The probability a study will detect a real effect if one truly exists.
Statistical power is the likelihood that a study's statistical test will correctly reject the null hypothesis when a true effect of a given size is present, commonly set at a threshold of 80%. It is determined by sample size, effect size, variability in the data, and the significance level (alpha) chosen. A study with low power risks a false-negative (Type II) error, meaning a genuinely effective treatment or true difference may be missed rather than disproven.
This guide was auto-drafted and is pending editorial review.