Working Scientifically — Fair Tests
In 2003, CSIRO researchers at Narrabri, NSW ran a 4-year field trial with 12,000 wheat plants — changing only one gene variable — to fairly test whether drought resistance boosted yield by more than 20%.
Printable Worksheets
Print or save as PDF — or build a custom worksheet from any module's questions.
Q1 · You want to test "Do plants grow faster with more light?" What would you change, measure, and keep the same?
Q2 · If you repeated an experiment 5 times and got very different results each time, what does that tell you about your experiment?
● Know
- The names and definitions of independent, dependent and controlled variables
- The difference between reliability and validity
- The five steps of a fair-test plan
● Understand
- Why only ONE variable should change at a time in a fair test
- Why a result can be reliable but still not valid
- How to write a testable hypothesis using "If… then… because…"
● Can do
- Identify the IV, DV and controlled variables in any investigation
- Write a hypothesis for a given question
- Plan a fair test with method, variables and results table
- Independent variable
- Dependent variable
- Controlled variable
- Reliability
- Validity
- The result you measure in an experiment
- Getting similar results when you repeat the experiment
- The one thing you deliberately change
- Measuring what you actually intend to measure
- Something you keep the same throughout
In a race, it would be unfair if one runner got a head start, wore lighter shoes AND ran on a flat track while others ran uphill. Science works the same way — only ONE thing should change at a time.
Every fair test has three types of variables:
- Independent Variable (IV) — the ONE thing you CHANGE on purpose. This is what you're testing.
- Dependent Variable (DV) — the thing you MEASURE. It "depends" on what you changed. This is your result.
- Controlled Variables — everything ELSE you KEEP THE SAME. Without controls, you can't be sure what caused any change you see.
Example investigation: "Does the surface area of sugar affect how quickly it dissolves?"
| Variable type | In this experiment |
|---|---|
| IV (what you change) | Surface area of sugar — whole cube vs crushed sugar |
| DV (what you measure) | Time (seconds) for sugar to fully dissolve |
| Controlled variables | Mass of sugar (same), temperature of water (same), volume of water (same), stirring rate (same) |
If you accidentally changed the water temperature between trials, you wouldn't know whether any difference was due to surface area or temperature. That's why controls matter so much.
- Water temperature
- Time for salt to dissolve
- Mass of salt used
- Volume of water
- Dependent variable (DV) — what is measured
- Controlled variable — kept the same
- Independent variable (IV) — what is changed
- Controlled variable — kept the same
Two questions matter for every experiment:
- Reliability — if you repeated this experiment, would you get similar results? Reliable experiments reduce random error. You improve reliability by:
- Doing multiple trials (at least 3) and averaging the results.
- Using the same equipment each time.
- Following the same method steps in the same order.
- Validity — are you actually measuring what you intend to measure? A valid experiment:
- Changes only the IV (has good controls).
- Uses a method that directly tests the research question.
A result can be reliable but not valid. Imagine your scales are broken and always read 10 g too heavy — you'd get the same wrong answer every time (reliable), but you're not measuring the true mass (not valid).
A result can also be valid but not reliable — you test the right thing, but your results jump around because of uncontrolled factors. The goal is to be both.
| Results consistent? | Actually testing the right thing? | |
|---|---|---|
| Reliable AND valid | Yes | Yes ✓ (the goal) |
| Reliable, not valid | Yes | No (e.g. broken scales) |
| Valid, not reliable | No | Yes |
An experiment is when you get similar results each time you repeat it. An experiment is when you are actually measuring what you intend — this requires good variables. You need for a trustworthy result.
Use these five steps every time you plan an investigation:
- Question — Write your research question in the form: "Does [IV] affect [DV]?"
Example: "Does the surface area of bicarb soda affect how quickly it reacts with vinegar?" - Hypothesis — A testable prediction with a reason: "If [IV changes], then [DV will increase/decrease] because…"
Example: "If the surface area of bicarb soda increases, then the reaction with vinegar will be faster, because more particles are exposed to react." - Variables — State your IV, DV, and at least 3 controlled variables.
- Method — Write step-by-step instructions including amounts, equipment and safety. Anyone should be able to follow your method and get the same results.
