Statistical Power
Statistics & analysis
The probability (conventionally targeted at 0.80) that a test detects an effect that genuinely exists — driven by sample size, effect size and significance level.
Statistical power is the probability that a study will detect an effect of a given size when that effect genuinely exists — formally, 1 minus the probability of a Type II error. Power of 0.80, meaning an 80% chance of detection, is the conventional minimum in social-science research.
Power depends on four linked quantities: sample size, effect size, significance level and power itself. Fix any three and the fourth is determined, which is what makes a priori power analysis possible — specify the smallest effect worth detecting, set alpha at 0.05 and power at 0.80, then solve for the sample size you need.
G*Power is the standard free tool for this calculation. Run it before collecting data and report it in your methodology chapter to justify your sample size. Post-hoc power computed from an observed non-significant effect is widely criticised as uninformative, so reviewers expect the a priori version.
Where it's used
- A priori power analysis to justify sample size
- Explaining non-significant results in small samples
Software used
Related concepts
Guides that use this concept
All guidesNot sure where to start?
Book a free 15-minute consult. We'll map your next three steps — no obligation.