Blog

Science at the Kitchen Table, With Things You Own

2026-10-18 · 9 min · English

Also in: Español · Português

Homeschool science does not get postponed because it is hard to teach. It gets postponed because it looks expensive. Search for elementary science and you get microscopes, kit subscriptions, goggles in three sizes, a monthly box of sealed packets, and the strong impression that a dining room is not a good enough place to do it.

It is a good enough place. For kindergarten through grade 5, the equipment list covering nearly everything worth doing is this: a jug of water, a kitchen scale, ice, measuring spoons and cups, a window, some paper, and a bit of patience. The microscope can wait until there is a reason for one.

Elementary science is a way of asking, not a list of facts

What makes a nine-year-old good at science is not knowing that water boils at 100°C. It is being able to ask how would I find out? and then set up something that finds out.

Facts matter and they accumulate. But a student who has memorised the parts of a plant and never planted a bean in a jar has learned vocabulary, not science. A student who grew two beans — one on the windowsill, one in the cupboard — and wrote down nine days of results has done the real thing, and the vocabulary sticks afterwards because it describes something they watched.

This is why the household-items approach is not a budget compromise. The apparatus being ordinary is an advantage: the student can set it up themselves, repeat it tomorrow, and change it on a hunch — none of which a kit with one packet of each ingredient allows.

The one idea that carries all of K–5: a fair test

If your student leaves grade 5 with one thing, make it this: a fair test changes one thing and keeps everything else the same.

The one thing you change is the variable you are testing. Everything else — the amount of water, the room, the time of day, the container, who is measuring — is held still on purpose, so that whatever happens can be blamed on the one thing you changed.

Students break this constantly, and the breaking is the lesson. When a test goes wrong, the question is never "what is the right answer?" It is "what else was different?"

Worked example, K–1: which paper towel holds more water?

Two squares of paper towel, two different kinds. Put ten drops of water on each with a spoon and watch which one soaks it up without leaking through.

The one thing changed is the towel. Kept the same: ten drops each, same spoon, same table, both lying flat. A five-year-old student will immediately want to put fifteen drops on the one they like better. That moment — "then it isn't a fair test any more" — is the whole curriculum in one sentence.

What the student notices: one towel darkens and stays whole; the other lets water through. They describe it; they do not need to explain absorbency.

Worked example, 2–3: does a heavier car roll further?

A board or stiff book propped on a stack of books makes a ramp. Roll a toy car down it and mark where it stops with a scrap of paper. Then tape two coins to the car and roll it again from the same spot.

The one thing changed is the car's weight. Kept the same: the ramp angle (do not bump the books), the starting line, the floor, and no pushing — released, not thrown. Run each version three times and measure with a tape measure or a piece of string.

What the student notices: three runs of the "same" car do not land in the same place. That spread is the point: the question becomes whether the gap between light and heavy is bigger than the gap between one light run and the next.

Worked example, 4–5: does salt make ice melt faster?

Two identical saucers, one ice cube each from the same tray, same size. Sprinkle a quarter teaspoon of salt on one. Set both side by side on the same counter, away from the window, and check every five minutes.

The one thing changed is the salt. Kept the same: cube size, saucer, location, air temperature, start time. A grade 5 student can go further and ask what they forgot to hold constant — did the salted cube get touched more? Should the saucers be swapped and the test rerun?

What the student notices: the salted cube melts faster, and makes liquid water at a temperature where the other is still solid. They can explain it as salt lowering the temperature at which water freezes, which is also why it goes on roads.

The other half: engineering design

Elementary science is not only investigation. The other strand — engineering design, which our curriculum covers as ETS1 — is about making something work, and it follows a different loop:

1. Define the problem. What has to happen, and how will we know it worked? "A bridge" is not a problem. "A bridge across this 20 cm gap that holds a full mug" is.
2. Generate more than one solution. Two sketches minimum, before any building. Everyone skips this step and it does the most work, because a student with only one idea defends it instead of testing it.
3. Build it. Paper, tape, cardboard, dry spaghetti, string, foil.
4. Test it against the criteria from step 1.
5. Improve one thing and test again. Same rule as the fair test: change one thing, so you learn what the change did.

Three problems that work at a kitchen table: a paper bridge across a 20 cm gap that holds a mug; a foil boat that carries the most coins before sinking; a cover that keeps an ice cube solid longest on a warm windowsill.

Eight activities with what to change and what to notice

Everything here uses things in an ordinary kitchen. Nothing here involves fire, boiling water, cleaning products, blades or anything sealed under pressure. Where an adult's hands are needed, it says so.

What to write down, and why that is the actual skill

Doing the experiment is the fun part. Writing it down is what makes it science, and it is the part homeschool science most often drops. A notebook page needs five lines, in this order, every time:

• The question. In the student's own words, ending in a question mark.
• What I changed. One thing.
• What I kept the same. The list. At K–1 this is spoken and the parent writes it.
• What happened. What was observed — a drawing, a number, a time, a measurement. Not a conclusion.
• What I think it means. The conclusion, kept separate from the observation on purpose.

Keeping those last two apart pays off for the next decade. "The cube with salt was gone at 11:20 and the other at 11:50" is an observation. "Salt melts ice" is a claim built from it. Students who write both, in that order, stop confusing what they saw with what they expected.

A spiral notebook is enough. Date every page. By May you have a portfolio without having assembled one — the easy half of record keeping.

What the grade bands should look like

K–1: observe and describe. The goal is noticing, in detail, out loud. Sort, compare, draw, use the words more, less, faster, rougher. One variable, two choices, no measuring tools beyond counting. Most of the writing is in the parent's hand.

2–3: compare and measure. Numbers arrive — the scale, the tape measure, the thermometer, a timer. Repeat a test three times and discover the three results differ. Record in a table instead of a sentence.

4–5: control variables and explain. The student names what they are holding constant before starting, predicts with a reason, and explains a result with a mechanism rather than a restatement. They should be able to look at someone else's test and say what is unfair about it.

What a worksheet can and cannot do

We publish free printable science practice sheets for every grade from kindergarten to grade 5 — one page per grade, 50 questions with the answers printed next to them, no sign-up. The hub is at /practice/science.html, and you can go straight to kindergarten, grade 2 or grade 5.

Here is what they are for, and what they are not.

A sheet checks whether the idea landed. After the ice-and-salt test, a question asking which variable was changed tells you in ten seconds whether your student understood the test or merely enjoyed it. That takes two minutes and is worth a lot.

A sheet is not the experiment. Fifty questions about plants does not replace a bean in a jar on a windowsill for nine days. If the sheets are the whole of your science, your student is learning to answer science questions, which is a narrower skill than it looks. Use them after a test, or on the day nobody has energy for a setup — not instead.

Everything free is listed at /free.html, including reading, English and writing, and a maths drill generator at /practice.html.

When you want the lessons as well

The free sheets are practice; they assume the teaching already happened. If you would rather not invent the teaching, the science book for a grade is $9.99 at /store.html: the lessons that teach each concept, worked examples, what to say out loud, the common misconception and how to catch it before it sets, a parent tip, review built in, and a question bank. Samples are at /workbooks.html, and what a lesson actually looks like walks through the shape of one.

A three-pack of the core subjects is $24.99, one subject across K–5 is $39, all subjects for one grade is $49, and everything is $349. For how that sits beside the rest of a budget, see how much homeschooling costs.

But start with the ice. The ice is free, and the student who has argued with you about whether a test was fair has already learned the thing the kit was going to sell you.

Questions: hello@uponlylearning.com.