Learning how to write a hypothesis is an essential skill for UK university students. A hypothesis is a clear, testable prediction about the relationship between variables. It turns a research question into something you can test with data, and it underpins most quantitative research. This complete UK guide explains what a hypothesis is, the difference between null and alternative hypotheses, how to write a testable one, and how hypotheses relate to research questions and variables.
How to write a hypothesis: Step-by-Step Guide
What Is a Hypothesis?
A hypothesis is a testable statement predicting a relationship between variables — for example, that one factor affects another. It is specific and falsifiable: you can gather data to support or reject it.
For further guidance on how to write a hypothesis, visit the Prospects UK dissertation guide — a trusted resource for UK students and graduates.
Null vs Alternative Hypotheses
Quantitative research uses two. The null hypothesis (H0) states there is no effect or relationship; the alternative hypothesis (H1) states there is. Statistical testing assesses whether you can reject the null in favour of the alternative.
Independent and Dependent Variables
A good hypothesis names the variables clearly. The independent variable is what you change or compare; the dependent variable is what you measure. The hypothesis predicts how the first affects the second.
Writing a Testable Hypothesis
✓ State a clear prediction.
✓ Name the variables.
✓ Make it specific and measurable.
✓ Ensure it is falsifiable.
A vague or untestable statement is not a hypothesis.
Hypothesis vs Research Question
A research question asks what you will investigate; a hypothesis predicts the answer in testable form. Quantitative studies often use hypotheses; qualitative studies usually use questions. See our research question guide.
Common Mistakes and Tips
✓ A vague, untestable statement.
✓ No clear variables.
✓ Predicting more than one relationship at once.
✓ Confusing it with a question. Tip: make it specific, measurable and falsifiable, with clear variables.
How Projectsdeal Helps
Dissertation writing service, research paper service and assignment help.
Directional vs. Non-Directional Hypotheses
An important distinction in hypothesis formulation is between directional and non-directional hypotheses. A directional (one-tailed) hypothesis predicts the direction of the expected effect: “Higher levels of exercise will be associated with lower levels of depression” (specifying that higher exercise leads to lower, not just different, depression). A non-directional (two-tailed) hypothesis predicts that there will be a difference or relationship but does not specify the direction: “There will be a significant relationship between exercise levels and depression.”
Directional hypotheses are appropriate when prior theory or evidence clearly predicts the direction of the effect. Non-directional hypotheses are more conservative and appropriate when the direction is uncertain. For statistical testing: directional hypotheses use one-tailed tests; non-directional hypotheses use two-tailed tests. Most UK dissertation supervisors recommend two-tailed testing unless there is strong theoretical justification for a directional prediction.
Hypotheses in Different Research Traditions
Hypotheses are primarily a feature of quantitative, deductive research designs. In the positivist tradition, a dissertation begins with a theoretical framework that generates specific, testable hypotheses, which are then tested against data. The results either support or fail to support (note: not “prove” or “disprove”) the hypothesis.
In qualitative and inductive research, formal hypotheses are typically not used. Instead, qualitative dissertations formulate research questions (“How do…?”, “What are…?”) rather than testable predictions. Attempting to force qualitative research into a hypothesis-testing framework is a common and significant methodological error.
Frequently Asked Questions
What is a hypothesis?
A clear, testable prediction about the relationship between variables.
What is the null hypothesis?
A statement that there is no effect or relationship between the variables.
What is the alternative hypothesis?
A statement that there is an effect or relationship between the variables.
What is an independent variable?
The factor you change or compare in a study.
What is a dependent variable?
The outcome you measure, which the independent variable may affect.
How do I write a testable hypothesis?
State a specific, measurable, falsifiable prediction naming the variables.
What is the difference between a hypothesis and a research question?
A question asks what you will investigate; a hypothesis predicts the answer in testable form.
Do all studies need a hypothesis?
No — quantitative studies often do; qualitative studies usually use research questions.
