How Do You Know If Your Child Actually Understands?

School will tell you a lot about your child's performance and surprisingly little about their understanding. Those are different things, and only one of them lasts.

The good news is that understanding is much easier to check than most parents assume. You don't need to be a teacher, you don't need to know the subject well, and it takes about ten minutes.

Here are five checks, roughly in order of how much they reveal.

1. Can they teach it back?

The most informative question available to you, and the least used.

Ask your child to teach you something they've just learned — properly, as though you don't know it. Then be a slightly difficult student. Ask why. Ask what would happen if. Ask them to slow down at the bit you didn't follow.

What you're listening for isn't fluency. It's whether the explanation has structure: does one thing lead to another, or is it a list of remembered steps in a fixed order?

The tell is what happens when you interrupt. A child who understands can pick up from anywhere, take a detour, and come back. A child who has memorised a sequence usually has to start again from the beginning, because the sequence is all there is.

2. Can they handle a change of costume?

Take the same idea and dress it differently.

If they've done fractions with pizzas, try it with a measuring jug, a bar of chocolate, or a number line. If they've learned about supply and demand with an example about ice cream, ask about concert tickets.

Understanding survives a change of surface. Memorisation doesn't — because what was memorised was the pizza, not the fraction.

This is worth doing gently. A child who fails this check hasn't been lazy; they've learned exactly what they were taught, which was one example. The fix is more examples in more clothing, not more repetitions of the same one.

3. Can they spot a mistake?

Show them a worked answer with an error in it and ask them to find it.

This is a genuinely different task from producing an answer, and much harder to fake. To recognise that something has gone wrong, you need a working model of how it should go. Pattern-matching gives you no such model, so a child relying on it will usually either miss the error entirely or declare the whole thing wrong without being able to say where.

An excellent variant: get it wrong yourself, deliberately, and see if they catch you. Children love this and it removes all the pressure of being tested.

4. Does it survive a month?

The most uncomfortable check, and the most honest.

Go back to something from four to six weeks ago, unannounced and unrevised, and ask about it.

Be prepared for it to be shaky, because forgetting is the normal behaviour of human memory, not a sign of failure. Almost everything fades without retrieval — that's true of you as well.

So the thing to watch isn't whether they recall it perfectly. It's how fast it comes back. Genuinely understood material re-forms quickly from a small prompt, because there's a structure to rebuild onto. Memorised material has to be built again from nothing, because nothing was ever built in the first place.

Fast recovery: they learned it. Starting from scratch: they passed a test once.

5. Can they ask a good question about it?

The subtlest check, and the one that tells you most about depth.

A child who genuinely understands something can find its edges. They'll ask what happens in the weird case, or how it connects to that other thing, or why it isn't the other way round.

A child working from memorisation has no edges to find, because they don't have a model — only an answer. Their questions, if any, tend to be procedural: is this going to be on the test, do I have to show working.

You can prompt this directly: "What's still confusing about this?" or "What would you want to ask the person who invented it?" A blank is not a bad sign in itself, but a child who can consistently generate real questions about a topic understands it well.

What if my child fails these checks?

First: almost every child will fail some of them on some topics, including children doing extremely well at school. That's not a crisis. It's just the ordinary gap between performing and understanding, and now you can see it.

Second: be careful with your reaction. If checking becomes a thing that produces disappointment, your child will start performing for the checks too, and you'll have lost your only honest instrument. Keep it curious and low-stakes. You're finding out together, not auditing them.

What actually helps, once you've found a gap:

• Go back further than feels necessary. Gaps are nearly always upstream of where they show up. A child stuck on algebra is often really stuck on what the equals sign means.
• Ask for the "why" rather than more practice. Another twenty questions strengthens the pattern. One good explanation builds the model.
• Let them struggle a bit before you help. Attempting something before being told the answer makes the eventual explanation stick considerably better, even when the attempt fails.
• Space it out. Revisit deliberately after gaps rather than cramming. This feels inefficient and is one of the most robust findings in learning research.

Why doesn't school do this?

Mostly arithmetic rather than indifference. A teacher with thirty children and a curriculum to cover cannot run a five-minute teach-back with each of them per topic. Written tests scale; conversations don't.

So school measures what's measurable at scale, and what's measurable at scale is performance. That's not a scandal — it's a constraint. But it does mean the report you get home describes something narrower than you probably assume, and the gap is yours to notice.

Ten minutes at a kitchen table, done occasionally and without ceremony, will tell you more about what your child actually knows than a term of marks.

How does Skoolio check for understanding?

By building these checks into how learning works rather than bolting them on at the end.

Teach Me mode has the child explain a concept back — check one, every time. Debate requires holding and defending a position, which needs a model rather than a memory. Concepts return after deliberate gaps, so retention is tested by design rather than by luck. And mastery isn't awarded for one correct answer; it's awarded when understanding holds up across time and across formats.

Which is why the parent dashboard doesn't show you a score. A score would answer the question school already answers. It shows you what your child understands, where things are shaky, and what they're curious about.

Not where did they place. Do they know this — really know it?