What makes an investigation controlled?
In a controlled investigation, a scientist changes one thing on purpose, measures what happens, and keeps everything else the same. This lets them link a cause to an effect.
Suppose a student notices that her bean plants near the window grow faster. Is it the light? The warmth? The fact that she waters those first? A controlled investigation sorts this out.
Start with a question and hypothesis
A testable question names what will be changed and what will be measured: "How does the amount of light affect how tall bean plants grow?"
A hypothesis is a testable explanation. A clear form is the if–then–because statement:
If bean plants get more hours of light, then they will grow taller, because plants use light energy to make food.
A hypothesis must be something that can be shown wrong by evidence.
The three kinds of variables
| Variable | Definition | Bean plant example |
|---|---|---|
| Independent variable | What you change on purpose | Hours of light per day (0, 4, 8, 12) |
| Dependent variable | What you measure | Plant height after 2 weeks (cm) |
| Controlled variables | What you keep the same | Seed type, soil, pot size, water, temperature |
A helpful sentence: "I will change the independent variable and measure the dependent variable."
Control groups
Many experiments include a control group that does not receive the treatment. It shows what happens normally, so you have something to compare against.
In a fertilizer test:
- Experimental groups: plants given 5 mL, 10 mL or 15 mL of fertilizer
- Control group: plants given water only
If all groups grow the same, the fertilizer probably has no effect.
Writing a procedure
A procedure is a numbered list of steps that someone else could follow exactly. Good procedures:
- list all materials and amounts
- give exact measurements ("100 mL of water every 2 days," not "some water")
- say how and when to measure
- include safety steps
If another scientist follows your procedure and gets the same result, your experiment is replicable.
Sample size and repeated trials
One plant might grow poorly because of a bad seed. A larger sample size reduces the effect of these chance differences. Use several plants in each group, or repeat each trial at least three times, then calculate an average.
| Light (h/day) | Plant A | Plant B | Plant C | Average height |
|---|---|---|---|---|
| 0 | 3 cm | 2 cm | 4 cm | 3 cm |
| 4 | 9 cm | 11 cm | 10 cm | 10 cm |
| 8 | 15 cm | 17 cm | 16 cm | 16 cm |
| 12 | 19 cm | 18 cm | 20 cm | 19 cm |
The data support the hypothesis: more light led to taller plants.
Ethics and safety
Scientists must plan safe and ethical investigations. That means wearing protective gear, treating living things with care, and getting permission before involving people.
Common mistakes
- Vague variables. "Plant growth" is vague; "height in cm after 14 days" is measurable.
- Changing more than one independent variable.
- No control group when one is needed.
- Tiny sample sizes, such as one plant per group.
Try it yourself
Design a controlled investigation for: "Does the temperature of water affect how fast sugar dissolves?"
- Write an if–then–because hypothesis.
- Name the independent, dependent and three controlled variables.
- Write a five-step procedure.
- Sketch a data table with room for three trials.
Quick recap
- A controlled investigation changes one variable and holds others constant.
- A hypothesis is a testable if–then statement.
- Independent variable = changed; dependent variable = measured; controlled variables = kept the same.
- A control group gives a baseline for comparison.
- Larger samples and repeated trials make results more reliable.