A veteran statistician argues effect size matters more than significance
In short: In a reflective essay, psychologist Cohen distilled decades of experience applying statistics in psychology and other sociobiomedical sciences. He recommended simpler studies with fewer variables, clear graphs, and single weights for combined measures, and warned against common misunderstandings of Fisherian null-hypothesis testing. Above all, he argued researchers should judge how large an effect is, not merely whether it is statistically significant.
The essay reflects a broader move within psychology toward weighing power and effect size rather than relying only on p-values, and it stresses that interpreting data still needs an investigator's informed judgement, and that understanding takes time. This is a personal essay of accumulated experience and opinion, not a new empirical study, so it offers no new data or statistical tests of its own.
Why it matters: Judging how large an effect is matters as much as whether it is statistically significant.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
- Summary by Intellect · source: Research: Cohen (1990). Things I have learned (so far). American Psychologist.
To cite this card, link to https://intellectplus.co.uk/app/history-research-lessons-in-applying-statistics-to-the-sociobiomedical-scienc.
How we source, check and correct cards: Editorial standards.
General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.
Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?
Related
More in History · Latest · A–Z · All topics
