How to Write a Research Hypothesis

A research hypothesis is a specific, testable prediction about the relationship between two or more variables — not a topic, not a question, and not a hope about what the data will show. Writing one that actually holds up under scrutiny means making it falsifiable, operationalizing every variable it names, and choosing the right structural form (null, alternative, directional, or non-directional) before a single data point is collected. This guide walks through what makes a hypothesis testable, the standard formats reviewers and statisticians expect, how a hypothesis differs from the research question and theoretical framework that surround it, the mistakes that get hypotheses flagged in review, and a practical template for writing one.

What makes a hypothesis testable and falsifiable

The philosopher of science Karl Popper argued that what separates a scientific claim from a non-scientific one is not that it can be verified, but that it can, in principle, be shown to be false. A hypothesis that no conceivable observation could ever contradict is not testable, no matter how reasonable it sounds. This is the falsifiability criterion, and it is the first filter any hypothesis needs to pass.

In practice, falsifiability comes down to two requirements:

Every variable must be operationalized. “Improves wellbeing” is not testable; “increases scores on the WHO-5 Well-Being Index” is. A variable is operationalized when it is defined in terms of how it will actually be measured or manipulated, not just named in the abstract. See operationalizing variables for the general definition.
The predicted relationship must be specific enough to fail. A hypothesis that predicts “some kind of effect” or “a difference of some sort” can never be wrong, because almost any result confirms it. A testable hypothesis commits to a direction, a comparison, or a threshold that a specific pattern of data would contradict.

A useful check: write down, before collecting any data, what result would prove the hypothesis wrong. If you cannot describe that result, the hypothesis is not yet in testable form.

Null vs. alternative, directional vs. non-directional

Quantitative hypotheses are almost always written as a pair, and then evaluated statistically against each other:

Null hypothesis (H0): states that there is no effect, no difference, or no relationship between the variables — this is the default position a statistical test is set up to try to reject.
Alternative hypothesis (H1 or Ha): states that an effect, difference, or relationship does exist. This is usually the researcher’s actual prediction and the reason the study is being run.

A significance test never proves either hypothesis true. It either rejects H0 (evidence is inconsistent with “no effect”) or fails to reject H0 (the data don’t provide enough evidence to rule out “no effect”) — it does not confirm Ha as correct. For a full side-by-side comparison of how the two are written, tested, and reported, see Null Hypothesis vs. Alternative Hypothesis.

The alternative hypothesis itself can take one of two forms:

Directional (one-tailed): predicts the direction of the effect — e.g., that a variable will increase or decrease a specific outcome. A directional hypothesis is only appropriate when there’s a genuine prior basis (existing evidence or theory) for expecting the effect to run one way, and it changes how the statistical test is set up (a one-tailed test concentrates the rejection region on one side of the distribution).
Non-directional (two-tailed): predicts that a relationship exists without specifying its direction — e.g., that a variable will be associated with a change in an outcome, without committing to increase or decrease. This is the more conservative default when prior evidence doesn’t clearly support one direction over the other.

How a hypothesis differs from a research question, an aim, and a theoretical framework

These four terms get used loosely and interchangeably in early drafts, but they answer different questions and belong in different places in a manuscript or protocol:

Research question is what the study is asking — an open question the study is designed to answer. Not every study needs a hypothesis: exploratory and much qualitative research is question-only, with no predicted answer stated in advance. See How to Write a Research Question for how to get from a broad topic to a specific, answerable question.
Hypothesis is a specific, falsifiable, predicted answer to that question, stated as a testable claim about variables and their relationship — it only applies where a study design (typically quantitative or experimental) actually calls for a stated prediction to test.
Aim or objective is a statement of intent or purpose (“to examine,” “to determine,” “to evaluate”) — broader and less committal than a hypothesis, and doesn’t itself require a predicted direction or measurable threshold. A study can state an aim without stating a hypothesis, but a hypothesis without a corresponding aim rarely makes sense.
Theoretical or conceptual framework is the underlying structure of theory and assumed relationships that explains why the hypothesis is plausible in the first place — it’s the reasoning scaffold the hypothesis is drawn from, not the testable prediction itself. If you’re not sure which of those two framework terms applies to your study, see Conceptual Framework vs. Theoretical Framework.

A simple ordering: the framework explains why you’d expect a relationship to exist; the question asks whether it does; the hypothesis states, specifically and falsifiably, what you predict the answer is; the aim describes what the study sets out to do about all of that.

