Definition · Plain-language
Research hypothesis
A research hypothesis is a clear, specific, testable and falsifiable predictive statement about the relationship between two or more variables, usually derived from theory or prior evidence.
The step most authors miss
Doing CRediT right? Don’t stop at the statement.
A CRediT statement credits you inside one paper. The recognition CRediT was built for happens when those roles are tied to you, persistently. Sign in with your ORCID — free — and claim your CRediT contributions on casrai.org, the home of the standard. They become a verified, portable part of your identity, not a line that disappears into one PDF.
Free: claim your contributions, then export a journal-ready CRediT statement, schema.org structured data, JATS XML, CSV or BibTeX — and preview your public profile. A membership publishes that profile publicly and verifies the journals you serve.
What makes a hypothesis a hypothesis
A research hypothesis is not a vague hunch but a disciplined prediction. To qualify, it must be specific (it names the variables and the population), testable (you can collect data that bear on it), and falsifiable (there is a conceivable result that would prove it wrong). It usually grows out of theory or prior findings rather than appearing from nowhere, which is what distinguishes a reasoned hypothesis from a guess. A useful template is the "if–then" form: if the independent variable changes, then the dependent variable will change in a stated way.
Directional and non-directional
Like its statistical counterpart, a research hypothesis can be directional or non-directional. A directional hypothesis predicts the way the relationship runs — for example, that students who sleep more before an exam score higher. A non-directional hypothesis predicts only that a relationship exists, without committing to its direction, which is appropriate when theory is unsettled. Choosing between them depends on how much prior evidence supports a specific direction; either way, the prediction should be fixed before data collection, ideally through preregistration, to prevent it being quietly rewritten to fit the results.
Hypothesis versus research question
A research question asks what you want to find out ("Does sleep affect exam performance?"), whereas a research hypothesis commits to a predicted answer ("More sleep before an exam improves performance"). Exploratory, qualitative, or early-stage work often proceeds from questions rather than hypotheses, because there is not yet enough theory to justify a confident prediction. The hypothesis is the sharper tool: by staking a claim that data can refute, it sets up the formal null and alternative hypotheses used in statistical testing.
Key facts
At a glance
- Definition: a testable, falsifiable prediction about a relationship between variables
- Key traits: specific, testable, falsifiable, grounded in theory
- Common form: "if [independent variable], then [dependent variable]"
- Types: directional (one-tailed) or non-directional (two-tailed)
- Not the same as: a research question (asks) or a statistical hypothesis (H0 / H1)
- Why it matters: turns a question into a claim that evidence can refute
Common misconceptions
What people often get wrong
Often heard: A research hypothesis is just an educated guess about anything.
Actually: A guess is not enough. A hypothesis must be specific, testable and falsifiable, and is normally derived from theory or prior evidence rather than plucked from intuition.
Often heard: A research hypothesis and a research question are the same thing.
Actually: A research question asks what you want to know; a hypothesis predicts the answer. Exploratory work often uses questions, while hypothesis-driven studies stake a refutable claim.
Often heard: A good hypothesis is one that cannot be proven wrong.
Actually: The opposite is true. Falsifiability is essential — a hypothesis earns scientific value precisely because there is an observable result that would refute it.
