Cognitive Biases · Belief

Simpson's paradox

Also called: Yule-Simpson effect

A trend that shows up in every subgroup can shrink or reverse when the groups are combined, because the groups differ in size or mix.

Evidence grade: Statistical. How we grade evidence

For example

A hiring report shows one office with a lower offer rate than another. Within each job type, that office actually offers at a higher rate, but most of its applications are for the hardest-to-win roles. The combined number points to the wrong conclusion.

DECIDE BETTER® · Rule 3Data answers your questionEvery data set answers somebody's question. Make sure it's yours, or you end up precisely wrong and only generally right.

Spot it in your week

Three quick questions. Your answers stay on this page.

  1. In the last week, did you read a combined average that seemed to conflict with the results for each group?
  2. Did a meeting this week draw a conclusion from totals without looking at the groups inside them?
  3. Did a report you read this week compare groups of very different sizes or mixes using one overall figure?

Your DecisionDNA™ can tell you which biases you're most exposed to.

Find your DecisionDNA™

Sources

Related entries

In public budgets

A statewide average can point one way while every district points the other, because districts differ in size.

See it on Colorado Works Better⁠ (opens in a new tab)

Get the countermeasure

Get the countermeasure and the decisions where this bites, free in the Volley app.

See the Volley app