06.5.4 · Skill in Statistical reasoning
Statistical power
The chance a test detects an effect that really exists.
Power depends on sample size, effect size and noise. Low-powered tests miss real effects and exaggerate the ones they find. Planning for adequate power is part of designing an experiment.
Resources
2- videoStart hereStatistical power, clearly explained ↗StatQuest with Josh Starmer
A short, visual explanation of what power is and why sample size changes it.
- videoStart herePower analysis, clearly explained ↗StatQuest with Josh Starmer
How to work out the sample size a test needs before running it.
Tools
1- ToolFreeEvan's Awesome A/B Tools
Free calculators for sample size, significance of results and sequential tests.