Power Rules: Practical Advice for Computing Statistical Power

Recent work emphasizes the importance of statistical power in the social sciences and shows that power tends to be very low. In this paper, I offer simple rules to compute power for balanced, between-subjects designs. For balanced, between-subjects designs (e.g., common for survey experiments), these rules are sufficient for a complete, compact, and compelling power analysis for treatment effects and interactions. Most importantly, these rules help researchers develop a sharp intuition about statistical power, lower the barrier to entry for researchers new to thinking carefully about statistical power, and help researchers design powerful, informative experiments.