Common mistakes when writing a hypothesis

Untestable or unfalsifiable claims. Predictions that no possible result could contradict — often because they invoke unmeasurable constructs, are stated as universal truths, or are so vague that any outcome “counts” as support.
Vague or unoperationalized variables. Naming a construct (“stress,” “quality of life,” “engagement”) without specifying how it will actually be measured leaves the hypothesis untestable in practice, even if it sounds testable in principle.
Circular reasoning. A hypothesis that is true by definition of its own terms — e.g., predicting that a “more effective” intervention will produce “better outcomes,” where effectiveness and better outcomes are defined identically — tests nothing.
Restating the research question as a statement. Turning “Does X affect Y?” into “X affects Y” is not yet a hypothesis if it still doesn’t specify direction, magnitude, population, or measurement — it’s a question wearing a period instead of a question mark.
Bundling multiple predictions into one hypothesis. A hypothesis that predicts effects on several unrelated outcomes, or via more than one mechanism, at once is difficult to test cleanly, because a mixed result (confirmed for one outcome, not the other) doesn’t clearly support or refute the hypothesis as a whole. Split multi-part predictions into separate, individually testable hypotheses.
Value-laden or unmeasurable terms. Words like “better,” “worse,” “improve,” or “harm” need an explicit, measured definition attached — otherwise the hypothesis imports a judgment call that the data can’t actually adjudicate.
Writing the hypothesis after seeing the results. Presenting a post hoc explanation of observed data as though it were an a priori prediction is a distinct and well-documented problem in its own right — see HARKing (Hypothesising After Results are Known).

A practical template for writing a hypothesis

A workable quantitative hypothesis generally states: a specific population or sample, the independent (predictor) variable and how it’s operationalized, the dependent (outcome) variable and how it’s measured, and the predicted relationship between them (and, if directional, which way it runs).

A simple template: “Among [population], [operationalized independent variable] will be [associated with / lead to an increase or decrease in] [operationalized dependent variable].”

Illustrative examples (constructed to demonstrate the structure — not drawn from or attributed to any specific real study):

Directional: “Among adults aged 18–35, completing a 10-minute guided breathing exercise before a timed cognitive task will be associated with a lower self-reported anxiety score (State-Trait Anxiety Inventory, state subscale) immediately afterward, compared with a no-intervention control group.”
Non-directional: “Among first-year undergraduate students, weekly hours spent in a structured study group will be associated with end-of-semester GPA.”
Null (paired with the directional example above): “There will be no difference in self-reported state anxiety scores between adults who complete the guided breathing exercise and adults in the no-intervention control group.”

Before finalizing, check the hypothesis against the failure test from the falsifiability section above: what specific result would prove each of these wrong? If that result is easy to state, the hypothesis is in testable shape. For how the hypothesis fits into the broader introduction of a manuscript, see Structuring the Introduction of a Research Paper.

Frequently asked questions

What’s the difference between a null and an alternative hypothesis?

The null hypothesis (H0) states there is no effect or relationship; the alternative (H1/Ha) states that one exists. Statistical tests are built to reject or fail to reject H0 — they don’t test Ha directly. See Null Hypothesis vs. Alternative Hypothesis for the full comparison.

How specific should a hypothesis be?

Specific enough that a defined, plausible result would prove it false — that means naming the population, operationalizing every variable, and stating the predicted relationship (and direction, if directional) rather than leaving any of those open-ended.

Can a hypothesis be proven?

No — not in the sense of being established as permanently true. Statistical testing can reject a null hypothesis (treating that as evidence consistent with the alternative) or fail to reject it, but it does not “prove” the alternative hypothesis correct. In the Popperian sense, hypotheses can only be falsified or survive attempts at falsification; they’re never verified outright.

Does every study need a hypothesis?

No. Exploratory, descriptive, and much qualitative research proceeds from a research question without a predicted answer stated in advance. A hypothesis belongs in a study designed to test a specific predicted relationship, typically (though not exclusively) in quantitative and experimental work.

What’s the difference between a directional and non-directional hypothesis?

A directional hypothesis predicts which way an effect runs (an increase or a decrease); a non-directional hypothesis predicts that a relationship exists without committing to its direction. Use a directional hypothesis only when prior evidence or theory genuinely supports expecting the effect to run one specific way.

Where does a hypothesis go in a manuscript or protocol?

Typically at the end of the introduction, immediately after the research question and gap statement, and often restated in the methods section alongside the statistical analysis plan that will test it.