Related Study Guides
How to Write a Research Question • How to Write a Methodology • How to Write a Dissertation • How to Write a Lab Report
UK students who master how to write a hypothesis gain a significant advantage in their academic career. Whether you are in your first year or final year, understanding how to write a hypothesis thoroughly will improve your overall academic performance and help you achieve better grades.
Types of Hypothesis in Academic Research
Understanding the different types of hypothesis used in academic research helps you formulate the most appropriate type for your study and articulate it correctly in your methodology chapter.
A directional (or one-tailed) hypothesis predicts not only that a relationship exists but the specific direction of that relationship. For example: “Students who receive structured writing feedback will achieve higher essay grades than students who receive no feedback.” Directional hypotheses are appropriate when existing theory or evidence gives you a strong basis for predicting the direction of the effect.
A non-directional (or two-tailed) hypothesis predicts that a relationship exists but does not specify its direction. For example: “There will be a significant difference in essay grades between students who receive structured writing feedback and those who do not.” Non-directional hypotheses are appropriate when the existing evidence is inconclusive or contradictory about the expected direction.
The null hypothesis (H₀) states that there is no relationship or difference between the variables being studied. Statistical hypothesis testing formally tests whether the observed data are consistent with the null hypothesis or whether they justify rejecting it in favour of the alternative hypothesis (H₁). In your dissertation, you will typically state both the null hypothesis and the alternative hypothesis for each research question that is addressed statistically.
A research question rather than a formal hypothesis is often more appropriate for qualitative or exploratory research. Research questions are open-ended and do not predict a specific outcome; they are appropriate when you are exploring a phenomenon rather than testing a predictable relationship. In mixed-methods or multi-strand studies, you might use both research questions (for the qualitative strand) and hypotheses (for the quantitative strand).
Formulating and Testing Your Hypothesis
A well-formulated hypothesis should be: clearly stated in simple, unambiguous language; specific about the variables being studied and the predicted relationship between them; testable using the data available to you; grounded in existing theory or evidence; and appropriately scoped to what your research design can actually test.
Before finalising your hypothesis, conduct a preliminary review of the existing literature to ensure it is plausible and adequately supported. A hypothesis that is contradicted by the bulk of existing evidence without a strong theoretical reason requires justification in your methodology. A hypothesis that simply restates an already well-established finding is unlikely to produce a meaningful contribution.
When testing your hypothesis statistically, the choice of statistical test depends on the type of variables involved and the distribution of your data. Parametric tests (such as t-tests, ANOVA, and Pearson correlation) assume that the data are normally distributed; non-parametric equivalents (such as the Mann-Whitney U test, Kruskal-Wallis test, and Spearman’s rho) are used when this assumption is not met. Your methodology chapter should specify and justify your choice of statistical test for each hypothesis.
Interpreting the results of hypothesis testing requires care. A statistically significant result (p < .05) means that the probability of obtaining the observed data if the null hypothesis were true is less than 5%—it does not prove that the alternative hypothesis is true. Effect size measures (such as Cohen’s d or partial eta squared) are also important, as statistical significance without practical significance is a common limitation of studies with large sample sizes. If you need expert guidance on hypothesis formulation, statistical testing, or dissertation methodology writing, professional academic support from research specialists can help you produce a statistically sound and well-written methodology chapter.
Need Expert Academic Help?
ProjectsDeal provides trusted dissertation, thesis, and essay writing support for UK university students. Get matched with a specialist in your subject area.
How To Write A Hypothesis: Key Insights for UK Students
UK students who master how to write a hypothesis gain a significant advantage. Understanding how to write a hypothesis thoroughly improves academic performance and helps achieve better grades at UK universities.
When developing skills in how to write a hypothesis, consistency is key. Practise regularly, seek tutor feedback, and use academic resources to strengthen your knowledge of how to write a hypothesis.
For further guidance on how to write a hypothesis, visit the Prospects UK dissertation guide — a trusted resource for UK students.