Working scientifically
How to write a testable scientific prediction
Connect a deliberate variable change to an expected measurable result and explain why. A useful prediction guides the method and can be checked against evidence.
Published

Start with the question being investigated
A prediction belongs to a specific enquiry. First identify the independent variable that will be changed and the dependent variable that will be measured. If those variables are vague, the prediction will also be vague.
For a question about how light intensity affects plant growth, light intensity is the planned change and a defined measure of growth is the outcome. “The plant will grow better” is difficult to test until better is replaced by height change, leaf number, dry mass or another agreed measure. Those measures describe different aspects of growth and cannot be treated as interchangeable.
The plant-investigation scene is AI-generated and portrays fictional Indonesian people. It does not represent our staff, a client, an actual teaching session or a school partnership.
Write change, outcome and reason
A useful structure is: if the independent variable changes in this way, then the dependent variable is expected to change in that way, because of this scientific idea. The wording can vary, but all three parts should be visible.
A complete practice prediction could read: within the light-limited range being tested, matched groups receiving more light are expected to gain more dry mass over fourteen days because light supports photosynthesis. It names an outcome and timescale, limits the claim and gives a reason; it does not report a result that has already happened.
Shoot height alone is not a dependable measure of biomass gain. A seedling in low light can become taller and spindlier. Use the stated outcome consistently, and choose a method with the teacher that can actually measure it.
The reason matters. It distinguishes a scientific prediction from an unsupported guess and reveals the model or prior knowledge being tested. The explanation should match the learner’s course level and avoid claiming certainty before evidence has been collected.
| Part | Question to answer | Plant example |
|---|---|---|
| Variable change | What will be deliberately changed? | Increase light intensity |
| Measurable outcome | What result will be recorded? | Greater increase in mean dry mass in matched groups over fourteen days |
| Scientific reason | Why might that pattern occur? | Within a light-limited range, more photosynthesis may increase biomass when other factors are not limiting |
| Boundary | What limits the claim? | Only within the tested range and conditions |
Make the method capable of testing the claim
Check that the investigation can produce the evidence named in the prediction. The dependent variable needs a measurement method and time point. Relevant conditions should be controlled or recorded so the effect of the planned change can be interpreted.
A prediction that compares temperature groups needs defined temperatures, the same outcome measure and a practical range. A prediction about concentration needs units and a clear direction. Precise variables make later graphs and conclusions more meaningful.
Avoid promising a perfect result

This AI-generated image illustrates enquiry planning with fictional people; it is not an actual tutoring session.
Scientific predictions are not guarantees. Biological variation, measurement uncertainty and uncontrolled conditions can produce overlap or an unexpected pattern. Use language such as “is expected to” rather than “will definitely”.
If the system may have a limit or optimum, state the predicted range. An enzyme may respond differently beyond its optimum, and plant growth may stop increasing when another factor becomes limiting.
Review the prediction after collecting data
Compare the observed pattern with the predicted direction and reason. State whether the evidence supports the prediction within the tested conditions. Do not rewrite the original prediction to match the result.
An unsupported prediction can still lead to valuable science. Look for limitations, alternative explanations and a next test. Keep the distinction between the data observed, the conclusion drawn and a possible explanation.
Get help planning a fair scientific enquiry
If a learner knows the topic but struggles to connect variables, predictions, methods and conclusions, bring the exact course requirements and a recent practical task to a science tutoring discussion.
Families seeking in-person support in west-central Bali can review the Tabanan tutoring area page. For identifying variables first, see independent, dependent and control variables explained.
Frequently asked questions
Is a hypothesis the same as a prediction?
Courses use the terms in different ways. A prediction states the expected measurable outcome for a particular test; a hypothesis often gives a broader proposed explanation. Follow the terminology required by the learner’s syllabus.
Can a scientific prediction be wrong?
Yes. A prediction is evaluated against evidence. An unexpected result should be reported honestly and examined for limitations, alternative explanations and useful next tests.
Related
Message us on WhatsApp: +62 858 6969 6869 · info@privatetutoringbali.com · We reply within one working day.