Simpson's Paradox
A trend appears in different groups but disappears or reverses when groups are combined.
"The aggregate can lie about the parts. The parts can lie about the aggregate."
What is Simpson's Paradox? A trend appears in different groups but disappears or reverses when groups are combined. Aggregated metrics can hide important subgroup effects.
A treatment helps men and women separately but appears harmful overall.
Aggregated metrics can hide important subgroup effects.
Analyze data at multiple levels of aggregation.
Use the model. Pick the move.
A trend appears in different groups but disappears or reverses when groups are combined. You've just seen this: A treatment helps men and women separately but appears harmful overall. Which lever does the model recommend?
Pick a lever. There are no neutral ones — every incentive funds a behavior somewhere.
Pick a reaction to Simpson's Paradox
One tap. We'll point you at the most useful next surface based on how this hits.
The full taxonomy entry
Every concept in the Atlas uses the same structure — so Simpson's Paradox can be compared, recombined, and cited like an element on a periodic table.
- Business
- Leadership
- Government
- Healthcare
- Education
- Sales
- Marketing
- AI
- Negotiation
- Media
- Public Policy
- Relationships
- Where in our org would Simpson's Paradox most often show up unnoticed?
- Which metric, ritual, or contract clause quietly rewards Simpson's Paradox?
- If we removed every payoff for Simpson's Paradox, what behavior would replace it?
- Who benefits when Simpson's Paradox persists — and who pays the cost?
- People defend the status quo using the language of simpson's paradox.
- Decisions cluster around the easiest narrative rather than the strongest evidence.
- New data changes the slide deck but not the decision.
- Anyone naming the pattern is treated as the problem.
Every Atlas entry is a node in a knowledge graph. See the related rail below to follow the connections.
See Simpson's Paradox through this lens
Each layer of the Incentives OS reframes this concept with its own thinkers, vocabulary, and diagnostic question.
Do you actually know Simpson's Paradox?
Three quick questions. Result is saved into your review streak — come back when the term is due to lock it in.
Which best describes Simpson's Paradox?
Worked example, counter-example & concept map
On-demand AI analysis grounded in the Lab's research. Cached on your device after first run.
When you encounter Simpson's Paradox, your prefrontal cortex has to do extra work to override the automatic response — and that override budget is finite.
Executive control, planning, impulse override, working memory, System 2. First thing to go offline under stress, fatigue, or low blood sugar. Why your 4pm decisions are worse than your 9am ones.
See Prefrontal in the Brain Atlas →Picked for you, from the Atlas
Ranked by shared learning paths, overlapping chips, and what you've saved.
Ignoring general statistics in favor of specific, vivid information.
Posterior = (likelihood × prior) / evidence.
Update beliefs in proportion to the strength of new evidence.
Start with a prior; update with new evidence.
Novices experiencing early success, often due to variance and small samples.
High-impact, hard-to-predict, retrospectively explainable events.
Send the card, not just the link
A pre-rendered social card with the title, eyebrow, and URL. Copy the link, post it anywhere, or download the SVG for slides.
More definitions to follow
Every term in the Atlas connects to a dozen others. Pick any of these and see where it takes you.
Decisions made to be defensible, not to be right.
Comprehensive AI regulation in the EU.
Howard Gardner's theory of distinct intelligences (linguistic, logical-mathematical, musical, bodily-kinesthetic, spatial, interpersonal, intrapersonal, naturalist).
Authenticity becomes a performance the moment it becomes a paycheck.
Knowing what you know — and how confidently you know it.
Worst-cases assumed as base-cases because they 'feel responsible.'
Output amplifies input — virtuous or vicious cycle.
People stay quiet when they sense their view is minority — even when it isn't.
Below-cost water rights drive overuse in arid agricultural regions.
Teams built around AI from day one vs. teams adding it to existing workflows.
Passion is inversely proportional to the amount of real information available.
Pausing System 1 to engage System 2.