What is recommended instead of performing post-hoc power analyses?

• A. Increasing the sample size after the study.
• B. Rerun the study with a revised protocol.
• C. Use a more lenient significance threshold.
• D. Report observed effect sizes and confidence intervals instead.

Answer: (D)

Post-hoc power analyses aren’t a reliable way to interpret a study’s findings. Power is the probability of detecting a true effect before data are collected, given a specific effect size. After data are in, calculating power using the observed effect size makes power depend on what happened to be seen in this sample, which doesn’t tell you much about the real effect and can mislead conclusions.

The recommended approach is to report the observed effect size and its confidence interval. The effect size shows how large and in what direction the effect is, while the confidence interval communicates the precision of that estimate and the range of plausible values for the true effect. This gives readers a clearer sense of practical significance and uncertainty, regardless of whether the study was formally powered.

Why the other options aren’t appropriate: increasing the sample size after the study isn’t feasible because it would mean collecting new data. rerunning the study with a revised protocol is something for future research planning, not a way to interpret the completed results. using a more lenient significance threshold inflates the risk of false positives and is not a valid remedy for interpreting findings.