We’re all different

We’re all different. Tiny variations in our DNA before and at conception. Our parents smoked/didn’t smoke, drank/didn’t drink, followed ideal/inadequate diets, enjoyed robust health or had food intolerances, allergies, viral infections. We were brought up in middle class suburbia/a tower block and attended inadequate/outstanding schools. All those factors contribute to who we are and what we are capable of achieving.

That variation, inherent in all biological organisms, is vital to our survival as a species. Without it, we couldn’t adapt to a changing environment or form communities that successfully protect us. The downside of that inherent variation is that some of us draw a short straw. Some variations mean we don’t make it through childhood, have lifelong health problems or die young. Or that we become what Katie Ashford, SENCO at Michaela Community School, Wembley calls the ‘weakest pupils‘.

Although the factors that contribute to our development aren’t, strictly speaking, random, they are so many and so varied, they might as well be random. That means that in a large population, the measurement of any characteristic affected by many factors – height, blood pressure, intelligence, reading ability – will form what’s known as a normal distribution; the familiar bell curve.

 The bell curve

If a particular characteristic forms a bell-shaped distribution, that allows us to make certain predictions about a large population. For that characteristic, 50% of the population will score above average and 50% below average; there will be relatively few people who are actually average. We’ll know that around 70% of the population will score fairly close to average, around 25% noticeably above or below it, and around 5% considerably higher or lower. That’s why medical reference ranges for various characteristics are based on the upper and lower measurements for 95% of the population; if your blood glucose levels or thyroid function is in the lowest or highest 2.5%, you’re likely to have a real problem, rather than a normal variation.

So in terms of general ability that means around 2.5% of the population will be in a position to decide whether they’d rather be an Olympic athlete, a brain surgeon or Prime Minister (or all three), whereas another 2.5% will find everyday life challenging.

What does a normal distribution mean for education? Educational attainment is affected by many causal factors, so by bizarre coincidence the attainment of 50% of school pupils is above average, and 50% below it. Around 20% of pupils have ‘special educational needs’ and around 2.5% will have educational needs that are significant enough to warrant a Statement of Special Educational Needs (recently replaced by Education Health and Care Plans).

Special educational needs

In 1978, the Warnock report pointed out that based on historical data, up to 20% of school pupils would probably have special educational needs at some point in their school career. ‘Special educational needs’ has a precise but relative meaning in law. It’s defined in terms of pupils requiring educational provision additional to or different from “educational facilities of a kind generally provided for children of the same age in schools within the area of the local education authority”.

Statements of SEN

The proportion of pupils with statements of SEN remained consistently at around 2.8% between 2005 and 2013 (after which the SEN system changed). http://www.publications.parliament.uk/pa/cm200506/cmselect/cmeduski/478/478i.pdf https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/225699/SFR30-2013_Text.pdf

It could be, of course, that these figures are an artifact of the system; anecdotal evidence suggests that some local authorities considered statutory assessments only for children who scored below the 2nd percentile on the WISC scale. Or it could be that measures of educational attainment do reflect the effectively random nature of the causes of educational attainment. In other words, a single measure of educational attainment can tell us whether a child’s attainment is unusually high or low; it can’t tell us why it’s unusually high or low. That often requires a bit of detective work.

If they can do it, anyone can

Some people feel uncomfortable with the idea of human populations having inherent variation; it smacks of determinism, excuses and complacency. So from time to time we read inspiring accounts of children in a school in a deprived inner city borough all reading fluently by the age of 6, or of the GCSE A*-C grades in a once failing school leaping from 30-60% in a year.  The implication is that if they can do it, anyone can. That’s a false assumption. Those things can happen in some schools. But they can’t happen in all schools simultaneously because of the variation inherent in human populations and because of the nature of life events (see previous post).

Children of differing abilities don’t distribute themselves neatly across schools. Some schools might have no children with statements and others might have many. Even if all circumstances were equal (which they’re not) clustering occurs within random distributions. This is a well-know phenomenon in epidemiology; towns with high numbers of cancer patients or hospitals with high numbers of unexpected deaths where no causal factors are identified tend to attract the attention of conspiracy theorists. This clustering illusion isn’t so well known in educational circles. It’s all too easy to assume that a school has few children with special educational needs because of the high quality of teaching, or that a school has many children with SEN because teaching is poor. Obviously, it’s more complicated than that.

What helps the weakest pupils?

According to Katie what ‘the weakest pupils’ need is “more focus, more rigour and more practice if they are to stand any chance of catching up with their peers”.   Katie goes on to unpack what she means. More focus means classrooms that aren’t chaotic. More rigour means expecting children to read challenging texts. More practice means practicing the things they can’t do, not the things they can.

