A hypothesis is a clear, testable prediction about the relationship between two or more variables. It turns your research question into a statement you can support or reject with evidence. This guide shows you how to write a hypothesis in six steps, with examples of each type and the mistakes students most often make.
Key takeaways
- A good hypothesis is specific, testable and based on existing research.
- Name your independent variable (what changes) and dependent variable (what you measure).
- In statistical testing you write a null hypothesis (no effect) and an alternative hypothesis (an effect).
- You never “prove” a hypothesis; you find evidence that supports it or you fail to reject the null.
What is a hypothesis?
A research question asks something open, such as “Does caffeine affect reaction time?” A hypothesis predicts the answer: “Students who drink coffee before a test will react faster than students who don’t.” The prediction must be testable, meaning you could collect data that shows it is wrong.
Hypotheses are most common in quantitative research, such as experiments and surveys. Qualitative studies, which explore experiences and meanings, often use research questions instead.
Variables in a hypothesis
| Variable | What it means | Example |
|---|---|---|
| Independent variable | The factor you change or compare | Caffeine intake (200 mg or none) |
| Dependent variable | The outcome you measure | Reaction time in milliseconds |
| Control variables | Factors you keep the same so they don’t affect the result | Time of day, sleep the night before, test used |
Types of hypothesis
- Null hypothesis (H0): there is no effect or relationship. “Caffeine intake has no effect on reaction time.”
- Alternative hypothesis (H1): there is an effect or relationship. “Caffeine intake affects reaction time.”
- Directional hypothesis: predicts which way the effect goes. “Caffeine reduces reaction time.”
- Non-directional hypothesis: predicts a difference without saying which way. “Reaction time differs between students who have caffeine and those who don’t.”
- Simple and complex hypotheses: a simple hypothesis has one independent and one dependent variable; a complex one has more than one of either.
How to write a hypothesis in six steps
- Ask a focused research question. Make it specific enough to test within your time and resources.
- Read the existing research. Previous studies tell you what to expect and give your prediction a reason.
- Identify your variables. Decide what you will change or compare and what you will measure.
- Write your prediction. State the expected relationship in one clear sentence. An “if… then…” structure can help: “If students consume 200 mg of caffeine, then their reaction times will be faster.”
- Make it specific and testable. Define the group, the measure and the comparison. Replace vague words like “better” with something measurable.
- Write the null and alternative versions. If you will use a statistical test, state both hypotheses so your analysis has a clear decision rule.
Worked example
- Research question: Does caffeine affect reaction time in university students?
- Alternative hypothesis (H1): University students who consume 200 mg of caffeine 30 minutes before a reaction-time test will respond faster than students who consume none.
- Null hypothesis (H0): There is no difference in reaction time between university students who consume 200 mg of caffeine and those who consume none.
What makes a good hypothesis?
- It is based on theory or previous research, not a guess
- It names the variables and the population clearly
- It can be tested with the data you are able to collect
- It could be shown to be wrong
- It is written as a statement, not a question
Common mistakes to avoid
- Writing a question instead of a prediction
- Using vague terms such as “improves” without saying how you will measure it
- Testing too many variables at once
- Saying the results “prove” the hypothesis
- Changing the hypothesis after seeing the results without explaining why
Hypothesis writing: FAQs
What is the difference between a hypothesis and a research question?
A research question asks what you want to find out. A hypothesis predicts the answer in a way you can test.
Do all dissertations need a hypothesis?
No. Experimental and statistical studies usually do. Qualitative and exploratory studies often use research questions or objectives instead. Follow your department’s guidance.
Can a hypothesis be wrong?
Yes. Results that fail to support your hypothesis are still valid findings, as long as your method was sound. Discuss why the results differed from your prediction.
Should I write a null hypothesis?
If you are using a statistical test, yes. The test assesses whether the evidence is strong enough to reject the null hypothesis.
Planning a dissertation? Read 12 common dissertation mistakes and how to avoid them. If you want feedback on your own research plan, our dissertation coaching and proofreading services work within university rules and never write assessed work for you.
