Imagine that two researchers, Bill and Ted, independently carry out clinical trials to test the same null hypothesis: H0: COX-2 selective inhibitors (used to treat arthritis) have no effect on the risk of cardiac arrest. They use the same population for their study, but Bill uses a sample size of 60 randomly selected subjects and Ted uses a sample size of 100 randomly selected subjects. Assuming all other aspects of their studies are the same, answer the following: Required:
a. Whose study has the higher probability of a Type-II error?
b. Whose study has higher power?
c. Whose study has the higher probability of a Type-I error?
d. Should their alternative hypotheses be one-sided or two-sided? Explain
e. Suppose Bill’s analysis results in p-value = 0.03. What should be his conclusion?

A) The 60-subject study will have the higher probability of a Type-II error.

B) 100-subject study will have higher power.

C) they will have the same probability of a Type-I error

D) Two tailed because either of the two results would be of interest

E) If p-value = 0.03, he should reject the null hypothesis if this p-value is less than the significance level. But if it's more than the significance level, he will fail to reject the null hypothesis.

Step-by-step explanation:

A) A type II error is an error that occurs when we accept the null hypothesis even though it is actually false.

Now, Smaller sample sizes usually would have a lesser chance of rejecting the null hypothesis even though it's false. Thus, the 60-subject study will have the higher probability of a Type-II error.

B) power = 1 - Pr(Type II error)

Since the 60-subject study has a higher probability of type II error it means it will have lesser power. Thus, 100-subject study will have higher power.

C) Type I error is when we incorrectly reject a true null hypothesis.

Pr(Type I error) = significance level

Now, since we are told that other aspects of their study are the same it means their significance levels are the same, it means they will have the same probability of a Type-I error

D) It should be two tailed because either of the two results would be of interest. It's either the drug was associated with more cardiac arrests or less potential treatment option.

E) If p-value = 0.03, he should reject the null hypothesis if this p-value is less than the significance level. But if it's more than the significance level, he will fail to reject the null hypothesis.