More On Plagiarism

The brouhaha from earlier today reminded of this piece I read in the NYT a couple of weeks ago. It basically talks about the burgeoning cottage industry of websites and online repositories that feature essays, research papers, solutions to past exams and problem sets.

But as companies with playful names like Cramster, Course Hero, Koofers and SparkNotes are transforming the way undergraduates like Mr. O’Connor study, some professors and ethicists are questioning whether such Web sites encourage cheating and undermine the mental sweat equity of day-to-day learning by seducing students with ready-made solutions and essays.

On Course Hero, for example, students can type in a college name and course number to unearth the previous semester’s particle physics final exam. They can find examples of research papers on, say, the causes of World War I. For homework, Cramster supplies step-by-step solutions to problems in more than 200 college-level math and science textbooks.

I think there is a fine line here between learning from other people's mistakes and actually cheating. With respect to problem sets and computational (i.e. mathy/econy/sciency) exams, I've been in classes (and TAed classes) where the professor voluntarily gives the students material from previous years. This actually serves the students well, by alerting them to potential pitfalls and the like. The only profs who have a lot to worry about in this regard are the lazy ones who refuse to update their exams and problem sets, or the ones who directly lift questions from the course textbook. If you make the system that easy to game, the solution should focus on changing the system, not wringing one's hands at the gamers. Professors need to do a better job of keeping students on their toes.

And honestly, the whole idea of literally downloading and submitting someone else's research paper is a bit of an overblown concern. When you're a professor or a TA, and you're grading students' papers, you have a pretty good idea before you pick up the paper of what you're going to get from a given student. There are some surprises to be sure, but on the whole, it's pretty consistent.

Another point is that every class has its own DNA, with the professor's personal views, readings, organization, and class discussion. Again, you can get a pretty good idea of which papers have been produced by good honest work as a result of the exposure to the class material, and which have been wholly imported.

That said, I completely accept the point that faculty must be more accepting of changing the old way of doing things. One option is to have more take-home exams based on the class reading material, and nothing else. Assign the question the morning the exams are due, have them due back five or six hours later, and voila, the cheaters have no option but to do things the right way.

With more technical stuff (math, econ, physics), I think more quizzes and fewer assignments would be good. A weekly 5-10 minute quiz ensures that you're keeping up with the material, and leaves less room to cheat.