LECTURE HUB
Lecture 8 - Hypothesis tests and what p means
🚦 Status🌟 Mastered
🧠 Mastery🦅 Adept
📅 Last Review25 May 2025
🎯KEY TAKEAWAYS
- A 🔤 p-value answers one narrow question: if there were truly no difference, how often would data at least this lopsided turn up?
- It is not the probability that there is no difference. It is not the size of the effect. It is not the chance I am wrong.
- 0.05 is a habit, not a boundary in nature. 0.059 and 0.041 are the same evidence.
- Stayed for the lecture after getting a 68 at the door. That is the takeaway I am proudest of.
⚡QUICK JOTS
- 18:28. paper back at the door. 68. not now, think about it later
- null = the boring assumption. nothing is going on
- then ask: how odd is my data if the boring assumption is true?
- 1.5 / 0.79 = 1.89 standard errors
- p ≈ 0.06. six times in a hundred by luck alone
- p is about the DATA given the null. not about the null given the data. (lecture 4 again, the swap)
- 0.05: a habit that got promoted to a law
- "not significant" ≠ "no difference"
- reading list: the statisticians' statement on p-values
- quiz next week
☑️HOMEWORK
- Write 🔤 p-value: what it is and the three things it is not
- Read 🔬 The ASA Statement on p-Values
- Prepare for 📌 Statistics I - Quiz 3 - Reading a p-value next week
- Find one sentence at work where someone says significant and check what they meant