IDEA FOUND // IDEA 131
Regression to the Mean
“Cases selected for being extreme often look less extreme next time, even without an intervention.”
01 / PLAINLY
What it means, plainly
When cases are chosen because one measurement was extreme, the next measurement is often less extreme if random variation helped create the first value. That can look as though something caused an improvement or decline.
02 / CONTEXT
A little more
When a measurement combines a fairly stable signal with random variation, selecting extremes favors observations that also had extreme noise. On repetition, that noise is usually less extreme.
03 / WHY IT MATTERS
Why it matters
It keeps a reward, treatment, or punishment from receiving credit for a change that might follow from repeating an extreme measurement.
04 / EXAMPLE
A familiar example
People are rewarded after their worst week and many improve; part of the change would have happened without the reward.
05 / LIMIT
What it does not mean
It is not a force pushing individuals toward average, nor does it guarantee every case will move closer.
06 / NOTICE
Notice it in your day
Pick the highest value in a varying series and ask how your interpretation changes if it was selected precisely because it was extreme.
FINAL NOTE
The idea worth keeping
When improvement starts from an extreme, compare against a group or several prior measurements.
QUESTIONS / 02
Questions people still have
Why might an extreme case look less extreme later?
Random variation may have helped create the first value and may not repeat as strongly.
Does regression to the mean move every case toward average?
No. It is not a force and does not guarantee an individual movement.