Data analysis
Read rates carefully and refuse easy causal leaps from tidy charts.
Thinking practices that matter here
- Evidence & Epistemology
Base rates
A striking result still sits inside how common the thing was to begin with.
- Evidence & Epistemology
Correlation is not causation
Things that move together, or one after the other, are not yet a cause.
- Evidence & Epistemology
Confirmation bias
A hunch can steer you toward the evidence that flatters it.
- Logic & Reasoning
Validity
A valid argument is one whose conclusion must follow, not one whose sentences merely sound true.
A sample scenario
A dashboard shows a sharp rise. What base rate and comparison group would change how you tell the story?