I'm giving my first midterm today. A little indication of how that impacts student behavior can be seen in the screen shot below. The number in the middle column is the number of hits for that post. The number in the left column is the number of comments. The class now has 28 students.
The post with the high numbers is one which linked to last year's midterm and where I responded to student questions on it. In contrast, the other posts indicate little interest. Further, most of the hits on the post with last year's midterm happened either yesterday or earlier today. Yesterday morning, that post only had 16 hits.
Some of this behavior we (the University and especially the instructors) induce by having exams in all classes clustered around the same time period. This is the simple consequence of dividing up the semester into chunks, which you must do if you are giving two midterms and a final. Viewed this way, it would better for students to take fewer classes at any one time, which would happen under a quarters system or if classes were only for a half-semester.
But I think much of this is because the students don't "turn it on" till near when an exam is approaching. Until then the vast majority of students are in passive mode. I do try to counter this by having weekly writing of blog posts and in most weeks other homework in Excel. I am underwhelmed by the effort I see from the students in doing this work.
You don't learn nearly as much by occasional bursts of energy followed by longer fallow periods. You learn a lot with a sustained intensity. The system doesn't seem to encourage that and I believe students are so habituated into their routines that the efforts of an individual instructor to counter this behavior will produce modest results, at best.
There is a by now an old argument on extrinsic reward versus intrinsic motivation in work and in learning. In that argument, intrinsic motivation wins, but only when it is likely to be present. The little evidence that I'm presenting here suggests that extrinsic reward is winning, in practice, and intrinsic motivation for students is rare. I don't know how we'd measure that more broadly, but if we did try to do this and if the conclusions from such looking more or less concurred with what I am saying here, then at least we'd have a problem statement.
We need that. Everything is not hunky-dory.