Power of the Two-Sample t-Test

The power of the two-sample t-test is the probability of detecting a specified difference between two treatment means when that difference is real.

In STA305, its power is driven by four inputs:

• the effect size to be detected
• the standard deviation
• the sample size per group
• the significance level $\alpha$

R

power.t.test(delta = delta, sd = sd, n = n_per_group,
             sig.level = 0.05,
             type = "two.sample",
             alternative = "two.sided")

Related

• STA305
• Power of a Statistical Test
• Sample Size Determination
• Two-Sample t-Test