What would happen if we abolished traditional exams?Posted: August 22, 2016
Finland is fashionable among educationalists because it is seen as proof that you can have a great education system without neoliberal high-stakes testing and accountability. Finland gained this status on the back of its PISA results and even though these results have significantly declined since the start of the century, Finland remains as popular as ever. Pundits laud its progressiveness.
So if anywhere has avoided exams, it’s Finland, right? Wrong. It’s true that Finland doesn’t have the number of standardised tests that characterise Australia, England or the U.S. but it’s worth noting that students in the academic stream at the end of High School sit a set of six-hour exams. Tim Oates explains:
“As for “no tests” – all Finns understand the importance of doing well in the Finnish Abitur –the university-oriented ‘finishing’ examination taken by 19-year-olds. The academic pathway (upper secondary to university) is considered of higher esteem than the vocational pathway at 16, into which over 40% of pupils go. On scrutiny, the Abitur examinations are just like English A Levels, although pupils may study seven or eight subjects, they only take four subjects – one of these in native language. The other subjects are just like A Levels – six hour, nationally-moderated tests in individual subjects. And these exams have been in place, and relatively unchanged in form, since the end of the 19th century.”
So if Finland can’t have an education system without traditional exams, where can? Is it even possible? What would it look like?
I had an interesting discussion on Twitter about this. Chris Woldhuis was at a conference where Eric Mazur was speaking. It appears as if Mazur made a claim, consistent with previous comments on the topic, that, “Cheating in an assessment is not a problem with the students, it is a problem with the assessment.” I disputed this and Chris and I had a discussion. Chris’s view was that if it is possible to cheat then the form of assessment is not right – we’re not asking students to solve real problems. My view was that any form of assessment is vulnerable to cheating and so I asked Chris for an example of the sort of problem he had in mind. He suggested, “Students run a cafe, make a profit.”
If we moved away from exams in favour of assessing projects of this kind then this would cause a number of practical problems. How can every student do something like run a cafe? Would the cafe be run as a team? If so, how can we be sure that everyone is playing an equal role? Perhaps we don’t really mind whether they are playing equal roles but, if so, how can we form any sort of assessment of the learning of individuals?
You might respond by asking me why I want to judge the students. And that’s a fair call. As a teacher I’m mainly interested in formative assessment – where is this student right now and what is needed for improvement? This doesn’t require us ranking the students relative to each other.
But I’m not the reason that we have formal exams. The Finnish Abitur does not exist to provide formative evidence for teachers. It’s not the teachers that want a summary of individual performance, it’s universities and employers.
If we abandoned examinations, universities would be left with problems such as how to select students for prestigious and challenging courses. In Australia, we have already moved away from Year 12 exams as the basis of entry into medicine. Presumably, these tests don’t give enough information to medical schools and so students have to study for something called the ‘Undergraduate Medicine and Health Sciences Admission Test’ (UMAT) in addition to their other exams.
If we moved to the use of projects to assess high school students then one consequence might be a mushrooming of such parallel tests. How much time would students then invest in running their cafes?
Suppose we took a Stalinist stance and forbade universities from asking students to sit such tests. Instead, we required universities to use the authentic projects completed in school. Would we then have a fair system based upon real and authentic assessment?
If something important like university entrance depends upon your ability to run a cafe and make a profit then the possibilities for gaming this system are endless. Wealthy parents could, for instance, buy a lot of coffee. The authentic nature of the project means that it is not sealed-off from the outside world and so our rich kids could perhaps pay for advice from successful cafe owners, accountants, advertising creatives and so on.
We can extend this to arguments made by Mazur himself:
“Let’s mimic more real life in our assessment practice. If you look at students taking an assessment, they’re set out in rows, separated by a gap and isolated from any source of information. Once you get your diploma, you never face that situation again. At work you can call whoever you want, you can google anything, yet this is how we assess students. So we should let people talk to each other, we should let people work together. We consider it cheating in that setting but not cheating when they are doing their jobs. So why create this artificial environment?”
Why not create an artificial environment? What’s so great about real-life?
It is true that, at work, you can call whoever you want, but if you call someone for advice then you are going to have to pay them. Googling the way to solve a mathematics problem might be of some benefit but it would be far more advantageous to have an expert instructor on the other end of Skype who you can work through the problem for you. So if you can afford it then that’s what you will do. Which is certainly an authentic replication of the inequalities of the real world, if that’s what you want.
Traditional exams are, in fact, far more equitable. Students sit these tests naked of their privilege and have to solve problems on their own rather than have this done for them. No, I’m not arguing that this takes away the advantages of the rich completely. They still have access to learning resources that poorer students don’t have. But I can’t think of a better way of minimising these advantages than with a traditional test.
When you also take account of the fact that exams are subject to less bias than the kind of teacher assessment you would need for a project, the equity case is overwhelming.