Katie’s post is based on the assumption that the weakest pupils can and should ‘catch up with their peers.’ But it’s not clear what she means by that. Does she mean the school not needing a bottom set? All pupils attaining at least the national average for their age group? All pupils clustered at the high end of the attainment range?  She doesn’t say.

In a twitter discussion, Katie agreed that there is variation inherent in a population, but

katie ashford bell curve

I agree with Katie that there is often room for improvement, and that her focus, getting all children reading, can make a big difference, but improvement is likely to entail more than more focus, more rigour and more practice. In an earlier post Katie complains that “Too many people overcomplicate the role of SENCO”.   She sees her role as very simple: “I avoid pointless meetings, unnecessary paperwork and attending timewasting conferences as much as possible. Instead, I teach, organise interventions, spend lots of time with the pupils, and make sure teachers and support staff have everything they need to teach their kids really, really well.

Her approach sounds very sensible.  But she doesn’t say what the interventions are. Or what the teachers and support staff need to teach their kids really, really well. Or what meetings, paperwork and conferences she thinks are pointless, unnecessary and timewasting. Katie doesn’t say how many children at Michaela have statements of special needs or EHCPs – presumably some children have arrived there with these in place. Or what she does about the meetings and paperwork involved. Or how she tracks individual children’s progress. (I’m not suggesting that statements and EHCPs are the way to go – just that currently they’re part of the system and SENCOs have to deal with them).

What puzzled me most about Katie’s interventions was that they bore little resemblance to those I’ve seen other SENCOs implement in mainstream schools. It’s possible that they’ve overcomplicated their role.   It could be that the SENCOs I’ve watched at work are in primary schools and that at secondary level it’s different. Another explanation is that they’ve identified the root causes of children’s learning difficulties and have addressed them.

They’ve introduced visual timetables, taught all pupils Makaton, brought in speech and language therapists to train staff, installed the same flooring throughout the building to improve the mobility of children with cerebral palsy or epilepsy and integrated music, movement and drama into the curriculum. They’ve developed assistive technology for children with sensory impairments and built up an extensive, accessible school library that includes easy-to-read books with content suitable for older kids for poor readers and more challenging texts with content suitable for younger kids for good readers. They’ve planted gardens and attended forest schools regularly to develop motor and sensory skills.

As a child who read avidly  –  including Dickens – I can see how many hours of reading and working through chapters of Dickens could improve the reading ability of many children. But I’m still struggling to see how that would work for a kid whose epilepsy results in frequent ‘absences’ of attention or who has weak control of eye movements, an auditory processing impairment or very limited working memory capacity.

I’m aware that ‘special educational needs’ is a contentious label and that it’s often only applied because children aren’t being taught well, or taught appropriately. I’m utterly committed to the idea of every child being given the best possible education. I just don’t see any evidence to support the idea that catching up with one’s peers is a measure of educational excellence, or that practicing what you’re not good at is a better use of time than doing what you are good at.

Section 7 of the Education Act 1996 (based on the 1944 Education Act) frames a suitable education in terms of an individual child’s age, ability and aptitude, and any special educational needs they may have. The education system appears to have recently lost sight of the aptitude element. I fear that an emphasis on ‘catching up’ with one’s peers and on addressing weaknesses rather than developing strengths will inevitably result in many children seeing themselves as failing to jump an arbitrary hurdle, rather than as individuals with unique sets of talents and aptitudes who can play a useful and fulfilling role in society.

I’d be interested in Katie’s comments.


34 thoughts on “We’re all different

  1. Brilliant. Thank you.

    I wonder what would have become of Picasso if he’d focussed only on what he wasn’t good at? Or my local millionaire corner shop owner? Or Beyoncé?

    Re: ‘clustering occurs within random distributions. This is a well-know phenomenon in epidemiology; towns with high numbers of cancer patients or hospitals with high numbers of unexpected deaths where no causal factors are identified’
    This is quantum. I am no physicist, you probably know better than me, but everything, including ‘the human condition’, comes in waves. Everything. I think maybe being human or even just being alive means finding ways, and for teachers and parents and ‘society’, showing children how to find ways to surf those waves to the best of our ability? I suppose Michaela think that’s what they are doing?

