3RD GRADE SCIENCE • ENGINEERING DESIGN

Planning Fair Tests — Plan a fair test for a model or prototype.

Learn how to set up a test that is fair so you can find out if your design really works.

Why Do We Need Fair Tests?

People have been building and testing things for a very long time. Imagine you built a paper airplane and wanted to know if it flies farther than your friend's airplane. You would need to throw both planes the same way, from the same spot. If you threw yours really hard and your friend's gently, that would not be fair! A fair test is a test where you only change one thing at a time so you can trust your results.

1400s
Leonardo da Vinci Tests Designs
The famous inventor Leonardo da Vinci drew plans for flying machines and bridges. He tested small models before building big ones.
1800s
Engineers Test Bridges
Engineers began carefully testing bridge designs. They learned they had to keep tests fair to know which design was strongest.
1903
The Wright Brothers Test Airplanes
Orville and Wilbur Wright tested over 200 wing shapes. They changed only one thing at a time to find the best wing design.
Today
Scientists and Engineers Use Fair Tests Every Day
From testing new sneakers to building rockets, fair tests help people figure out what works best.

All of these inventors had the same question: How do I know my design is actually better? The answer is to plan a fair test. If a test is not fair, you cannot trust the results. Let's learn how to plan one!

Key Ideas Behind Fair Tests

Before you run a fair test, you need to understand a few important ideas. These ideas help you plan your test so that you can learn something real about your model or prototype (a first version of something you are building).

1

One Change at a Time

In a fair test, you change only one thing. This is called the variable you change. Everything else stays the same.
2

Keep Everything Else the Same

The things you keep the same are called controlled variables. They make the test fair because nothing extra changes.
3

Measure What Happens

You watch and measure what happens because of the change. This is called the result or outcome. You write it down so you can compare.
4

Test More Than Once

Do your test several times. If you get the same result each time, you can be more sure it is correct. Scientists call these repeated trials.
KEY TAKEAWAY
Think of a fair test like a race. In a running race, all runners start at the same line, run the same distance, and start at the same time. If one runner got a head start, that would not be fair! In a fair test, everything is equal except the one thing you change.

See How a Fair Test Works

Let's look at an example. Imagine you built two ramps and you want to test which ramp surface makes a toy car roll farther. The picture below shows how to set up a fair test.

Both ramps are the same height and use the same car. The only thing that is different is the ramp surface. You measure how far the car rolls on the floor after it leaves each ramp.

Look at the diagram above. Notice how both ramps are the same height. The same toy car sits at the top of each one. You let go of the car the same way each time — no pushing! The only difference is the ramp surface. That is the one thing you change. This makes the test fair.

Steps to Plan a Fair Test

Planning a fair test is like following a recipe. If you follow the steps in order, your test will be set up the right way. Here are the steps you should follow every time.

Step-by-Step Guide

  1. Step 1 — Ask a question. What do you want to find out about your model or prototype? For example: "Does a taller ramp make the car go farther?"
  2. Step 2 — Pick the one thing you will change. This is your variable you change. In our example, it is the ramp height.
  3. Step 3 — Decide what you will measure. This is the result. In our example, you measure how far the car rolls.
  4. Step 4 — List everything you will keep the same. These are your controlled variables. Same car, same floor, same way of letting go.
  5. Step 5 — Plan to test more than once. Do at least 3 trials so your results are trustworthy.
  6. Step 6 — Record your data. Write down your measurements in a table so you can compare them.
⚠️ Remember!
If you change more than one thing at the same time, you will not know which change caused the result. That is why we always change only one thing!

Understanding the Three Types of Variables

When scientists talk about fair tests, they use the word variable a lot. A variable is anything in your test that can change. There are three kinds, and you need to know all three to plan a fair test.

The three boxes show the three types of variables. The variable you change (pink) is on the left. The result you measure (yellow) is in the middle. The controlled variables (purple) are on the right.

Look at the three boxes in the diagram. Every fair test has all three types of variables. If you can fill in all three boxes before you start your test, you are ready to go! Try this: whenever you plan a test, draw three boxes on your paper and label them "Change," "Measure," and "Same." Fill them in, and your test will be fair.

Worked Example: Testing a Parachute

Let's walk through a real example. Your class built parachutes out of plastic bags. You want to find out if a bigger parachute falls more slowly than a smaller one.

