Saturday 3 November 2012

Exams: testing, testing, testing

Every time a new exam is invented (eg Ebacc, SPAG) or an exam scandal (last summer's GCSE fiasco), the exam junkies appear. These are people who really believe in exams and testing and who love the apparatus and supposed science of them. They appear at meetings and in the media throwing around the jargon of the testing system as if the jargon itself was evidence of how worthwhile the whole thing is.

My thoughts:

1). We should never lose sight of the fact that exams are a system for sorting, selecting and segregating. This means that we have to ask ourselves, why? For what purpose? Are they really (as the test junkies claim) a means of creating a true meritocracy, where only the most able and most qualified rise to the top?

The problem with this scenario is that it ignores the fact that we live in a society that is already rigged to ensure that those in power and with wealth stay in power and keep their wealth, but also pass their power and wealth on to their offspring. This is what this society calls freedom.

The consequence of this freedom is that the powerful and the wealthy create a school system and an exam system that ensures that they either succeed in it; if they fail in it, they have the wealth to pay for more and more education until they succeed; or in the event of none of that working out, there are other routes to keeping that power and wealth - through relations and friends.

Taking the first of these: how can an exam system that seems fair, ensure that certain kinds of people succeed? It's about language and culture. In a previous blog, I gave the example of the sample of the spelling, punctuation and grammar test that is up on the DfE website. You could try an experiment, take two children - one who has a wide, deep reading habit, the other who can read but hardly ever does. Check that neither child has had any formal grammar teaching - or if they have, it's about the same. Now sit down with them and go through the sample grammar questions.

These questions as they stand at the moment, have a form which states the question, gives an illustration of a correct answer and then asks the candidate to do several on their own. What happens to the child who has the reading experience behind her/him, is that she/he can look at the given, correct answer, derive a principle and apply it to the questions. This is made easier by the fact that the shape of the sentences or phrases follows the shape of the sentence or phrase in the given correct answer. The reason why the examiners do that is because they know that children of this age don't really understand the grammar being asked of them. They learn how to spot patterns, similarities and cues and apply them - rather as I do maths, routines that I don't understand.

The reason why the child with the reading experience can do these questions is for language-and-culture reasons. By spending several hours a week reading, they have not only engaged with the feelings and ideas in the books, they have engaged with the 'vehicle' that conveys them - language. That's to say, without noticing, they have in their heads the structures, grammar, techniques, processes that formal standard English uses. So, seeing and hearing sentences shown as, say, 'active' and 'passive' might throw them for a moment if they don't know those terms, but the moment they see the examples, they can recognise them - even though most of the examples given in these tests are ludicrous sentences that no one ever says. In other words, the reading child has a kind of 'shadow sense' of the structures that lie behind the sentences given.

The non-reading child (the one who can read but doesn't) finds this very difficult. He or she doesn't have the experience of text, doesn't have that 'shadow sense' to help them spot the pattern laid down in the given correct answer.

Now, take that forward to a point, let's say, a year later, after the kind of exercises and drills that will be coming into schools, all being done in the name of fairness and ultimately that mythic meritocracy. We will tell ourselves that doing the drills and exercises will enable everyone to do the test; they will all  have had the same amount of grammar teaching.

Quite clearly, they have started that grammar teaching from a different linguistic and cultural base.

What conclusions might we draw from this?

a). Rather than spending more and more time doing less and less (ie more drills, more exercises) we should be doing everything we can to even up that linguistic and cultural base - ie getting every child reading for several hours a week.

b). We should recognise that such tests in the present set-up are largely a means of confirming a specific set of linguistic and cultural experiences. They are not about 'ability' or 'intelligence'.

c). I have focused on one kind of test, and one kind of question. This is only one tiny fraction of what goes on between class, home and school. If we look at the whole curriculum, the whole test and exam apparatus, we can look to see how the kinds of class/home experience slot into (or do not slot  into)the school experience and how this is reinforced and cemented in by the exam system - no matter how the test junkies justify it.

Another example: a few years ago in the Key stage 2 SATs paper, the main passage for comprehension was an article from the Sunday Times on caving. I'm of the view that all tests show a linguistic and cultural bias of some kind - even if it's sometimes hard to spot - but this was so blatant as to be laughable. On the one hand, there were children doing that test whose parents read the Sunday Times, and/or have the time and money and experience to do 'outdoor pursuits', and/or know people who do either or both of these things - and spend time talking and reading about such things (brochures, planning etc), and on the other, those who would have no idea what was going on, who've never been anywhere near a cave, or that kind of outdoor pursuit.

There wasn't even the pretence that what was being tested was equal or fair for everyone. It was just a piece of blatant class and culture politics enacted through the exam system. The exam system does the job that ultimately the wealthy and powerful want: ie that the population is segregated out for them into the different layers whilst ensuring that their own position (and that of their offspring) is maintained.

2). If all this sounds too much like a conspiracy, try this: consider what kinds of activities (which might include testing) would favour the unpowerful, the unwealthy? What experiences do the children of the unpowerful and unwealthy have which could in some way be part of a day's activity and which the wealthy and powerful would find as difficult as the unpowerful would have found that SAT paper on caving?

