Those details are what will make the new CourseSmart service tick. Say a student uses an introductory psychology e-textbook. The book will be integrated into the college’s course-management system. It will track students’ behavior: how much time they spend reading, how many pages they view, and how many notes and highlights they make. That data will get crunched into an engagement score for each student.When asked if this isn't creepy, Mr. Devine says "Not if it helps you succeed." But surely ends do not always justify means (water torture might work even better, after all).
The idea is that faculty members can reach out to students showing low engagement, says Sean Devine, chief executive of CourseSmart. And colleges can evaluate the return they are getting on investments in digital materials.
The comment thread is interesting. Some object because, they say, it won't work -- students will figure out how to game it. (But tracking eye saccades, so the professor can get word by word data for each student rather than page by page, will probably be next.) Others defend the practice on the basis that students can opt out (though that itself is arguably a privacy issue). Others complain that time spent reading is a poor measure of anything since students simply read a different speeds. Another just says "Relax and welcome to the 21st century," going on to explain some other electronic learning tools that help both instruction and assessment.
I think I react to this in other terms: dignity, maturity, independence. To the extent colleges behave as though they believe students are lab rats, to be trained through operant conditioning, they aren't doing anything to create intellectual excitement or an inclination to continue learning when unsupervised. And if higher education does not aspire to such things, we might as well be replaced by MOOCs, without those expensive dormitories, libraries, and student laboratories. If students aren't doing their reading, maybe we need to look deeper at why rather than resorting to surveillance in order to force them.
My first reaction was ``thats creepy''
ReplyDeleteBut upon reflection and reading the rest of the post
my second reaction was ``thats really creepy''
Frankly I don't care how the student learns the material
(off the course text, of some other text, in class,
in study groups, using water torture to get another student
to explain it to him) so long as the student learns it.
Testing if they know the material is fine.
Getting into testing their mechanism is... creepy.
In the paper about nb, our online annotation tool (http://nb.mit.edu/) we discuss what I think are beneficial applications of such monitoring. By looking at *aggregate* reading data, we can identify which parts of a text are "hanging up" students in general. It's natural to have a creepy first reaction, but remember that we already do a lot of "invasive monitoring" by asking students to submit homework that tells us what they understand and by keeping track of their attendance. So I don't automatically assume such monitoring is *bad*. On the other hand, the way it's described, it seems that they've chosen a mode of monitoring (and incentivizing) that is *silly*, since there are obvious ways to game the system.
ReplyDeletePretty interesting contrast between the last two comments, both by CS professors I believe. Unknown would argue, it seems, that the only problem with CourseSmart is its ineffectiveness; it would be fine if it actually worked and could not be gamed. Which, if I am not mistaken, would make Gasarch feel even more creeped out.
ReplyDeleteI actually accept that the technology may have its uses -- for schoolchildren for example. And I suppose I am to think that an e-book is not really analogous to a book, but to Rosetta Stone or something interactive like that.
I am waiting for Amazon to improve my reading experience by instrumenting War and Peace ....