Judging an Effect by Comparing It to the Trait's Own Scale
The practical skill that makes the previous two nodes usable: put every reported effect on one ruler. The paper's own numbers give the ruler — the sex separation is about 5 SD, women's pitch falls about 0.8 Hz per year from 223.3 Hz at ages 20 to 30 to 197.4 Hz at 60, and the lead allele contributes 2.1 Hz per copy, 4.2 Hz for a homozygote. Same trait, same units, wildly different magnitudes.
Questions this Concept answers
- Why is a per-allele effect of 2.1 Hz best judged beside the other things that move a voice?
Effect size - Wikipedia
Effect size is a quantitative measure of the magnitude of a phenomenon, reported alongside significance because a p-value says nothing about how large or important a result is. The page distinguishes…