TechWorkRamblings

by Mike Kalvas

202208211331 Simpson's paradox

Simpson's paradox is a phenomenon in probability and statistics in which a trend appears in several groups of data but disappear or reverses when the groups are combined.1

simpsons-paradox.gif

The paradox can be resolved when confounding variables and causal relations are appropriately addressed in the statistical modeling.

The paradox has been used to illustrate the types of misleading results that can be produced through the misuse of statistics.2 In particular, 202304031254 Small careful samples are better than large poor ones. If we measure a homogenous group and take an average, we'll likely learn something. If we measure heterogenous groups and take the average, we'll likely see something misleading. For example:

The average adult has one breast and one testicle, but that does not represent the average person in our society.2


  1. Wikipedia contributors. (2022). Simpson’s paradox. In Wikipedia. https://en.wikipedia.org/w/index.php?title=Simpson%27s_paradox&oldid=1096740859

  2. Hamming, R. W. (2020). The art of doing science and engineering: Learning to learn (pp.251). Stripe Press. https://www.goodreads.com/book/show/53349431-the-art-of-doing-science-and-engineering 2