The big idea
🎲
Four ideas, no mathematics
Significance, effect size, base rates, and multiple comparisons explain most of the gap between studies and headlines.
Scientific Thinking · Lesson 3
Statistical Thinking for Everyone
beginner⏱ 18 min · 7 cards
Key term
The most misread number
What does p = 0.03 actually mean?
tap to reveal ↻
p-value
Data this extreme would appear 3% of the time if there were no real effect. Not the probability the hypothesis is true, and silent on how large the effect is.
Try it
Significance vs Size
“p = 0.001, blood pressure lowered by 0.2 mmHg, n = 100,000.”
Significance says it is probably real. Is it big enough to matter?
Decide which results would actually change a decision.
Ask these
Before being impressed
Tap each.
Nuance
Found by reproduction
Maester Note
A widely cited debt-and-growth threshold was weakened when a graduate student obtained the spreadsheet and found an averaging error. Sharing data is what made the correction possible.
Quick check
When is a truncated y-axis legitimate?
You did it
Nice work
Mark this lesson complete to track your progress.