Planning a Fair Test for a Parachute
1
Step 1 — Ask a QuestionDoes a bigger parachute fall more slowly than a smaller parachute?
2
Step 2 — Identify What You Will ChangeThe size of the parachute is the variable you change. You will test a small parachute (20 cm × 20 cm) and a large parachute (40 cm × 40 cm).
3
Step 3 — Identify What You Will MeasureYou will measure the time it takes to fall from the top of a staircase to the floor. You will use a stopwatch.
Result to measure: fall time in seconds
4
Step 4 — List Controlled VariablesYou keep these the same: same material (plastic bag), same string length, same weight attached, same drop height, same person dropping it.
5 things stay the same
5
Step 5 — Plan Repeated TrialsYou will drop each parachute 3 times. Then you find the middle number to get a good answer.
3 trials for each size
6
Step 6 — Record Data in a TableMake a table with columns for the parachute size, trial number, and fall time. After testing, you can compare the times.
Sample data from the parachute fair test
Parachute SizeTrial 1Trial 2Trial 3
Small (20 cm × 20 cm)2.1 seconds1.9 seconds2.0 seconds
Large (40 cm × 40 cm)3.5 seconds3.3 seconds3.4 seconds

The large parachute took about 3.4 seconds to fall, while the small parachute took about 2.0 seconds. Because the test was fair, you can be confident that the bigger parachute really does fall more slowly!

Fair Tests vs. Unfair Tests

How can you tell if a test is fair or unfair? Let's compare some examples side by side.

FeatureFair Test ✅Unfair Test ❌
Things changedOnly one thing is differentTwo or more things are different
Controlled variablesEverything else stays the sameSome things are different and you didn't notice
Number of trialsTest is repeated 3 or more timesOnly tested once
ResultsYou can trust the resultsYou cannot be sure what caused the result
ExampleTest two paper airplanes with same paper, same throw, but different wing shapeTest two paper airplanes with different paper, different throw, and different wing shape
KEY TAKEAWAY
Think about baking cookies. If you want to know whether chocolate chips or raisins taste better, you should use the same recipe and the same oven temperature for both batches. The only thing you change is the mix-in (chips vs. raisins). If you also changed the oven temperature, you would not know if the taste was because of the mix-in or the oven!

From Fair Tests to Real Engineering

Planning fair tests is a big part of the engineering design process. Engineers design something, test it, learn from the results, and then improve it. When you get older, you will plan even bigger and more complex tests. But the basic idea stays the same: change one thing, keep everything else the same, and measure the result.

What You Learn NowWhat Comes Next
Change one thing at a timeDesign experiments with many variables
Do 3 trialsUse math to figure out averages and patterns
Write results in a simple tableMake graphs and charts to show data
Test a model or prototypeTest real products for safety and quality

Every time you plan a fair test now, you are practicing a skill that real engineers and scientists use every single day. Keep it up!

Practice Problems

PROBLEM 1CONCEPTUAL
What does it mean for a test to be "fair"? In your own words, explain why changing only one thing at a time is important.
PROBLEM 2BASIC
Maya wants to test if a longer paper airplane flies farther than a shorter one. She plans to use the same paper, throw both planes the same way, and test in the same room. What is the variable she changes? What does she measure? Name two controlled variables.
PROBLEM 3INTERMEDIATE
Carlos built two toy boats. He wants to test which shape moves through water faster. He tests Boat A in a bathtub and Boat B in a swimming pool. He pushes Boat A gently and Boat B hard. Is this a fair test? Explain what Carlos should fix.
PROBLEM 4APPLIED
Your class built two bridges out of craft sticks. Bridge A uses 20 sticks and Bridge B uses 40 sticks. You want to find out which bridge holds more weight. Write a complete plan for a fair test. Include: your question, the variable you change, what you measure, at least three controlled variables, and how many trials you will do.
PROBLEM 5CRITICAL THINKING
Emma tested her windmill prototype three times. Her results were: Trial 1 = 5 spins, Trial 2 = 12 spins, Trial 3 = 6 spins. Trial 2 is very different from the other two. What might have gone wrong? What should Emma do next? How does this connect to the idea of a fair test?

Lesson Summary

A fair test is an experiment where you change only one thing (the variable you change) and keep everything else the same (the controlled variables). You measure the result to see what happens. This helps you know for sure what caused the outcome. Engineers and scientists have used fair tests for hundreds of years, from the Wright Brothers testing airplane wings to engineers testing bridge designs.

To plan a fair test: (1) ask a question, (2) pick the one thing to change, (3) decide what to measure, (4) list what stays the same, (5) plan repeated trials, and (6) record your data in a table. Remember: if you change more than one thing, you will not know which change caused the result. Keep your tests fair, and your results will be trustworthy!

Varsity Tutors • 3rd Grade Science • Planning Fair Tests — Plan a fair test for a model or prototype.