A healthcare consultant wants to compare the patient satisfaction ratings of two hospitals. Before collecting the data for a 2-sample t-test, the consultant uses a power and sample size calculation to determine the sample size required to detect a difference of 5 with a probability as high as 90% (power of 0.9). Previous studies indicate the ratings have a standard deviation of 10.
To detect a difference of 5 with a power of 0.9, the consultant needs to collect a minimum sample size of 86. Because the target power value of 0.9 results in a sample size that is not an integer, Minitab also displays the power (Actual Power) for the rounded sample size.