Research methods guide
How to interpret a statistically non-significant result
When a result falls short of statistical significance, use the effect estimate and confidence interval to describe the range compatible with the analysis and compare it with a meaningful difference defined in advance.
In a synthetic comparison, the prespecified meaningful improvement is 4 units. The estimated mean difference is 2 units with a 95% CI from −1 to 5. Interpret the interval against the threshold.
Reporting the same estimate in Results and Discussion
For a difference, the null value is zero. Use the original outcome units.
| Item | Worked example | Verification |
|---|---|---|
| Results sentence | “Mean difference 2.0 units (95% CI −1.0 to 5.0).” | Confirm subtraction direction, population, model, and table. |
| Practical importance | The interval includes effects near zero and improvements beyond the prespecified 4-unit threshold. | Cite the prespecified threshold or identify the interpretation as exploratory. |
| Discussion sentence | “The estimate was imprecise and remains compatible with both small effects and a potentially meaningful benefit.” | Match the claim to the design. An ordinary non-significant result cannot establish equivalence. |
Apply the method to your materials
Put the threshold, estimate, and interval together
Specify direction and the threshold’s source. For ORs or HRs, use 1 as the null and interpret the interval on the reported scale.
Check where uncertainty comes from
Verify the prespecified analysis, sample size, and loss to follow-up. Distinguish random error from bias and measurement limitations.
Questions that arise in practice
Does a narrow interval establish equivalence?
An equivalence claim needs prespecified margins and a design and analysis suited to that question. An ordinary failed superiority test does not establish equivalence.
Verified sources
The official sources below were checked on 2026-11-12. Access conditions and update dates are rechecked before publication.
RigorMed workspace
Start with your research materials
Bring your research question, available materials, and the steps you want to check.
