Skip to content

Instantly share code, notes, and snippets.

What would you like to do?
[Subject to minor updates]
Science-Driven Decision Making: Syllabus
1. Setup & intro
2. What the news sound like
- Find and assess the keywords
- Find and assess the metrics
3. Association
- What is "observational" data?
- What does association really mean: average differences
- What does it usually look like behind the scenes: oh, the variance!
- Is it "real"?
- The traps of small sample sizes
- What does "accurate to plus or minus three percentage points, 19 times out of 20" really mean?
- What it doesn't mean
- The importance of prior knowledge
- How does association come about?
- What's the problem? Simpson's paradox
- When is association useful for decision making?
4. Causality
- Why run experiments
- How to run experiments - What is A/B testing
- The importance of sampling and randomization
- What do "blind" and "double blind" mean?
- What is the "Placebo effect"?
- What can experiments tell us?
- Deja vu? More averages, variance, tests and prior knowledge
[Short Break]
5. More buzzwords explained
- Prediction: association, causality, both?
- What is "Machine Learning"?
- Does "Big Data" make all our problems go away? (spoiler: no)
- Patterns, patterns everywhere
6. Media and society
- The "positive results" bias
- The "news" bias
- The "sales" bias
7. Comprehensive article analysis exercise
8. Roadmap ahead
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment