacy

Deliverology - a science?

"You get an ology, you're a scientist," said Maureen Lipman in the BT ads. But what if you invent one?  Deliverology is a phrase conjured up by a civil servant to describe the methods of the Prime Minister's Delivery Unit (PMDU), set up by Tony Blair in 2001 and led by Sir Michael Barber, who describes its work and principles in two books, Instruction to Deliver (2007) and Deliverology (2010, chiefly for the US market.) 

The PMDU was the key element in Blair's idea of "joined up government". It was a small group, operating with Blair's full personal authority, designed to provide "gentle but unrelenting" pressure on departments to do as Blair wished - ie improve what he saw as delivery of services. Its remit included policing, transport - "I only want you to sort out the railways" - health and education, and Sir Michael's books make it clear that he sees the approach as a success. I think this is only partially true. 

Let's begin with the successes, the first of which is unequivocal. Sir Michael was part of a small group set up by Gillian Shephard as secretary of state to find a solution to the disaster of Hackney Downs School. The outcome of closing the school and starting again led to Mossbourne Community Academy, a breakthrough in state education in inner cities, and a major achievement on the part of all involved in it, including Sir Michael and Tony Blair.

The work on policing led to a reduction in recorded crime figures that persists, but that may or may not reflect a real reduction in crime, as no-one who does not have a vested interest believes them. The legal aid bill may be a better indication of crime rates. The state of the railways is a matter of debate - public subsidy has been massive and the fare structure for long journeys is ludicrous, with some tickets virtually given away, and others adding the cost to a hotel room to what might be seen as a normal fare. Can Sir Michael really claim to have "sorted out the railways"?

Has health improved? We have new hospitals, but not enough nurses, and those we have do not do enough nursing. On the other hand, the NHS is beset by health tourism and population changes that make it very difficult for it to function as we wish it to, and it would not be fair to blame this on deliverology, unless this were seen as the solution to all problems. I recently heard a spokesman for the Royal College of Surgeons lament at that organisation's exclusion from the monitoring of hospitals, which he felt would have dealt with the situation in Staffordshire more effectively. Management without detailed knowledge of the field is to an important degree operating in the dark.

So to education, which is the field Sir Michael knows most about, and particularly to the National Literacy Strategy, for which he feels personally responsible. Sir Michael notes a rapid improvement in literacy up to around 2000, followed by a stall and failure to meet targets, which he sees as a key background factor in the fall of Estelle Morris. Sir Michael sees the failure to meet the target as stemming from his taking his foot off the pedal, and herein lies the error. Literacy standards improved from 1997  because schools were paying specific attention to English teaching rather than, as its director, John Stannard, put it, "piggybacking" it onto other subjects. They stopped improving because of basic weaknesses in the methodology of the strategy that neither Sir Michael nor David Blunkett were in a position to identify. These included over-rigid and excessive lesson plans (I described the approach in The Guardian as "Cromwellian"), the "searchlights" approach to reading, and formulaic importation of grammatical terminology from the adult discpline of linguistics, irrespective of whether it was the best way to teach it. Similar errors beset every element of the strategy's work, including its over-scripted plans for children with learning difficulties, which prevented teachers from properly considering the child in front of them.

So, the strategy ensured consistency, which is better than chaos. It did not lay the foundations for further progress, any more than its companion in maths did. There is a similar weakness in the managerial approach of McKinsey and Company, to which Sir Michael moved after the 2005 election, and which relies overwhelmingly on data and movement of data as evidence of success. In an activity such as railways, it is not too difficult to gather reliable data. Train services can be measured. The rise of accountancy to its current dominant position in management consultancy is based on using information that accountants had and that other people did not. In business, it works, or the accountants would not be coining it as they do.

In education, though, data are based on analysis of performance. This analysis provides a buffer zone between the facts and the data, and this buffer zone allows manipulation. So, to take an example from Deliverology, you can increase the number of graduates by increasing university admissions (either by lowering grade requirements or by giving out more high grades) and lowering the requirements for graduation. The figures will then show that you are making improvements. An examiner wrote to The Guardian recently that the government had instructed his chief to reduce all grade boundaries by five per cent. This would improve the pass rate, and similar manipulation took place in the late 90s, when the QCA ordered the Cambridge board to reduce its boundaries after it had completed its trials. These shenanigans make test scores as reliable as the figures in Stalin's five-year plans. Onwards and upwards.

