Life scientists answer questions with experiments. In math you met dependent and independent variables; an experiment uses the same idea. A test is fair when only one thing can explain the result.
Key idea
Change one thing on purpose (the independent variable). Measure one thing (the dependent variable). Keep everything else the same (the controlled variables), and compare with a control group that gets the usual treatment or none. Repeat each group several times.
Why it works
If two things change at once, either one could have caused the difference, and you cannot tell which. Holding the rest steady leaves only one possible cause. Repeating trials matters because living things vary: one plant may simply be a strong seedling.
Worked examples
Worked example: Name the variables
Leo asks, "Does fertilizer change how many tomatoes a plant makes?" He gives three groups 0 g, 5 g or 10 g each week.
- He changes the fertilizer: that is the independent variable.
- He counts tomatoes: that is the dependent variable.
- Water, sunlight, pot size and the kind of tomato stay the same: controlled variables.
- The 0 g group is the control group.
Answer: Independent: fertilizer. Dependent: number of tomatoes.
Worked example: Find the problem
Maya puts plant A in a dark closet and plant B in a sunny window. She also waters B twice as often. B grows taller. What can she conclude?
- Two things differ: light and water.
- Either one could explain the taller plant.
Answer: Nothing certain. She must change only the light and water both plants the same.
Worked example: Use the mean
Three tomato plants with 5 g of fertilizer make 12, 15 and 12 tomatoes. What is the mean?
- Add: $12 + 15 + 12 = 39$.
- Divide by the number of plants: $39 \div 3 = 13$.
Answer: 13 tomatoes.
Common mistake
Changing two things at once (more light and more water), or leaving out the control group. Check: list everything that differs between groups. The list should hold exactly one item.