Confidence interval

A confidence interval gives a range of plausible values for an effect, rather than a single point estimate, and conveys how precise that estimate is. A narrow interval signals a precise result; a wide one signals uncertainty, often from a small sample. The standard interpretation is about the procedure: across many repeated studies, 95 percent of the intervals produced would contain the true value. When reading a study, look at the whole interval, not just the point estimate. Check whether it crosses the value of no effect (zero for a difference, one for a ratio), which signals the result is compatible with no effect, and check whether both ends would be practically meaningful. For example, a drug reported to cut risk by 30 percent with a 95 percent interval of 5 percent to 50 percent shows a benefit that could be modest or large, whereas an interval of 28 percent to 32 percent would be far more precise. Wide intervals warn against over-reading a headline number.