    I say let them get on with it, but let their fans stop having a go at people who do things differently. There are plenty of other schools doing well ‘in deprived areas’ who don’t do what Michaela does. Not that I take OFSTED as our great redeemer, but last time they came to a school I was in – in a far more deprived area than Michaela’s – they couldn’t believe how well children behaved. They compared registers with the roll to see if the disruptives had been sent out for the day. They ate in the dinner room and watched the playground through the windows before, during and after school. They couldn’t believe it. There are plenty of schools where behaviour is not out of control. I believe this was the root of the success of the London challenge. But if other educators want to do what Michaela does, they can go and watch and emulate them. They are always inviting, entreating us to do so.

    I don’t understand all this all-pervading edublogger stuff implying ‘my methods are good because yours are bad.’ It has nothing to do with education? It has everything to do with politics, politics being a fancy word for ‘how we share the money out.’

  2. Yes. Each individual is unique: we’re all different. But, no. No individual is uniquely unique; we’re all the same. This superficial paradox is resolved by our understanding of the determinants of the phenomenon at issue. Only when the determinants are multiple and complex do we observe a bell-shaped/normal/Guassian distribution. That is, the “causes are out of control” and are said to be “due to chance.”

    In EdLand there is a predilection to forcing a matter into a normal distribution even when it is counter-productive to do so. For example, we do this in standardized test construction, and we do it in many aspects of “special educational needs.” Then we are disappointed when it is so difficult to “eliminate gaps” or (as you say) “catch up.”

    To reliably achieve any educational intent, inherently involves bringing the determinants under control and shifting the normal distribution to one in which all of the observations pile up at the “strong end,” eliminating the “weak end”–or explaining the determinants of the remaining weakness.

    The achievement of the intent can be done only by specifying the minimum prerequisites required of all students involved and then insuring that all children involved have these prerequisites, irrespective of their other differences. That is, at the start, you focus on aspects that are “all the same.” If you focus on aspects in which they are “all different,” they will remain “all different.”

    To take the logic out of the abstract into a concrete example, consider reading instruction. Children entering Reception/Kindergarten are known to vary widely in vocabulary and other language characteristics–with normal curve distributions observed. The popular stance is to promote preschool and poverty-elimination initiatives to “reduce the gap.” Not only is this impractical, it’s unnecessary. (That’s not to argue against these initiatives. Only to say that the initiatives are not prerequisite to accomplishing any instructional intent.)

    The smart stance is to focus on the known fact that all children entering Reception/Kindergarten, with few exceptions can speak in whole sentences and participate in everyday conversation. That is, they all have the minimum prerequisite for reading instruction, even though a few will “smart as a whip” and a few will be “dumb as dirt.” By the end of Grade/Year 1 instruction, the “normal distribution can be skewed into a distribution where the observations pile up up at the top, and SEN “specific reading disability” has been eliminated. This has been done in over 700 schools in England and is accomplishable in other UK schools and in all English-speaking schools throughout the world.

    The same logic that applies to individuals applies to schools: they’re all diffiferent–and they’re all the same.

      • Unfortunately, the schools remain anonymous at this point. No analysis of the Screening Check data has yet been done at the schooi level. The investigations that have been done have not delved below the National and LEA level. The National results support the logic that I posted. The LEA results show wide variability among LEA’s, that is not a function of demography; it’s in the instruction at the school and classroom level. As I’ve said in this venue and in a lot of other venues, the methodology for teasing out “what’s going on” is easy and inexpensive. To date though, there has not been the will

        With all the drum-beating for “evidence-based decision making,” “Research-ED,” and such, it seems odd. “Logical incrementalism” seems to be dragging its feet.

      • Local authority areas vary hugely. Surrey isn’t directly comparable to Sunderland.

        And the current trend for ‘evidence-based decision-making’ is more about theory than practice. There’s simply too much political interference.

  3. Hmm. The thing is, the brains of the littlies in Surrey are directly comparable to the brains of the littlies in Sunderland. And all of the children with few exceptions have the prerequisites to be taught how to read by the time they enter reception. The other thing is, performance on the Alphabetic Code [Phonics] Screening Check at the end of Yr 1 and Yr2 does not closely track with the differences the demographic differences among the LEAs.

    “Political interference” is too facile an explanation. Certainly politics, economics, psychology, and the other sciences are in play, as they are in every other aspect of life. The evidence of the Screening Check is both theoretically and practically relevant. The thing is, analysis at the school and class level would refute educational beliefs that benefit powerful interests.

    • No large population in one area is directly comparable to another if they have different geographical or socio-economic characteristics. That’s not how populations work.

      And assuming that all of the children with few exceptions have the prerequisites to be taught how to read is just that; an assumption. Like many people, you have blithely written off children in special schools and those that scored 0 in the check as ‘a few exceptions’.

      Also, you appear to have assumed that I am suggesting that socio-economic deprivation is the cause of low scores. I’m not. I’m saying that the causes of children’s performance in school are many and varied. Many statistical methods are based on the variation inherent within large populations.

