What is regression to the mean in simple terms?

Unusually good or bad results often involve some luck, so the next result usually lands closer to your typical level.

Does regression to the mean mean everything ends up average?

No. It doesn't mean the spread of the whole group shrinks. It means cases selected for being extreme tend to measure closer to the average next time, and the same regression shows up when you look in the reverse direction.

Regression to the Mean

🔬 Natural Sciences By CampusDown Wiki Editorial Team Last updated October 27, 2026

Regression to the mean is the tendency, when two measurements are less than perfectly correlated, for cases that were extreme on the first measurement to be closer to the average on the next. Francis Galton first described it in 1886 while studying the heights of parents and children, and the statistical term regression comes from his work. It is a classic statistical trap that distorts causal interpretation by making change appear where no real cause exists.

Contents

1. Galton's discovery and the statistical principle
2. Cognitive illusions and the Secrist fallacy
3. Research design and remedies
4. FAQ
5. References

Galton's discovery and the statistical principle

Galton (1886) found that the children of very tall parents tend to be shorter than their parents and the children of very short parents taller, and he called this regression toward mediocrity. He first read it as a hereditary phenomenon, but in Natural Inheritance (1889) he realized it is a statistical property that appears whenever two variables are imperfectly correlated. As Stigler (1986, 1997) recounts, the insight led to the birth of correlation and regression analysis. An observed value can be seen as the sum of a stable true level and a temporary chance component. An extreme value is likely to reflect chance pushing hard in one direction, and since that luck does not repeat on the next measurement, the value moves back toward the mean. The lower the correlation, the stronger the regression.

Cognitive illusions and the Secrist fallacy

Kahneman and Tversky (1973) showed that people tend to ignore regression to the mean when making predictions. A famous example, recounted by Kahneman (2011), involves Israeli Air Force flight instructors. They believed punishment worked better than praise because pilots praised for a great flight did worse next time and those scolded for a bad one improved, but they were mistaking the natural regression that follows extreme performance for cause and effect. In economics, Secrist (1933) wrote The Triumph of Mediocrity in Business after seeing highly profitable firms converge toward the average over time, but Hotelling (1933) countered that this was merely statistical regression and proved nothing about competition. Friedman (1992) warned that the same fallacy keeps recurring in studies of growth convergence across countries.

Research design and remedies

Campbell and Kenny (1999) showed that regression artifacts can systematically distort the conclusions of before-and-after studies. For example, if you pick only the students with the lowest test scores for extra tutoring, their scores will appear to rise on the next test even if the tutoring has no effect. Nesselroade, Stigler and Baltes (1980) addressed the problem in the study of change, and Bland and Altman (1994) and Barnett, van der Pols and Dobson (2005) proposed remedies such as using a randomized control group or selecting subjects based on the average of several measurements. Schall and Smith (2000) confirmed that extreme early-season batting averages in baseball also regress toward the mean. Along with Berkson's paradox and survivorship bias, regression is a key phenomenon to watch for when interpreting selected samples.

FAQ

What is regression to the mean in simple terms?: Unusually good or bad results often involve some luck, so the next result usually lands closer to your typical level.
Does regression to the mean mean everything ends up average?: No. It doesn't mean the spread of the whole group shrinks. It means cases selected for being extreme tend to measure closer to the average next time, and the same regression shows up when you look in the reverse direction.

Sources reviewed: 2026-10-26

References

Galton, F. (1886). Regression towards mediocrity in hereditary stature. Journal of the Anthropological Institute of Great Britain and Ireland, 15, 246–263.
Galton, F. (1889). Natural Inheritance. London: Macmillan.
Stigler, S. M. (1997). Regression towards the mean, historically considered. Statistical Methods in Medical Research, 6(2), 103–114.
Stigler, S. M. (1986). The History of Statistics: The Measurement of Uncertainty before 1900. Cambridge, MA: Harvard University Press.
Kahneman, D., & Tversky, A. (1973). On the psychology of prediction. Psychological Review, 80(4), 237–251.
Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus and Giroux.
Secrist, H. (1933). The Triumph of Mediocrity in Business. Evanston, IL: Bureau of Business Research, Northwestern University.
Hotelling, H. (1933). Review of The Triumph of Mediocrity in Business by Horace Secrist. Journal of the American Statistical Association, 28(184), 463–465.
Friedman, M. (1992). Do old fallacies ever die? Journal of Economic Literature, 30(4), 2129–2132.
Campbell, D. T., & Kenny, D. A. (1999). A Primer on Regression Artifacts. New York: Guilford Press.
Barnett, A. G., van der Pols, J. C., & Dobson, A. J. (2005). Regression to the mean: What it is and how to deal with it. International Journal of Epidemiology, 34(1), 215–220.
Bland, J. M., & Altman, D. G. (1994). Regression towards the mean. BMJ, 308(6942), 1499.
Nesselroade, J. R., Stigler, S. M., & Baltes, P. B. (1980). Regression toward the mean and the study of change. Psychological Bulletin, 88(3), 622–637.
Schall, T., & Smith, G. (2000). Do baseball players regress toward the mean? The American Statistician, 54(4), 231–235.

Reports and materials that use this theory (Korean)

There are no materials dedicated to this theory yet, but you can search related reports in the CampusDown library (Korean).

Find more reports · Search '평균으로의 회귀' on CampusDown →

Regression analysisWhat regression analysis is, how least squares fits a line, how to read coefficients and R², the differences among simple, multiple, and logistic regression, Galton's origin of the term, and common pitfalls.➗ Mathematics & Statistics Sample size determinationSample size determination, from the 1936 Literary Digest poll fiasco to Cohen's power analysis: the survey formula (about 385 for ±5% at 95%), sizing experiments with effect size, α, and power, A/B test samples, and tools like G*Power.➗ Mathematics & Statistics Survivorship biasSurvivorship bias is drawing conclusions from cases that survived a selection process while overlooking those that failed and disappeared.🧠 Psychology Confirmation BiasConfirmation bias explained: seeking, interpreting and remembering information that fits what you already believe, Wason's 2-4-6 task, Lord, Ross and Lepper's biased assimilation study, filter bubbles, and ways to counter it.🧠 Psychology Berkson's ParadoxWhat Berkson's paradox is: two independent traits appear negatively correlated because of how a sample was selected. Covers Berkson's hospital example, colliders in causal graphs, Heckman's sample selection bias and examples from economics, sociology and data science.🔬 Natural Sciences

Latest ArticlesView all ›

New guides for college study and assignments (Korean)

This article was last updated on October 27, 2026. It is based on widely recognized original works and textbooks; when citing it in a paper, please check the original sources listed in the references.