Science experiments at home that actually teach
The fizz is the easy part. A guess before, one change at a time and a why after are what turn a kitchen experiment into real science.

The internet is full of science experiments for kids: baking soda volcanoes, walking water, an egg sucked into a bottle. Nearly all of them work, in the sense that something fizzes, climbs or pops. Whether your kid learns anything from the fizz is a different question.
The good news is that you don’t need a better experiment. You need a few small things around the one you already have: ask for a guess before you start, change one thing at a time, and ask why afterward. Then bring it up again a day or two later. That turns a trick into science, and it works with whatever is in your kitchen.
The short answer
- Start with a question, not a recipe. “Why do some things float?” beats “let’s make a lava lamp.”
- Ask for a guess first. A prediction gives your kid a stake in what happens.
- Change one thing at a time. That’s what makes it a test rather than a show.
- Ask why, and let them explain. Putting it in their own words is where it connects to what they already know.
- Ask again tomorrow. Recalling it later does more for memory than seeing it again.
Start with a question
Every experiment answers a question, even when the instructions never say which. Find it first, and say it out loud. In Tinkerworks Lab, every room starts with one, and they make good models for home:
- Density Pool: Why do some things float?
- Static Electricity: Why does a balloon stick to your hair?
- Shadow Theater: Where do shadows come from?
- Pendulum Wave: Can swinging make a wave?
A question gives the experiment a point. Your kid isn’t watching a trick. They’re finding something out.
Ask for a guess first
Before anything happens, ask: what do you think will happen, and why? Write the guess down, or just say it out loud so it’s on the record.
This one isn’t from a study we can point you to. It’s practical. A kid who has guessed has a stake in the result. If they’re right, they want to know why. If they’re wrong, they really want to know why, and that’s usually the conversation worth having.
Don’t tell them the answer first. The whole point is that the experiment does the telling.
Change one thing at a time
This is the idea that makes science science. If you want to know whether a heavier weight changes how a pendulum swings, change the weight and keep the string the same. Change both at once and you can’t tell which one did it.
It’s easy to practice at home. Pick one thing to change, keep everything else the same, and run it again. Ask your kid what you’re keeping the same, and why that matters.
The Lab is built around the same move. Each room runs a model of the real thing, so a kid can change one thing, like the length of a pendulum or the angle of a wing, and see what it does.
Ask why, and let them explain it
When it’s over, ask why it happened, and then let your kid explain it out loud, in their own words. Resist the urge to fill the silence with the right answer.
There’s research behind this one, with a limit. A large review of ten study techniques gave a moderate rating to two that look a lot like it: elaborative interrogation, which roughly means asking why something is true, and self-explanation, which roughly means explaining it in your own words. Both came out behind practice testing and spreading practice out, which were rated highest (Dunlosky et al., 2013). The review was about students studying, not kids at a kitchen table, so treat it as a reason to try it, not a guarantee.
If the explanation is wrong, that’s fine. Ask what would show whether it’s right. Then run it again.
Ask again tomorrow
The next day, at breakfast or in the car, ask: remember what we floated yesterday? What happened, and why?
That little question does more than it looks like. In two experiments with prose passages, people who were tested on a passage remembered more of it two days and a week later than people who reread it (Roediger & Karpicke, 2006). The same review that rated asking why as moderate put practice testing, which means recalling something instead of looking at it again, among its two most useful techniques (Dunlosky et al., 2013).
Say what the model leaves out
Here’s a step most home experiments skip: no model is the whole truth. A pendulum on a string has air pushing on it. A toy car on a ramp has friction slowing it down. Saying so out loud teaches something real about how science works. We simplify on purpose, and we say what we simplified.
Every room in the Lab wears a label, in its own words, saying what its model leaves out:
- Orbits: Simplified: one planet, no air
- Roller Coaster: Simplified: one car, no air
- Build an Atom: A model, not to scale
Four to try this week
Each of these uses things you already have, and each one matches a room in the Lab if your kid wants more. Ask for a guess before every one. We’ve left out what happens on purpose. Balloons and grapes are choking hazards for little ones, so stay close.
- What floats? Fill the sink or a big mixing bowl with water on a Saturday morning. Collect five things from the kitchen: an orange, a spoon, a grape, a cork and an apple. Guess which will float, then test them one at a time. Then peel the orange and try it again. (In the Lab: Density Pool.)
- The sticky balloon. Rub a blown-up balloon on your kid’s hair or a wool sweater, then hold it near a wall, some tiny bits of paper, or a thin stream of water from the tap. What changes if you rub it for longer? (In the Lab: Static Electricity.)
- Shadow sizes. After dark, shine a flashlight at a bedroom wall and hold a toy between them. Guess what happens to the shadow as the toy moves toward the light, and then away. (In the Lab: Shadow Theater.)
- The fair pendulum. Tie a washer or a key to a piece of string and let it swing. Count the swings in 10 seconds, three times, and take the middle number. Now change one thing only: a longer string, or a second washer. Which change makes a difference? (In the Lab: Pendulum Wave.)
Where the Lab fits
Tinkerworks Lab is a workshop of 22 rooms, each a model of something real that a kid can take apart: wind over a wing, sand on a vibrating plate, a coaster on its first hill. Professor Funke runs it, and every one of his 276 facts is checked against a published source. Each room ends with Test your learning, which asks a kid for the answer instead of showing it again.
Two things it doesn’t do, so you know. It doesn’t bring ideas back on a schedule the way Times Machine brings back facts, so the ask-again-tomorrow part is yours. And it isn’t a curriculum. It’s a lot of science to play with, not a year of it, and we haven’t studied whether it raises science grades.
The best way we know to use it is the same as the kitchen: let your kid play with a room, then ask them at dinner what they found out.
Sources
- Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Psychological Science in the Public Interest, 14(1), 4–58.
- Test-enhanced learning: Taking memory tests improves long-term retention Roediger, H. L., & Karpicke, J. D. (2006). Psychological Science, 17(3), 249–255.
Every source here is one our science page leans on, with what we checked in it and where its evidence stops.


