“Check for understanding” is a common phrase in education. Seems simple — you check to see if students have learned something, and then do something about it.
I think what typically gets called a check for understanding is actually a few different things, and those are worth disentangling.
Check for Understanding
To me, a check for understanding is "did what I just tried to teach make sense." Students are supposed to learn something, then I ask them a question or give them a problem to see if they understand it. If they get the question wrong or can't solve the problem something went wrong. This type of thing is important, but it’s also prone to false positives. A student might seem to understand an idea because they just heard an example, but that doesn’t mean it will stick. Here's an example. For a long time I couldn't remember the difference between affect and effect. I would look it up, then use the correct word in whatever I was writing. This is a lot like a check for understanding. It looks like I understand the difference, I wrote it correctly! But I wouldn't retain it. I do this teaching all the time. "Here's how to solve this type of equation, now solve one just like it." I’m not saying this is bad, it’s just insufficient. It's not enough to make sure learning sticks.
Check for Retrieval
One reason affect/effect didn't stick for me is that I wasn't retrieving it from my long-term memory. I read "affect is the verb, effect is the noun, except for blah blah" and used the right one. I wasn't doing the thinking, I was just letting that information pass from the source through my brain to my document. In the same way, I need to check that students can retrieve what they learned when they don't have the guidance of a teacher. This isn't, "ok we did one, now do ten more" and it also isn't "ok remember what we just learned, the steps are this and that, try this problem." Checking for retrieval means giving students a bit of mixed practice where they need to think about what they're doing and remember which strategy applies to a problem. That thinking and reaching into the newly formed memory consolidates the learning and helps it to stick. As I’m defining it, a check for retrieval happens during the initial lesson. The difference between check for understanding and check for retrieval is the level of guidance and support, to see if students can do the math on their own.
Check for Remembering
This last one isn't very catchy but it's probably the most important. Students understood what they learned in the moment, and they could do a bit of retrieval in class immediately afterward. That doesn't mean the learning will stick days or weeks later. If that learning is important in a future lesson I need to check if they have remembered it. A question at the end of a lesson to check for understanding is pretty common. What's much less common is a quick check for understanding at the start of a future lesson where that knowledge is necessary. That’s a check for remembering. Often when I do this I realize that way fewer of my students than I thought remember what something. There’s nothing wrong with that. Forgetting is part of learning, and a quick reminder will often bring things back. A check for remembering is a chance to see if that reminder is necessary, and to make sure I’m not assuming students remember something that they don’t.
Coda
One word I hear over and over again recently is "mastery." Teaching to mastery, making sure students have mastered this skill. I don't think mastery is a real thing. There is lots of math I've been pretty good at, and then stopped using and forgotten. I taught calculus for six years and got very good at all the ins and outs of Taylor polynomials. By any reasonable definition I mastered those skills. But I haven't used them in a few years and I would need a refresher on a lot of it if you gave me a Taylor polynomial problem today. I had a student in my homeroom once who had incredible knowledge of flags of countries around the world from playing the FIFA soccer video game. I spent a weekend practicing and quizzing myself and I beat him in a flag naming competition. I could probably name 180 different flags that day. Two weeks later he challenged me again and I'd forgotten half of them.
Forgetting is part of learning. Mastery doesn't exist. Learning fades if we don't use it. A big part of teaching is checking for understanding in lots of different ways and at different intervals, making connections between old ideas and new, and building on prior learning. I’m distinguishing between these types of checks for understanding because if I skip one of these steps the learning is a lot less likely to stick around for the long term.
"Mastery doesn't exist". Hmm, that's a little strange. After all, we can observe examples of mastery in sports, the arts, technology, . . . The list goes on. So I'm this is the best formulation.
"Learning fades if we don't use it." I agree absolutely: use it or lose it. And you provide examples of this from your own experience, with calculus or FIFA flags. But for a while you really did have mastery!
I think we underestimate just how quickly we lose mastery. It happens much faster than we think! There's a famous quote about musical practice (attributed to many different people): "If I don't practice for one day, I notice it. If I don't practice for two days, the critics notice it. If I don't practice for three days, the public notices it."