In the problems below, assume that a significance level, a, of 0.05 and a power, (1 — /3), of 0.8 are desired unless otherwise stated.
A clinical trial is to be conducted to investigate and compare the effects of a new cholesterol-lowering drug with the effects of a standard drug. Determine the sample sizes necessary for the analysis scenarios in parts (a) and (b) below, a. A two-tailed, two-sample t test will be performed to compare the average cholesterol levels of hypercholesteremic patients treated with the new drug and hyper-cholesteremic patients treated with the standard drug. Under the homogeneous assumed to be 25 mg/dl. It will be important to detect an absolute difference in average true cholesterol levels of 25 mg/dl or larger, b. A test comparing the proportions of new-drug-treated patients and standard-drug-treated patients who complain of side effects is to be performed. Previous studies suggest that 25% of patients treated with the standard drug complain of side effects. It is important to detect an absolute difference in population proportions of 0.1 or more.