      For the past three decades, central government has had significant control over the English education system in a way it had never had previously. And has micromanaged it ever since. It micromanages the evidence too. Political interference isn’t just ‘in play’.

      Which powerful interests do you mean?

  4. The 2015 Yr 1 performance of Surrey and Sunderland on the Screening Check was exactly the same–78% pass. And Yr 2, almost the same–92% in Surrey and 91% in Sunderland.

    • They are still not directly comparable. You can’t conclude that because scores were the same it must have been due to tuition any more than you could conclude that social deprivation caused low scores. That’s not how complex causality works.

      • 78% Sunderland = 78% Surrey. That’s not much to brag about in either LEA. By the end of Yr 2, the % is 91-92% The increase wasn’t due to less “social deprivation.” It was due to “tuition”–of some sort. We don’t know what the tuition was, or how it differed from that in schools–some of which could be in Surrey or Sunderland–who taught all their children to pass the Check. But we could easily find out. That’s how science works to untangle complex matters.

  5. I think we need to stop taking it as read that passing the phonics screener check is the equivalent of ‘literacy’.
    It is not.
    Teaching phonics means children learn phonics which is a fraction of learning to read and write, but no more than a fraction.. There is far more to reading and writing than phonics. The 700 schools mentioned above are those that had 95% or more (or was it 100%?? Did Nick Gibb send out 700 congratulations letters?) pass the screener in year 1 and their ‘success’ is attributed to the use of famous synthetic phonics schemes. If those 700 schools are also outstanding in their literacy attainment, lots more is being done than just phonics. “This school is outstanding; this school does SP, therefore SP is outstanding”, is a narrative that seems to be bamboozling too many people in positions of power. It needs to be resisted. I was refused an information request for the names of the 700, but I have looked at several data dashboards for schools that have had their phonics screener ‘success’ lauded in their local papers. Their demographics deserve a bit of candid analysis, in my opinion.
    I could show you today several year 2 children who ‘failed’ the screener last year. They have marked difficulties with oral language – EAL together with S&L difficulties – and are still at early stages of learning to read and write. But they have had an extra year of having phonics drummed into them, and have passed the screener this year.
    The screener does not equal literacy. Blending a mix of 20 words and 20 pseudo GPC clumps that are not words equals literacy? How inappropriate of certain SP proponents and Nick Gibb to suggest that it does. But let’s not argue about the pseud words. We need to challenge the notion that phonics equals literacy. It does not.

      • Well, since you insist.

        You’re absolutely right “that passing the phonics screener check is NOT the equivalent of ‘literacy’.” I don’t recall Gibb or any SP proponent claiming it is, and I certainly didn’t, but that’s not the point. We could also quibble about the cut score and items in the Screen, but that too would miss the point.

        The Check is a valid means for identifying individuals who require further instruction in handling the English Alphabetic Code; individuals who can read all of the items on the Check do not require any further instruction in reading per se. They now have the prerequisite to learn and be taught not only more “literacy” but other matters using their reading capability.

        The risk in not using such a Screen is that some children otherwise “fall through the cracks.” Not only do they lack the prerequisite for subsequent “literacy” instruction, but they are eventually termed “dyslexic,” mistaking student-difference for mal-instruction.

      • “…individuals who can read all of the items on the Check do not require any further instruction in reading per se…”
        Absolutely untrue. Many children can read all the items on the check are still very much in need of further instruction. I’m not sure what you mean by per se?
        This comment demonstrates to me that you don’t really know about the practicalities of teaching children to read; that you don’t do it.

        “The risk in not using such a Screen is that some children otherwise “fall through the cracks.” Not only do they lack the prerequisite for subsequent “literacy” instruction, but they are eventually termed “dyslexic,” mistaking student-difference for mal-instruction.”
        This sounds to me that you think knowing all my phonics is a prerequisite for ‘literacy’ instruction; that I need my phonics first and then I learn to read and write. Again, absolutely untrue. The PSC is at the end of year 1. Do I not do any reading of books or writing before then?
        Prerequisite for literacy instruction? Another arbitrary term that shows me you don’t do this teaching of phonics or reading and writing. It shows me you have read or heard and decided to agree with statements made by commercial SP scheme sales people.

        Would you like to reveal your identity? Most serious bloggers and commenters are not ashamed to tell others their names. “Fall through the cracks”? Sounds like Nick Gibb? Are you he?

  6. Ooops. My first in this thread “revealed my name.” I stopped paying any attention to how I got “signed in” after that, and my name somehow got lost in the drift. Sorry, I meant to post “in person” throughout.

