Open to question

A couple of weeks back the head of OCR ventured his opinion that eventually students would be allowed to use Google in the exam room for their exams. I commented here on why he was probably misguided, but I did say i would comment later on why it wasn’t a completely stupid idea.

So here goes.

My view of the current exam system is that the exams themselves are limited. The cost of marking exams tends to lead to quite restrictive questions, designed to elicit a limited range of responses. This works to the advantage of the exam boards in that they can anticipate most likely responses in a mark scheme and then use have them marked quickly and cheaply.

Now, there is an argument that this is ok for some qualifications as the main focus should be to establish if the student has been able to recall knowledge that they should have at that age. If you believe that a qualification should show if a student has been able to take that knowledge, process it and then use the processed knowledge in some way then you could argue that both GCSE and A-Level can be quite limiting.

This is what I would like to see.

All qualifications should consist of three papers.

  1. A multiple choice question paper.
  2. A short answer question paper.
  3. A long answer question paper.

All these paper types will be familiar to anyone involved with the current exam system. But I would like to see a couple of subtle (not really) differences.

The first difference is that both the multiple choice paper AND the short answer question paper should be machine marked. The technology exists to do this and i see no reason why it should not be used. It will lead to more consistent marking and it will also be cheaper and faster. I see no real downsides to this approach.

The second difference, which is the big, controversial one, is that the long answer question paper should be an open book and open computer paper. This paper would necessarily be marked by humans. Also it would be marked against a very limited mark scheme which would require fairly expert (and consequently expensive) markers. This paper would be all about how the student can take their knowledge they have about the subject (which will have been tested in the MCQ and short answer papers) and combine it with other information sources.

As an aside, practical subjects also fit easily into this structure with the practical part of the exam essentially being the long answer question paper.

There would be another big difference to the qualification structure.  The marks available to the student for the third paper would be scaled according to the scores they had achieved on the first and second. It would be impossible to obtain full marks on the third paper if full marks were not obtained on at least the first. This would prevent a student “just looking it up” and using the open book part of the exam to boost their grade.  They would have to have proved they knew the stuff in exam conditions before they benefited from it in the open book paper.

Because the third paper requires a more subjective approach to marking it is less dependent on the actual questions asked. This leads to the suggestion that not all students need to be answering the exact same questions which would enable a wider range of sitting options allowing the necessary access to limited technology.

This, like our existing system, is clearly not a perfect answer. And I am sure that many will claim that their particular subject cannot be tested in this way.

In spite of that I do think it has some merit.


One thought on “Open to question

  1. I can’t see much wrong with this idea for discursive or content heavy subjects. Practical subjects, including science, are more problematic but I’m keen to see an improvement in the use of simulations in assessments.

Comments are closed.