So during this class, we discussed homework. The research on who actually does it (not the kids who need it), the effect it has on inequality in education (makes it way worse), and generally why it should be knocked off its high pedestal.

Now, honestly, a lot of us teachers kind-of-sort-of know this but then assign “homework” anyway, but give time in class to complete it. Does this fix the problem? Well … there’s the matter of what students actually *do* when you assign homework.

If homework is “marked” in any way – including a simple check-for-completion! – you get the graph on the left (based on Peter’s research observations in multiple classrooms; image grabbed from a ppt from the course). ‘Getting Help’ means that someone like a tutor or a parent did the work ‘with’ them – and they probably can’t do it on their own afterwards.

If you remove homework marks completely, the cheating *almost* disappears and is split roughly evenly between Did It and Didn’t Do It. So it’s an improvement … that’s still only helping around 1/3 of your students. That’s, uh, not great numbers.

So what’s the fix? For starters, recognize that there really is enough time in-class for students to learn things. You might need to recover some of that time from things that aren’t students-doing-the-thinking, but it’s possible.

The next step is to replace homework with “check your understanding” opportunities … wait, no, bear with me, yes I *know* your textbook calls its question sets “check your understanding” and that hasn’t helped, just hold on before you leave. It’s not about replacing the *questions* or just what we call them – it’s about replacing the *mentality* and the motivation for doing them.

So the example we discussed was, a set of CYU questions are put out for students to work on together, labelled as to what topic they’re checking and/or what difficulty. Students are *not* required to complete them all. They’re there solely as a way for students to check whether they *can* do them. They’re there to give some indication of where the student is going and where they’re currently at.

Now, a weird effect that’s been observed is that even though textbooks *have* stuff you can use for this, you almost definitely are better off not giving students something in a textbook format – like not even a photocopy of a textbook page. Textbooks have too strong a “complete this busywork” association tied to them already, such that it poisons the well.

The *where they are* and *where they are going* guideposts are the most important part to any approach to this. Homework as ‘practice’ more-or-less doesn’t work: those who need it, don’t get it (see above), and unlike physical practice where you actually *can* get even better at something like weightlifting by continuing to lift over and over and over, math skills are not that transferable. Once you know how to factor quadratics, you reeeeeeally don’t need to do 50 more. Like, okay, maybe a few. But not 50.

But if someone isn’t yet there, like maybe they can factor the easy ones but don’t know how to handle the harder ones, they need some way to know that. Solving 50 easy ones won’t actually build up mental “muscles” that suddenly enable them to solve hard ones! That’s nonsense. They need to *know* that there’s more to learn and roughly how to get there.