    I didn’t say that children who pass the check didn’t need further instruction. They can read any text with understanding equal to that were the communication read to them. That’s what I meant by “per se.”

    I didn’t say anything about “Phonics.” Actually, the Screening Check is mis-titled. That’s why I refer to it as an Alphabetic Code Screening Check –with “Phonics” in brackets. I said that passing the Check is prerequisite to “MORE ‘literacy’ instruction,” no to “literacy instruction. The term, “literacy” has come to be little more than an honorific term instructionally, but that’s a whole nother story.

  7. What are you claiming for / about the 700? What has been in those schools you allude to that you cannot name?

    • To repeat what I said in my first comment on this thread: “By the end of Grade/Year 1 instruction, the “normal distribution can be skewed into a distribution where the observations pile up up at the top, and SEN “specific reading disability” has been eliminated. This has been done in over 700 schools in England and is accomplishable in other UK schools and in all English-speaking schools throughout the world.”

      The blog is titled, “We are all different.” My point was/is: “We are all the same” also holds. Reliable instruction inherently converts a bell-shaped distribution into a highly skewed distribution with as thin a tail at the low end as possible. On a national level, Screening Check performance has improved each year in this regard from 2011 to 2015, but the data have not been analyzed at the school/class level–where the action is.

      • I am asking what is this instruction you speak of, that has been done in 700 schools and eliminated ‘specific reading disability’?
        I would like to know:
        What is the instruction?
        How are effects of this instruction measured?
        How are these measurements interpreted?
        What is the ‘specific learning disability’ that has been eliminated?

  8. “…they don’t need to blend the phonemes? Blending is a big problem for children with low auditory/visual working memory capacity.”
    Not sure what you mean by not needing to blend phonemes? That’s what the PSC is meant to check?
    I agree it is a big problem for children with low auditory/visual working memory capacity.

  9. All human beings (with few exceptions) have a working memory capacity of 4-7 elements; The crux is what is in long term memory to be drawn upon. The action in reading is not “blending the phonemes. It’s in the Grapheme/Phoneme CORRESPONDENCES that constitute the Alphabetic Code. In languages such as Finnish, Turkish, and Spanish where the Correspondences are all one-to-one, children can be taught to handle the Code in a couple of months. The English Code is more complex and instruction takes much longer. “The end of Yr 1” is a reasonable time expectation to finish the job.

    The PSC is a valid means for identifying individuals who need further instruction in handling the Correspondences, not the phonemes. Children have been “blending phonemes” since infancy when they learned to talk.

    • I have to give an out-of-the-box response, here:
      When Rowan Williams was Archbishop of Canterbury, his statements and responses to questions and events were so convoluted, so vague, so full of sophistry, that I knew instinctively that he did not fully understand nor believe in his church’s fundamentals.

    • When children begin to talk, they are not blending phonemes, they are mimicking patterns of sound with symbolic value. We know they start with words/phrases and it takes time for them to identify the components of those words/phrases via implicit learning. It’s the same process as someone learning a foreign language by immersion rather than formal instruction.

      The whole point of SSP is that many young children struggle to identify the components of words at the phoneme level or below. Making that identification, and grapheme-phoneme correspondences, explicit, gives children the prerequisites of reading, as you say.

      But SSP is an explicit, conscious process, unlike learning to talk which is largely implicit and subconscious. In learning to read, children need to be able make the GP correspondences, then blend the phonemes the graphemes represent, and this depends on how well their working memory is working.

      Young children tend to have a smaller working memory capacity than adults. This is hardly surprising since working memory is essentially a bunch of neural loops that briefly replay sensory signals, and children’s neural networks are still developing.

      SSP is clearly very beneficial for the majority of children. Some of the children who had low scores on the Check probably had poor tuition. But some of them are going to be the ‘few exceptions’ that you keep mentioning in passing. It’s not safe to assume that we know how few ‘exceptions’ there are – because as far as I’m aware, we don’t know that.

  10. What you don’t have is what Dehaene is saying in these two papers. He doesn’t say anything about “blending,” “SSP,” “limited working working memory,” “implicit learning,” or other elements of your depiction here. And he adds important elements that you omit.

    You’re close, but not “exactly.”

    • No because he’s describing what happens at the neurological/brain area level, not at the behavioural level.

      Of course he adds elements that I omit. I was commenting on a discussion about the phonics check, not writing a discourse on the mechanism of reading.

      • Hmm. Your comment deals with the “mechanism of reading” up until the last paragraph where you deal with the Screening Check. You have the last paragraph “exactly.”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s