Exam Standards: Good News and a suggestion

I have written here previously about my anger at the decline in exam standards, particularly at GCSE level and particularly in Science. And most particularly in Physics. My concern is that children from state schools which ‘teach to the exam’ will be denied the possibility of careers in Science. The Qualifications and Curriculum Development Authority oversaw a perncious system which systematically drove down exam standards year after year and penalised anyone – teachers, publishers or schools – who tried to object.

So it is great to report that the QCDA is moribund, and I am delighted to find that Ofqual has finally found the teeth to object to this decline. After rejecting the next syllabus revision as too low in standards today I read that they have sent back the exam boards revised syllabi as still not challenging enough. Hurray! Ideally, exam standards could rise year upon year as OfQual drove standards higher and higher. I am not hopeful, because the same political pressure which drove the previous government to devalue educational qualifications still exist. However I do have a suggestion.

My Suggestion

When I took my O level exams, the grade awarded was determined not by any absolute standard, but by where one came in the ranked exam order. Thus an A indicated that one had achieved a mark in the top 10% (say, I don’t know the exact fraction used) of the exam cohort. Nowadays an A indicates a mark exceeding 70% (I think) and can be achieved by any fraction of the exam cohort. It thus no longer serves to discriminate among exam candidates. And discrimination amongst candidates is – like it or not – a key purpose of exams.

If the old procedure was reintroduced it would automatically condemn a certain fraction of the cohort to ‘failure’, no matter what they achieved. However there is no need to use the system universally. A mixed system could be used: An A grade could still indicate the top 10%, B would indicate the next 10%, and C would indicate a norm-based pass above some nominal pass mark. There would be no limit to the number of people who could get C. This allows for improved teaching to result in improved pass rates without a guaranteed fraction ‘of failures’ while allowing employers or universities to look for academic high achievers.

• Is there something wrong or unfair with this?

GCSE Standards: I begin to learn

Exam Hall

Having shouted from the rooftops about GCSE standards I thought I should learn more about them. So this evening I went to the Edexcel Web Site, clicked around and downloaded a couple of papers. Specifically I downloaded:

• Edexcel GCSE Physics P3 Ref 5049/01 from 11 June 2008
• London Examinations GCE ‘O’ Level Ref 7540/01 from 25 January 2007

Educationalists could probably describe the differences between the papers more succinctly but I will have a go. I don’t have the right to reproduce the questions, but  here are my impressions

• The questions each last for a nominal 6.8 minutes, but the O level has 11 questions and the GCSE has 9
• The GCSE paper has a formula sheet for even the most basic formulae e.g. frequency = 1 /period. The O level paper has no formula sheet. Calculators are allowed in both exams.
• The O level paper questions are abstract and numerical. The student has to be familiar with more theory than for the GCSE paper so as to recognise the relevant theory in an unfamiliar – and often abstract – context.
• One GCSE Question shows a diagram of a cathode ray tube and the student is asked to label the cathode (1 mark) and anode (1 mark). There are then 4 multiple choice questions:
  • The electron beam is produced by the the heated (a) cathode (b) anode (c) screen
  • The inside of the tube contains (a) air (b) a gas (c) a vacuum
  • When electrons hit the screen (a) light waves (b) microwaves (c) radio waves are produced
  • When the heater current is increased the number of electrons produced will (a) decrease (b) increase (c) stay the same
• This question has no numerical component; students can guess from a choice of just three answers,in general one of which is pretty obviously wrong; the third question is wrong because although it seeks the answer (a) the correct answer is actually (a) (b) and (c).
• There is no direct equivalent O level question. But one related question involves the student placing Radio waves, X-rays, Infra Red, A missing component, Visible light, microwaves, gamma rays in the correct of wavelength and identifying the missing component – ultra violet light I think. The speed of light in a vacuum is then given and the students asked
  • State one other property common to all electromagnetic waves ( I think they are seeking the fact that they are transverse waves, or that they can be polarised)
  • They are the asked to calculate the frequency of radio waves with a wavelength of 2 x 10^3 metres. They need to remember velocity = frequency x wavelength and do a sum involving exponents

In the  GCSE question there is no place for the student who has the numerical ability – an ability which is key to all practicing physicists – to demonstrate their insight and skill. In contrast the O level actually tests abilities and knowledge that, frankly, I use every day in my work.

This comparison has not been exhaustive, and the GCSE paper I looked at was not without merit. But its questions were vague and qualitative in comparison with the straightforward simplicity of the O level questions. The O level was a real test of physics, while the GCSE physics tested only a basic level of comprehension and a little general scientific knowledge. Based on this – admittedly limited – investigation, I feel that my concerns are justified. These exams don’t really examine the ability to do physics at all.