significant
Meaning of significant: A word that promises far less than it sounds like it does: a result unlikely to be chance, not a result that matters.
Definition of significant
Significant, in the statistical sense, describes a difference between groups that is larger than chance alone would comfortably explain. A study calculates how often a gap that size would appear if the groups were really the same, and when that probability falls below an agreed threshold the result is called statistically significant. The label says nothing about how large or how useful the difference is.
What is actually being tested?
Two groups in a study almost never come out exactly equal, even when the treatment does nothing at all, because the people in them differ. The test asks a narrow question: if there were truly no difference, how often would sampling alone throw up a gap this wide? That answer is the p value. A small p value means the observed gap would be an odd coincidence, so the finding is labeled significant. The threshold is a convention agreed in advance, not a natural boundary, and a result just above it differs barely at all from one just below.
Why is significant not the same as important?
The word carries a promise in ordinary English that it does not carry in statistics. A trial large enough will detect a difference too small for anyone to notice, and that difference will still be called significant. A small trial can miss a large and useful effect entirely and report no significance, which is not the same as showing there is no effect. Size of effect and strength of evidence are separate questions, and this label answers only the second.
How do the related terms line up?
| Term | What it means |
|---|---|
| Statistically significant | A gap this large would be unusual if nothing were going on |
| Clinically significant | Large enough to change how someone feels or fares |
| Not significant | Evidence too weak to rule out chance, not proof of no effect |
| Confidence interval | The range of effect sizes the data are compatible with |
Where does the word mislead?
Reporting outside research papers routinely reads significant as meaning large or proven. The problem compounds when many comparisons are made at once, since testing enough questions guarantees that some clear the threshold by luck alone. Fields that lean heavily on the label, including psychological research and studies of hereditary risk, have moved toward reporting the size of an effect and its range of uncertainty alongside the p value, precisely because the single word conceals both.
Also known as
Used in a sentence
The difference in response rate between the two arms was statistically significant.
Good to know
A significant result may describe a difference too small to matter to anyone, and a result reported as not significant may reflect a study too small to detect a real effect rather than the absence of one.