Why Post-hoc Power Doesn't Matter When Tests Aren't Significant
HTML-код
- Опубликовано: 11 сен 2024
- This video tutorial tackles two important points about post hoc power analysis:
1. Why Post-Hoc Power with Observed Effect Sizes Is a No-Go
I'll explain why calculating power based on observed effect sizes after obtaining non-significant results is redundant. It doesn't offer new insights beyond your p-value.
2. The Alternative: Power Calculation with Meaningful Effect Sizes
Discover the smarter way to conduct post-hoc power analysis using theoretically or practically meaningful effect sizes. This approach adds value, particularly when your sample size wasn't within your control (e.g., in a secondary analysis).
Uncover the importance of meaningful post-hoc power analysis and how it enriches result interpretation.
References:
Gelman, A. (2018, September 24th). Don’t calculate post-hoc power using observed estimate of effect size. Statistical Modeling, Causal Inference, and Social Science. statmodeling.s...
Hoenig, J. M., & Heisey, D. M. (2001). The abuse of power: The pervasive fallacy of power calculations for data analysis. The American Statistician, 55(1), 19-24.
Lakens, D. (2014, December 19th). Observed power, and what to do if your editor asks for post-hoc power analyses. The 20% Statistician. daniellakens.b...