The Great Harvard Cheating Scandal of 2012 rumbles on, with more details emerging which make it seem to me that it was neither “cheating” nor was it an “examination”. Â Regardless of what the official course syllabus stipulates, contradictory guidance from a professor and 4 teaching assistants, along with accepted practice and failures to catch this sooner, lead me to believe that Harvard will have to drop the allegations. I doubt if we will ever know the full detail of this specific case, but it touches on general issues of course design which I want to comment on.
For those who missed it the brief summary is that about half the students on a final year class have been accused of cheating because they collaborated on a take home final examination.  As someone who has taught in a university for nearly 20 years, that situation is all wrong.
First, a take home is not an examination.  In my book, an exam is when all the class have a limited time, in a hall under supervision, to answer a set of unseen questions.  It is a tried and tested means of assessing students, causing them fear and terror, and testing their knowledge under stress. It is the only way to be sure you minimise collaboration.  It is administratively convenient, but in the 21st century I don’t think it develops useful or healthy knowledge work skills which is why I never use it any more.  If you hand out a list of questions and allow students to take them away to work on, you must expect discussion and collaboration.
Second, collaboration, in the 21st century, is not “cheating”; it is an essential skill which universities need to foster, not condemn as collusion.  Universities have a dangerously ambiguous attitude to the collaboration/collusion issue which is actually unhealthy.  In real life, no one, apart from the odd idiosyncratic genius, works alone.  There is a clear case the some creative work can only be done alone, and I accept that, but in most jobs from NASA down to McDonalds, people work in teams. Learning how to do that effectively – divide work, make best use of the skill mix in the group, fairly reward effort, helpfully give and take supportive criticism and final alternative roles for people who can’t hack it – these are all vital skills.
I set group assignments, and even where I’m setting individual papers, I tell my students it is ok to collaborate as long as the final paper reflects their own work, and shows evidence of their mind at work – I expect them to assert their intellectual ownership of the material.  I minimise the risk of freeloading collusion because I require students to submit, during the course, reading summaries, mindmaps, essay plans and drafts which they will collect and  turn in at the end along with the essay.  Apart from making explicit the research and writing process, it means there is a trail of evidence to show student engagement with the topics – if you are really worried about “collusion” then design assessments which reduce the risk of people cutting corners. If you don’t design assessments to reduce collusion,  you can’t complain when a class cuts corners on the way to an A.
And why is anyone surprised that the work submitted by half the class shows similarities? We have not yet been told how alike the answers are, so it is hard to judge this point.  (Most likely, we will never see enough detail in public to judge this point in this case, since student work is usually treated as confidential unless the students themselves chose to publish it.)  However, in any given class, led by one academic who has chosen a set of topics, readings and paper titles, would you not expect the class to come up with similar solutions?  If the class came up with 279 different answers, would one not ask what was the teacher doing? Even in something like a creative writing class, the approach taken by the instructor will tend to shape the class in particular directions. In any topics in the arts, humanities and social sciences, there are a number of central theories, often embodied in the same key readings which turn up on course reading lists all over the world. If students work over these readings in 10 person tutorials, a consensus will emerge – and if it doesn’t, then, again, you really must ask what were the teaching team doing if they did not shape the course to cover those key issues?
I tend to avoid lecturing, favour discussion and try to convey the impression that I do very little work and let my students explore problems, but really even with that approach, I’m still herding them to explore paths I know to be safe – I picked the readings and topics, so I’ve defined the territory for the course. At the start of term, I can predict which questions will be most popular, and I have a pretty good idea which will hit my desk at the end. In my War, State and Society, for example, many will do the Guerilla War topic and they will all, in various ways, deal with T.E. Lawrence, Mao, Che and a few others, because those are unavoidable in that topic. They will have used the same readings, and since they come from similar backgrounds, they will process that material and serve it back in broadly similar language.  The reason we pay for universities is that academics are expected to use their knowledge to shape learning experiences. A result is that, particularly in large classes, half the class will often submit similar answers to the same questions.
I also wonder why this is only coming up now, after many, if not all, of that class is finished, graduated and at work.  If there was substantial collusions, possibly amounting to plagiarism in some form, how was that not detected during the grading and before the marks for the course were finalised?  What is the process at work? I have taught classes of 280-300, with 4 grad students grading papers, and we, as a matter of course, cross checked samples of the work. In most of courses in my dept, a random sample of papers are moderated by another full member of staff, and then sent to an external examiner from another university before the marks are approved at a final board. We have  robust process to monitor standards,  and in many UK universities with better staff-student ratios, their processes are even more rigorous.  So what happened in Harvard? How is it that in one of the worlds top universities, where people pay substantial fees, that the process, whatever it is, allowed this to slip through until after many of these students were gone out the door with degrees?
I have quite a few questions about this scandal, which the media have jumped on with great alacrity. When the final account emerges, it will undoubtedly be more complex than the headlines so far. The process, which rightly must allow all the students and the teaching team to present their case, will take time, and when it is done, the media will have moved on looking for other bad news stories.
Leave a Reply