Confidence interval
Short definition
A confidence interval is a range of values, calculated from a study's data, that indicates the precision of an estimate and the effect values compatible with those data, usually at 95%.
In plain language
A study never measures the exact effect, only an estimate. The confidence interval shows the margin of uncertainty: if it is narrow, the estimate is precise; if it is wide, as happens in small studies, the true effect may be very different from the one observed. If the interval for a difference between groups includes zero, the data are compatible with no effect.