StatQuest with Josh Starmer on YouTube · Open on YouTube
Core skills: SQL depth, statistics, Python and dashboards · Advanced beginner · Lesson 20 of 99 · 15 min video, then about 5 hours of practice
Hypothesis testing, p-values and practical significance
After this you can state null and alternative hypotheses, run and interpret a t-test or proportion test, and distinguish statistical from practical significance.
Why it matters
Stakeholders will constantly ask whether a change is 'real'. Learn what a p-value is (the probability of data at least this extreme if nothing were going on) and is not (the probability the hypothesis is true). Then learn the more important habit: always report the size of the difference and its uncertainty, because with enough data a trivially small difference becomes 'significant'.
How to use this resource
StatQuest on hypothesis testing and the null hypothesis, clear about what the test is and is not telling you.