What does 'statistically significant' mean, and does it mean the result matters?
Updated: 2026-09-09 · Safety and evidence
Short answer
Significant means the result would be unlikely if there were no real difference at all; it is a statement about chance, not about size. A large study can detect minute differences and call them significant. Whether it matters depends on the effect size, what was measured, and in whom.
In plain English
- A p-value speaks to the probability of seeing that result by chance, not to how useful the effect is.
- With very large samples, tiny differences come out significant.
- With small samples, a real effect can fail to reach significance.
- Confidence intervals are usually more informative: they show the range of values compatible with the data.
What we know
- Statistical significance and clinical relevance are distinct concepts in research methodology.
- Confidence intervals convey magnitude and precision, not just a yes/no.
What we don't know yet
- A p-value does not tell you whether a result will replicate, or hold in another population.
- It also says nothing about whether the measured change is something a person would notice.