Question: a doctor is performing a clinical trial of a research...
A doctor is performing a clinical trial of a research medication for easing symptoms of emphysema. She recruits 50 patients with the disease and randomly assigns 25 to the treatment group and 25 to the placebo group. After 6 weeks, she measures each patient’s blood oxygen level and compares the sample means of the two groups. She calculates a test statistic of 4.38 and a p-value of 0.0001. She concludes that, since 0.0001 < 0.05, the medication significantly raises blood oxygen levels compared to the placebo. She calculates that the power of her test is 0.7.
What would constitute a type-I error in this scenario and what is its probability? What would constitute a type-II error in this scenario and what is its probability?