Two Concepts of Power: A Commentary on Pek et al. (2024)

Statistical power remains a central yet confusing concept in psychology. Pek et al. (2024) define power as the probability of rejecting the null hypothesis when the alternative is true, a view consistent with many textbook accounts. We argue, however, that Cohen (1962, 1988) advanced a broader definition: the probability of obtaining a significant result, without conditioning on a specific alternative effect size. This broader perspective allows power to serve two functions—planning studies with hypothetical effect sizes and evaluating completed studies with plausible effect size estimates. The broader concept is useful to evaluate single studies and to estimate publication bias. We conclude that power analysis of published research is an important tool for the credibility revolution of psychological research.