We might as well start with exams
In case anyone missed it, one of countless examples of mass cheating in exams is being reported quite widely, such as at http://www.ctvnews.ca/world/hundreds-expelled-in-india-for-cheating-on-pressure-packed-exams-1.2289032.
The videos are stunning (Chrome and Firefox users – look for the little shield or similar icon somewhere in or near you browser’s address field to unblock the video. IE users will probably have a bar appearing in the browser asking if you want to trust the site – you do. Opera, Konqueror and Safari users should be able to see the video right away), e.g.:
As my regular readers will know, my opinions of traditional sit-down, invigilated, written exams could not be much lower. Sitting in a high-stress environment, unable to communicate with anyone else, unable to refer to books or the Internet, with enormous pressure to perform in a fixed period to do someone else’s bidding, in an atmosphere of intense powerlessness, typically using a technology you rarely encounter anywhere else (pencil and paper), knowing your whole future depends on what you do in the next 3 hours, is a relatively unusual situation to find yourself in outside an exam hall. It is fair enough for some skills – journalism, for example, very occasionally leaves you in similar conditions. But, if it actually is an authentic skill needed for a particular field, then it should be explicitly taught and, if we are serious about it, it should probably be examined under truly authentic conditions (e.g. for a journalist, in a hotel room, cafe, press room, or trench). This is seldom done. It is not surprising, therefore, that exams are an extremely poor indicator of competence and an even worse indicator of teaching effectiveness. By and large, they assess things that we do not teach.
If that were all, I might not be so upset with the idea – it would just be weird and ineffective. However, exams are not just inefficient in a system designed to teach, they are positively antagonistic to learning. This is an incredibly wasteful tragedy of the highest order. Among the most notable of the many ways that they oppose teaching are that:
- they shift the locus of control from the learner to the examiner
- they shift the focus of attention from the activity to the accreditation
- they typically punish cooperation and collaboration
- they typically focus on content rather than performance
- they typically reward conformity and punish creativity
- they make punishments or rewards the reasons for performing, rather than the love of the subject
- they are unfair – they reward exam skills more than subject skills.
In short, the vast majority of unseen written exams are deeply demotivating (naysayers, see footnote), distract attention away from learning, and fail to discriminate effectively or fairly. They make the whole process of learning inefficient, not just in the wasted time and energy involved surrounding the examination itself, but in (at the very least) doubling the teaching effort needed just to overcome their ill effects. Moreover, especially in the sciences and technologies, they have a strong tendency to reinforce and encourage ridiculous content-oriented ways of teaching that map some abstract notion of what a subject is concerned with to exercises that relate to that abstract model, rather than to applied practices, problem solving and creative synthesis – i.e. the things that really matter. The shortest path for an exam-oriented course is usually bad teaching and it takes real creativity and a strong act of will to do otherwise. Professional bodies are at least partly culpable for such atrocities.
There is one and only one justification for 99% of unseen written exams that makes any sense at all, which is that it allows us to relatively easily and with some degree of assurance (if very expensively, especially given the harmful effects on learning) determine that the learner receiving accreditation is the one that has learned. It’s not the only way, but it is one of them. That sounds reasonable enough. However, as examples like this show in very sharp relief, exams are not particularly good at that either. If you create a technology that has a single purpose of preventing cheating, then cheats (bearing in mind that the only thing we have deliberately and single-mindedly taught them from start to finish is that the single purpose of everything they do is to pass an exam) will simply find better ways to cheat – and they do so, in spades. There is a whole industry dedicated to helping people to cheat in exams, and it evolves at least as fast as the technologies that we use to prevent it. At least twenty percent of students in North America admit to having at some point in the last year cheated in exams. Some studies show much higher rates overall – 58% of high school students in Canada, for example. It is hard to think of a more damning indictment of a broken system than this. The problem is likely even worse in other regions of the world. For instance, Davis et al (2009) reckon a whopping 83% of Chinese and 70% of Russian schoolkids cheat on exams. Let me repeat that: only 17% of Chinese people claim never to have cheated in an exam. See a previous post of mine for some intriguing examples of how that happens. When something that most people believe to be wrong is so deeply endemic, it is time to rethink the whole thing. No amount of patching over and tweaking at the edges is going to fix this.
