Consider a design for an exploratory therapeutic trial of a new antihypertensive drug compared with placebo. It has been agreed that a between group difference in mean change from baseline (that is, treatment effect) of at least 20 mmHg in SBP would be considered clinically meaningful. 

The primary hypothesis will be tested at a=0.1. If the standard deviation for the between difference in mean change from baseline is 40 mmHg, what is the required sample size per treatment group for a test with power of 80% (do by hand and round up to the nearest whole number)?

Answer:

Question:2

Consider a design for an exploratory therapeutic trial of a new antihypertensive drug compared with placebo. It has been agreed that a between group difference in mean change from baseline (that is, treatment effect) of at least 20 mmHg in SBP would be considered clinically meaningful.

The primary hypothesis will be tested at a=0.1. If the standard deviation for the between difference in mean change from baseline is 40 mmHg, what is the required sample size per treatment group for a test with power of 90% (do by hand and round up to the nearest whole number)?

Answer:

Question: 3

Which test is more likely to commit a type-II error:

A. A test that is powered at 80%

B. A test that is powered at 90%

QUESTION:4

Which test will have the largest sample size (all else being constant):

A. A test with alpha = 0.05

B. A test with alpha = 0.01

C. A test with alpha = 0.10