How to foster honesty in those writing exams online

Let’s be honest. Most parents help their children to do well with their school projects. Parents with more money, formal education, technical skills, and time available can make that school project count for much more in the school-based assessment than if the pupil did it on his/her own. We know that.

This past weekend I conducted one of those Twitter polls and asked the question: Would you as a parent help your child struggling with an online exam done from home? Of the 681 votes cast, a solid third of the respondents answered “of course”.

Not very funny when it comes to high-stakes tests and examinations such as the online assessments forced on us by the pandemic lockdown. Still stuck at home, the question of fairness and honesty in examinations-from-a-distance is cause for growing anxiety, especially among university academics.

A student getting a degree in internal medicine or clinical psychology based on dishonest exam outcomes can have catastrophic consequences for society. More than that, cheating is simply wrong and unfair on those who play by the rules.

So, can universities enforce cheat-free online examinations? The answer is no. It is, quite frankly, impossible. This does not mean that some universities do not sometimes go to extreme measures to try to police these exams — such as keystroke verification software and e-proctoring of online assessments. Such measures are both expensive and difficult to implement with tens of thousands of students writing from home. I therefore want to propose the following seven steps for building integrity into the assessment system.

First, make it clear that the university expects honesty in examinations. This should be written boldly in the heading of the first page of the examination sheet. I know dishonest students will ignore this, but the standard would be set in print as a reminder of what the university stands for and expects from its students. At least some drivers slow down when they see a big speed limit sign and many more do so when they see a speed camera reminding them of consequences.

Second, build integrity into the pedagogic relationship between lecturer and students. Teaching is a transaction of trust. When students don’t trust their lecturer to act in their best interests, but rather to try to catch them out with so-called smart questions, or encourage guessing which questions will appear from a “scope” briefing — then a cheating culture is being cultivated. Teaching should take the desperation out of assessment.

Third, remove all questions that require reliance on memory; recalling facts is quite simply poor assessment design, regardless of whether students write in a hall teeming with invigilators or from the comfort of their bedrooms. Good assessments try to probe for understanding, and especially a deep sense of the subject matter. It is hard to cheat when set problems demand that you show understanding. An openbook examination then gives you no advantage.

Fourth, replace memory questions with ones that call for application, comparison, contrast, and personal reflection. The question “what do you think about the legal, social and research questions about the reopening of schools?” requires students to express an informed opinion rather than restate what others have already said.

Fifth, provide students with second and third opportunities in subsequent exam opportunities. I know it increases the workload for those doing the assessments. But it takes away the pressure of an all-or-nothing sit-down examination; in other words, it lowers the chances of cheating.

Sixth, tell students a week in advance what the examination questions will actually be about. If these are smart questions that are intellectually demanding, it makes no difference to the distribution of scores in the final assessment. But the notice period directs students to read, think and discuss questions without the need for guessing.

Seventh, give students full feedback on how they performed before giving an additional opportunity to those who scored, say, 50% or less on the first or second attempt. If the purpose is education and not condemnation, then good assessment should enable the former. None of these suggestions are made in the abstract. I teach a class of about 280 students in a postgraduate certificate class for student teachers with a first degree from different disciplines (science, economics, languages and many others). I do my own marking. I call students personally when they achieve full marks. And I give generic feedback in writing and, on request, individual feedback as well as Zoom-feedback opportunities where students can ask questions about an assessment they struggled with. It is hard work.

But the goal of my online assessments is not to fail students (though some inevitably will) but to use the data from those assessments to enable them to achieve the highest standards in my field of teaching. In the process, the pressure to cheat is minimised and will, in any case, not help students at all.