liveBook · Manning

Statistical power

As explained in Timeless Algorithms, The Foundational Ideas

Statistical power is the probability that a test rejects the null hypothesis when a specified alternative condition is true.

Statistical power is the probability that a test rejects the null hypothesis when a specified alternative is true. Sample size and the magnitude of an improvement influence power; a small sample can leave a genuine, modest effect undetected. section 4.2.3 section 10.2.3

A small clinical trial

When a trial’s result does not meet its significance threshold, limited power may mean a modest treatment effect went undetected. The result is suggestive, not proof of effectiveness or ineffectiveness; a larger, adequately powered follow-up is warranted. section 4.2.3

How Timeless Algorithms, The Foundational Ideas covers it

How deep it goes
Defines rejection probability under an alternative and links power to sample size and effect magnitude.
Examples
Uses a small clinical trial with a potentially undetected modest treatment effect and a proposed larger follow-up.
What you can do after
Evaluate ambiguous test results by considering whether low power could hide a genuine, modest treatment effect.
How it connects
Connects power to null and alternative hypotheses, sample size, and Type II error.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Timeless Algorithms, The Foundational Ideas. The section links point to the parts of the book it draws on.