So, did the PMDU deliver nothing?  No. It delivered more consistent management and a focus on figures. Its emphasis on unreliable data may, though, have been responsible for the calamity that hit the school inspection service in 2005. Sir Michael does not mention this in his book, and it would be interesting to find out. I'll try a FoI request.

 

 

 

Spanish lesson using ipad.

Thanks to Joe Dale for this link to an excellent account of a Spanish lesson in which students wrote and recorded a simple story using ipads. The detail is important.

Why Johnny still can't read. Updated, 7pm 31st July

Last year, I posted twice about a six year old who had not made a start in learning to read, and who had made immediate progress using my basic approach after two very strenuous lessons. A note is here.  In the interim, he had had one hour per week from his school, focusing on general discussion rather than reading, and some extra phonics work from a retired teacher. Neither school nor teacher was interested in what I had done with the pupil, and the school did not acknowledge a letter I sent  describing the work.  The pupil, whom we will call Johnny, has just started Y3 with his reading assessed as L1c, ie barely started, and his carer brought him to see me again. 

The assessment was accurate, and showed no progress from the time of our last lesson. Why? The answer is that, whatever teaching Johnny was given, it had not been followed through to ensure that the words that had been studied had been practised and securely lodged in his long-term memory. As a result, he was having to try to work out each word from scratch each time, and could not do it, partly because blending sounds was a very slow and tortuous process, and partly because the high incidence of irregularity in common words made the process unreliable. So, the teaching he was given - such as it was - did not have a chance to take root.

We started again, with Not Now, Bernard (NNB).  Johnny could read not, which was an improvement from our starting point last year, but not  now, which we had to work on. He read Bernard as Ben. So, he had problems with any letter combination beyond a three-letter word, and with any word of more than one syllable. His second error showed that he was using information from letters, but not all of it, and that he was guessing.  We looked at Richard, which I unpacked and explained, and then Bernard. 

On page 1, Johnny could not read Hello, whieh we worked on using the letters as I could not think of a word with similar patterns and the same spelling, but managed Dad and said from memory. Johnny was very attentive to the picture on page 2. When I asked what showed the father's feelings, he said, "You have three options" - pointing to the hammer hitting the finger, the face turning green, and the shout. He then noticed two more "options" which were that the father was jumping, and had his eyes screwed up. I hadn't noticed all of these myself, but was more pleased when Johnny read father correctly.

We proceeded in the same way until P10.  Johnny had real problems with ate, which he pronounced et, a local variation which is some way removed from the letters, and indoors, an irregular spelling, which only has floor as a directly comparable word. However, we worked on these during a second session by telephone - Johnny is now happy to work in this way - and he got them right. 

This posting will be kept updated. The key lesson is that there are dimensions of thinking and learning beyond the surface of what are generally seen as teaching methods that are crucial to establishing a base for children in Johnny's position. For some children, recall of what they have learned is virtually automatic, and they can then generalise. For those like Johnny, learning to read is much more like learning music, and requires constant practice if they are to make the mental adjustments required in moving from spoken language to print.

 

Update 1, 27th July, 7pm.  Johnny's carer (C) tried reading with him this morning, and found it difficult. Johnny is used to people starting words for him if he can't read them, and the support I use is different - essentially, moving to a word with the same structure, working on it, showing more examples, and then moving back. So, I phoned him this evening, and he read went as walked, picking up the wrong idea from the shared first letter, and the pic of Bernard walking down the stairs.  I asked C to write bent, then sent. Johnny found it hard to see the connection, so I worked on these words a little more, then added Kent, which he read correctly. He then read p7 of NNB, which had two other words he had previously struggled with, ate and every, without hesitation, and the next page which in cluded indoors, which had also held us up in the previous lesson. I then asked C to write soar, and discussed with Johnny how a bird might soar upwards. We moved to boar, and Johnny got it first time, though he didn't know what it was. He had, though, seen a wild pig, and we talked about it. I asked C to write down roar, the first word of the next page, and Johnny read it. We then had to work on behind, using kind and find as models. Johnny got this right, but went back to an old error, substituting mum for mother. With some persuasion, he read it again and got mother right, then the next page.

