Tuesday 19 June 2018

Exams are designed to confirm the capabilities of those who devise them.



I believe that exam systems are designed to reinforce and confirm the capabilities or self-perceived capabilities of those who devise them. They then make this equal ‘education’.

That's to say:  their content, their ways of asking questions, the way they are marked, the kind of preparation needed to do well in them, the amount of preparation needed to do well in them, their ways of being marked, the immediate outcomes of having succeeded or failed at them, the ways in which certain exam subjects are prioritised over others, the way certain subjects, kinds of knowledge, fields of knowledge are not included in the exam system - all this, and probably more, are designed and put into practice in order to confirm, affirm and reinforce the actual capabilities or the self-perceived capabilities of those who design the exams, set them, mark them, big them up as being necessary, good, essential.

This exam system is ultimately what is the 'knowledge' that schools are asked to transmit to pupils. This has become more and more the case as the preparation required for the exams has increased and squeezed out those school activities which are not examined. The testing and exam system not only contains the knowledge required, it shapes that knowledge, it shapes how it is acquired, how it is taught, what activities go on in relation to that knowledge so that it is transmitted, how all this creates a particular kind of behaviour in relation to teaching, learning and knowledge. In fact, there is a good argument for saying that there comes a point at which it is hard for any learner or teacher to fully distinguish between what is learned and how it is learned. As a consequence of the way I was taught 'mental arithmetic' in 1955-1957, I find it almost impossible to separate out the speed with which I am 'supposed to' perform an act of arithmetic from the sum itself. If you ask me what is 9 times 8, I am immediately in a state of rush. 9 times 8 is not simply the content 72. It is the pattern of behaviour of trying to say back to  you (or myself) 72 as fast as I can. Incidentally, I have no other strategies of trying to work that out, other than to go to a calculator. Rationally, while I am writing this blog, I know that I could say to myself that 10 times 8 is 80, and so 9 times 8 is 8 less than that and as I'm 'fast' on base ten calculations and number bonds of 10, but this is not about rationality. The reinforced behaviour pattern of Miss Williams patrolling the class, bent at the knee, trying to find backsliders who haven't learned their times tables, pointing at the failing children (me), shouting the question at us, under threat of failing the 11-plus, is way, way more important in my mind than the rationality of using base ten to solve things. 

The people in charge of education - at the very top - are people with a very narrow 'formation'. They are all university educated, predominantly Oxbridge, many are privately or part-privately educated.  They work to assumptions based on the kind of knowledge that they rate as important. Inescapable from their outlook are such matters as their route to the top of the state power tree, what kinds of schools they put their own children through. They do hardly anything to widen this 'constituency' out in order to weigh up what other possibilities there are in terms of education systems, exam systems, or views of what constitutes knowledge. As they are 'successful', they make assumptions about what constitutes success, what is the best route to what they think of define as success. 

So, when we break the exam system down, we can see that it involves certain kinds of behaviour and a certain kind of stance towards exam-knowledge, and of course a certain kind of exam-knowledge. Because we have high-stakes, centrally organised, norm-referenced exams, there is an in-built lack of questioning of the system itself. The system is the system is the system. There is no way within the system to raise questions about the system itself. It just 'is' the best way, the only way. And yet every year, there are gaffes within the system but because the system is infallible, the only people who are allowed to be fallible are the students and teachers. 

Further, exams can only test what is examinable, on the only combination of people who can be examined ie the individual. The fact that knowledge does not belong to an individual but is created by groups and society as a whole, the fact that we acquire knowledge through language (which is itself a socially created practice), the fact that we acquire knowledge socially (in classrooms, families, labs, etc) is all ignored because we have created the notion that knowledge 'belongs' to the individual - even though it doesn't! Further to this, is the way in which a 'subject' - let's take my own, English, has to be boiled down into examinable units. A play, like say, 'Romeo and Juliet' has to be reduced to what are examinable (ie markable) 'points'. English teachers are required to make lists and schemes of these markable points about 'Romeo and Juliet' and by various systems of instruction, convincing talk, get the students to repeat these points within fixed time limits on a specific day at a specific time, a year or so later. In a sense, 'Romeo and Juliet' is industrialised into a form of mass production. It is split up under various categories of production, so that it can be reproduced at this particular time. That time - the exam day - is in effect when the product (the finished exam) rolls off the production line. 

However, this 'product' is not strictly speaking the individual student. The product is that year's cohort's exam papers. That product is then 'finished' by being marked, graded and then norm-referenced. On this morning's Woman's Hour on BBC Radio 4, we heard with clarity from the Schools Minister, Nick Gibb exactly how this norm-referencing goes on. He said that the old GCSEs were too easy, so Michael Gove made them harder. Because they are harder, he said, they had to lower the pass mark, in order to achieve 'comparable outcome' ie to make the hard one's 'outcome' (product) match the easier ones. This meant that the 'bell curve' or distribution curve of the grading system, must always be the same. A few outright fails, loads of people in the middle, a few top grades. This is how people in power who run and control education view children and ultimately the population as a whole. This is data. This, they believe, is the reliable and valid way to think about us! It also means that ultimately, schools, teachers and pupils are helpless in the face of this norm-referencing. Whatever happens, they will impose that bell-curve on the results. They will even remark papers in order to secure that curve, pushing people down or up in order to achieve it. It's called 'moderation' and indeed, as Nick Gibb said, achieving 'comparable outcomes'. 

