effect size

Terms from Statistics for HCI: Making Sense of Quantitative Data

Effect size is how large the effect is, as opposed to how likely it is to be valid or reproducible. For example, if we had a huge data set we might be able to assess that a really tiny effect is statistically significant. It is important to give some indication of the effect size so that you can assess the importance of the result. Note that some measures called 'effect size' are actually about the ratio between the raw size of the effect and other factors. Effect size is one of the three elements in the noise–effect–number triangle that can be manipulated to increase the statistical power of a study or experiment.

Used on pages 12, 65, 106, 107, 108, 109, 114, 134, 143

Also known as size of an effect, size of effect, size of effects, size of the effect