Regression Toward the Mean

Regression toward the mean is the statistical phenomenon in which an extreme result on one measurement of a random variable tends to be followed by a result closer to the average on the next measurement of that same variable. It becomes especially visible when many variables are sampled and only the most extreme results are singled out for attention, since a second round of sampling on those same standout cases will typically look less extreme, closer to the overall average, even with no real change having occurred. The strength of the effect depends on whether the variables being compared are drawn from the same underlying distribution or from genuinely different ones, and because people who select cases for their extremeness often misread this natural drift back toward average as a meaningful change, failing to account for it is a documented cause of false conclusions drawn from follow-up data in experiments, tests and everyday judgments alike.