Principle of Analytic Graphics
- Principle 1: Show Comparisons
- always comparative (compared to what)
- randomized trial - compare control group to test group
- evidence for a hypothesis is always relative to another competing hypothesis
- Principle 2: Show causality/mechanism/explanation/systematic structure
- form hypothesis to evidence showing a relationship (causal framework, why something happened)
- Principle 3: Show multivariate data
- more than 2 variables because the real world is multivariate
- show as much data on a plot as you can
- example
- slightly negative relationship between pollution and mortality
- when split up by season, the relationships are all positive \(\rightarrow\) season = confounding variable