Via Michael Nielsen on Twitter, a Wired article and a research group website for the Stanford Study of Writing. As the Wired piece reports, the group has done a large study of student writing, and finds that modern college students write more and are better writers than students in the past.
This is a little hard to square with my personal experience (he says, procrastinating from grading a depressingly large stack of student lab reports), and that of many of the commenters at Wired. There are enough caveats in the description of the study that these needn’t be contradictory, though that’s not necessarily flattering for the students. My favorite sentence of the Wired piece is: “The Stanford students were almost always less enthusiastic about their in-class writing because it had no audience but the professor: It didn’t serve any purpose other than to get them a grade.” Yeah, because students twenty years ago really enjoyed their in-class writing…
It occurs to me, though, that this is a question that it ought to be possible to settle with Science. Many faculty have files of old papers and exams going back years– it ought to be possible to dig up some representative samples of student writing from ten or twenty years ago, and compare them with modern papers on comparable topics. This is probably more feasible in the humanities and social sciences than in the physical sciences, as the standard lab equipment has changed dramatically over the last decade or two, but even in physics this could probably be done.
All we would need would be a few old faculty files, and some faculty willing to assign the same questions to modern students. And some faculty who would be willing to do some double-blind grading of papers selected from those two groups.
Of course, this probably runs afoul of all kinds of research ethics rules, so it’ll never happen. Which is a pity, because I’d really like to know the answer (my own files don’t go back far enough to check. I do have a box containing some papers I wrote as a student, but that’s not exactly representative…).