Am halfway through the second set of papers: close reading of an early modern poem ten to forty lines in length, no interpretative or critical sources allowed, use of the OED highly encouraged. Two plagiarism cases so far (well, three, but the third one plagiarized a single sentence that wasn't even germane to her analysis, and once you get into single-sentence stuff, I'm inclined to dock the grade and move on rather than go through the hassle of filing a full report). I am disappointed, but not especially surprised; that is, in one case I'm both surprised and disappointed by WHO turned out to be a plagiarist, but not surprised that there were plagiarists. I mean, if you Google, say, "death be not proud poetry analysis," you WILL find many, many examples of what you are looking for. And as Angelo says in Measure for Measure, we are all frail.
There's a school of thought that says you're not supposed to give assignments like this. If you go on the Chronicle forums (which are a hotbed of judgmentalism, but somehow I can't tear myself away from them), you will find lots of people who will tell you that if you don't craft highly specific, unplagiarizable assignments that are unique to each class, you're just asking for trouble. As my older, Catholic-schooled relatives would say, you are giving students an Occasion Of Sin, and part of the moral responsibility is yours if they fall.
I don't agree. For one thing, I don't believe there's such a thing as an unplagiarizable assignment. (In my very first lit class, back at the University of Basketball where the Honor Code had teeth, I came up with a paper topic that I thought was unique -- "imagine you've invited two of the playwrights we've read this semester to dinner at your house and write a dialogue between them" -- and one kid STILL plagiarized. I doubt that anything will discourage the truly determined.) I'm also not convinced that it's a good idea to design your classes around the possibility that students will cheat. They probably will, but it gives the cheaters too much power.
Also, it seems to me that if the assignments are too specific, it's less likely that students will be able to transfer the skills they learn to their other classes. (I wonder if this is why the conferences I held last week involved one student after another telling me they'd never had to write a poetry analysis paper before. The simple, generic assignment that allows students a fair amount of latitude to make their own choices seems to have fallen out of favor.)
Still, I'm feeling a touch of nagging guilt; I have to remind myself that several of the other papers are excellent. At their best, the students picked poems that they liked and had a lot to say about, and a couple of them told me in conference that they were surprised at how much they were enjoying the assignment. (A few of them even chose to write about poems that they weren't required to read, since I told them they could pick any poem from the anthology if it met the requirements for the assignment -- it will give me a break from endless papers on "Death be not proud." Yay for people discovering Herbert and Vaughn on their own!)
I do think I need to do more scaffolding and really teach students how to focus closely on language, although this may not be feasible if I'm teaching two large-ish sections again; I'm not sure a short OED assignment plus class discussion are enough to prepare them for a longish analytical paper. One of the students showed me a cool color-coding method one of her high-school teachers had taught her; it involved identifying patterns of words and highlighting the ones that fit together in different colors. I might try this next time around.