Hypothesis testing
Content for Thursday, January 28, 2027
Goals
By the end of this topic, you should be able to:
- Carry out the four steps of a hypothesis test
- Say what a p-value is, and what it isn’t
- Tell Type 1 from Type 2 errors
- Explain power, and what raises it
- Report an effect size, and judge it against real education research
Reading
- Before session 1: sections 8-1 to 8-3.
- Before session 2: sections 8-4 to 8-6. Read 8-6 for the idea of power. We don’t compute power by hand, so you can skip working through Examples 8.6 and 8.7.
NoteThe book and our slides
- The book: says in one place that the p-value is the probability of a Type 1 error. Our slides: α is that probability, set before we look. p is how surprising our sample would be if H₀ were true. It isn’t the chance that H₀ is true, or that we made an error.
- The book: says we fail to reject H₀. Our slides: say we maintain H₀. Both mean the same thing: the data didn’t rule H₀ out.