[hlnx1]

A study result can be statistically significant and still be too small to matter in practice. Effect size helps answer the question “how big is the difference?”

• Statistical significance mainly tells you whether a result is unlikely to be due to chance alone
• Effect size describes the magnitude of the difference
• Common conventions describe effects as small, medium or large, but context matters
• In elite sport, even small effects can be meaningful, while in health contexts larger effects may be needed

Example: creatine for endurance athletes.

Do you look at effect sizes when reading research?

[hlnx1]

A worked example: a supplement improves time-trial performance by 0.5% on average with a small effect size. For a recreational runner, that may be barely noticeable. For an elite athlete in a close race, it could matter.

[Author]

Posts