 

This was a tricky session on the phone, as Johnny would sometimes respond to me by nodding, which of course I could not see, so C had to mediate. Nevertheless, we got two pages read accurately, as well as some words that had previously caused trouble, and Johnny is beginning to see how the letters will help him. C agreed with me when I said that it looked as if people had not really followed teaching through to the point at which he fully understood and remembered the words he had learned. It is worth emphasising, though, that in words like behind we are dealing with vowels that do not represent their most frequent sound, and roar, with a group of two vowels that need to be practised.  In both cases, the information conveyed by letters needs to be interpreted to understand how it operates in the word.

 

Update 2, 31st July, 7pm. A difficult session this morning. C had tried to work with J yesterday, and he had not co-operated. However, J phoned at 10 am and asked me to ring him back for a lesson. When I did so, he found it hard to focus on the book and to look at the words C was writing down when he had a problem - eg put. I decided after 15 minutes to break off the lesson, and did so kindly but firmly, pointing out to J that he had rung me and asked for a lesson, and that I had stopped what I was doing to teach him, and that he needed to be fair to me, with which he agreed.

At 6.30 this evening, I found another phone message. J said he was sorry and asked me to ring back. He completed NNB,  with work on difficult words - taken (originally read as took), I've and I'm, which are difficult abbreviations for a weak reader. However, he read the words he'd previously struggled with accurately, and I asked C to write all the words he'd had trouble with, from the beginning, on cards, and to have him work on speeding up his reading of them. I finished by asking J if he'd like a new book tomorrow. He said that he'd prefer more work on this one, "So I'll get better and better," and that's what we'll do.

Some thoughts on the new National Curriculum.

The slimmed down NC has been generally welcomed for its scope and lack of prescription. It also represents a revolution for the initial stages of language teaching, as it requires a balance of reading, writing, speaking and listening in an area where spoken language has previously been predominant. The main reason for this lies in the evidence from brain research of the overlap and interconnectedness of brain areas active in spoken and written language, as set out by Matt Davis in his ALL article earlier this year. It also reflects HMI's finding in Modern Languages Achievement and Challenge, that good practice in reading and writing are seldom found in primary and early secondary work.

The requirement that children should write from memory rather than copy in the initial stages is also new, and is based on trials, first in Essex (published by Strasbourg U. In 1995) and later in Hackney, that have shown that it is possible for beginners to enjoy writing and to take a pride in writing what they want to say accurately in their new language.  Writing is not necessarily more difficult than speaking - we can take our time, review it, add things and cross things out - and, provided children understand what they are doing, each can and should contribute to the other.  If we can see things and read words as well as hear them, we are using two channels of communication rather than one. Written language will also stand still for us to look closely at it, in a way that spoken language won't. Not everyone can "catch the wind" of spoken language, and insistence on their doing so is a major  cause of failure.

So, it is no exaggeration to say that the new NC represents a new beginning for language teaching in our schools, and one that leaves all of us free to develop effective approaches using whatever we find works, including the good parts of all previous national guidance. There are lots of blogs and sites available at little or no cost - as well as the expensive ones - and there is no harm in listing them.  We also need more sites that will help teachers improve their skills, not from some predetermined starting point, but from where they happen to be. The internet makes these virtually free to operate and to run, and so, given goodwill, is research. We don't need hundreds of thousands of pounds to try out ideas with a class or in a cluster and to report what we find, and the emphasis on expensive studies must not be allowed to dominate us. Joseph Lister started with one boy with a broken leg, Vygotsky and Fernald with their observations in their own schools and nurseries.  Research and reporting should be built into our normal work, and we should share the results on websites, free of charge.

 

Designing small scale research in languages - first draft.

This file is the first draft of a template for constructing small-scale research projects in modern languages, based on the principle of best evidence rather than absolute truth. There has been much more experimental work in reading than in languages, and hence many more mistakes. It would be straightforward to provide a reference for each of the errors I've listed, but the point here is not to criticise the studies themselves - see my notes on randomised controlled trials as a starting point for such criticism - but to avoid repeating the errors. Correspondence is, as ever, most welcome.

Download Some Principles of Small-scale Research Design.