What is a Hypothesis Statement?
A hypothesis statement is a specific, testable prediction about the relationship between two or more variables, written before the data are collected. "Students who sleep at least eight hours will recall more words than students who sleep five hours" is a hypothesis: it names the variables, the expected relationship, and something you could measure.
Hypothesis vs Research Question vs Aim
- Aim: what the study sets out to do ("to investigate the effect of sleep on memory").
- Research question: what it asks ("Does sleep duration affect recall?").
- Hypothesis: the predicted answer, precise enough to test.
Hypotheses are expected in experimental and other confirmatory quantitative research. Exploratory, descriptive and qualitative studies usually work from research questions instead.
Types of Hypothesis
- Null hypothesis (H₀): no effect or difference ("Sleep duration has no effect on recall").
- Alternative or experimental hypothesis (H₁): there is an effect or difference.
- Directional (one-tailed): predicts the direction ("more sleep, higher recall"), used when theory or earlier findings support it.
- Non-directional (two-tailed): predicts a difference without its direction ("recall will differ").
- Associative or causal: variables are related, or one causes changes in the other.
How to Write a Hypothesis
- Start from the research question and what the literature already shows.
- Identify the independent and dependent variables.
- Operationalise them: say how each will be measured ("recall = number of words remembered from a list of 20").
- State the predicted relationship, with direction if justified.
- Write the matching null hypothesis.
Sentence Starters
- "If ... then ...": "If students sleep eight hours, then they will recall more words than if they sleep five."
- "There will be a significant difference in ... between ... and ..."
- "... will be positively (or negatively) correlated with ..."
What Makes a Good Hypothesis
- Testable and falsifiable: some result could show it wrong.
- Specific: clear variables, population and direction.
- Grounded: based on theory or earlier findings, not a guess.
- Simple: one prediction per hypothesis.
After the Study
A hypothesis is never proved. Researchers say it was "supported" or "not supported" by the data; statistically, the test rejects, or fails to reject, the null hypothesis. A hypothesis that is not supported is still a valid, reportable result. It shapes the whole research design, especially in experimental studies.
Full guide: Hypothesis Examples: Science, Null and Published Studies
Use case: AI for research paper writing