It is of course absolutely vital that pupils do not understand this process. In terms of 'knowledge', it's interesting to note that education is full of processes e.g. discipline, setting, and this exam system itself which do not constitute what is commonly called 'knowledge'. An obvious example of this is that the framing and language of exam questions are clearly absolutely crucial to how examinees answer questions. In fact, we can all tell of examples of ourselves and/or pupils who clearly knew the knowledge required of a question but were unable to answer the question because they didn't realise or know that that is what the question was asking. In fact, I had an example of this when trying to help my then 11 year old do a sample KS2 SATs English question which asked 'Explain the use of language in....'. I couldn't. I couldn't explain the use of the language in Geraldine Kaye's 'Comfort Herself'. All I could say was that she used a particular kind of language. But I couldn't 'explain it'. Next day, the teacher in charge of setting this particular sample SATs-type paper 'explained' to me what the question meant. He said it just meant that my son had to say 'author intention'. In other words when it says 'Explain', he had to guess (in fact, read off the formulaic stuff about the supposed 'effects' of a particular kind of language) what the author's intention was. Whether this really was the author's intention, and whether this really did have the effect that the formula says it has, is really quite beside the point. This is about producing the required answer ('fact') to a question which is itself written in a particular kind of exam code which you have to learn in order to produce the kind of answer which the system says in important knowledge. The fact that it's unverifiable stuff, created out of and for exams doesn't matter. The fact that how that passage in that novel affected the reader, some readers, many readers, is again - beside the point. 

Now again, according to Nick Gibb, this morning all this is resulting in 'closing the attainment gap' by 10%. I'm not sure how he came up with this figure, but surely it cannot come from within the exam system itself because, as he himself pointed out, they are rigged to achieve comparable outcomes. The bell curve is the bell curve is the bell curve. Is it perhaps one of the international tests? 

There are several insuperable methodological problems here. The international tests do not ask the same questions as the ones asked of national pupils. In other words, the international tests act as some kind of international court of justice, with their own rationale of what constitutes important knowledge and the right way for pupils to demonstrate this. For the moment, this doesn't matter whether it is better or worse than our national system. The point is that it's different. So any 'attainment gap' judged by this system is not determined by the absolute golden rule of all educational testing - compare like with like. 

Further, we now know that the sampling procedures of those selected to sit these tests produces invalid results. Remember, a nation's educational standards are being assessed (produced) by testing a relatively tiny group of people. Take a couple of examples: Nick Gibb always fails to point out that in at least some of the English schools' samples there are children who are being educated in private schools. Children in private schools follow an education process which is not 100% determined by Nick Gibb and the state education system. Another: there are biases in the sampling in terms of e.g. urban and rural. In China, this is crucial because of patterns of migration and what is happening to rural children being 'left behind' in their villages. A statement is made by people like Nick Gibb about the superiority of the Chinese system with no reference to the fact that rural children are not being sampled by these international tests. In other words national outcomes are being mapped on to tiny unrepresentative (perhaps they will always be unrepresentative) samples. Once again this is a product. A result of tests being used as part of a mass production industry, producing results, grades, and outcomes. 

Where does this leave us? It leaves us fundamentally with a new wave of 'thinking' about education, knowledge, teaching and learning which claims to be talking objectively about such matters but which in fact are entirely intertwined and interwoven with these matters of testing, (the 'what' and the 'how' of testing and the effect of how the exam system words its way back through the system. 

In my previous blog about 'what is important about texts', the sub-text (!) is really about which of those alternatives is selected as exam-worthy (and therefore self-evidently 'important' according to the system) as opposed to those that get excluded, overlooked or marginalised as not-important. The next step in that investigation is to ask why is one kind of knowledge important and another not. And what are we missing, what are we leaving out and why,

In relation to English, the core feature which is increasingly being marginalised as the main reason why we read in the first place (and indeed the main reason why writers write!) which is to affect readers. This is not the be-all and end-all of what we can do with texts in schools, but for it to be marginalised or even excluded strikes me as perverse. I know of examples of e.g. the killing of Nancy from 'Oliver Twist' or of the falling in love of Romeo and Juliet which are treated simply as exemplifications of something and/or a series of facts which can be learned about those scenes. 

Whatever literature is for - which will often - perhaps mostly - be about affecting audiences, inviting people to ponder, wonder, muse, reflect, and relate this to the possibilities and events one's own life and society, are being pushed out the door by this method, in favour of important knowledge (?), which is perversely and in a strange self-confirming cycle, the knowledge that is tested! Important knowledge, then,  is that which is tested - but how did we arrive at such a narrow, reductive view of what's important, and what constitutes wisdom and what constitutes the stuff we need to make this world a better place?! We arrived at it because those in charge of it design a system which matches their own or their own self-perceived capabilities. 'It made us, so it must be good.'