Variables in Change Investigations
In 2018, CSIRO researchers changed just 1 variable — temperature — and doubled the speed of a fertiliser reaction, proving that controlling variables is the key to understanding cause and effect.
Printable Worksheets
Print or save as PDF — or build a custom worksheet from any module's questions.
● Know
- Experiments have three main variable types: independent, dependent and controlled.
- Only one thing should change at a time — the independent variable.
- Controlled variables are kept the same so they can't sneak in and affect results.
● Understand
- Changing more than one variable makes results impossible to interpret.
- A 'fair test' is one where only the IV is varied; everything else is controlled.
- Results need to be both reliable (repeatable) and valid (measuring the right thing).
● Can do
- Identify the IV, DV and CVs in any experiment.
- Design a fair test for a given research question.
- Spot when an experiment isn't fair and explain why.
You run 2 identical bread slices side by side for a week: one sits in a warm pantry and one sits in the fridge. After 7 days the pantry slice is green with mould and the fridge slice is perfectly fine. Something you changed — temperature — produced a visible difference in outcome. Now the question is: was it really the temperature, or could it have been the light, the humidity, the bread brand? To be sure, every other condition must be kept identical. Every experiment has three kinds of variable. The independent variable (IV) is the one you deliberately change to test its effect. The dependent variable (DV) is what you measure or observe as a result. Everything else that could influence the outcome must stay the same — these are the controlled variables (CVs).
Scientists use this framework to isolate cause and effect. If you only change the IV and keep all CVs constant, then any change in the DV can be fairly blamed on the IV. This is the skeleton of every fair test. Without controlled variables, you cannot be sure what caused the observed effect, which makes your conclusion unreliable.
Remember: the IV is what you change, the DV is what you measure, and the CVs are what you lock down.
A student asks, “Does temperature affect how fast sugar dissolves?” The IV is the temperature of the water. The DV is the time taken for the sugar to fully disappear. The CVs include the amount of sugar, the volume of water, the type of sugar, and the stirring rate.
At CSIRO, agricultural scientists test new fertilisers by changing only the fertiliser type (IV) while keeping soil type, water, and sunlight (CVs) identical across plots. This ensures any difference in crop yield (DV) is caused by the fertiliser, not by hidden factors.
Many students think that anything you measure in an experiment is the independent variable. In fact, what you measure is the dependent variable. The independent variable is the factor you deliberately change.
A fair test changes only one variable at a time. If you change two things at once — for example, both the temperature and the stirring rate — you cannot tell which one caused the difference in results. The experiment loses its power to show cause and effect.
Controlled variables solve this problem. By keeping every other factor identical, you isolate the effect of the independent variable. This is why scientists write long lists of controls before they begin any investigation. A well-controlled experiment gives you confidence that the change you measured is real and not just noise.
Imagine two plant pots: one gets more light and more water, the other gets less of both. If the first plant grows taller, you cannot tell whether light or water was responsible. To make it a fair test, you should change only the light while keeping water, soil, and pot size the same.
The Australian Sports Anti-Doping Authority (ASADA) runs drug tests under strictly controlled conditions. The time of day, food intake, and sample handling are all kept identical so that any detected substance can be fairly linked to the athlete, not to random variation.
Some students believe that doing more repeats fixes a poorly controlled experiment. Repeats improve reliability, but they cannot fix a confounded variable. If two factors change at once, every repeat will still be unfair.
A hypothesis is a testable prediction that links the independent and dependent variables. It is not a random guess; it is based on prior knowledge or observations. A good hypothesis uses the format: “If [IV] changes, then [DV] will… because…”.
Writing a clear hypothesis forces you to identify your variables before you start. It also gives you a way to evaluate your results at the end. If the data supports the prediction, the hypothesis is supported — though not absolutely proven. If the data does not support it, you revise your thinking and test a new idea. Strong hypotheses are specific, measurable, and grounded in the science you have already learned.
“If the concentration of acid increases, then the reaction time with magnesium will decrease, because there are more acid particles available to collide with the metal surface.” This hypothesis names the IV (acid concentration), the DV (reaction time), and the reason (collision theory).
CSIRO climate scientists form hypotheses about how rising temperatures affect rainfall patterns across Australia. They then test these predictions using decades of Bureau of Meteorology data, refining their models as new evidence emerges.
Students often say a hypothesis is “just a guess.” A guess has no reasoning behind it. A hypothesis is an educated prediction backed by scientific understanding and must be testable through experiment.
Even experienced scientists make mistakes with variables. The most common error is changing more than one independent variable at a time. Another is confusing the independent variable with the dependent variable — measuring the thing you meant to change, or changing the thing you meant to measure. A third is forgetting to list obvious controlled variables such as room temperature, the volume of liquid, or the starting mass.
These mistakes break the fair-test rule and make your conclusion unreliable. The best defence is a simple checklist: identify your IV, DV, and every CV before you touch any equipment. Write them down and review them with a partner. This small step saves hours of wasted work and prevents you from drawing false conclusions.
A student investigates how temperature affects dissolving time but uses 100 mL of water at 20°C, 50 mL at 40°C, and 200 mL at 60°C. They changed both temperature and volume, so they cannot tell which factor caused the difference. A fair test keeps the volume identical.
At ANSTO, researchers running nuclear experiments must control temperature, pressure, and radiation shielding with extreme precision. Even tiny uncontrolled variations could invalidate an entire study, so every variable is logged and monitored automatically.
Many students skip listing “obvious” controls like room temperature because they think everyone knows they should stay the same. In science, nothing is assumed. Every controlled variable must be recorded so others can repeat your experiment exactly.
Here's a student's working. One line has an error — click it.
- A student investigates how temperature affects dissolving time.
- They use 100 mL of water at 20°C, 50 mL at 40°C, and 200 mL at 60°C.
- They conclude that higher temperature always makes dissolving faster.
Designing a fair test is the foundation of good science. Start by choosing a clear, testable question. Then identify your independent variable, your dependent variable, and every controlled variable you can think of. Write a hypothesis that links the IV and DV with a reason.
Next, plan your method so that only the IV changes. List your equipment, describe your steps in order, and decide how many repeats you need. Finally, sketch a results table with columns for the IV, each repeat, and the mean. A well-planned investigation tells a story that others can trust and repeat.
A student wants to find out if acid concentration affects how quickly magnesium reacts. The IV is concentration, the DV is reaction time, and the CVs include the mass of magnesium, the volume of acid, and the temperature. Keeping these controls identical ensures any change in rate is caused by concentration alone.
BlueScope Steel controls variables such as ore composition, furnace temperature, and oxygen flow when refining iron. Small changes in any of these can alter the quality of the final steel, so strict control is essential for consistent production.
Some students think you can skip controlled variables if the result seems obvious. Science does not work on “seems obvious.” Without controls, you have no evidence that your IV caused the change, and your conclusion cannot be trusted.
At the start of this lesson, you read about CSIRO researchers who doubled a reaction's speed just by changing temperature — and discovered that changing one variable at a time is the only way to know what actually caused the result.
Now that you've practised identifying and controlling variables, look back at your initial ideas. Did you already understand why changing one thing at a time matters, or did the lesson shift your thinking about what makes an experiment fair?
1. In an experiment to see how the amount of bicarb soda affects the height of a volcano reaction, what is the independent variable?
2. What is a controlled variable?
3. Why is a fair test important in science?
4. In an investigation of how surface area affects rusting, which would be a controlled variable?
5. Which statement describes a hypothesis?