Setting Homework That's Worth the Time It Costs You

There is a particular kind of homework that exists only because it is expected. You set it on Thursday because the calendar says Thursday. You spend part of Friday finding out who hasn't done it. And at some point over the weekend you look at a pile of it and quietly accept that you are not going to learn much from what's in front of you. The students spent time on it. You spent more. And the honest answer to what it changed about next week's teaching is usually nothing at all.

It's worth saying the uncomfortable part plainly, because most conversations about homework skip it. Homework is the only work in the job that you set, chase, mark, and then rarely use. Every piece of it costs you three separate times — once when you decide what it should be, once when you follow up the students who didn't do it, and once when it comes back and has to be looked at. The students carry the first cost. You carry the other two, for every class, every week, for the whole year.

So before deciding how much to set, it helps to be strict about what a piece of homework is actually for. There are really only two jobs it does well. It can give practice on something you have already taught properly, so a skill becomes automatic. And it can pull back something you taught weeks ago, before it fades. Both work because the student can do the task alone — without you, and without help at home. Almost everything else, including new material they haven't been taught and projects that need a printer, a device and a quiet room, measures the house a child goes home to more than it measures their learning.

Once it is set for one of those two jobs, the shape can stay the same all year, and that is where the time comes back. The same structure every week: a small number of questions on what you taught this week, and a couple on something from a month ago. Students stop asking what the homework is, because the homework is always the same shape. You stop designing it from scratch, because you are only changing the content. And the adults at home, who are often trying to help and rarely told how, finally know what the routine looks like.

The chasing and the marking take most of the rest of the time, so decide both rules once and stop re-deciding them. One consequence for missing work, applied the same way to everyone, costs a fraction of what a case-by-case judgement costs — not because the judgement is wrong, but because making it thirty times a week is what wears you down. And don't mark practice the way you mark assessed work. If the task is retrieval, it can be checked in the room in five minutes, by the students, against a sheet. What you need from it isn't a book full of ticks. It is the two or three things the class got wrong, and that, read as a signal rather than as compliance, is worth more than a weekend of marking.

This is also the sort of work worth handing over, because it is mechanical rather than professional. Since aime works from your curriculum and what your classes have actually covered, it can build the week's homework out of the right prior material — this week's practice, plus retrieval on the topic you taught a month ago — with the answers alongside it so the checking takes minutes. You still decide what matters and what to do about what it shows. You are just not writing the same shape of task from scratch, fifty times a year, for every class you teach.

Set less than you think you should, set it for one job, and read what comes back. If you wouldn't use what comes back, the task was set for the calendar rather than for the class, and it is worth redesigning rather than repeating. Homework doesn't have to be another evening of work for you. Shaped properly, it is the cheapest information you will get all week.

If you want that homework — and the planning, retrieval and marking around it — built from what your class has actually covered, that's what aime is for. It's free to start: 14 days of everything, no card needed, at teachaime.ai.