The steps
- Ask a question. Make it something you can test, such as "Does a longer string make a pendulum swing slower?"
- Find out what you already know. Talk about it, look it up together, or think about times you have seen something similar.
- Make a prediction. Say what you think will happen and why: "I think the long string will swing slower, because it has farther to go."
- Plan a fair test. Decide what you will change, what you will measure and what you will keep the same.
- Run the test and record. Write results as you go, and repeat each test more than once.
- Draw a conclusion. Did the results match the prediction? It's fine if they didn't.
- Share it. Tell someone, draw a poster, or write it up in a science notebook.
Real scientists don't always go in a straight line. They often loop back, ask a sharper question and test again. Kids can do the same.
What makes a test fair
A fair test changes one thing at a time. Scientists call the parts of an experiment variables.
| Variable | What it means | In the pendulum test |
|---|---|---|
| Independent | The one thing you change on purpose | String length |
| Dependent | The thing you measure | Number of swings in 20 seconds |
| Controlled | Everything you keep the same | Same washer, same starting angle, same timer, same person counting |
If you change the string length and the weight at the same time, you can't tell which one made the difference. That is the whole idea of a fair test in one sentence.
A worked example
Here is how one family might run the pendulum test.
| String length | Try 1 | Try 2 | Try 3 | Middle result |
|---|---|---|---|---|
| Short (about a hand long) | Write it here | |||
| Medium (about an arm long) | ||||
| Long (as long as the table is high) |
Copy a blank table like this into a notebook before you start. Counting swings is easier with two people: one watches the timer and calls "go" and "stop", the other counts. After three lengths, most kids spot the pattern on their own. Shorter strings swing more times.
Helping kids make predictions
Young children often say "I don't know" when asked to predict. Offer two choices instead: "Will it sink or float?" Older kids can add a reason using "because". Write the prediction down before the test, so nobody can quietly change it afterward. Praise good reasoning, not right answers. A thoughtful wrong prediction teaches more than a lucky right one.
What's the difference between a hypothesis and a prediction?
A hypothesis is a possible explanation, and a prediction is what you expect to see if that explanation is right. For young children, "prediction" is plenty.
What if the results are messy?
Messy results are normal. Repeat the test, check whether something else changed, and talk about what might have caused the difference.