Monday, 17 September 2012

Clegg O-level to 'raise standards' - NOT.

Yes, there will be a new exam system for 16 year olds. It will be the very old exam system. More or less like the one I sat in 1962: a two-hour pen and paper test. Nick Clegg claims here:

that this will 'raise standards'.

Where is the evidence that two-hour pen and paper tests 'raise standards'?
(Hint: nowhere)
What 'standards' are being raised?
(Hint: the standards required to pass increasingly less useful, less worthwhile exams)

All that is happening is that they are trying to increase the 'reliability' of the test ie making it easier to mark, and easier to come up with consistent grades. This will decrease the 'validity' or worth of the exam.


1. It will call upon teachers, parents and students to spend yet more time developing 'exam-technique' ie what I call Knowledge 2 (K2). These are the pointless, useless games of eg
a) second-guessing what questions will be asked in the exam
b) teaching the mark-distribution across the paper so that 'weaker' pupils concentrate on the 'mugging-up' sections in order to earn certain marks. These nearly always require some kind of rote-learning, which can be done without any understanding.
c) exam-taking techniques to do with 'timing  your answers' eg moving on from things  you do know in order to finish the paper
d) rubric-drilling - learning the language of the instructions of exams and learning how to gamble with the instructions in terms of getting the marks. Some people who are quite capable of answering questions find that figuring out the rubric under exam conditions is hard. Some years, the rubric changes without anyone being forewarned. This penalises those who learn rubric off by heart without understanding it. This has nothing to do with the knowledge of the subject (K1).

2. The Knowledge being tested (K1) will become much more based on what is purely testable. That's to say, the subject in question will have to be broken down into reliably and consistently markable units. This will not 'raise standards'. It will if anything lower them as it makes more and more of the tasks being asked become mechanical, and easily learned through rote-learning without understanding.

A good deal of the testing in the exams will therefore involve 'retrieval', 'inference' and 'regurgitation'. There will be much less on 'interpretation' because interpretation is open to interpretation (!)  ie less 'reliably' or 'consistently' marked but of course much more 'valid' or worthwhile in terms of what we want young people to achieve in school and for life afterwards; much more useful knowledge and skills to take into real life.

[glossary: 'retrieval' is what they test when they say, 'Bobby had a blue hat.'  and ask: What colour was his hat?
'inference' is what they test when they then add, 'It was raining.' and ask 'Why was Bobby wearing a hat?'

Note: If you answer: 'He was wearing a blue hat because he supports Chelsea' that is a 'wrong answer' even though you were using 'retrieval' from  your body of knowledge about hats and the colour blue; and 'inference' from your knowledge of who wears blue hats and why.

3. In short, the new O-level style exam will narrow down even further what was a narrow, national test. It will do this by narrowing K2 and increasing the amount of time spent on K1.

4. It will also be a norm-referenced exam. In other words, politicians and their minions at Ofqual (or whatever takes its place) will jiggle the marks the markers give in order to make the overall results' graph look the way they want it to. As John Tomsett points out here:

norm-referencing means that schools end up competing with each other for a place on the graph. This a) demotivates teachers and students who learn that they are not only or simply being tested for what students know, but that there may well be a politically motivated moving of the goal posts which may well move you from the grade that will let you do more education to one that won't; b) will ensure that one school won't help another because ultimately 'your lot' might push 'my lot' down the graph.

This latter point is again pure politics. It says that standards are raised through schools competing for customers and then getting the customers in  your school to compete with the next school.

This replicates the market - winners and losers. But we're not talking about selling sugar and cars here. We're talking about the lives of a generation of students going through a school, the livelihoods and commitment of hundreds of teachers.

All that will happen is that as a school with a large percentage of borderline students who are norm-referenced down the graph (as has happened this year) will bring down upon themselves punitive measures, increased drilling in exam-passing techniques and possible closure, disruption, more lives wrecked.

However, this suits politicians (not pupils and teachers) because they can keep up their narrative of decline crap about bad teachers, bad students which they will (with some other crack-brained reform) change. Yet, this new exam will be the cause of 'failure'.

5. This exam will lower standards. This exam will lower standards. This exam will lower standards.

In the meantime, instead of bringing in even more mind-numbing testing (and the teaching-to-the-test) at 16, they should be raising the age of fulltime education and training to 18 and abolishing most of the 16 plus nationally moderated testing, leaving it to locally devised tests, portfolios and interviews.

6 PS - please note that the above does not challenge the whole notion that the only way to assess worth is to give it a mark! I do challenge that. In classrooms, labs and workshops, it is quite possible to assess worth by other means than marks and grades. We are so inducted into thinking that the 'quality' of a piece of work has to have a number that we forget that in life, for much of the time we assess things on the basis of eg their usefulness in a particular time and place (not for all time); their likeability or how we are 'affected' by it and/or we are capable of taking on board the idea that things are 'patchy' ie some of it better than others.