Open the Build a testable hypothesis room
This article opens a matching guided lab. Follow the on-screen prompts, make a prediction, and connect the model result back to this topic.
01 // What a hypothesis is
A hypothesis is a specific, testable prediction about what will happen in an experiment — not just any guess. It names the variable you plan to change (the independent variable), the variable you plan to measure (the dependent variable), and it explains why you expect that result based on what you already know.
A good hypothesis is falsifiable: there has to be a possible result that would prove it wrong. "Plants are affected by light" is too vague to test. "If a bean plant gets 12 hours of light a day instead of 6, then it will grow taller after two weeks, because more light gives the plant more energy for photosynthesis" is specific enough to test and specific enough to fail.
02 // The if-then-because template
Fill in three blanks: what you change, what you expect to happen, and the reason you expect it.
| If (independent variable) | a bean plant gets 12 hours of light per day instead of 6 |
| Then (dependent variable + direction) | its height after two weeks will be greater |
| Because (the reasoning) | more light gives the plant more energy for photosynthesis |
Formula If [I change this], then [this will happen], because [this is why].
03 // What makes a hypothesis weak
- No predicted direction. "Light will affect growth" does not say taller, shorter, faster, or slower — it can never be wrong, so it is not testable.
- Two variables changed at once. "More light and more water will make it grow taller" — if it grows taller, you cannot tell which one caused it.
- No reasoning. The "because" clause forces you to connect the prediction to real science, and it is usually the part a rubric checks first.
- A hypothesis is not the same as a research question. The question is "does more light make plants grow taller?" The hypothesis is the specific if-then-because answer you predict.
04 // Make each part of an if–then–because statement testable
The 'if' names the condition or independent variable the investigator will change. The 'then' predicts a measurable response, such as plant height after two weeks or time for a pendulum to complete ten swings. The 'because' gives a scientific reason for the predicted relationship. A sentence such as 'If plants get more light, then they grow better because light is good' is vague: it does not say how much light, what 'better' means, or which process explains the prediction.
A stronger version might say, 'If identical seedlings receive six rather than three hours of light each day, then their dry mass after four weeks will be greater because light supplies energy for photosynthesis.' That prediction identifies a measurable outcome and a plausible mechanism. It still needs a fair test, repeated plants, and controls for water, soil, species, and temperature before the conclusion can be trusted.
05 // Use background research to justify a prediction
A hypothesis is informed by observations, prior knowledge, and research; it is not a random guess. Before writing one, identify what is already known and what remains uncertain. Use sources appropriate to the topic, then explain the scientific connection in your own words. A citation supports the rationale but does not replace the student's explanation of why the factor should affect the measurement.
Some questions support a directional prediction, such as expecting a longer pendulum to have a longer period. Other questions are exploratory because the direction is uncertain. In that case, state the expected relationship without inventing certainty, or write a non-directional hypothesis that predicts a difference. The key requirement is that the prediction can be checked against data and that the variables are clearly identified.
06 // A hypothesis can be supported, not proven
When results match a prediction, say they support it under the tested conditions. One classroom experiment cannot prove that the explanation is true in every setting. A small sample, measurement error, or an uncontrolled factor may create the same pattern. When results do not match, report that honestly and inspect the data and method before revising the explanation.
Do not rewrite the original hypothesis after seeing the results and pretend it was the prediction all along. Keep a dated copy, then write a revised hypothesis for a new test. This preserves the difference between prediction and explanation. It also helps students learn that unexpected results can be scientifically useful, especially when they reveal a missing variable or a weak assumption.
07 // Check the hypothesis before collecting data
Underline the condition in the 'if' clause, circle the measurable outcome in the 'then' clause, and mark the scientific reason in the 'because' clause. If one part is missing, revise it before beginning. Ask whether another student could tell exactly what to change and what to record. If not, operationally define the variable, response, duration, and units.
A useful hypothesis is concise, but the experiment plan carries the detail. It should not include multiple unrelated factors, a desired outcome, or a claim about what the class hopes to see. Write the prediction before testing, preserve it with the data, and return to it when interpreting results. That sequence makes the comparison between expectation and evidence meaningful.
08 // Keep the prediction separate from the experiment’s result
Write the hypothesis before running the test and save the original wording in the notebook. Afterward, compare the observed pattern with the prediction instead of rewriting the hypothesis to make it appear correct. If the data do not support the prediction, that is still a useful outcome: inspect the procedure, consider another explanation, and design a new test rather than labeling the experiment a failure.
A strong hypothesis names the population or sample, the condition being changed, the measured outcome, and—when evidence allows—the predicted direction. Avoid claiming a mechanism as fact when the activity cannot observe it directly. For example, a height prediction can be tested in a short plant study, while a claim about a cellular pathway may require instruments and controls beyond a classroom setup.
09 // Questions people ask
What is the difference between a hypothesis and a theory?
A hypothesis is a single testable prediction for one experiment. A theory is a broad, well-tested explanation supported by many experiments and observations over time, like cell theory or the theory of plate tectonics.
Does a hypothesis have to turn out correct?
No. An experiment that disproves a hypothesis is still a successful experiment, as long as it was a fair test. Many important discoveries started with a hypothesis that turned out to be wrong.
What is a null hypothesis?
The null hypothesis predicts no effect or no difference — for example, "the amount of light will not change plant height." Scientists often test against the null hypothesis to see if their data shows a real effect or just random variation.