Power and variability
Section 2.7 on power analysis says:
"The statistical power of a test (and, hence, the Type 2 error rate) depends on the effect size, the sample size, and the alpha level."
What about the variability of the effect size? This factor seems to be missing in this section. If it's an intended simplification for pedagogical purposes, perhaps consider adding a footnote or a cross-reference to a section where variability is accounted for?
Further suggestion: Use "proportion" rather than "percentages" in:
"The power of the test is the percentages of the distribution on the right that is larger than the critical value."
I will incorporate the suggestions in a few weeks when my schedule allows, but please keep adding them when you see them.
fixed