01 / 20

Analyze Experimental
Design

Learn to identify variables and spot design flaws that make an experiment unreliable.

GED Science

What Is an Experiment?

An experiment is a planned test designed to answer a question by changing one thing and measuring the result. Scientists use experiments to find out whether one factor actually causes a change in another.

The Three Key Variables

Independent
The factor the scientist changes on purpose. Cause.
Dependent
The factor that is measured as a result. Effect.
Controlled
Factors kept the same for every test group.

Variables in Action

Independent Amount of sunlight Plant controlled: water, soil, pot Dependent Plant height

The Control Group Rule

A reliable experiment needs a control group: a group that does not receive the treatment being tested. It gives a baseline to compare the experimental group against.

Comparing Two Groups

Control Group No treatment (baseline) Experimental Gets treatment (the variable) vs

Correlation vs Causation

Correlation
Two things change together, but one may not cause the other.
Causation
One factor directly causes the change in another.

How to Analyze a Design

1
Find the question. What is the experiment trying to test?
2
Identify variables. What is changed (independent), measured (dependent), and kept constant (controlled)?
3
Look for a control group. Is there a baseline to compare against?
4
Check for flaws. Too few subjects? More than one variable changed? Biased sample?

Why Sample Size Matters

A larger sample size (number of subjects or trials) makes results more reliable. Testing only one plant or one person could give a result that was just luck.

Worked Example 1

A student waters one plant with plain water and another with fertilizer water. After 2 weeks, the fertilized plant is taller. Identify the variables.

Independent
Type of water (plain vs fertilizer)
Dependent
Plant height after 2 weeks
Controlled
Sunlight, soil, pot size, plant type

Worked Example 2

A scientist tests a new fertilizer. The treated plants get more fertilizer AND more sunlight than the control plants. The treated plants grow taller.

Worked Example 3

A company surveys 5 of its own employees about a new energy drink. All 5 say it boosts focus. The company claims the drink works for everyone.

✕ Too small a sample
5 people cannot represent everyone.
✕ Biased sample
Employees may favor their employer.
✕ No control group
No group tested without the drink.

Common Mistakes

✕ Confusing variables
Mixing up what is changed with what is measured.
✕ Changing too much
Altering more than one factor ruins the test.
✕ Correlation = cause
Assuming linked things must cause each other.
✕ Tiny samples
Drawing big conclusions from few trials.

Design Checklist

Good Design Flawed Design
Variables changedOnly oneSeveral at once
Control groupPresentMissing
Sample sizeLarge enoughToo small
Sample typeRepresentativeBiased
RepetitionRepeated trialsSingle trial

Question 1 of 5

A student tests how temperature affects how fast sugar dissolves in water. Which is the independent variable?

Question 2 of 5

An experiment compares plants given a new fertilizer to plants given nothing. The plants given nothing are the:

Question 3 of 5

A study finds that towns with more libraries have lower crime rates. What is the BEST conclusion?

Question 4 of 5

A researcher gives the treated group more medicine AND a special diet than the control group. Why is this a flaw?

Question 5 of 5

A claim is based on testing just 3 people. What is the main problem with this design?

You Can Now Analyze Designs

1
Spot the independent, dependent, and controlled variables.
2
Check for a control group and a fair comparison.
3
Watch for flaws: multiple variables, small or biased samples.
4
Remember: correlation does not equal causation.