Power of study as a function of sample size.

<p>This figure shows the power of a study with two independent groups as a function of sample size for different levels of effect size (assuming that alpha, 2-tailed, is set at.05). For a small effect size (d = .2) we would need two samples of 393 participants to yield a power of 80%. This means that there is 80% chance of finding a significant difference between the groups, given that an effect of this size exists at the population level. For a medium effect size (d = .5) we would need two samples of 64 participants to achieve this level of power. For a large effect size (d = .8) we need 26 participants per group.