An effect size measures the magnitude of a treatment effect independent of sample size. It allows for the comparison of studies with varying sample sizes within a subject area. This method of comparing treatment effects is particularly useful when conducting meta-analyses that compare many studies. The classical calculation for effect size is the difference between the means of the treatment and control group divided by the standard deviation of either of the groups or the pooled standard deviation, assuming that the two groups are homogenous.

An effect size can be thought of as the standardized difference between the two groups. An effect size of 1 means that the treatment group outscored the control group by a full standard deviation.

One of the statisticians who devised the effect size referred to an effect size of .2 as small, .5 as medium, and .8 as large.