Yesterday I attended a very interesting discussion on the problem of student collusion in assignments. It's a really grey area that is particularly prevelant in the sciences and engineering. This is the kind of thing we mean:
Student A and Student B discuss their assignment one evening. Together, they think through what processes are needed to solve the problems, and how to work through these processes. They then go away and write their answers individually. Unsurprisingly, they end up submitting very similar work. Their lecturer refers their work to the student discipline committee and the students are very aggrieved because they don't think they've done anything wrong.
Is this a form of cheating? As it turns out from our discussion, it depends on who you ask. Some would say no – each student has handed in an individually written assignment – but some would say yes – the ideas behind the assignment have been shared (or copied from one student to another) and that is collusion and unacceptable.
If the teaching staff can't agree on this kind of thing, what message do the students get? We often say things like "We encourage you to discuss the assignment and learn from each other, but what you submit needs to be your own work", but the students are left not knowing what this really means. The complaint is "on the one hand you are telling us to work together, and on the other hand you're telling us not to work together. What do you want?" This is unsurprising since it appears that we (as a group of lecturers) can't agree on what it means. At the discussion yesterday a lecturer from the Faculty of Law told us that Law students are often so paranoid that they might be accused of collusion that they go to great lengths not to talk to each other about assignments. That's probably unhelpful.
Anyway, is it actually a problem? One could argue that, if students are learning, then that is what we wish to achieve, and the fact that two assignments are rather similar is not a problem. I think that's a reasonable viewpoint, but then one has to ask the question, "How do you know that the student has actually learnt anything?" Consider the following scenario.
Students A, B, and C meet to work on their assignment. Student A is a strong student. Student B is not so strong, but is keen to learn and do well. Student C is keen to come out of the degree with a bit of paper that says 'BSc' on it. How he gets there and what he learns or doesn't learn on the way is immaterial. Students A and B have a good conversation about the assignment. Student B learns a lot from student A. Student B asks student A some really tough questions, which gets A thinking too and engaging with the material even more deeply. Student C pretends to listen. After this, students A, B and C prepare their answers, then have a look at what each other have written. Students A and B are happy to leave their assignments as they are; but student C then goes and changes his work to bring it in line with what the other two have written.
As a consequence, Students A, B and C hand in very similar work. Students A and B have learned a lot by doing this assignment. Student C has learned nothing.
The lecturer can't tell, from looking at the assignments, the extent of the student learning. They are all similar. How can the lecturer know then if the students have learned anything? As a result, all three get sent to the student discipline committee. The only way then is to ask the students, which is something that the student discipline committee is having to do very frequently.
If students aren't learning, the obvious place to look is the assessments. They are probably bad assessments. A good assessment task should drive a student to learn, not drive him or her to copy someone else's. One might argue that if this scenario is occuring, the lecturer needs to have a careful look at the sort of assessment tasks that are being set.
There are ways to go forward. There are on-line systems where every student gets a slightly different assignment to do (but addressing exactly the same learning outcomes), that is marked automatically. If student A copies student B, then their answers will be wrong, since student A has a different assessment to student B. (E.g. student A might have to evaluate the centripetal acceleration of an object at 3.5 m/s speed, moving in a circle of radius 2.0 m; student B might have to evaluate the centripetal acceleration when the object moves at 2.5 m/s around a radius of 3.0 m. It's the same learning involved, but the answers are different.)
Or, if we genuinely want students to collaborate closely, then we should set assignment tasks that demand a close collaboration – e.g. group tasks. The realilty is that almost no scientist or engineer works on their own – projects often have large teams of people on them. Team assignments then are what a student will face when they are in the workplace. So why aren't we giving them more of them as undergraduates?
The issue is a thorny one with no clear-cut answer. We can expect, therefore, a continual plethora of complaints from teaching staff, and counter-complaints of heavy-handedness and double-think from the students. Or, perhaps, we can think about our assessments more carefully.