Which factors influence statistical power?

• A. Power is not affected by alpha.
• B. Effect size, sample size, and alpha all influence power.
• C. Only the sample size determines power.
• D. Only the effect size determines power.

Answer: (B)

Power is the probability of detecting a real effect when one truly exists. It depends on three main factors: how big the true effect is, how many participants you have, and how strict you set the criterion for calling something significant (the alpha level).

A larger true effect makes it easier to notice, so power increases as the effect size grows. More participants reduce sampling error and make the test statistic more likely to cross the rejection boundary under the alternative, so increasing sample size also boosts power. The alpha level defines how easy it is to declare significance; a higher alpha (more lenient) increases power because it makes rejection more likely, but it also raises the risk of a false positive.

If the effect is small, you’ll need a larger sample to achieve adequate power; if the effect is large, you can reach high power with a smaller sample. Power analyses use these factors to plan studies and determine the needed sample size or detectable effect size.