But it’s not just exams
This is part of a much broader problem, and it is a really simple and obvious one: if you teach people that accreditation rather than learning is the purpose of education, especially if such accreditation makes a massive difference to what kind and quality of life they might have as a result of having or not having it, then it is perfectly reasonable that they should find better ways of achieving accreditation, rather than better ways of learning. Even most of our ‘best’ students, the ones that put in some of the hardest work, tend to be focused on the grades first and foremost, because that is our implicit and/or explicit subtext. To my shame, I’m as guilty as anyone of having used grades to coerce: I have been known to annoy my students with a little song that includes the lines ‘If a good mark is what you seek, blog, blog, blog, every week’. Even if we assume that student will not cheat (and, on the whole, mature students like those that predominate at Athabasca U do not cheat, putting the lie to the nonsense some have tried to promote about distance education leading to more cheating) it challenges teachers to come up with ways of constructively aligning assessment and learning, so that assessment actually contributes to rather than detracts from learning. With skill and ingenuity, it can be done, but it is hard work and an uphill struggle. We really shouldn’t have to be doing that in the first place because learning is something that all humans do naturally and extremely willingly when not pressured to do so. We don’t need to be forced to do what we love to do. We love the challenge, the social value, the control it brings. In fact, forcing us to do things that we love always takes away some or all of the love we feel for them. That’s really sad. Educational systems make the rods that beat themselves.
Moving forwards a little
We can start with the simple things first. I think that there are ways to make exams much less harmful. My friend and colleague Richard Huntrods, for example, simply asks students to reflect about what they have done on his (open, flexible and learner-centred) course. The students know exactly what they will be asked to do in advance, so there is no fear of the unknown, and there is no need for frantic revising because, if they have done the work, they can be quite assured of knowing everything they need to know already. It is a bit odd not to be able to talk with others or refer to notes or the Web, but that’s about all that is inauthentic. This is a low-stress approach that demands nothing more than coming to an exam centre and writing about what they have done, which is an activity that actually contributes substantially to effective learning rather than detracting from it. It is constructively aligned in a quite exemplary way and would be part of any effective learning process anyway, albeit not at an exam centre. It is still expensive, it still creates a bit more stress for students who have learned to fear exams, but it makes sense if we feel we don’t know our students well enough or we do not trust them enough to credit them for the work they have done. Of course, it demands a problem- or enquiry-based, student-centred pedagogy in the first place. This would not be effective for a textbook wraparound or other content-centric course. But then, we should not be writing those anyway as little is more certain to discourage a love of learning, a love of the subject, or a satisfying learning experience.
There are plenty of exam-like things that can make sense, in the right kind of context, when approached with care: laboratory exercises, driving tests, and other experiences that closely resemble those of the practice being examined, for example, are quite sensible approaches to accreditation that are aligned with and can even be supportive of the learning process. There are also ways of doing exams that can markedly reduce the problems associated with them, such as allowing conversation and the use of the Internet, open-book papers that allow students to come and go as needed, questions that challenge students to creatively solve problems, exams that use questions created by the students themselves, oral exams that allow examiners to have a useful learning dialogue with examinees, and so on. There are different shades of grey and not all are as awful as the worst, by any means. There are other ways that tend to work better – for instance, badges, portfolios, and many other approaches that allow us to demonstrate competence rather than compliance, that rely on us coming to know our students, and that allow multiple approaches and different skills to be celebrated – but not all exam-like things are as bad as the worst of them.
And, of course, if we avoid exams altogether then we can do much more useful things, like involving students in creating the assignments; giving feedback instead of grades for work done; making the work relevant to student needs, allowing multiple paths, different evidence; giving badges for achievement, not to goad it, etc, etc. There’s a book or two in what we can do to limit the problems though, ultimately, this can only take us so far because, looming at the end of every learning path at an institution, is the accreditation. And therein lies the rub.
