How to write a research hypothesis

By Dr. Rafiq Muhammad, MD, PhD · Updated June 2026

A research hypothesis is a clear, testable, falsifiable prediction about the relationship between variables — stated before you collect data. The test that matters: could the data prove it wrong? If not, it isn’t a hypothesis yet. Note up front that hypotheses belong to confirmatory quantitative work — exploratory and qualitative studies use research questions instead.

Null vs alternative

Statistical tests are built around the null: you gather evidence to reject it, or you fail to reject it. You never prove the alternative directly — a subtlety that trips up a lot of discussion sections (see what a p-value actually means).

Directional vs non-directional

A directional (one-tailed) hypothesis predicts the direction — “group A scores higher than group B.” A non-directional (two-tailed) hypothesis predicts a difference or relationship without saying which way. Only go directional when theory or prior evidence genuinely justifies predicting the direction; otherwise two-tailed is the honest default.

Build it from your variables

  1. Name the variables. Your independent and dependent variables — what you change and what you measure.
  2. State the expected relationship between them, grounded in theory and the literature.
  3. Operationalise. Define each variable so it can be measured — a hypothesis about vague constructs can’t be tested.
[Independent variable] is associated with [change in dependent variable] in [population].
e.g. “Adults who complete a structured telehealth programme show a greater reduction in HbA1c at three months than those receiving usual care.”

The make-or-break test: is it falsifiable?

A hypothesis that no possible data could contradict is useless. Each one must name measurable variables, predict a specific relationship, and be capable of being wrong. That’s also what makes it testable with a defined statistical test — if you can’t name the test that would evaluate it, the hypothesis isn’t specified tightly enough.

The free Hypothesis Testability Checker flags hypotheses that are vague, non-falsifiable, or missing a measurable variable — before your supervisor does.

Get the free Research Proposal toolkit

A proposal-readiness checklist and a variables-and-hypotheses worksheet covering every section reviewers expect — from Research Proposal Writing Simplified. We’ll email you the download link.

One email with your download, then occasional research tips. One-click unsubscribe, anytime. We never sell your data.

Get Research Proposal Writing Simplified

Frequently asked questions

What is a research hypothesis?

A clear, testable, falsifiable prediction about the relationship between variables, stated before data collection.

Null vs alternative?

The null says there’s no effect; the alternative says there is. Tests reject or fail to reject the null.

Directional vs non-directional?

Directional predicts the direction of the effect; non-directional predicts only a difference. Go directional only when theory justifies it.

Do all studies need a hypothesis?

No — confirmatory quantitative studies do; exploratory and qualitative studies use research questions instead.

Conceptual framework → Open the Testability Checker →