Let's start with the bilingual children. Imagine, if I waved my wand and said that bilingualism was one of the most important skills/capabilities that anyone could have and so the school system must reward this. (This would only be a version of what ministers say when they talk about, say, 'spelling' be an essential skill or some such. Again, if you think 'bilingualism' is an odd thing to choose, consider the case of Singapore (much vaunted by Gove et al), most people in Singapore are bilingual. One of the reasons for their success in the international league tables might just be that the population is bilingual! There are good reasons for thinking that might be the case based on what kinds of linguistic and cultural work are required to be bilingual...
I digress!)

Back with my magic wand scenario: I would hire the kind of docile hack who sits in the DfE doing what the minister bids them to devise bilingualism 'benchmarks' and tests. These might be, say, a standardised test or role play where a health practitioner explains in English to a bilingual child that the child's parent who is monolingual in a non-English language that the parent must come into hospital...or some such. I'm sure you could think of a better one or variants of it to suit the variety of linguistic experiences that children have.

So, the only children who would be able to do some kind of bilingual test in England (Wales is a different case altogether!) would be children who are themselves migrants or who have at least one migrant parent or carer, or some specific religious or cultural reason for having additional languages eg Jewish children being able to read ancient Hebrew, Muslim children (migrant or not) being able to read Arabic.

(Consider the situation in the US where children with Spanish is part of their linguistic range. Does the education system reward that kind of dual language facility right from the start? Or only when it becomes a matter of high order literacy at, say 15 or 16?)

So back with my scenario in England. The effect of rewarding bilingualism in English schools - let's say, adding it into SATs scores or some such, would completely alter what 'success' is in the schools. The profile of 'abilities' or 'achievement' or 'attainment' or  'success' (or whatever it's called this week) would be completely different.

Now, this scenario ain't gonna happen. Bilingualism isn't rated in this country even though it is becoming increasingly clear that there are many jobs going to people who are and even though there are strong pyscho-linguistic reasons for thinking that bilingualism - when given space to thrive - has repercussions on learning and thinking way beyond whether you can ask the way to the shops in two languages or not!

But - and this is crucial - the reasons why it isn't going to happen are cultural and political. They are nothing to do with education and nothing to do with 'attainment'. The supposed objectivity of 'assessment' gets locked into the political decisions as to what is assessed and why.

I give bilingualism as one obvious example. But there are many others.

Our leaders make decisions all the time about the relative ratings they give to skills, achievements and capabilities across the range of human activity. Sometimes this relates directly to what is generally accepted as useful and sometimes not. In the last couple of weeks it's been revealed that there is only one trained scientist in the House of Commons! This tells  you all you need to know about how the political elite is trained. Small wonder then, when the little cabals concerned with education get in their huddles or shouting at party conferences they end up talking about Shakespeare and not homeostasis. Small wonder that the word 'engineering' hardly ever passes their lips...and yet the world that we live in, the world that has been made before we got here, and is being made even as I write this, relies on science and engineering.

Primary schools remain biased towards such things as spelling, good words, times tables - a massive amount of surface correctness. Consider the possibility that science and engineering are based on problem solving, actual and theoretical. Consider the idea that a classroom could be a place where at least half the time and space was taken up with problem-solving, actual and theoretical. The school-building and site are themselves  a site of many problems, some already solved, some not. Many of the parents are engaged in problem-solving every day. Imagine the possibility that a whole chunk of schooling was about finding and trying to solve problems (I don't mean psychological!) that arise in the everyday life of the school and the every day life of the parents. Imagine if part of the job of a school would be to draw the parents (and friends/relations etc) into the school to share these.

Suddenly, what constitutes 'knowledge' or 'skills' would alter. It would have to be clear that a 'problem' could be anything from very small everyday tasks like moving furniture, mending leaks, poaching an egg to bigger ones that crop up at work in, say, the building trade, or the health service, or the post office or to even bigger ones like building bridges, digging tunnels, processing waste and so on.

People who face these problems and frequently have to solve them wouldn't necessarily be the same people whose lives, work and cultural background are favoured by the kind of curriculum we  have in schools, the kind of curriculum which is assessed by tests which mostly just confirm that kind of cultural background.

Suddenly, people who cook, clean, build, dig, measure and the like would be those with something to say to the curriculum.

I say this as an indication of how the exam system just works with the cultural and political bias of the curriculum which is a matter mostly of socially matching the make-up of the elite in society.

3,) Further thoughts in next blogs on exam stuff:
a)the bias of favouring individualism over social thinking and doing
b)the jargon of reliability and validity needs unpicking
c) exams as ways of reducing people for the convenience of bureaucrats
d) the knowledge of how to do the exams is as important if not more important than the knowledge or skills being tested
e) exams as the police system of schooling
f) the teaching to test blame game
g) the exam-league table hoax g)the norm-reference/criterion reference lie
h) 'levering up standards' - what are better ways than exams for doing this?
i) what would constitute a useful form of testing?