A short, clear, peer-reviewed article explaining “why it is nonsensical to use retrospective power analyses to conduct a postmortem on your study”.
Link: Why post-hoc power analyses are nonsensical
What is this resource and what does it offer: This resource is an easy-to-read and easy-to-understand peer-reviewed article that describes why retrospective power analyses (i.e. power analyses that are done after the study has been completed and the database has been finalized) are useless and nonsensical. The author states that retrospective power analyses are really analyses that should be used to plan a future study, and not analyses to describe why the just completed study did not produce the desired results. The article is clear and concise, and requires only a few minutes to read.
Citation: Jiroutek, Michael R., and J. Rick Turner. “Why It Is Nonsensical to Use Retrospective Power Analyses to Conduct a Postmortem on Your Study.” The Journal of Clinical Hypertension 20, no. 2 (2018): 408–10. https://doi.org/10.1111/jch.13173.
Who created this resource: Michael R. Jiroutek, DrPH, is an associate professor of biostatistics at Campbell University in North Carolina. He transitioned to academia in 2012, after over 10 years’ experience as a biostatistician in the pharmaceutical industry. J. Rick Turner, PhD, DSc, now deceased, was a respected consultant, educator, and clinical trialist with particular interests in cardiology and clinical pharmacology. His experience included work for the FDA, Quintiles, GlaxoSmithKline and Campbell University.