« September 2005 | Main | November 2005 »

October 20, 2005

Cheaters Never Prosper, Or Do They?

I read a very interesting article about student cheating in Communications of the ACM recently that reminded me that finding ways to subvert the system is still very much a part of the mindset for many students and led me to wonder how much effort we can and should put into discouraging it.

I never really thought very much about cheating among computers science students until the late 1980's when I was working in a computer science department at a very large university. When discussing this issue, the faculty tended to fall into two camps: the "perpetrators should be punished" camp and the "boys will be boys" camp. What surprised me, however, was that more of the faculty tended to fall into the latter than the former group.

What I found particularly irksome was the opinion among these folks that somehow computer science students were different or should be treated differently than other students in the university. Being a fine old institution, our university had a history of being particularly harsh in matters relating to plagiarism. A student in the English department caught passing off a couple of borrowed sentences as her or his own in an essay would be publicly disgraced and dismissed from the program and from the university. Why should the case be different for computer science students?

Keep in mind too, that this was back before the days when we began to look at our teaching methodologies in light of industrial practices relating to software development. There was no groupwork as part of the curriculum. There was just stealing, and the magnitude and creativity behind it was almost staggering. In those days people trooped off to the Computing Center to run and print their programs. Printouts were stolen with great regularity. Some students became seasoned dumpster divers, rifling through the cast off paper in the garbage for bits of useable code they could steal.

Sometimes, desperate times call for desperate measures. In order to curb what he perceived as a growing tendency toward "unofficial collaboration" one colleague of mine adopted a particularly successful strategy called "one cheats, two fail". When he found duplicate code on individual assignments, he called both students into his office and told them that both of them would fail unless the copier confessed. Both students were humiliated, the student who did the original work learned to protect it more carefully, and rough justice was often administered to the cheater who refused to admit to the act.

In the 1990's I also worked for an educational publisher and I would frequently receive email from students posing as teachers requesting copies of textbook teacher guides so that they could have the answers to the class assignments. Over time, I developed an almost uncanny ability to spot the pretenders. Usually it was their appalling grammar.

These days, students simply comb the Internet for snippets, applets, or entire applications to submit as original work. Some folks still defend this as justifiable on the basis that code reuse is a highly efficient and effective use of programmer time. Others remain vigilant.

How about you?


Posted by cstephenson at 06:03 PM | Comments (3)