A study too small to reliably detect a real effect if one exists.
An underpowered study has a sample size too small to give adequate statistical power (conventionally ≥80%) to detect a true effect of the size expected, raising the risk of a false-negative (Type II error) result. A non-significant finding from such a study should not be read as evidence of 'no effect' — the study may simply have lacked the numbers to detect it. Power depends on sample size, expected effect size, and variability, so judgements of adequacy require checking the study's own power calculation or effect-size confidence intervals rather than the p-value alone.
This guide was auto-drafted and is pending editorial review.