Variables & Method

In science, experiments are used to investigate the relationship between two measurable factors called variables. To properly test a hypothesis, a scientist changes one variable and observes how it affects another. 

The variable that the scientist changes on purpose is called the independent variable. It must have a suitable range of values (for example, different temperatures or concentrations) so that any pattern or trend can be clearly seen in the results. The variable that is measured is called the dependent variable. It changes because of what happens to the independent variable. 

To make sure that the results are trustworthy, all other factors that could affect the outcome must be kept the same. These are called control variables. There can be many control variables in an experiment, and identifying the most important ones is a key part of planning. If relevant control variables are not held constant, the results may be biased or invalid. A valid experiment — also known as a fair test — is one where the only variable that changes is the independent variable.

The variable that the scientist changes on purpose is called the independent variable. Only one independent variable is used per experiment. It must have a suitable range of values, wide enough to show any patterns or trends in the results — typically at least four different values. The range should be appropriate for what is being tested (for example, temperatures that are suitable for an organism being studied).

Choosing and testing the range often happens during trialling. Planning must include how the independent variable will be altered and what equipment is needed. The independent variable must only change due to the planned adjustments — not by accident or due to uncontrolled conditions.

The variable that is measured is called the dependent variable. It changes because of what happens to the independent variable. 

Again, only one dependent variable is used per experiment. It must be measured accurately using correct scientific units, and the method of measurement must be clearly explained so the investigation is repeatable and reliable.

It must be very clear exactly what is being measured. For example, in plant studies, it is important not to mix up growth with germination because they are different outcomes. In animal behaviour studies, scientists must measure behaviour in an objective way — not based on assumptions about what the animal likes or feels — to avoid bias.

To make sure that the results are trustworthy, all other factors that could affect the outcome must be kept the same.

Control variables are all the other variables — apart from the independent and dependent variables — that could affect the results. They must be identified and kept constant to ensure a fair test. There may be many control variables in a single experiment. It is important to focus on controlling the ones most likely to significantly influence the outcome.

Examples of control variables include:

• Environmental factors like temperature, light intensity, or soil moisture

• Amounts of materials like substrate or chemicals

• Biological factors such as age or gender of animals

If control variables are not properly managed or described, the experiment can become biased and the results invalid. Some factors may be outside the investigator’s complete control; these become assumptions, which must still be acknowledged and justified.

A method is a detailed, step-by-step description of how an investigation will be carried out. It explains exactly how the independent variable will be changed, how the dependent variable will be measured, what equipment will be used, how control variables will be kept the same, and how many repeats will be performed.

A good method is detailed enough that another person could repeat the investigation and obtain similar results without needing additional explanation. Methods often include diagrams of the experimental setup, which must be accurately drawn and labelled. A well-written method ensures the experiment is valid (measures what it is meant to measure) and reliable (can be repeated to produce similar results).

To design a valid method, the experiment must measure what it is meant to measure. So the method must be a fair test, meaning:

• Only the independent variable is changed

• The dependent variable is measured

• Control variables are kept constant, to prevent them affecting the results.

So the first step in designing a valid method is to make clear decisions about how the fair test will be carried out. You must first plan how the independent variable will be changed and select an appropriate range — usually at least four values spread from low to high. You then decide exactly how the dependent variable will be measured, including suitable scientific units (such as seconds, centimetres, or degrees Celsius). Then, you must identify control variables and describe how each one will be kept constant, such as controlling temperature, light, pH, or the amount of material used.

You will also need to consider how many repeats or trials will be carried out. Repeating each measurement makes your results more reliable, because a single result could be a mistake or affected by random variation. These multiple trials and measurements for the same variable allow descriptive statistics (e.g. the mean and standard deviation) to be calculated. The more trials and data gathered, the more confident you can be of the final results. 

Scientists must also think about the amount of time the full investigation will take and whether the equipment available is appropriate. During planning, assumptions may be made about things that cannot be fully controlled but are unlikely to affect the results significantly. Assumptions are things you assume to be true, but do not test. They should be based on sound scientific principles, and should be acknowledged and justified. In fact, you should be able to justify all the decisions you make in your experimental methodology.

Trialling (testing) your method before doing the full experiment is an important part of scientific practice. It allows you to check whether your plans are workable, whether the range of the independent variable is suitable, and whether your measurements are clear and consistent. Trials may show that parts of the method need to be changed or improved — this is normal and expected in science. All changes and reasons for changes should be recorded in your logbook. Persistence is a major part of designing good investigations!

If possible, investigations should be carried out multiple times, so that you can be sure that the data collected is reliable. All of these considerations in designing a valid method apply when evaluating the design and methods of the work of others. 

A repeatable method must include enough detail that someone else can follow it exactly the same way. It should be written using a logical structure, such as numbered steps or well-organised paragraphs. The method is also normally written in past tense and passive voice (e.g., “The solution was heated…”). Details must be quantitative, meaning you include exact amounts or conditions such as how long, how much, how many, how hot, or what concentration - with units!

To ensure consistency, you must clearly state:

• What will be measured

• When it will be measured

• How it will be measured

• What units will be used

• How the data will be recorded and processed

Your method must specify the range of values for the independent variable and describe how the dependent variable will be accurately measured. It should also state how each control variable will be kept the same. If diagrams are included, they must be labelled. The diagrams should also support your written instructions, not replace them. When writing your final report, reviewing and editing your method helps ensure clarity, accuracy, and high scientific quality. Ideally, someone unfamiliar with your work should be able to follow your method and produce the same pattern of results — this is the ultimate test of a strong scientific investigation.