Athabasca University | AU Student/Staff Login | Invited Guest Login
Are final examinations on the way out at Harvard? | Harvard Magazine Jul-Aug 2010
Excellent.
One down, a few million to go, it's time to end this barbaric, anti-learner practice. I exaggerate. Exams per se are not wrong and in some cases may be a very sensible way to accredit people - driving tests, for instance. I can even imagine ways that they might be constructively aligned with the learning intentions (sporting contests, for instance), though very rarely in an academic setting. But, like the lecture, it's an idea of very limited applicability that has become a plague in our systems and it is high time that we had to provide a real, strong, learner focused justification for using them. And no, not the dumb old argument about knowing it is all their own work (palpable nonsense in most settings in higher education). And not the one about efficiency (whose?).
- Some meandering thoughts on 'good' and 'bad' learning
December 20, 2022 - 7:39pm
Jon Dron - Can GPT-3 write an academic paper on itself, with minimal human input?
July 2, 2022 - 12:48pm
Jon Dron - Can GPT-3 write an academic paper on itself, with minimal human input?
July 2, 2022 - 12:51pm
Jon Dron - Brunel University's Integrated Programme Assessment - a neat way to decouple learning and...
November 24, 2022 - 11:22am
Jon Dron
Jon Dron
Comments
What would be your perspective on exams as part of professional certifications? Would you consider exams appropriate or would you suggest other approaches?
Exams might be efficient but, most of the time, completely lack any authenticity. The ability to answer questions in a weird and unrealistic setting may have little correlation with the ability to perform your professional duties, unless they are usually performed under exam conditions. Most professionals perform their functions in a rich community, supported by the tools and information sources of their trade, in a specific context. Exams of the sort we usually provide in academia do not allow that and, if they do, are not the things I am complaining about. Portfolios or similar tools would usually be far better, especially mapped to evidence. Fairer, richer and more revealing. Viva voces would work in some circumstances. So would descriptive essays in others. So would evaluations by accredited peers of work done in a professional context. There is no hard and fast rule about what works in a given situation but, whatever system you use, it should show that the person being accredited can perform the job they are being accredited to do. Exams seldom show that. If you want to show that you are a computer professional, show that you can do that under the conditions computer professionals work under. If you claim to be a surgeon, prove it by performing surgery (supervised by real surgeons of course :-). And prove that you can do it again under different conditions.
I am fully in favour of summative assessment in most areas, but there are many ways to do that which work better than unseen written exams of the sort academia has been infested with for the past couple of hundred years. In the Western world, they were invented because it was really hard to prove geometric knowledge in a viva (China had them over a thousand years before that). They persisted and gained traction because they were cheap and easy for academics to administer and showed *some* correlation with actual skill. But they are a million miles from being a universal solution for all situations and the correlation with professional ability is not that strong. Many people fail them not due to lack of skill in the profession but due to lack of skill in taking exams, and vice versa. And they are not even that cheap in many situations. At Athabasca, for instance, they are incredibly expensive, stressful and time consuming, and attack the very people we most wish to support - typically, working people who, for one reason or another, did not follow the standard academic path when others did.
I don't want to get rid of them any more than I want to get rid of lectures or learning management systems or drill and practice approaches to teaching, though all have major and incontrovertible weaknesses. I just want us to choose when we use them wisely and reflectively. That will make them a lot rarer than they are.