Dondorp et al. (2010) carried out a large clinical trial that compared effectiveness of two drugs, artesunate and quinine, for treatment of African children having severe malaria. They used the results to test the null hypothesis that the proportion of children dying was the same in the two treatments against the alternative hypothesis that one treatment was better than the other. Which single answer below is correct? Explain your answer.

The null distribution used in the test of the null hypothesis . . .

a. describes the probability distribution of possible true benefits of each of the two drug treatments.

b. describes the possible probability distribution of true benefits and costs of one drug treatment or the other drug treatment.

c. describes the probability distribution of possible observed differences between the treatment groups if there were truly no difference between drug treatments.

d. describes the probability distribution of possible observed differences between treatments given that one of the drugs really is better than the other.