[the blog with the woofing dog] How Not To Design An Examinations System, Part One: Grade Inflation

[Blogger]

Yesterday, the Government announced their new examinations system to replace the GCSEs.  They released this (rather thin) consultation document.  It's not very good.  In fact I'd say it's woeful.  They do not back their reasons for change with compelling evidence, and the proposals they are making are sketchy at best.

But who am I to criticise?

Well, I've been teaching Maths in this country since 2005.  I don't think that GCSEs are not "challenging", or that students' hard work is "in vain", or that their grades are "worthless".  They're the best measure we have at the moment for determining students' abilities.  Students deserve what they get.  (Mostly.)

However, the system has numerous faults, many of which I'll outline below.  I think we can do much much better.  So I'm not going to blindly defend it in the face of change.  In fact, I'd love to see GCSEs replaced.  Just not by the English Baccalaureate Certificate.

Grade Inflation

"This consultation sets out the Government’s plans to restore rigour and confidence to our examination system at age 16, which has been undermined by years of continued grade inflation."

[Paragraph 1.1]

In the 1988 Olympics (the year the first students sat GCSEs), 2.9% of the men's 100 metre runners ran (legally) faster than 10 seconds.  In the London 2012 Olympics, that figure had increased to a staggering 11.0% - nearly four times the proportion of athletes meeting the same standard!

Does this mean that the men's 100 metre sprint is an easier race than it used to be?

No, of course it doesn't.  One hundred metres is still the same length it used to be, and clocks still run at the same speed.  The improved standards are down to increased professionalism in the sport, the greater use of technology in the track, the shoes, the monitoring and regulation of the athletes' fitness, and the drive in all of them to do a little bit better than the standard set the previous year.

But the improvement in GCSE results year on year is rarely given the same logical analysis, and in this report it's no different.  The only "evidence" offered is in this one statistic:

"The public recognise this to be true. 60% of those surveyed in a recent YouGov poll believe that GCSEs have got easier, while only 6% think that they have got harder."

[Paragraph 3.4]

Oh, well, if the public think exams have got easier, then that must be right.  Although any member of the public over the age of 40 is likely to have ever taken a GCSE exam, and very few people indeed will have taken a GCSE around 20 years ago, and then taken a GCSE in the same subject again in the last few years.

In fact, the public's exposure to this is rather likely to have been influenced by the papers reporting every single August that exams are getting easier.  This was illustrated brilliantly by this article in 2005 by the wonderful Ted Wragg, a lecture by whom I was lucky enough to attend whilst teacher training, shortly after he wrote the article, and shortly before he died.

Those ready to point to the annual comparison the papers give between exam questions would do well to note that they usually compare the most difficult O Level questions with easy GCSE questions.  They rarely compare old GCSE questions with new GCSE questions.  But even if they did, they'd not necessarily be comparing questions at the same grade.

For the record, I do believe there has been some grade inflation, but it's impossible to tell how much.  The improvement in GCSE results is large, but to blame it all on easier exams is to close your eyes and ears to logic and the world around you.

Any exam system is likely to see some improvement in results over time, simply because schools will get used to the system and get better at teaching it to the pupils.  With every year there is an extra practice exam paper with which students can hone their skills.

Since the introduction of school league tables, there is an ever present pressure to do better than the year before, to overtake competing local schools, or face the consequences.  In the few years I've been teaching, I've definitely seen an increase in after-school and holiday revision sessions, one-to-one and small-group interventions, the use of consultants brought in from outside, students being pulled off timetable to focus on getting their coursework finished.

Nothing is left to chance.  This is driving an improvement in examination results at least as much as any creeping grade boundaries.

Teaching has changed.  Since those first GCSE students in 1988, the Internet has been invented.  THE INTERNET.  It's changed our world.  It's changed the teaching profession.  Resources can be shared like never before, they can be interactive and engaging like never before.  Homework can be tracked and tailored to individual students so much more easily.

And this is assumed to have no effect on exam results?

It's curious that a Government supposedly so committed to build an examination system fit for the 21st century, should so lazily lay the changing results at the door of grade inflation.

Two flags are raised in my mind, however, by the increasing exam results.

(1) The more students achieve the highest grades, the harder it is to differentiate the most able and, more importantly, the harder it is to motivate those students to go on and improve even further.  Where do you go from the top grade?

I find it interesting then that the proposal is to make Grade 1 the top grade in the new EBacc/EBC.  Firstly, it's potentially confusing, as it means that students will go from the National Curriculum Levels, where 1 is the bottom grade, into a system where 1 is suddenly the top grade.

Only mildly confusing, you might say, but also it leaves no room to extend the grading upwards to allow the students to progress as far as they can.  Couldn't Grade 1 be the starting point, and then you can have, in theory, a limitless scale up which students can climb?

It might seem like nitpicking, but there are an awful lot of these little details where the Government have appeared to have not thought things through as much as they might have done, and these little details really stack up.

(2) Back to the GCSEs, the improved results raise more questions in my mind about "teaching to the test" than they do about grade inflation.  If schools can get that much better at teaching students to pass the test, it implies that exams might be something you can learn to do and get a high score on, rather than something that can prepare you for real life.

Far more thought needs to go into how you set a test that gives those with really good thinking skills, rather than practising-and-repeating skills, something to get their teeth sunk into. The sort of questions that you can't simply rote-learn the method to.

So I find it curious that this is not mentioned when talking about teaching to the test.  The Government's hope seems to be that teachers won't worry so much about getting the students to pass the test if they only have one chance to take it at the end of the course:

"Together with Ofqual, we have already taken steps to discourage ‘teaching to the test’ and to tackle the culture of re-sits by ensuring that students are tested at the end of the course. This will give students the opportunity to gain a deep and broad understanding across the whole subject."

[Paragraph 5.11]

Trust me.  They'll worry about getting the students to pass the test.   Even more so.

--
Posted By Blogger to the blog with the woofing dog on 9/18/2012 03:33:00 PM