Moving forwards a lot
The central problem that we have to solve is not so much the exam itself as the unbreakable linkage of teaching and accreditation. Exams are just a symptom of a flawed system taken to its obvious and most absurd conclusion. But all forms of accreditation that become the purpose of learning are carts driving horses. I recognize and celebrate the value of authentic and meaningful accreditation, but there is no reason whatsoever that learning and accreditation should be two parts of the same system, let alone of the same process. It it were entirely clear that the purpose of taking a course (or any other learning activity – courses are another demon we need to think carefully about) were to learn, rather than to succeed in a test, then education would work a great deal better. We would actually be able to do things that support learning, rather than that support credit scores; to give feedback that leads to improvement, rather than as a form of punishment or reward; to allow students to expand and explore pathways that diverge rather than converge; to get away from our needs and to concentrate on those of our students; to support people’s growth rather than to stunt it by setting false goals; to valorize creativity and ingenuity; to allow people to gain the skills they actually need rather than those we choose to teach; to empower them, rather than to become petty despots ourselves. And, in an entirely separate process of assessment that teachers may have little or anything to do with at all, we could enable multiple ways to demonstrate learning that are entirely dissociated from the process. Students might use evidence from learning activities we help them with as something to prove their competence, but our teaching would not be focused on that proof. It’s a crucial distinction that makes all the difference in the world. This is not a revolutionary idea about credentialling – it’s exactly what many of the more successful and enlightened companies already do when hiring or promoting people: they look at the whole picture presented, take evidence from multiple sources, look at the things that matter in the context of application, and treat each individual as a human being with unique strengths, skills and weaknesses, given the evidence available. Credentials from institutions may be part of that right now, but there is no reason for that idea to persist and plenty of alternative ways of showing skills and knowledge that are becoming increasingly popular and significant, from social network recommendations to open badges to portfolios. In fact, we even have pockets of such processes well entrenched within universities. Traditional British PhDs, for example, while they are examined through the thesis and an oral exam (a challenging but flexible process), are examined on evidence that is completely unique to the individual student. Students may target the final assessment a bit, but the teaching itself is not much focused on that. Instead, it is on helping them to do what they want to do. And, of course, there are no grades involved at all – only feedback.
It’s going to be a long slow struggle to change the whole of the educational system across most of the world, especially as there’s a good portion of the world that would be delighted to have these kinds of problems in the first place. We need education before we can have cheating. But we do need to change this, and exams are a good place to start. It changed once before, with far less research to support the change, and far weaker technologies and communication to enable it. And it changed recently. In the grand scheme of things, the first ever university exam of the kind we now recognize as almost universal was the blink of an eye ago. The first ever written exam of the kind we use now (not counting a separate branch for the Chinese Civil Service that began a millenium before) was at the end of the 18th Century (the Cambridge Tripos) and it was only near the end of the 19th Century that written exams began to gain a serious foothold. This was within the lifetime of my grandparents. This is not a tradition steeped in history – it’s an invention that appeared long after the steam engine and only became significant as the internal combustion engine was born. I just hope institutions like ours are not heading back down the tunnel or standing still, because those heading into the light are going to succeed while those that stay in the shadows will at best become the laughing stock of the world.
On the subject of which, do watch the video. It is kind-of funny in a way, but the humour is very dark and deeply tragic. The absurdity makes me want to laugh but the reality of how this crazy system is wrecking people’s lives makes me want to cry. On balance, I am much more saddened and angered by it than amused. These are not bad people: this is a bad system.
I know some people will want to respond that the threat or reward of assessment is somehow motivating. If you are one of those, this postscript is for you.
I understand what you are saying. That is what many of us were taught to believe and it is one way we justify persisting despite the evidence that it doesn’t work very well. I agree that it is motivating, after a fashion, very much like paying someone to do something you want them to do, or hitting them if they don’t. Very much indeed. You can create an association between a reward/punishment and some other activity that you want your subject to perform and, as long as that association persists, you might actually make them do it. Personally speaking, I find that quite offensive, not to mention only mildly effective at achieving its own limited ends, but each to their own. But notice how you have replaced the interest in the activity with an interest in the reward and/or the desire to avoid punishment. Countless research studies from several fields have pretty conclusively shown that both reward and punishment are strongly antagonistic to intrinsic motivation and, in many cases, actually destroy it altogether. So, you can make someone do something by destroying their love of doing it – good job. But that doesn’t make a lot of sense to me, especially as what they have learned is presumably meant to be of ongoing value and interest, to help them in their lives. It is my belief that, if you want to teach effectively, you should never make people learn anything – you should support them in doing so if that is what they want to do. It is good to encourage and enthuse them so that they want to do it and can see the value – that’s a useful teacher role – but it’s a whole different ballgame altogether to coerce them. Alas, it is very hard to avoid it altogether until we change education, and that’s one good reason (I hope you agree) we need to do that.