Effect Size

An effect size measures how big a difference or relationship is, on a scale that doesn’t depend on sample size — for example, “the new drug lowers blood pressure by 8 mmHg,” or a standardised measure like Cohen’s d. It answers “how much?”, whereas a p-value only answers “is it distinguishable from zero?”

Reporting effect size is the cure for p-value tunnel vision. In a huge dataset a trivial effect can be highly significant yet practically meaningless; in a small one a large, important effect can fail to reach significance. Always pair statistical significance with practical significance — the real-world size of the thing you found.