My Homework Dilemma

I'm doing it again. Every year (except my first), by the first of the year, I have pretty much stopped checking to see if students have done their homework. Keeping tabs on this just isn't a priority for me, although pretty much everything else goes smoother when everyone has done their homework.

The question of how to deal with homework has been one on my mind since my first week in my MiT program. I certainly didn't want to check it everyday (maintaining daily routines isn't in my skillset) and I was reluctant to include homework in my grading scheme at all. After all, we were being rightly advised not to grade first practice efforts. And yet if they didn't have any external incentive, students were unlikely to even look at the assignment much less earnestly engage with the material (which is what we teachers *really* want).

I haven't come much further since then - I have conceived of a dozen different systems for dealing with the issue of homework, and all of them have failed in certain, fundamental aspects. My next big idea is the Problem Set, much like a university course.

My wife, in her graduate math courses, is doing Problem Sets nonstop these days. They're great - challenging, thought provoking questions which must be addressed and completed by a due date at which point they are collected, closely graded and returned as feedback. The collection/grading frequency is relatively low (which suits my tastes) and there is flexibility for the student to prioritize their work in terms of timing - as long as they get it in by the due date. It doesn't quite address the "first practice" issue, though. I might assign additional problems (from the book?) to be their first practices. Then the Problem Set would be a concise and robust set of questions of my own design - these are really the best problem set questions anyway. Then I've just transfered the homework dilemma to the book problems - we'd still have to talk about them in class.

I toyed around with a lesson schedule which completely ignored time in class to discuss the homework assignments until the end of a section - right before a summary assessment. There were other problems with the idea (trying to fit different topics into the same 5-day time frame) so I've abandoned the idea, but maybe the model of delayed and concentrated homework discussion can be adapted to other lesson schedules.

The trick is that besides the lessons, in which the students are exposed to the ideas and/or develop the important concepts and skill themselves, a good learning model should have some independent, reflective practice built in with time scheduled to debrief these experiences. So what exactly is it that we, as teachers, would like to get out of homework? Better put, what does the learning classroom need from homework? Before designing a system, what will be the systems purpose?

Like a 15-lesson CD for learning a new language (which has other important lessons for the classroom teacher - another post), the students need instructional/discovery experiences, in which new material is presented or old material developed further, and practice experiences in which no new material is presented so as not to distract the student from securing the old information as knowledge. The interactive environment of the classroom is best utilized for the former, leaving independent time outside of the classroom for the latter.

Again, I'm thinking of the university model in which there are lecture meetings (direct instruction of new material), section meetings (more interactive concept development) and office hours (for reflection/questioning of the material). It's no wonder I have difficulty with homework discussions - I don't have any office hours. Or do I? Can I dedicate a class period to homework questions with some sort of optional participation? Then I'm addressing student concerns and not forcing every student to sit through a response to a question they don't need answered. Maybe it could be an open structure day of work - I can be helping and answering questions from those who have been working on the problem set (I'm already sold on this idea) and the rest can be catching up on the work in groups or independently. Such a day should probably not be the day before any kind of assessment - that would set up the less motivated for failure. Somewhere in the middle, then, or maybe to impose a delay between the question session and the assessment I can spend the next day or two working on the next topic, although I don't like the sound of that idea - too disruptive of the lesson sequence.

I suppose there is a third type of experience - reviewing and reflecting on the practice time. Since this is another interactive experience, it would work well in the classroom, although I have considered moving the discussion online via a chat room, blog, or threaded discussion. I haven't found any user-friendly and free services that would adapt immediately to classroom use, but the online forum has been successfully implemented at the university level for years now. Hell, the kids are already doing it with instant messaging. I should probably explore IM services more since free threaded discussions are clunky and full of high-school inappropriate adds.

Of course, I haven't even mentioned the problem of constructing the Problem Sets.