- Results table — Draw a blank table before you start. Put the IV in the first column; add columns for each trial and an average column for the DV.
| Surface area of bicarb (IV) | Trial 1 — time (s) | Trial 2 — time (s) | Trial 3 — time (s) | Average (s) |
|---|---|---|---|---|
| Whole tablet | ||||
| Crushed (powder) |
A student is testing "Does the height a ball is dropped from affect how high it bounces?" She drops balls of three different sizes from three different heights, and measures the bounce height. Is her experiment a fair test? Predict and explain your reasoning.
How close was your prediction?
Nice — you spotted the extra uncontrolled variable.
Key insight: changing more than one thing at a time makes it impossible to know which variable caused the result.
For each investigation below, identify the IV, DV and two controlled variables.
- "Does the colour of light (red vs white vs blue) affect the rate of photosynthesis in aquatic plants?"
- "Does the concentration of vinegar affect the amount of gas produced when mixed with bicarb soda?"
Choose ONE of the following questions and write a complete 5-step fair-test plan:
- "Does the surface area of a soluble tablet affect how quickly it dissolves in water?"
- "Does the mass of a parachute affect how quickly it falls?"
Your plan must include: question, hypothesis (with "because"), IV, DV, at least 3 controlled variables, a 4-step method, and a blank results table.
Earlier you were asked: A student tests "does music help plants grow?" but plays music to one plant while watering it twice as much. Is that a fair test?
Now write a full explanation of what is wrong, using the terms independent variable, controlled variable, and validity.
Q1. A student investigates "Does the temperature of water affect how quickly salt dissolves?" Identify the IV, DV and two controlled variables for this experiment. (3 marks)
Q2. Explain the difference between a reliable result and a valid result. Give one example of each. (3 marks)
Q3. Plan a fair test for the question "Does the surface area of bicarb soda affect how quickly it reacts with vinegar?" Write a hypothesis, list all variables and outline the method in 4 steps. (5 marks)
Answers
▾MCQ 1
C — The dependent variable is what you measure — the thing that changes in response to the IV. Temperature is what you change (IV), volume of water is controlled, and time to dissolve is the result you measure (DV).
MCQ 2
C — Reliability is about getting consistent results when you repeat the experiment. The best way to improve reliability is to repeat trials (at least 3) and calculate an average to reduce the effect of random errors.
MCQ 3
B — Ball size is a variable that was not kept the same (not controlled). If ball size varies alongside drop height, you cannot know which of the two caused any change in bounce height — the results are not valid.
MCQ 4
B — Validity means you are actually measuring what you intend. This requires good controls — only the IV changes, and everything else stays the same. Getting the same result every time is reliability, not validity.
MCQ 5
C — A proper hypothesis identifies the IV and DV and includes a scientific reason ("because…"). Options A, B and D are vague predictions with no reason and no clear IV/DV structure.
Short Answer 1
Model answer: IV = temperature of water (e.g. 20°C, 40°C, 60°C). DV = time (in seconds) for the salt to fully dissolve. Controlled variables (any two of): mass of salt used, volume of water, type of salt, stirring rate, size of salt crystals.
Short Answer 2
Model answer: A reliable result is one you get consistently when you repeat the experiment — the data is reproducible. Example: measuring the mass of a rock three times and getting 45 g, 45 g, 44 g — very consistent, so reliable. A valid result is one that actually measures what you intend — your experiment is set up correctly with good controls. Example: testing whether drop height affects bounce height, keeping the same ball each time (controlled), so any difference is definitely due to drop height (valid).
Short Answer 3
Model answer: Hypothesis: If the surface area of bicarb soda increases (crushed vs whole tablet), then the reaction with vinegar will be faster, because more bicarb particles are exposed to vinegar at the same time. IV = surface area of bicarb (whole tablet vs crushed powder). DV = time (seconds) for bubbling to stop. Controlled variables: mass of bicarb (2 g each trial), volume of vinegar (50 mL each trial), concentration of vinegar (same bottle), temperature of vinegar (room temperature). Method: Step 1 — Measure 50 mL of vinegar into a beaker. Step 2 — Add 2 g of whole bicarb tablet and start a stopwatch. Step 3 — Stop the watch when bubbling stops; record the time. Step 4 — Repeat with 2 g of crushed bicarb powder; repeat each condition 3 times and average.